Liberal Parties in Europe 2018046598, 9780815372387, 9781351245500

189 51 5MB

English Pages [412] Year 2019

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Liberal Parties in Europe
 2018046598, 9780815372387, 9781351245500

Table of contents :
Cover
Half Title
Series Page
Title Page
Copyright Page
Contents
List of figures
List of tables
List of contributors
Acknowledgements
Introduction
PART I: Case studies
1. The Norwegian Left and the Finnish Centre:
What, no capital ‘L’ liberal parties?
2. The Danish Liberal Parties
3. The Centre Party and the Liberals: The Swedish members of the liberal party family?
4. Liberalism in the Netherlands: The VVD and D66
5. Belgian Liberals: Living apart together…
6. Diversity, unity, and beyond: The Swiss Liberals
7. Liberal Parties in Austria
8. It’s (not only) the economy, stupid?: Past and future of the German Liberal Party
9. The UK Liberal Democrats: Liberalism at a crossroads
10. Fianna Fáil: In the Liberals but not of the Liberals
11. Nuanced liberalism: The weakness of liberal parties in Spain
12. Liberal parties in Central and Eastern Europe: Between success and failure
13. Latvia, Lithuania, and Estonia
14. The Liberals in Europe: The alliance of Liberals and Democrats for Europe
PART II: Comparative perspective on liberal parties in Europe
15. Liberal parties and elections: Electoral performances and voters’ profile
16. Governmental participation and alliances of liberal parties in Europe
17. The liberal party family ideology: Distinct, but diverse
18. How liberal parties organise
Liberal parties in Europe: Conclusion
Index

Citation preview

Party Families in Europe

LIBERAL PARTIES IN EUROPE Edited by Caroline Close and Emilie van Haute

‘In a period of high party system volatility, this timely book focuses on a group of parties that includes some of Europe’s oldest electoral competitors. Its chapters present useful profiles and comparisons of contemporary liberal parties, and provide broader insights about the meaning and endurance of party family labels.’ — Susan E. Scarrow, University of Houston ‘With detailed case studies of the history, ideology, organization, electoral performance, and governmental participation of liberal parties in 20 European countries, plus at the EU level, this volume provides a most useful resource for students of party politics. In addition, the concluding comparative chapters provide the necessary analysis to consider whether there truly is a liberal party ‘family’ and the features that unite and distinguish these parties from their competitors.’ — William Cross, Carleton University

BK-TandF-9780815372387_TEXT_CLOSE-181001-FM.indd 1

14/01/19 2:08 PM

Liberal Parties in Europe

This book investigates how liberal parties have evolved over time as a party family, in a comparative perspective. Through a discussion of the applicability of the concept of party family to liberal parties, it gives a better picture of the development, challenges, and opportunities for liberal parties in Europe. The history of liberal parties in Europe is peculiar and the origins of the liberal family are not clearly defined. Liberal parties are still quite heterogeneous given the various meanings embraced in the idea of liberalism, including economic liberalism, cultural liberalism, progressivism, social-liberalism. Bringing together the best specialists engaged in the study of liberal parties, and with a two-levels perspective (comparative and case study), this book renews and expands our knowledge on the liberal party family in Europe. Four major themes are developed, linked to the four approaches of the concept of party family: electoral performances, participation to power, ideology and political program, and party organization. These themes are systematically developed in case studies, and in comparative chapters. Primarily aimed at scholars and students in comparative politics, this book should especially appeal to scholars in the fields of political parties and party systems, representation and elections, voting behaviour, and public opinion. Caroline Close is an Assistant Professor at SciencePo ULB, Centre d’étude de la vie politique (Cevipol), Université libre de Bruxelles (ULB). Her main research interests are party organization, intraparty politics, party ideology, legislative studies, elections, and democratic innovations. Emilie van Haute is an Associate Professor and Chair of SciencePo ULB and conducts her research at Centre d’étude de la vie politique (Cevipol), Université libre de Bruxelles (ULB). Her main research interests include party membership, intra-party dynamics, participation, elections, and voting behaviour.

Party Families in Europe Series editor: Emilie van Haute

The concept of party families is central to comparative party politics. Looking systematically at individual party families, their origins, development, ideology, policy positions, organisational structure, and/or sociological composition, this series investigates the nature of families of political parties. Themes are systematically developed through case studies and comparative chapters to consider key issues around: • • • •

Electoral performance and composition Participation to power Ideology and policy positions Party organisation

For more information about this series, please visit: https://www.routledge. com/Party-Families-in-Europe/book-series/ASHSER-1442 Green Parties in Europe Edited by Emilie van Haute Regionalist Parties in Europe Edited by Oscar Mazzoleni and Sean Mueller Liberal Parties in Europe Edited by Caroline Close and Emilie van Haute

BK-TandF-9780815372387_TEXT_CLOSE-181001-FM.indd 4

14/01/19 2:08 PM

Liberal Parties in Europe

Edited by Caroline Close and Emilie van Haute

First published 2019 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2019 selection and editorial matter, Caroline Close and Emilie van Haute; individual chapters, the contributors The right of Caroline Close and Emilie van Haute to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Names: Close, Caroline, editor. | Haute, Emilie van, editor. Title: Liberal parties in Europe / edited by Caroline Close and Emilie van Haute. Description: Abingdon, Oxon ; New York, NY : Routledge, 2019. | Series: Party families in Europe | Includes bibliographical references and index. Identifiers: LCCN 2018046598| ISBN 9780815372387 (hardback) | ISBN 9781351245500 (e-book) Subjects: LCSH: Liberalism—Europe—European Union countries— Case studies. | Political parties—Europe Union countries—Case studies. Classification: LCC JC574.2.E85 L524 2019 | DDC 324.2/16094—dc23 LC record available at https://lccn.loc.gov/2018046598 ISBN: 9780815372387 (hbk) ISBN: 9781351245500 (ebk) Typeset in Times NR MT Pro by Cenveo® Publisher Services

Contents

List of figures List of tables List of contributors Acknowledgements Introduction

x xii xvi xx 1

CAROLINE CLOSE AND EMILIE VAN HAUTE

PART I

Case studies

21

1. The Norwegian Left and the Finnish Centre: What, no capital ‘L’ liberal parties?

23

DAVID ARTER

2. The Danish Liberal Parties

44

KARINA KOSIARA-PEDERSEN

3. The Centre Party and the Liberals: The Swedish members of the liberal party family?

60

NIKLAS BOLIN

4. Liberalism in the Netherlands: The VVD and D66

77

GERRIT VOERMAN

5. Belgian Liberals: Living apart together…

95

VIVIEN SIERENS

6. Diversity, unity, and beyond: The Swiss Liberals OSCAR MAZZOLENI

113

viii  Contents 7. Liberal parties in Austria

129

LAURENZ ENNSER-JEDENASTIK AND ANITA BODLOS

8. It’s (not only) the economy, stupid?: Past and future of the German Liberal Party

146

SEBASTIAN U. BUKOW

9. The UK Liberal Democrats: Liberalism at a crossroads

166

ALAN WAGER AND TIM BALE

10. Fianna Fáil: In the Liberals but not of the Liberals

185

CONOR LITTLE AND DAVID M. FARRELL

11. Nuanced liberalism: The weakness of liberal parties in Spain

205

LIDIA NÚÑEZ

12. Liberal parties in Central and Eastern Europe: Between success and failure

225

BLAGOVESTA CHOLOVA AND JEAN-MICHEL DE WAELE

13. Latvia, Lithuania, and Estonia

241

DAUNIS AUERS

14. The Liberals in Europe: The alliance of Liberals and Democrats for Europe

259

WOUTER WOLFS AND STEVEN VAN HECKE

PART II

Comparative perspective on liberal parties in Europe

279

15. Liberal parties and elections: Electoral performances and voters’ profile

281

CAROLINE CLOSE AND PASCAL DELWIT

16. Governmental participation and alliances of liberal parties in Europe

310

JOHAN HELLSTRÖM AND DANIEL WALTHER

17. The liberal party family ideology: Distinct, but diverse

326

CAROLINE CLOSE

BK-TandF-9780815372387_TEXT_CLOSE-181001-FM.indd 8

14/01/19 2:08 PM

Contents ix 18. How liberal parties organise

348

STEFANIE BEYENS, EMILIE VAN HAUTE, AND TOM VERTHÉ

Liberal parties in Europe: Conclusion

364

CAROLINE CLOSE AND EMILIE VAN HAUTE

Index

377

Figures

3.1 3.2 5.1a 5.1b 5.2 7.1 8.1 8.2 8.3 8.4 10.1 10.2 10.3 10.4 10.5 11.1 11.2 11.3

Party positions on two dimensions, 1999–2014 Party membership and ratio membership/total electorate for election years 1960–2014 Electoral results of MR, Lower Chamber (federal), 1972–2014 (% vote share) Electoral results of OpenVLD, Lower Chamber (federal), 1972–2014 (% vote share) Positioning of Belgian parties in a two-dimensional political space (1999–2014) Policy positions and saliency scores for the general election in Austria in 2013 Electoral performance federal elections (German Bundestag) FDP’s electoral performances, 1949–2016 (European, federal, and Land levels) Parties in the German political space Formal party structure Electoral performance of Fianna Fáil – lower Chamber Positions of Fianna Fáil, the Liberals, and other party families on the economic left-right dimension Positions of Fianna Fáil, the liberal party family, and other party families on the EU integration dimension Positions of Fianna Fáil, the liberal party family, and other party families on the Gal-Tan dimension Fianna Fáil and party families compared on two measures of intra-party democracy Electoral performances of liberal parties in the general (2008, 2015) and European (2009, 2014) elections in Spain (in %) Evolution of voters’ self-declared ideology in Spain (in %) Left-right position of liberal voters versus rest of the electorate

BK-TandF-9780815372387_TEXT_CLOSE-181001-FM.indd 10

67 72 100 100 104 134 151 151 153 154 189 191 193 194 199 211 213 214

14/01/19 2:08 PM

Figures xi 11.4 Evolution of the positions of the Partido Popular on free market economy: Percentage difference with respect to the average number of favourable mentions on the topic in the period 1989–2011 11.5 Partido Popular’s positions on decentralisation: Percentage difference with respect to the average number of favourable mentions on the topic on the period 1989–2011 11.6 Party positions of Spanish parties on a two-dimensional political space 15.1 Distribution of voters’ aggregated ideological preferences 16.1 Participation of liberal parties in Europe, 1945–2015 16.2 Share of portfolio allocation to liberal parties in Europe 16.3 Coalition partners of liberal parties in Europe 16.4 Portfolios of liberal parties in Europe 17.1 Principal component analysis of issue positioning: factor plot after rotation (Promax) 17.2 Party families’ position on the economic dimension (CHES 2014) – Boxplot 17.3 Party families’ position on the cultural dimension (CHES 2014) – Boxplot 17.4 Liberal parties in the two-dimensional ideological space 17.5 Party families’ positioning on the general, economic, and cultural dimensions, 1945–2015 (Manifesto Project) 17.6 Classical, social, and conservative liberals’ positioning on the general, economic, and cultural dimensions, 1945–2015 (Manifesto Project) 17.7 Proportion of classical, social, and conservative liberals within the liberal party family, 1945–2015 (Manifesto Project) 18.1 Evolution of Members/Electorate ratio by party family, 1990–2010 18.2 Evolution party leadership selectorate, 1955–2012

BK-TandF-9780815372387_TEXT_CLOSE-181001-FM.indd 11

216 217 217 305 311 314 317 320 334 336 337 338 340 342 343 354 359

14/01/19 2:08 PM

Tables

0.1 1.1a 1.1b 1.2a

List of parties Venstre chronology Finnish Centre chronology Venstre’s electoral performance in the general elections, 1906–2013 (%) 1.2b The Finnish Centre’s electoral performance in the general elections, 1907–2015 (%) 1.3a Venstre participation in postwar Norwegian governments 1.3b Agrarian-Centre participation in postwar Finnish governments 1.4a Norwegian Venstre party leaders, 1945–2016 1.4b Finnish Agrarian-Centre party leaders, 1945–2016 1.5a Party membership in the Norwegian Venstre (selected years only) 1.5b Party membership in the Agrarian-Centre Party (selected years only) 2.1 Chronology of the main facts/events/developments of the party 2.2 Electoral performances of V, RV, and LA, Folketinget (% of valid votes) 2.3 Governmental experience of V, RV, and LA, 1901–2016 2.4 List of party leaders, results of the contest (if applicable), length of the leadership, and reason for resigning 2.5 Party membership figures and membership/total electorate for election years 1935–2015 3.1 Chronology of the main events of the Centre Party and the Liberals 3.2 Electoral performance of the Centre Party and the Liberals 1970–2014 3.3 Governmental experience of the Centre Party and the Liberals 1970–2014 3.4 Salience of major policy areas, 1998–2014

9 25 26 28 30 31 33 37 38 39 40 46 47 49 55 56 61 62 64 69

Tables xiii 3.5 4.1a 4.1b 4.2 4.3 4.4a 4.4b 4.5 5.1 5.2 5.3 5.4 5.5 5.6 6.1 6.2 6.3 7.1 7.2 8.1 8.2 8.3 8.4 8.5 8.6 9.1 9.2

Party leaders of the Centre Party and the Liberals in Sweden, 1946–2016 Chronology of the main developments of the VVD, 1948–2016 Chronology of the main developments of D66, 1966–2016  Election results for VVD and D66 in the Netherlands (Second Chamber), 1948–2012 Government coalitions in the Netherlands and liberal participation (1946–2016) Party leaders VVD, 1948–2016 Party leaders D66, 1966–2016 Party members, VVD and D66, 1948–2016 Chronology of the main developments of MR and Open VLD in Belgium Seats and seat share at federal level (Lower Chamber, 1971–2014) Governmental participation of MR and OpenVLD at federal level (1972–2014) CMP Scores on five dimensions: MR versus OpenVLD, 1974–2010 Leadership contests – MR-OpenVLD, 1971–2011 Party membership (OpenVLD-MR), 1971–2012 The formation of liberal parties in Switzerland – Chronology of key events Electoral strength of the main Swiss parties, National Council, 1848–1917 Electoral strength of the main Swiss parties, National Council, 1919–2015 Chronology of main developments in the history of liberal parties in Austria Results for LIF and NEOS in parliamentary elections (Nationalrat) Chronology of the main developments of the FDP, 1945–2016 Electoral performance federal elections (German Bundestag; party votes) Governmental experience of the FDP Left-right position and issue saliency, FDP 1949–2013 List of party chairs Party membership Chronology of the main developments of the Liberal party and the Liberal Democrats, 1859–2016 Electoral performance of liberal parties, 1918–2015

71 78 81 81 83 90 90 91 98 101 101 103 107 108 115 116 117 131 132 147 150 152 155 160 160 167 168

xiv  Tables 9.3 9.4 9.5 10.1 10.2 11.1 11.2 11.3 12.1 13.1 13.2 13.3 13.4 13.5 13.6 13.7 13.8 14.1 14.2 15.1 15.2 15.3 15.4 15.5 15.6 15.7 15.8

Relation of liberal parties towards participation in executive office List of Liberal Democrat leaders, 1988–2016 Party membership figures per year since the foundation of the Liberal Democrats, and ratio membership/total electorate for election years Fianna Fáil and Pedersen’s (1982) party lifespan model Fianna Fáil: party leadership and spells in government Main events related to liberal parties since the transition to democracy Positions among liberal voters and the rest of the electorate 2015 preelectoral survey: vote recall and vote intention of Liberal voters (in %) Electoral performance of liberal parties (% of the votes, national legislative election) Chronology of the development of liberal parties in the Baltic states Electoral performance of liberal parties in the Baltic states, 1992–2015 Government experience: Liberals in power in the Baltic States Liberal parties in the Baltic states’ CMP left-right scores Liberal parties in the Baltic states’ CMP freedom and human rights score Liberal parties in the Baltic states’ CMP market economy score Liberal parties in Baltic states’ CMP European Community/ union positive score Party membership in the Baltic states Seats of the liberal group in the European Parliament, 1979–2017 ALDE Party leaders Electoral performances of liberal parties in the Benelux countries, Austria, Germany, and Switzerland Electoral performances of liberal parties in Nordic countries Electoral performances of liberal parties in Ireland and the United Kingdom Electoral performances of liberal parties in Italy and Spain Electoral performances of liberal parties in the Baltic countries Electoral performances of liberal parties in CEE countries Selected cases and frequency distribution The determinants of the liberal party vote: Logistic models (odds ratio)

171 179 180 188 189 209 215 218 232 243 246 247 249 250 251 252 256 264 274 283 289 293 295 296 297 302 304

Tables xv 15.9 The diversity of liberal parties: Average positioning of their voters (final cluster centres) 16.1 Liberal parties in government in 27 European countries, 1945–2015 16.2 Single-party governments, 1945–2015 16.3 Appendix. Parties included in the comparative analysis 17.1 Case selection 17.2 Party families: Frequency distribution 17.3 Structure matrix 17.4 Manifesto Project left and right items 17.5 Party families’ positions and dispersion on the general, economic, and cultural left-right dimensions (CHES 2014) 17.6 Party families’ positioning and dispersion on the general, economic, and cultural left-right dimensions, 1945–2015 (Manifesto Project) 18.1 Mean staff members HQ and PPG (in full-time equivalent) 18.2 Mean share party revenue and campaign expenditure (in %) 18.3 % of parties with specific rights or guaranteed representation for groups (in %) 18.4 Distribution of costs, benefits, and alternative affiliation by party family 18.5 Comparative competitiveness leadership elections

306 313 316 323 331 332 334 335 336 341 351 352 356 357 360

Contributors

David Arter is Research Director in the School of Management, University of Tampere, Finland. He has written extensively on aspects of Nordic politics and government over many years and the third edition of his widely used text Scandinavian Politics Today was published in 2016. Daunis Auers  is Associate Professor of Comparative Politics at the University of Latvia and Research Director at the Certus think tank in Riga, Latvia. He studied at the London School of Economics and defended his PhD at University College London, UK. He has been a Fulbright Scholar at the University of California, Berkeley, USA. His book on the Comparative Politics and Government of the Baltic States was published by Palgrave Macmillan in 2015. Tim Bale is a Professor in the School of Politics and International Relations at Queen Mary University of London, UK. He specializes in political parties and elections in the UK and Europe. Recent publications include the fourth edition of European Politics: A Comparative Introduction, the second edition of The Conservative Party from Thatcher to Cameron, and Five Year Mission: The Labour Party under Ed Miliband. Stefanie Beyens  is an Assistant Professor and Postdoctoral Researcher at Utrecht University, the Netherlands. She has worked on new political party survival and elections, recently expanding her research agenda to include successful public organisations and their effectiveness and legitimacy. Anita Bodlos is a Predoctoral Researcher at the Department of Government at the University of Vienna, Austria. Her research interests comprise party behaviour across communication channels, online campaigning, and quantitative content analysis. Niklas Bolin  is Senior Lecturer in Political Science at the Mid Sweden University, Sundsvall. His research interests include political parties, party system change, and parliamentary government with a special focus on Scandinavia. Previously he has published in journals such as Journal of Elections, Public Opinion and Parties, Party Politics, Scandinavian Political Studies, and West European Politics.

BK-TandF-9780815372387_TEXT_CLOSE-181001-FM.indd 16

14/01/19 2:08 PM

Contributors xvii Sebastian U. Bukow is Research Fellow at the Düsseldorf Party Research Institute (PRuF) and Head of Politics and Party Research of the Heinrich-Böll-Foundation Berlin, Germany. He is chair of the standing group on political parties of the German Political Science Association (DVPW) and member of the DVPW executive committee. His research interests are parties as multilevel organizations, the German party system, political sociology, and democratic institutions. Blagovesta Cholova  is an Associate Researcher at the Université libre de Bruxelles (ULB), Belgium. Her research focuses on right-wing parties and populist movements in Central and Eastern Europe, and more specifically in Bulgaria. She has published several articles in SouthEastern Europe, Europeana, Revue Internationale de politique comparée, Baltic Worlds, and Slovak Journal of Political Science. Caroline Close  is an Assistant Professor at the Centre d’étude de la vie politique (Cevipol), Université libre de Bruxelles (ULB), Belgium. Her main research interests are party organization, intraparty politics, party ideology, legislative studies, elections, and democratic innovations. Pascal Delwit  is Professor of Political Science at the Université libre de Bruxelles (ULB) and member of the Centre d’étude de la vie politique (Cevipol), Belgium. His research interests include political parties and electoral processes in Europe. He recently edited Du parti libéral au MR. 170 ans de libéralisme en Belgique (Editions de l’Université de Bruxelles, 2017) and published Les gauches radicales en Europe (Editions de l’Université de Bruxelles, 2016). Jean-Michel De Waele  is Professor of Political Science at the Université libre de Bruxelles, Belgium. His research focuses on political parties in Central and Eastern Europe, the political regimes of that region, leftwing parties in Europe, and the relationship between sport and politics. He has edited several books with Palgrave, Armand Colin, Bruylant, and contributed many other book chapters on these topics. Laurenz Ennser-Jedenastik  is Assistant Professor at the University of Vienna’s Department of Government, Austria. His research focuses on parties, elections, bureaucracy, and social policy. David M. Farrell  is the Head of the School of Politics and International Relations, University College Dublin, Ireland. He is a member of the Royal Irish Academy, and co-editor of Party Politics. His research interests include political parties, elections, electoral systems, representation, and constitutional deliberation. His most recent books are The Act of Voting: Identities, Institutions and Locale (co-edited with Johan Elkink; Routledge, 2016) and A Conservative Revolution? Electoral Change in Twenty-First Century Ireland (co-edited with Michael Marsh and Gail McElroy; Oxford University Press, 2017).

BK-TandF-9780815372387_TEXT_CLOSE-181001-FM.indd 17

14/01/19 2:08 PM

xviii  Contributors Johan Hellström  is an Associate Professor in Political Science at Umeå University, Sweden. His research focuses on political institutions and democratic representation in Europe. Hellström is currently principal investigator for two research projects on coalition politics and party government, representative democracy in Europe (RDE), and Party Government in Europe Database (PAGED). Karina Kosiara-Pedersen is an Associate Professor of Political Science at the University of Copenhagen, Denmark, and the Director of the Center for Voting and Parties (CVAP). Her research focuses on party membership, political recruitment, campaigns, and elections. Conor Little  is a Postdoctoral Researcher at the Department of Political Science, University of Copenhagen, Denmark. His research interests include political parties, political careers, the comparative politics of climate change, and Irish politics. He was awarded a PhD for his doctoral thesis – a comparative study of Green parties in governing coalitions – by the European University Institute in March 2014. He currently holds an Individual Postdoctoral Fellowship from the Danish Social Sciences Research Council (FSE). Oscar Mazzoleni  is Professor in Political Science and Director of the Research Observatory for Regional Politics at the University of Lausanne, Switzerland. He has published several chapters, articles, and a book on Swiss parties in a comparative perspective. He recently co-edited Understanding Populist Organization: The European Radical Right Parties, MacMillan-Palgrave (with R. Heinisch) and Regionalist Parties in Western Europe: Dimensions of Success, Routledge (with S. Mueller). Lidia Núñez obtained her PhD in the Université libre de Bruxelles, Belgium, and she is an Honorary Postdoctoral Researcher (F.R.S.-FNRS) at the same institution. Her main research interests include electoral systems, political parties, elections, and institutional reforms. Her work has been published in West European Politics and in the European Journal of Political Research among others. Vivien Sierens  is a PhD candidate at the Université libre de Bruxelles (ULB) and Vrije Universiteit Brussel (VUB), Belgium. He is a member of the Centre d’étude de la vie politique (Cevipol). His main research interests include party membership, intraparty cohesion, and political participation. Emilie van Haute  is an Associate Professor and chair of SciencePo ULB and conducts her research at Centre d’étude de la vie politique (Cevipol), Université libre de Bruxelles (ULB), Belgium. Her main research interests include party membership, intra-party dynamics, participation, elections, and voting behaviour.

BK-TandF-9780815372387_TEXT_CLOSE-181001-FM.indd 18

14/01/19 2:08 PM

Contributors xix Steven Van Hecke is Assistant Professor in Comparative and EU Politics at the Faculty of Social Sciences of the University of Leuven, Belgium. Tom Verthé is Postdoctoral Assistant at the Centre for Local Politics at Ghent University (UGent), Belgium, and Guest Professor at Vrije Universiteit Brussel (VUB), Belgium. His research focuses on parties, elections, and electoral behaviour. Gerrit Voerman  is historian and Director of the Documentation Centre Dutch Political Parties of the University of Groningen, the Netherlands, and, since 2011, Professor of ‘Development and Functioning of the Dutch and European Party System’ at this university. He has published widely on political parties, especially on party history; organization, membership, and identity; and on candidate selection. Since 2008, he has served as editor of a series on the political parties in the Netherlands. Alan Wager  is a Doctoral Researcher in the School of Politics and International Relations at Queen Mary University of London, UK, where he teaches politics. He is writing a PhD on coalitions, political realignment, and interparty politics in the British context. Daniel Walther is a PhD candidate in Political Science at Umeå University, Sweden. Daniel got his BA from the University of Birmingham, UK, and his MSc from Lund University, Sweden. He is currently based at Umeå University where he does research primarily on comparative politics and coalition governments. His main research interests include cabinet stability, party behaviour, political polling, and elections. Wouter Wolfs is PhD Fellow of the Research Foundation – Flanders (FWO), Belgium, and prepares a dissertation on the finance of European political parties at the Global Governance Institute (University of Leuven, Belgium).

BK-TandF-9780815372387_TEXT_CLOSE-181001-FM.indd 19

14/01/19 2:08 PM

Acknowledgements

This book is the continuation of the conference on ‘Liberal parties in Europe’ held in Brussels (Belgium) in March 2016. We would like to thank the participants to this event, as well as its sponsors: the Fonds National de la Recherche Scientifique (FRS-FNRS), the Association belge francophone de science politique (ABSP), and the Faculté de Philosophie et Sciences Sociales of the Université libre de Bruxelles (ULB). We are grateful to the Routledge team for their professional and human collaboration, and to the referees for their very constructive remarks.

Introduction Caroline Close and Emilie van Haute

The history of European modern democracies is intrinsically linked to that of liberal parties. Liberalism has indeed inspired modern constitutions in the 19th and 20th centuries, as it promoted individual rights and freedom, the rule-of-law, and parliamentarianism. Despite the universal appeal of liberal values, liberal parties have known diverging fates. In some ‘old’ European democracies, liberalism – incarnated by one or several parties – has constituted and still represents a major force (e.g., Belgium, Denmark, Finland, the Netherlands). In southern European countries, no liberal party has ever broken through and these parties have remained quasi absent from the political game (e.g., France, Spain), although recent successes of liberal organisations challenge this view (Ciudadanos in Spain, En Marche in France). In other cases, liberal parties have for long played the role of pivotal party, such as in Germany, or that of third party, such as in the United Kingdom. After the fall of the USSR, liberal parties have known relative successes in the Baltic States and across Central and Eastern Europe, depending on the national context. At the European level, liberal parties gathered in the Alliance of Liberals and Democrats for Europe (ALDE) constituted the third ideological family in terms of seats in the European parliament until the 2014 elections, when they were relegated to the fourth place. This volume reflects on the evolution of liberal parties in Europe. It fills a major gap in the literature, as very few volumes have been dedicated to this party family (De Winter 2000; Delwit 2002a; Kirchner 1988). Comparative knowledge on the liberal party family is quite diffuse and can be found in occasional publications that analyse differences and congruence between party families in ideological (Camia and Caramani 2012; Ennser 2012; Freire and Tsatsanis 2015), sociological (Knutsen 2008; van der Brug et al. 2009), or organisational terms (Poguntke et al. 2016). This lack of scientific attention is quite paradoxical given the prominent role of political liberalism in the development of European constitutional democracies (Roussellier 1991) as well as in the building of the European Union. This relative paucity can be attributed to several factors, among which the noticeable heterogeneity of this family, and thus to the difficulty of identifying its boundaries

2  Caroline Close and Emilie van Haute (Delwit 2002b). This volume aspires to overcome this complexity, by providing an up-to-date, comparative, and systematic analysis of liberal parties across Europe. It extends the geographical scope of existing volumes by including both old and new democracies across Europe and uses the concept of party family as an analytical framework. It raises two main questions: (1) Do liberal parties constitute a family? (2) How can we characterise liberal parties today?

The concept of party family as a framework for analysis The concept of party family is central in comparative party politics, as it constitutes a convenient tool to categorise parties across time and space. In their seminal work, Mair and Mudde (1998) present four approaches to classify parties into families: name; affiliation to transnational federations; origins and sociology (generic approach); and policy and ideology. In this volume, we combine these approaches and expand them to identify liberal parties, highlight their common characteristics and differences, and assess their belonging to one single party family. Name and transnational federations One of the most straightforward ways to identify liberal parties is to look at their name or at their transnational affiliation. Although these two approaches are easy to apply and reveal a great deal of the party’s self-perception and identity, they carry some problems. Regarding party labels, they might be used strategically: either by appealing explicitly to a popular ideology but implicitly belonging to a somewhat different obedience; either by belonging to one ideology but avoiding to refer to it directly in the party’s name. Regarding transnational federations, not all parties develop transnational bonds, especially the smallest and newest ones. Furthermore, some transnational federations today regroup very diverse political groups and philosophies. In the same way that labels can be used strategically, transnational affiliations can also result from strategic considerations and negotiations. In the case of liberal parties, labels can be misleading and contribute to the heterogeneous image of the liberal party family. Most parties of this obedience seem reluctant to use ‘liberal’ in their denomination. The term ‘liberal’ is explicit in a minority of cases, remarkably for 19th- and 20th-century parties (e.g., in Belgium, UK, Switzerland) but also for (newer) organisations willing to reaffirm liberal values (Liberal Forum in Austria; Liberal Alliance in Denmark; the Liberals in Sweden – previously known as Peoples’ Party). Overall, liberal parties rather appeal to a plethora of other concepts that reflect core liberal values, among which ‘freedom’ and ‘democracy’ prominently figure: Movement for Rights and Freedoms (DPS, Bulgaria); Free and Democratic Party (FDP, Germany); People’s Party

Introduction 3 for Freedom and Democracy (VVD, Netherlands); Democrats 66 (D66, Netherlands); Freedom Union (UW, Poland); Freedom and Solidarity (SaS, Slovakia); Union, Progress and Democracy (UPyD, Spain), etc. Concepts such as ‘civic’ and ‘citizenship’ are also put to the front, such as for the ‘Alliance of the New Citizen’ (ANO in Slovakia) or Ciudadanos (Citizens) in Spain; as well as the idea of ‘reform’ (Belgian Reform Movement; Estonian Reform Party). Some labels also participate in the fluid ideological identity of the liberal family, suggesting a placement on the ‘left’ (Venstre in Denmark and Norway) or at the ‘centre’ (Centre Party in Finland, Sweden, Estonia; Centre Union in Lithuania). Finally, we can notice a certain avoidance of the term ‘party’: ‘movement’, ‘union’, or ‘alliance’ are preferably used, reflecting either a critical stance towards mass organisations or the fact that these liberal organisations often result from the convergence of disparate movements. Interestingly, the history of the European liberal party reflects these tensions (see Chapter 14). Labels can tell a lot about the values promoted by liberal parties and about their organisation but they are of little help when trying to identify them a priori. This approach is thus discarded. The strategy adopted in this volume is to use parties’ affiliation to the European liberal party (ALDE – Alliance of Liberals and Democrats for Europe) as a starting point to identify liberal parties. This European party gathers national parties along core liberal values drafted in a common manifesto for the European election, usually including individual freedom and civil liberties, open market and pro-European integration stances. Besides, the ALDE parliamentary party group behaves in a quite cohesive way in the European parliament, also suggesting a common ground between the national parties affiliated. In a few cases, we use affiliation to the Liberal International. We also consider new or recent parties that have not yet had the opportunity to become a member of any transnational federation, provided that there are signs that the party engages in the process of becoming a member. Dead parties that used to belong to a liberal transnational federation are also included in the longitudinal analyses. Although transnational affiliation is used as a starting point, the belonging of these parties to the liberal party family is assessed and discussed by analysing other criteria, which are described below. We will see that in some cases, the belonging of some ALDE members to the liberal family is questionable when looking at these other criteria. Origins and sociology According to the Mair and Mudde’s (1998) conclusion, the generic approach constitutes one of the most useful and promising tools to study party families, along with the ideological approach. By relying on the classical cleavage theory of party politics (Lipset and Rokkan 1967; Seiler 1980), Mair and Mudde propose to categorise parties by linking them to the structural

BK-TandF-9780815372387_TEXT_CLOSE-181001-Introduction.indd 3

14/01/19 2:11 PM

4  Caroline Close and Emilie van Haute conflict from which they originated. The origin of parties on a specific line of division would then be reflected in the parties’ sociological composition. Hence, each party would represent and organise one or several specific segments of society. The generic approach to the classification of parties is praised for its theoretical anchorage. However, its geographical scope is quite limited: The cleavage theory in which it is based was developed in the context of 19th-century Western Europe, and does not easily apply to younger democracies. Besides, it is sometimes considered as too static to take into account changes in party systems, such as electoral realignments or the emergence of new parties. The generic approach is challenging to apply to liberal parties. The link between liberal parties and the four founding cleavages varies across countries and makes them less discernible than other families on this criterion (Seiler 2002). In some contexts, liberal parties have emerged along both the religious and socioeconomic cleavages (e.g., Benelux countries); in other contexts they were associated with only one of these cleavages. In Nordic countries, they developed closely to the agrarian movements, thus on the rural side of the territorial cleavage. In addition to having various origins, liberal parties have realigned in diverse and sometimes opposite directions. Furthermore, new liberal actors have emerged, either in consolidated democracies or in younger and postcommunist democracies, and these are less easily classifiable around the classical four cleavages. These fluctuating origins and developments result in a lack of distinctiveness of the liberal electorate across Europe (van der Brug et al. 2009; Vandermotten and Médina Lockhart 2002). Despite the drawbacks of the generic approach exposed above, it still constitutes a useful tool to characterise and understand the sociology of liberal parties in their different historical and geographical contexts. Looking at the origins of liberal parties and at their sociological composition will allow mapping the internal diversity of this alleged party family, and discussing its very existence. Ideology and policy positions The fourth approach pointed by Mair and Mudde (1998) to classify parties into families is based on their ideological coherence. The approach directly relates to the generic approach, which considers parties built on similar sides of structural conflicts as sharing a common societal project and as promoting the interests of similar segments of society. This approach has the advantages of being easily operationalised through the measurement and coding of party policy statements. The systematic collection of this type of data across countries has considerably increased, either through manifesto content analysis or through expert survey questionnaires (Klingemann et al. 2006; Laver and Hunt 1992). Methodological

BK-TandF-9780815372387_TEXT_CLOSE-181001-Introduction.indd 4

14/01/19 2:11 PM

Introduction 5 concerns relate to the variety of sources and materials that can be used, but more importantly, pertain to the contextual nature of these types of data, showing great variations across elections and national contexts (Dinas and Gemenis 2010; Mair and Mudde 1998: 218). In the case of liberal parties, data availability allows for long-term and cross-national comparisons, especially for older generations of liberal parties. Comparative analyses of party families’ ideology rely on multiple types of data: manifesto content (Camia and Caramani 2012), expert surveys (Freire and Tsatsanis 2015), or candidate surveys (Ennser 2012). These studies point to three main findings regarding the liberal party family. First, at the aggregate-level, its ideological placement can be described as ‘ambivalent’ (De Winter and Marcet 2000; Smith 1988), given its central position on the general left-right scale, its rightist position on socioeconomic issues and its centre-leftist or progressive stances on cultural issues (Camia and Caramani 2012; Elff 2013; Ennser 2012). Second, as far as ‘new’ cleavages are concerned (Kriesi et al. 2006), liberal parties are usually described as pro-integration, although not much more than the two other traditional families, the Christian and Social-Democrats (Marks et al. 2002). Third, these aggregate patterns hide great internal variations. The liberal party family usually appears as the most heterogeneous (Ennser 2012; Freire and Tsatsanis 2015), although this conclusion has been reappraised (Camia and Caramani 2012). Many of these existing studies were centred on Western Europe and on ‘old’ liberal parties, with some exceptions (e.g., Camia and Caramani 2012). Newer liberal actors, either emerging in transition contexts or in consolidated European democracies, remain understudied. Until recent breakthroughs, newer parties have faced the problems of minor parties: Given their relatively small size and parliamentary representation in the first years of existence, data were not collected systematically. In postcommunist contexts, major problems pertain to the high volatility and instability of most parties in these countries both in terms of electoral performances and organisational relevance. This volume applies the ideological approach to analyse both old and young liberal parties in diverse geographical spaces, henceforth providing an up-to-date perspective on the ideological characteristics of the liberal party family, and a critical assessment of its boundaries. Party lifespan, participation to power, and organisation In addition to these classical approaches to study party families, this volume makes several adaptations and innovations to address some of the difficulties and limitations pointed by Mair and Mudde (1998). First, as in a previous volume edited in this collection dedicated to Green parties (van Haute 2016), this volume integrates a more longitudinal perspective to avoid the synchronic perspective on party origins induced

BK-TandF-9780815372387_TEXT_CLOSE-181001-Introduction.indd 5

14/01/19 2:11 PM

6  Caroline Close and Emilie van Haute by the cleavage theory. We investigate and discuss the origins as well as the evolution of liberal parties along existing or new cleavages, across geographical contexts. Besides, we look specifically at the development of individual parties over time, by using the party lifespan approach developed by Pedersen (1982). This approach allows identifying four phases of party development: (1) the threshold of declaration, that is, when the party first declared its intention to participate to election; (2) the threshold of authorisation, that is, when the party met the legal requirements to participate in elections; (3) the threshold of representation, that is, when the party entered the parliament; and (4) the threshold of relevance, that is, when the party gained influence over government formation and policy output. Second, we go a little further than the threshold of relevance as we also examine how liberal parties behave once they reach government. We look at the ability of liberal parties to occupy executive office, at the type of government and coalition partners they prefer, and at their strategies in terms of portfolio allocation. This approach innovates at several levels. First, it extends the concept of party family beyond origins and ideological stances, and takes into account parties’ strategies and behaviour. Second, it examines the extent to which ideology and origins translate into specific strategies. For instance, we might expect foundational cleavages and ideological positions to affect parties’ choice of partners (i.e., they opt for parties positioned on the same side or close to them on specific cleavages) as well as the type of portfolios they try to obtain (i.e., portfolios should reflect salient issues) (Budge and Keman 1990). Third, it offers a new way of assessing the existence and boundaries of a party family, as it potentially reveals common behavioural patterns (e.g., pragmatism vs. idealism) between members of the same family. The final adaptation is the discussion of liberal parties’ organisation. Whereas the ideological and sociological dimensions impinge a lot on the concept of party family, organisational types are more loosely associated with party families. Yet, parties’ core values are often reflected in the way parties manage their members and build their organisations (Close 2016; Enyedi and Linek 2008). Therefore, we integrate party organisation as an additional criterion on which to assess the existence of a liberal party family that would share common organisational patterns. We expect liberal values, such as the promotion of individual freedom and the aversion for abuses of state authority and centralisation, to be reflected in the way they organise. Following these adaptations, four major themes are developed in this volume: 1 Origins, development, and sociology: Following Lipset and Rokkan’s theory, as well as Pedersen’s lifecycle approach, we examine the birth of liberal parties, their electoral performances, and the sociology of their electorate. We attempt to identify common and divergent patterns in

Introduction 7 their origins, electoral development and profile of their voters, to stress whether liberal parties encapsulate similar segments of the electorate across national contexts. 2 Participation to power: In addition to examining the electoral performances of liberal parties, we investigate their experience of governmental participation over time. We try to highlight similarities and differences in their ability to exert power, the type of governmental coalitions or alliances that they favour, the type of portfolios they get and the type of policies they develop. 3 Ideology and policy positions: We analyse whether liberal parties share common policy positions and how these have evolved over time. We relate these to their electoral performances and governmental participation, which might affect programmatic choices; but also to their sociological bases, which evolve alongside political program. 4 Organisational structure: We investigate whether liberal parties, often associated with an elite or cadre-based type of organisation, can still be described as such. We examine the extent to which this model is shared by all members of this family, and how these organisational structures have evolved over time, notably in regards to the rights granted to individual party members.

Book outline and structure This volume includes two types of contributions. The first part consists in case studies for which contributors present the major developments of national liberal parties on the four aspects listed above. These country chapters cover 20 countries: 12 in Western Europe, 8 from Central and Eastern Europe (including the Baltic States). Additionally, one chapter is dedicated to the European liberal party. Most case studies are presented in a full chapter: Denmark (Chapter 2), Sweden (Chapter 3), the Netherlands (Chapter 4), Belgium (Chapter 5), Switzerland (Chapter 6), Austria (Chapter 7), Germany (Chapter 8), United Kingdom (Chapter 9), Ireland (Chapter 10), and the ALDE (Chapter 14). Interestingly, Chapter 11 discusses the absence of a relevant liberal party in Spain. Other cases are grouped together: Finland and Norway (Chapter 1); Bulgaria, Czech Republic, Poland, Romania, and Slovakia (Chapter 12, which discusses the difficulties of studying liberal parties in these Central and Eastern European countries); Estonia, Latvia, and Lithuania (Chapter 13, which highlights the relative success of liberal parties in some of the Baltic States). These country chapters adopt a similar structure. Each chapter starts by describing the main events or developments of the party since its foundation, chronologically. The chronology lists the main developments of the party in the party system, following the four phases identified by Pedersen (1982). It also includes the main intra-party events (relevant leadership contests or crises, main organisational changes, scissions, fusions, name changes),

8  Caroline Close and Emilie van Haute electoral performances (with a main emphasis on the results for the Lower Chamber), and the relation of the party towards participation to power at various levels. The second section presents the position of the party on the main cleavages or issues: socioeconomic issues, religious, moral, or cultural issues, centre-periphery issues (or structure of the state), environment, law and order, immigration, and Europe. It also discusses the main developments or shifts across time when necessary. The third section presents the party organisation and structure. It starts by describing the main party bodies and their competences (with a specific focus on leadership selection, candidate selection, and policy position formulation), and points evolutions over time when necessary. It then presents the evolution of party membership figures and elements on profile when available. Finally, it describes the formal and informal links between the party and social movements. Finally, the chapter concludes with a short analysis of the main challenges and opportunities for the party in the future. The second part of the book includes comparative chapters that analyse the major evolutions related to one of the four core themes presented above, hence contributing to the discussion of the concept of party family. These four chapters contribute greatly to the genealogic, sociological, ideological, and organisational approaches to the notion of party family. These comparative chapters cover the cases presented in the first part of the book, plus eight additional countries, which brings the geographical scope of the book to 28 countries. Overall, the book mentions no less than 87 liberal parties or movements (see Table 0.1) and provides more in-depth development and analysis for 48 of them. In Chapter 15, Caroline Close and Pascal Delwit discuss the electoral fate of liberal parties in Europe as well as the evolution of the profile of liberal party voters. The chapter focusses on national elections to the Lower Chamber, and covers a wide range of countries. It uncovers common patterns in the electoral development of liberal parties in specific geographical areas, but point to discrepancies between these areas. It also presents comparative elements regarding the composition of the liberal electorate (sociodemographic characteristics and political positioning), again highlighting the diversity of the liberal party electorate. In Chapter 16, Johan Hellström and Daniel Walther take a comparative approach and examine the patterns of government participation among liberal parties, and their success in getting both the Prime Ministership and other important ministerial portfolios. They investigate under which circumstances they are included in or excluded from government, which parties they form governments with and their success in obtaining important ministerial portfolios. Their analysis reveals that liberal parties seem to have an above-average ability to secure a place in government and to obtain ministerial portfolios. The authors highlight the importance of liberal parties’ ideological centrality in the party system, electoral success, and (historically determined) party system factors in explaining this success.

Table 0.1  List of parties Full name in original language

Country

Party acronym

Austria Austria

LIF NEOS

Austria

FPÖ

Austria

JuLis

Austria Belgium (FR)

JUNOS MR

Belgium (FL)

OpenVLD

Open Vlaamse Liberalen en Democraten

Open Flemish Liberals and Democrats

Bulgaria

DPS

Bulgaria

NDSV

Bulgaria

NLP

Dvizhenie za prava i svobodi Nacionalno dviženie za stabilnost i văzhod Narodna Liberalna Partia

Movement for Rights and Freedoms National Movement for Stability and Progress National Liberal Party

Liberales Forum NEOS - Das Neue Österreich und Liberales Forum Freiheitliche Partei Österreichs Junge Liberale Österreich Junge liberale NEOS Mouvement réformateur

English translation

Previous names

Chapters

Liberal Forum The New Austria and Liberal Forum

NEOS – Das Neue Österreich

Freedom Party Austria

(until 1986)

16

Young Liberals of Austria Young liberal NEOS Reform Movement

 

7

JuLis LP-PL (1945-1961); PVV-PLP (1961-1972); PLP (1972-1976); PRLW (1976-1979); PRL (1979-1993); PRL-FDF (1993-1998); PRL-FDFMCC (1998-2002) LP-PL (1945-1961); PVV-PLP (1961-1972); PVV (1972-1992); VLD (1992-2007)

7 5, 15, 16, 17, 18

7, 15, 16, 17 7, 15, 16, 17

5, 15, 16, 17, 18

12, 15, 16, 17 12, 15, 16, 17 12

(Continued )

Table 0.1  List of parties (Continued) Country

Party acronym

Croatia

HSN

Croatia

HSLS

Croatia

IDS

Croatia Czech Republic Czech Republic

LS ANO2011 ODA

Czech Republic

US-DEU

Czech Republic

LIDEM / VIZE 2014 RV LA V EK ER

Denmark Denmark Denmark Estonia Estonia

Full name in original language

English translation

Hrvatska narodna stranka - Liberalni demokrati Hrvatska socijalnoliberalna stranka Istarski demokratski sabor Dieta democratica istriana Liberalna stranka ANO2011 Občanská demokratická alliance Unie Svobody– Demokratická unie Řád národa

Liberal Party Yes2011 Civic Democratic Alliance Freedom UnionDemocratic Union Order of the Nation

Radikale Venstre Liberal Alliance Venstre Eesti Keskerakond Eesti Reformierakond

The Social Liberals Liberal Alliance The Liberals Estonian Centre Party Estonian Reform party

Previous names

Chapters

Croatian People’s Party – Liberal Democrats

15, 16, 17

Croatian Social Liberal Party Istrian Democratic Assembly

15, 16, 17 15, 16, 17

16, 17 12, 15, 16, 17 12 12, 15, 16, 17 16, 17

Merger of Rahvuslik Koonderakond Isamaa (Pro Patria National Coalition) and Eesti Liberaaldemokraatlik Partei (Estonian Liberal Democratic Party)

2, 15, 16, 17, 18 2, 15, 16, 17, 18 2, 15, 16, 17, 18 13, 15, 16, 17 13, 15, 16, 17

Finland

KESK

Suomen Keskusta

Finnish Center Party

Finland

Liberals

Liberals

Finland

SFP

Finland France

AC UDF

Germany Greece Hungary

FDP/Freie Demokraten Drassi SZDSZ

Liberaalinen kansanpuolue Suomen ruotsalainen kansanpuolue Svenska folkpartiet i Finland Åländsk Centern Union pour la Démocratie Française Freie Demokratische Partei Drassi Szabad Demokraták Szövetsége – a Magyar Liberális Párt

Hungary Iceland Iceland Ireland Ireland Italy

MLP FSF BF Fianna Fáil PD PRI

Italy

PLI

Magyar Liberális Párt Framsóknarflokkurinn Björt framtíð Fianna Fáil Progressive Democrats Partito Repubblicano Italiano Partito Liberale Italiano

Maalaisliitto (until 1965); Keskustapuolue (1965-1988)

1, 15, 16, 17 1, 15, 16, 17

Swedish People’s Party

15, 16, 17

Ålandic Centre Union for French Democracy Free Democratic Party

16, 17 16 8, 15, 16, 17, 18

Action Alliance of Free Democrats – Hungarian Liberal Party Hungarian Liberal Party Progressive Party Bright Future Fianna Fáil Progressive Democrats Italian republican Party

16, 17 15, 16, 17

Italian Liberal Party

15, 16

16, 17 15, 16, 17 15, 16, 17 10, 15, 16, 17, 18 15, 16, 17 15, 16

(Continued )

Table 0.1  List of parties (Continued) Country

Party acronym

Full name in original language

English translation

Italy Italy Italy

IdV ApI R

Italia dei Valori Alleanza per l’Italia Radicali Italiani

Italy of values Alliance for Italy Italian Radicals

Latvia

LC

Latvijas Ceļš

Latvia’s Way

Latvia Lithuania

LA LLS

Lithuania

LCS

Lithuania

LiCS

Lithuania Lithuania

LRLS PTT

Latvijas Attīstībai Lietuvos Liberalų Sąjunga Lietuvos Centro Sajunga Liberalų ir centro sąjunga Liberalų Sąjūdis Partija tvarka ir teisingumas

Latvia’s Development Lithuanian Liberal Union Lithuanian Centre Union Liberal and Centre Union Liberal Movement Order and Justice

Lithuania

NS

Luxembourg Netherlands Netherlands Netherlands

DP D66 LU PvdV

Naujoji Sąjunga (Socialliberalai) Demokratesch Partei Democraten 66 Liberale Unie Partij van de Vrijheid

New Union (Social Liberals) Democratic Party Democrats 66 Liberal Union Freedom Party

Netherlands

VDB

VrijzinnigDemocratische Bond

League of Free Democrats

Previous names

Radical Party (PRPartito Radicale) Latvijas Pirmā Partija (LPP- Latvia’s First Party)

Chapters 15, 16, 17, 18 16, 17 16, 17 13, 15, 16, 17 13, 15, 16, 17 13, 15, 16, 17 13, 15 13, 15, 16, 17

Liberalų Demokratų Partija (LDP-Liberal Democratic Party)

Liberale Staatspartij (until WWII)

13, 15, 16, 17 13, 15, 16, 17 13 15, 16, 17 4, 15, 16, 17, 18 4 4 4

Netherlands

VVD

People’s Party for Freedom and Democracy Left Liberal People’s Party

.N PD

Volkspartij voor Vrijheid en Democratie Venstre Det Liberale Folkepartiet Nowoczesna Partia Demokratyczna

Norway Norway

V DLF

Poland Poland Poland

RP

Ruch Palikota

Palikot’s Movement

Portugal

PLD

Portugal

PDR

Portugal

PSD

Liberal Democratic Party Democratic Republican Party Social Democratic Party

Romania

ALDE

Romania

PDL

Partido Liberal Democrata Partido Democrático Republicano Partido Social Democrata Alianța Liberalilor și Democraților Partidul Democrat-Liberal

Modern Democratic Party

Alliance of Liberals and Democrats Democratic Liberal Party

4, 15, 16, 17, 18  

1, 15, 16, 17, 18 1

  Unia Demokratyczna-UD; Unia Wolności-UW (1994-2005) Kongres LiberalnoDemokratyczny-KLD (1990-1994); Unia Wolności (1994-2005); Platforma Obywatelska-PO

12 12, 15, 16, 17 12, 15, 16, 17

16 16, 17 16, 17 12

 

12

(Continued )

Table 0.1  List of parties (Continued) Country

Party acronym

Romania

PNL

Slovakia

ANO

Slovakia

HZDS

Slovakia Slovakia

SaS SDKU

Slovenia

SMC

Slovenia

LDS

Slovenia

ZaAB

Slovenia Spain

DL .

Spain Spain

. .

Spain

.

Full name in original language

English translation

Previous names

Chapters

National Liberal Party

 

12, 15, 16, 17

Alliance of the New Citizen Movement for a Democratic Slovakia

 

12

Freedom and Solidarity Slovak Democratic and Christian Union

 

Partidul Național Liberal Aliancia Nového Občana Hnutie za demokratické Slovensko Sloboda a Solidarita Slovenská demokratická a kresťanská únia Stranka modernega centra Liberalna demokracija Slovenije Stranka Alenke Bratušek Državljanska lista Partido Liberal

Modern Centre Party

15, 16, 17

Liberal Democracy of Slovenia Party of Alenka Bratušek Civic List Liberal Party

15, 16, 17

Partido Moderado Derecha Liberal Republicana Derecha Liberal Republicana

Moderate Party Republican Liberal Party Liberal Republican Right

12 12 12

15, 16, 17 Partido Progresista until 1868; merged with Alianza Popular (with Union Liberal)

15, 16, 17 11

11 11 11

Spain

UPyD

Unión, Progreso y Democracia Ciudadanos, partido de la ciudadanía Convergència i Unió

Spain

Ciudadanos

Spain

CIU

Spain

UCD

Union de Centro Demócratico

Union, Progress and Democracy Citizens, party of the citizenship Convergence and Union. (An alliance formed by Democratic Convergence of Catalonia (CDC) and the Democratic Union of Catalonia (UDC) (Dissolved in 2015). Note that Convergencia run under the label Democracia y Libertad in 2015 elections) Union of the Democratic Centre

Spain

CDS

Centro Democrático y Social

Democratic and Social Centre

11, 15, 16, 17 11, 15, 16, 17 11, 16, 17, 18

Merge of the Federación de Partidos Demócratas y Liberales, the Partido Demócrata Popular and Partido Liberal (Enrique Larroque)

11

11, 15, 16

(Continued )

Table 0.1  List of parties (Continued) Full name in original language

English translation

Partido Reformista Demócratico Centerpartiet Liberalerna

Democratic Reformist Party Centre Party Liberals

FDP-PLR

FDP.Die Liberalen PLR. Les LibérauxRadicaux PLR. I liberali-radicali

The Liberals

Switzerland

PLS

Swiss Liberal Party

Switzerland

LdU-ADI

UK

Lib Dems

Liberale Partei der Schweiz Parti libéral Suisse Partito Liberale Svizzero Landesring der Unabhängige Alliance des Indépendants Anello degli Indipendenti Liberal Democrats

Country

Party acronym

Spain

PRD

Sweden Sweden

C FP

Switzerland

Previous names

11 Bondeförbundet 1) Folkpartiet Liberalerna 2) Folkpartiet Freisinnig-Demokratische Partei/Parti radical/ Partito radicale (FDP) until 2009 (then: merge with PLS)

Alliance of Independents

Liberal Democrats

Chapters

3, 15, 16, 17, 18 3, 15, 16, 17, 18 6, 15, 16, 17

6, 15, 16, 17

16, 17

Liberal Party (1859-1988); Social and Liberal Democrats (1988)

9, 15, 16, 17, 18

Introduction 17 In Chapter 17, Caroline Close assesses the existence of a liberal party family on the basis of the parties’ ideological or policy orientations. The chapter examines the distinctiveness and homogeneity of the liberal party family in a cross-sectional and longitudinal perspective, by using data from the Chapel Hill Expert Survey and from the Manifesto Project. The analysis shows that the liberal party family has developed a distinct ideology that combines a marked right-wing position on economic issues with centre-left preferences on cultural matters; but it also points out the great diversity of the family, and analyses the relative weight of different orientations of liberalism through time. In Chapter 18, Stefanie Beyens, Emilie van Haute, and Tom Verthé examine what differentiates the organisations of liberal parties from those of other party families. Their analyses show that in terms of degree of organisation, liberal parties tend to display an organisational structure largely similar to other party families. In terms of level of centralisation, liberal parties distinguish themselves by a specific combination of costs and benefits of membership and by more contested leadership selection processes. Using newly available datasets, this chapter conducts the first large scale comparative study of liberal parties’ organisational specificities. Based on both the country and comparative chapters, the conclusion reviews the initial discussion of the notion of party family, and discusses its application to liberal parties in Europe. As part of a series on party families in Europe, it relates this question to the general debate on party families. Furthermore, the conclusions debate on the future of liberal parties in Europe, and on their ongoing impact on politics and policies in Europe. By gathering the best specialists engaged in the study of liberal parties, and by combining in-depth case studies and comparative chapters, this volume fills an important gap in the literature on liberal parties, which remain to this day one of the least studied party family in Europe – despite its considerable importance in the development of European democracies and the European Union. In doing so, the book also greatly contributes to the discussion on the notion of party family in Europe.

References Budge I., Keman H. (1990), Parties and democracy: Coalition formation and government functioning in twenty states. Oxford: Oxford University Press. Camia V., Caramani D. (2012), ‘Family meetings: Ideological convergence within party families across Europe, 1945–2009’, Comparative European Politics 10, pp. 48–85. Retrieved from https://doi.org/10.1057/cep.2011.1 Close C. (2018), ‘Parliamentary party loyalty and party family: The missing link?’, Party Politics 24(2), pp. 209–219. De Winter L. (ed.). (2000), Liberalism and liberal parties in the European Union. Barcelona: Institut de Ciències Polítiques i Socials.

18  Caroline Close and Emilie van Haute De Winter L., Marcet J. (2000), ‘Liberalism and liberal parties in the European Union’. In: De Winter L. (ed.), Liberalism and liberal parties in the European Union. Barcelona: Institut de Ciències Polítiques i Socials, pp. 11–24. Delwit P. (ed.). (2002a), Libéralismes et partis libéraux en Europe. Brussels: Éditions de l’Université de Bruxelles. Delwit P. (2002b), ‘Le libéralisme politique: une famille politique complexe’. In: Delwit P. (ed.), Libéralismes et partis libéraux en Europe. Brussels: Éditions de l’Université de Bruxelles, pp. 7–15. Dinas E., Gemenis K. (2010), ‘Measuring parties’ ideological positions with manifesto data. A Critical evaluation of the competing methods’, Party Politics 16, pp. 427–450. Retrieved from https://doi.org/10.1177/1354068809343107 Elff M. (2013), ‘On the distinctiveness of party families’, Presented at the 71th Annual Conference of the Midwest Political Science Association, Chicago. Ennser L. (2012), ‘The homogeneity of West European party families. The radical right in comparative perspective’, Party Politics 18, pp. 151–171. Retrieved from https://doi.org/10.1177/1354068810382936 Enyedi Z., Linek L. (2008), ‘Searching for the right organization. Ideology and party structure in East-Central Europe’, Party Politics 14, pp. 455–477. Freire A., Tsatsanis E. (2015), ‘Party families, ideological distinctiveness and cohesion: A strong test of the heuristics of the concept of “familles spirituelles”’, Presented at the ECPR General Conference, Montreal. Katz R.S., Mair P. (eds.). (1994), How parties organize. Change and adaptation in party organizations in western democracies. London: Sage Publications. Kirchner E.J. (ed.). (1988), Liberal parties in Western Europe. Cambridge: Cambridge University Press. Klingemann H.-D., Volkens A., Bara J., Budge I., McDonald M.D. (2006), Mapping policy preferences II: Estimates for parties, electors, and governments in Eastern Europe, European Union, and OECD 1990–2003. Oxford: Oxford University Press. Knutsen O. (2008), Class voting in Western Europe. A comparative longitudinal study. Plymouth, MA: Lexington Books. Kriesi H., Grande E., Lachat R., Dolezal M., Bornschier S., Frey T. (2006), ‘Globalization and the transformation of the national political space: Six European countries compared’, European Journal of Political Research 45, pp. 921–956. Laver M., Hunt W.B. (1992), Policy and party competition. London: Routledge. Lipset S.M., Rokkan S. (1967), Party systems and voter alignments: Cross-national perspectives. New York: Free Press. Mair P., Mudde C. (1998), ‘The party family and its study’, Annual Review of Political Science 1, pp. 211–229. Retrieved from https://doi.org/10.1146/annurev. polisci.1.1.211 Marks G., Wilson C.J., Ray L. (2002), ‘National political parties and European integration’, American Journal of Political Science 46, pp. 585–594. Retrieved from https://doi.org/10.2307/3088401 Pedersen M.N. (1982), ‘Towards a new typology of party lifespans and minor parties’, scandinavian political studies 5, pp. 1–16. Retrieved from https://doi. org/10.1111/j.1467-9477.1982.tb00256.x Poguntke T., Scarrow S.E., Webb P.D. (2016), ‘Party rules, party resources, and the politics of parliamentary democracies: How parties organize in the 21st century’, Party Politics 22, pp. 661–678. Roussellier N. (1991), L’Europe des libéraux. Paris: Editions Complexe.

Introduction 19 Scarrow S.E., Webb P.D., Poguntke T. (eds.). (2017), Organizing political parties: representation, participation, and power, comparative politics. Oxford: Oxford University Press. Seiler D.-L. (1980), Partis et familles politiques. Paris: P.U.F. Seiler D.-L. (2002), Le paradoxe libéral: la faiblesse d’une force d’avenir. In: Delwit P. (ed.), Libéralismes et partis libéraux en Europe. Brussels: Éditions de l’Université de Bruxelles, pp. 37–56. Smith G. (1988), ‘Between left and right: The ambivalence of European liberalism’. In: Kirchner E.J. (Ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 16–28. van der Brug W., Hobolt S.B., De Vreese C.H. (2009), ‘Religion and party choice in Europe’, West European Politics 32, pp. 1266–1283. van Haute E. (ed.). (2016), Green parties in Europe. London & New York: Routledge. Vandermotten C., Médina Lockhart P. (2002), ‘La géographie du libéralisme belge et européen’. In: Delwit P. (ed.), Libéralismes et partis libéraux en Europe. Brussels: Editions de l’Université de Bruxelles, pp. 57–74.

Part I

Case studies

1

The Norwegian Left and the Finnish Centre What, no capital ‘L’ liberal parties? David Arter

Introduction The two parties that are the focus of this chapter, the Norwegian Venstre (Liberals – literally ‘Left’ – V) and Finnish Suomen Keskusta (Centre – KESK), originated before the achievement of national independence in Norway in 1905 and Finland in 1917 and both opposed the ‘imperial connection’ – in Norway the federative function vested in the Swedish crown and in the Finnish Grand Duchy the Russian Czar’s status as head of state. Both parties played a decisive role in the resolution of the constitutional question in the two countries. But there the similarity largely ends. The Norwegian Venstre founded in January 1884 is the oldest Norwegian party and emerged as a ‘challenger movement’ – more a ‘coalition of interests’ than a political party. Grounded in a standardised composite of rural dialects landsmål (now nynorsk) it worked to achieve universal suffrage, full freedom of religion, and an end to the Union with Sweden. The predecessor of the Finnish Centre, Maalaisliitto (the Agrarian Party) emerged in connection with electoral and parliamentary reform in 1906 to represent the interests of the numerous body of family-sized farmers in two main regions – the northwest (Pohjanmaa) and southeast (Karelia). It was a class party that, through a programme of land reform creating land for the landless (crofters, hillsiders, and agricultural labourers), built a mass base in the post-independence 1920s, giving the new Republic a solid bourgeois foundation in the process. The profile of the two parties could not be more different. Venstre has never changed its name; the Finnish Agrarian Party became the Centre Party (Keskustapuolue) in 1965 and then the Finnish Centre in 1988. The Finnish Centre has been the largest party in three of the four general elections between 2003 and 2015 and the largest party on five occasions since World War II. Venstre in contrast has not been the largest party since 1915 and gained a modest 4.9% at the last general election in 2017. Whereas the Finnish Centre has provided four prime ministers since 2003, 1935 was the last time there was a Venstre prime minister. Crucially, whereas the Finnish Centre, despite the existence of a strongly ‘Eurosceptic’ farmer wing, backed the country’s application to join the European Union in 1992 – endorsed in

24  David Arter a referendum two years later – Venstre split over Norway’s third application for European Economic Community (EEC) membership in 1972 and a pro-EEC minority formed a splinter party, which changed its name to Det Liberale Folkepartiet (Liberal People’s Party – DLF) in 1980. Venstre never really recovered from the division over European integration. The Finnish Centre applied for observer status in the Liberal International in 1983 and full membership in 1988. However, the argument advanced in this chapter is that the Finnish Centre cannot be regarded as a liberal party on the basis of its origination, ideology, or until the 1980s cross-national links. Rather, its claim to liberal credentials was part of a modernisation programme designed further to promote the party’s transformation from class to catchall party. Venstre has traditionally defined itself as a social liberal party but electoral decline, and the threat of falling below the 4% national qualifying threshold for top-up parliamentary seats, has prompted the search for a niche identity and to the current claim that it is the greenest of the non-Socialist parties.

Origins and development Until the completion of mass democracy (universal suffrage) and electoral reform (the introduction of proportional representation) shortly after World War I, Norway witnessed a system of tripartism. The (non-Socialist) Left (Venstre) was formed 28 January 1884, and notably under the leadership of Johan Sverdrup demanded the accountability of the political executive (nominated by the Swedish monarch) to the Norwegian parliament, the Storting. Despite the manifest reluctance of King Oscar II to give ground, the principle of parliamentarism was conceded in 1885 and at the general election that year Venstre gained 63.4% of the vote and 76 of the 123 Storting seats. The right (Høyre) was organised in response to the Liberal Associations (Venstreforeninger) that formed across the country to promote the demand for parliamentarism and it comprised those ‘establishment elements’ broadly supportive of the Union with Sweden. The Labour Party, which emerged in 1887, mobilised both the rural and urban workers and, under the French-style single-member, double-ballot system, threatened to emasculate the left and right as legislative actors. The shift to a proportional (PR) closed-list voting system in 1919, therefore, represented an historic compromise that recognised the magnitude of Labour’s electoral growth and sought to salvage at least a measure of legislative representation for the two old left-right parties. The introduction of PR in 1919 undermined the ‘movemental’ character of Venstre and accelerated the split into its component parts. In its earliest years Venstre represented the farmers, cottagers (husmenn), day labourers, and other workers but in addition to Labour’s rise, the electoral challenge to Venstre was heightened in 1921 by the emergence of a class-based Agrarian Party, which renamed itself Centre Party in 1959. The temperance

The Norwegian Left and the Finnish Centre 25 movement was Norway’s largest popular movement in the first decades of the 20th century and Venstre was the leading temperance party and instrumental in the introduction of prohibition following a referendum in 1919. Not all Venstre’s supporters, however, took an abstemious line on alcohol. In Oslo (then Kristiania) 78% voted against prohibition and the Venstre newspaper Dagbladet was a declared opponent (Johansen 2013: 60). Legislation ended prohibition in 1926 but in the temperance stronghold of Hordaland county in the southwest – where 80.8% voted in favour of prohibition – opposition to the opening of liquor outlets in Bergen (54.5% against prohibition in 1919) spawned the creation in 1933 of a Christian People’s Party and this became a nationally based party in 1945. The 20th-century Venstre fragmented further over the question of European integration. Venstre supported the first Norwegian application to join the Common Market in summer 1961 but it splintered over Norway’s third application in 1972, a pro-EEC minority forming the Det Nye Folkepartiet (New People’s Party), which changed its name to Liberal People’s Party in 1980. At the time of the rupture, 8 of the 13 Venstre MPs joined the new party but at the 1973 general election it gained only a solitary seat (Hordaland) and lost that four years later. In 1988, the Liberal People’s Party reunited with Venstre. The fragmentation of the old non-Socialist left first along functional lines (with the creation of the Agrarian Party), then along a secular-­confessional axis (the emergence of the Christian People’s Party), and finally over Europe (with the defection of a pro-EEC minority) was reflected in declining electoral results. In the five general elections between 1906 and 1918, Venstre averaged 35.4% of the vote and it was the largest party on three occasions. In the six general elections between the wars (1921–1936) Venstre averaged 18.2% of the poll. Over the 17 general elections between 1945 and 2013 Venstre averaged 6.2% and it fell below the threshold of representation over the period 1985–1993. By the end of World War II, the left-right bipartisanism of the 1880s had been replaced by an historic feature of all the mainland Scandinavian party systems, namely a fragmented non-Socialist configuration of parties consisting (in Norway) of a basic Liberal-Agrarian-Christian-Conservative quadripartism. The Venstre-derivative parties (Liberals, Agrarians, and Christian People’s Party) competed on, and for the centre ground and were prospective partners in coalition-building (Table 1.1a). See Table 1.1a at http://partyfamilies.eu/index.php/category/chapters/ chapter-1/ Unlike the disparate movemental character of the nascent Venstre, the Finnish Centre originated in 1906 as a typical small farmer-peasant party, largely antedating the mechanisation of the forest industry and the commercialisation of agriculture. It was a class party based on the numerous body of independent, family-sized farms concentrated in the northwest and southeast of the country. The party was a by-product of Czar Nicholas II’s

26  David Arter 1905 October Manifesto that facilitated the superimposition of mass democracy on an overwhelmingly rural society in the Finnish Grand Duchy. All men and women of age 24 or older were entitled to vote in elections to a new unicameral parliament Eduskunta. Grounded in the radicalism of the Finnish-speaking peasantry in the geographic peripheries and its visceral prejudice towards the largely Swedish-speaking and urban politico-economic elite (Arter 2001), the Agrarians remained a minor, regionally based party before independence in 1917. The crofters and landless rural proletariat were largely seduced by the messianic messages of the Social Democrats, which in 1916 gained the only absolute majority of seats ever gained in the unicameral Eduskunta. The Agrarians backed the White forces during the Finnish civil war that followed independence and then played a strategic role in the enactment of a republican constitution in July 1919. Legislation in the 1920s creating land for the landless – that is, a programme of social engineering – cemented the bourgeois foundations of the new state and contributed to the exponential growth in Agrarian support. By 1930 the Agrarians registered an all-time record poll of 27.3% (Table 1.1b). See Table 1.1b at http://partyfamilies.eu/index.php/category/chapters/ chapter-1/ Two wars with the Soviet Union – the Winter War of 1939–1940 and so-called Continuation War of 1941–1944 – led to the need to resettle a 400,000 refugee population from the Agrarian heartland of Karelia (conceded to the Russians), and to the imposition of a substantial war indemnity repayable largely in terms of heavy goods and machinery. The concomitant acceleration in the industrialisation process transformed the predominantly agricultural society and led as early as the 1950s to radical elements in the Agrarian Youth Organisation to push the need for an adaptive strategy. A new programme in 1962 presaged a change of name – from Agrarian Party to Centre Party – at an extraordinary party conference in October 1965. When the Agrarians changed their name they were the largest single party, polling 23.0% at the 1962 general election and its chair Johannes Virolainen led a non-Socialist coalition that included the Conservatives (Kansallinen kokoomus). In contrast, when the Centre Party became simply the Finnish Centre in 1988 it did not do so from a position of comparable strength; indeed, the early 1980s have been described as the party’s ‘danger years’ (Kääriänen 2002). First, it lost the presidency following the resignation through ill health of the long-serving, former Agrarian head of state, Urho Kekkonen. Second, it languished behind both the Social Democrats and Conservatives in the electoral arena, its support undercut by the populist challenge of former Agrarian Veikko Vennamo (and subsequently his son Pekka) and after the 1987 general election it was obliged to enter opposition for the first extended period since its change of name. Although written off as a ‘sunset party’ in the context of the transition to a predominantly service-based economy,

The Norwegian Left and the Finnish Centre 27 the Finnish Centre has defied the ‘doom-merchants’ to become the leading party in post-Nokia Finland. When the Agrarians became a Centre-label party in 1965, the Finnish electoral party system was characterised by a bifurcated left – two equally sized parties, the Communist-dominated Finnish People’s Democratic League (SKDL) and the Social Democrats (SDP) – and no less than five non-Socialist parties at, or to the right of centre. Even more than in Norway, Finland was marked by a fragmented, non-Socialist collection of parties. The Conservatives (National Coalition) and the Swedish People’s Party (Svenska Folkpartiet), the organ of the national language minority, formed part of the historic party system. The latter dated back to the first general election to the reformed legislature in 1907, whereas the Conservatives emerged from those who had canvassed a constitutional monarchy rather than a republican form of government in 1919. In addition, two newer parties were influenced by the Agrarians’ change of name. In December 1965, the Finnish People’s Party (Suomen kansanpuolue) and the Liberal League (Vapaamielistenliitto) joined forces to form the first capital ‘L’ Liberal Party in Finland (Liberaalisen kansanpuolue) that pledged to pursue ‘integrationist centrist politics’ (Borg 2015: 46). In 1966, the Finnish Smallholders’ Party (Suomen Pientalonpoikien Puolue), founded by the dissident Agrarian Veikko Vennamo, renamed itself the Finnish Rural Party (Suomen Maaseuden Puolue – SMP), not least in the hope of profiting from the defection of traditionalist elements in the Agrarian ranks disaffected with the party’s new designation and catchall electoral strategy. In short, when the Agrarians opted to become a Centre-label party, they faced competition to hold on to their rural electorate from the SMP (and in the north and east the Communists) and competition to win over an urban electorate from the Liberals (in addition, of course, to the Conservatives and Social Democrats). Following the example of the Swedish sister party in 1957 and the Norwegian Agrarians two years later, the Finnish Agrarian Party changed its name in 1965 primarily as an adaptive electoral strategy designed to expand into the towns and compensate for the rapid decline in the numbers employed in the agricultural sector. The manifest logic was the need to respond to social structural change, which was reducing the size of its core electorate – its classe gardée. However, implicit in the Centre-label, it may be presumed, was the pursuit of centrism as a political strategy and a concern to differentiate the party from, and position it between, left and right as an optimal office-seeking course.

Electoral performance The trajectory of Venstre’s support since its origins in the last quarter of the 19th century has been strongly downward. In 1885, Venstre gained 63.4% of the electorate – which comprised primarily freehold farmers – and was

28  David Arter the largest party. One hundred years later, it polled a miserly 3.1% and failed to cross the threshold of representation (Table 1.2a). Venstre’s original strength was in the rural municipalities in the south and west where the strength of local cultural traditions based on religious fundamentalism, the rural language landsmål and prohibition made it more difficult for Labour to penetrate when the franchise was extended to working men in 1900 (Kinsey 2006: 261–283). In the first Storting election after World War II in 1945, when Venstre polled 13.8%, the party, although weaker in the east, retained substantial support in areas of the south and west. Venstre split over Norway’s third EEC application in 1972 and has never really recovered. From 1977 it profiled itself as an environmental party and began to attract a younger cohort of supporters (Knutsen 2003: 132), but between 1985 and 1993, it fell below the threshold of representation and was no longer a legislative actor. See Table 1.2a at http://partyfamilies.eu/index.php/category/chapters/ chapter-1/ To compete for the 19 compensatory seats, a party has to win at least 4% of the national vote, and Venstre’s regular struggle to do so has created incentives for supporters of other (non-Socialist) parties to cast a ‘tactical coalition vote’. In 2005, Venstre advanced by two percentage points as a result of tactical voting; and before the 2013 Storting election, Venstre’s deputy leader publicly encouraged Conservative voters to vote tactically so as to oblige the Conservatives (the prospective election winners) to cooperate with Venstre and the Christian Democrats rather than the radical right Progress Party (Jensen 2016: 29). Jensen (2016) claims that 18.3% of Norwegian voters in 2013 cast a tactical vote, although there is survey evidence (albeit based on a very low N) that in 2013 27% of Venstre’s vote came from those who voted Labour in 2009 compared with only 11% who voted Conservative (Aardal, Bergh, and Haugsgjerd 2014: 13). Over the four general elections in the new millennium (2001–2017), Venstre has averaged 4.8% of the poll. According to the 2013 election study (Kleven et al. 2013: 23) three quarters of its vote derived from the salariat – public and private-sector employees – while two thirds were educated to university level. Although clearly battling to remain a parliamentary party, the evidence indicates an increase in party loyalty among Venstre voters. In 2009, 30% of Venstre supporters had backed the party in 2005, whereas in 2013, 59% of Venstre voters had backed the party four years earlier. In Oslo in 2013 Venstre’s poll was 8.2%, an increase of 1.8 percentage points compared with four years earlier. Unlike Venstre, the Agrarian-Centre Party in Finland has demonstrated remarkable electoral resilience, defying its portrayal in the 1970s and 1980s as a ‘sunset party’. Over the 20 general elections since World War II, the Agrarian-Centre has averaged 20.8% of the vote – making it one of the larger parties in the Finnish multiparty context – although for the two decades after its change of name in 1965 the redesignated party appeared threatened

The Norwegian Left and the Finnish Centre 29 with irreversible decline. In the six general elections between 1945 and 1962, the Agrarians averaged 23.2%; in the 14 general elections between 1966 and 2015, the Centre averaged 19.8%. Yet despite this fall in the average level of its support, the Centre has been the largest party on four occasions since the name change – polling its best postwar result of 24.8% in 1991 – while on two occasions (1972 and 2011) it received under 18% and plummeted to a postwar nadir of 15.8% in the last-mentioned year. Stated another way, since adopting the Centre label the party’s vote has become more volatile in line with the increased volatility of the electorate as a whole. In the six postwar elections as an Agrarian Party the average volatility (Pedersen index) for the entire electorate was 5.9%; over the 14 general elections since the change of name it has risen to an average 9.2%. In its traditional rural heartlands the Centre’s main electoral challenge has come from the populist SMP and its successor the True Finns (now simply ‘The Finns’ Party) (Perussuomalaiset – PS) and in this respect the 1970 and 2011 general elections were particularly traumatic. 1970 was the first of the Scandinavian ‘earthquake elections’ and the Pedersen volatility index climbed to 14.7%. The Centre was overtaken by the Conservatives as the largest non-Socialist party whereas the SMP claimed 10.5% of the vote – an advance of 9.5 percentage points on four years earlier when Veikko Vennamo was the party’s solitary MP. In the following three general elections between 1972 and 1979, the Centre averaged only 17.1% and languished in fourth place behind the Social Democrats, Conservatives, and Communists. At that stage the change of name was questioned as a serious tactical error and the fear was that the Centre Party was in terminal decline. In 2011, the PS advanced by an unprecedented 15 percentage points while the Centre plunged to its lowest vote since the October 1917 general election – two months before Finland declared its independence. The question of bailouts for Greece and Portugal were timely for the uniformly and strongly Eurosceptic PS – which ran a full slate of candidates in every mainland district for the first time (Arter 2013) – whereas the Centre as the prime minister’s party since 2003 looked stale and was obliged to support the proposed EU stabilisation mechanism despite strongly anti-EU sentiments within the party, particularly from the core farming elements. Even then the Centre remained the largest party in 2011 in its six stronghold districts in the centre and north of the country – albeit declining there by an average of 10.5 percentage points. But in the capital of Helsinki, only the Christian Democrats of the eight parliamentary parties polled a lower vote and in the neighbouring ‘dormitory’ district of Uusimaa, the largest of the 14 mainland districts (district magnitude = 35 in 2011), it lost 43% of its vote. The disastrous 2011 election result followed eight years of Centre-led coalition. In contrast, as an opposition party in 1991, the Centre achieved its best postwar result of 24.8% (Table 1.2b). This was against the backdrop of four years of Conservative-Social Democrat-led coalition, the imminent slide into the deepest recession in Finnish history and the election of a new

30  David Arter and highly media-friendly Centre leader Esko Aho. The result nonetheless pointed to familiar fault lines in the Centre’s electoral base. In the seven central and northern constituencies, the party averaged 38.2% in 1991; and in the Lapland constituency, it gained an absolute majority of the vote. In the seven southern districts it averaged only 17%. Moreover, although the Centre elected its first MP for Helsinki, its poll of 6% in the capital city was still well under half that of the Greens and although it doubled its vote in Uusimaa it remained smaller than the Swedish Peoples Party. See Table 1.2b at http://partyfamilies.eu/index.php/category/chapters/ chapter-1/ Whether the Centre may be said to have made the transition from class to catchall party depends very much on the criteria applied. Geographically, the Centre draws substantial support in the rural market towns; it has significantly less appeal in the southern cities. More specifically, in the 2015 election study, respondents were asked whether they lived in a city centre, suburb, municipality (commune) or outlying rural area. 20% of Centre voters reported living in a city centre, 17% in a suburb, 29% in a municipality, and 47% in an outlying rural area. In socioeconomic terms the Centre is overwhelmingly the strongest party among farmers (the original core class). In 2015, 79% of agricultural entrepreneurs backed the Centre (Westinen 2016: 252). Equally, the party was supported by 1 out of 5 entrepreneurs, administrative staff, and junior clerical workers. Overall, Finnish elections have witnessed a heightened incidence of vote-switching in the context of a growth in the service-sector middle class (salariat) and a decline in traditional working class jobs. The blue-collar workers in particular have become increasingly fickle in electoral terms, traditionally favouring the two left-wing parties, more recently the Conservatives and PS and now it seems the Centre. A state broadcasting poll (Yle Politiikka 8.7.2016) published in July 2016 made the Centre the most popular party among the blue-collar working population.

Participation in government The Norwegian Venstre has supplied seven prime ministers – albeit all of them before 1935 – and since World War II has participated in six governing coalitions (Table 1.3a). It is presently a ‘support party’ for the ConservativeProgress Party minority coalition formed after the 2017 general election. Since its name change, the Finnish Centre has provided eight prime ministers; it participated in all seven majority coalitions between 1966 and 1987, and since then it has led four majority coalitions either with the Conservatives or the Social Democrats (Table 1.3b). Legislative party strategies do not lend themselves to ready generalisation and inevitably reflect circumstantial elements. However, since World War II Venstre may be said to have pursued a non-Socialist bloc alignment strategy. The Finnish party

The Norwegian Left and the Finnish Centre 31 system has never been bloc-based, not least because of the size and strategic position of the Agrarian Party. Rather, during the years of electoral decline that followed its change of name, the Centre pursued a centrist bloc-building strategy both to strengthen its bargaining position in the government formation process and to challenge the domination of the Social Democrats. The Venstre bloc alignment strategy has taken the form of either an optimal version, that is a governmental association with all the parties of the centre-right (excluding the radical rightist Progress Party) or a suboptimal version comprising three of the four centre-right parties. In the first category there was the Lyng (1963) and Borten (1965–1971) four-party, non-Socialist coalitions; in the latter the Bondevik I centrist minority (1997–2000) and the Bondevik II in which the Conservatives replaced the Centre Party. Unlike the Danish Social Liberals (Radikale Venstre), the Norwegian Venstre has never sought (or been able) to pursue a ‘hinge party strategy’ (Arter 2016) of keeping government options open to both left and right. For a small centrist grouping in a Labour-dominant party system, the only realistic pathway to office has been via non-Socialist bloc cooperation in its optimal or suboptimal form. See Table 1.3a at http://partyfamilies.eu/index.php/category/chapters/ chapter-1/ Relative to its legislative size, Venstre has enjoyed relatively high coalition potential. Between 2001 and 2005, for example, it had only two Storting seats but boasted three cabinet ministers. This was because of the so-called ‘incompatibility rule’ that prevents ministers from serving as members of parliament. However, particularly since 1989 when the fiercely anti-EU Centre Party left the Syse coalition, non-Socialist bloc cooperation has been compromised by the question of European integration. The Centre and pro-EU Conservatives have not cooperated since the collapse of the aforementioned Syse cabinet, although in 2001 the Conservatives, Christian People’s Party, and Venstre agreed to end the Bondevik II coalition if the EU membership question was raised – the so-called ‘suicide clause’ (Sitter 2006: 579). The Centre ‘defected’ to become part of a left-centre majority coalition under the Labour leader Jens Stoltenberg between 2005 and 2013. A more recent obstacle to the formation of a non-Socialist bloc alternative has been the rise of the Progress Party, which in the six elections between 1997 and 2017 averaged 17.7% of the vote. Viewed as a racist, populist, and decidedly illiberal party, Venstre’s inclination has been to ostracise the Progress Party. In April 2008, the Venstre leader Lars Sponheim signed a written ‘guarantee’ in Norway’s largest newspaper VG undertaking not to form a coalition with, or act as a support party for the Progress Party (Allern 2010: 905). Since then, Venstre has moderated its approach and participated in a new form of non-Socialist bloc cooperation – a written agreement obliging the minority government of Conservatives and Progress Party formed in autumn 2013 to consult with its ‘support parties’, Venstre

32  David Arter and the Christian People’s Party, on a regular basis on major policy issues and all matters specified in the agreement (Allern and Karlsen 2014: 660). In January 2018, Venstre relented further to enter a minority coalition with the Conservatives and Progress Party. We noted earlier that implicit in the Agrarians’ adoption of the Centre label was the pursuit of centrism as a legislative strategy and a concern to differentiate the party from the left and right, while positioning it between the two as an optimal office-seeking course. However, for the two decades after its name change there were severe constraints on the Centre’s room for manoeuvre in the government formation process. First, electoral decline weakened the party’s strategic position. In 1966, the two left-wing parties won an overall legislative majority and the Social Democrats replaced the Centre as the largest party. By 1970, the Conservatives had become the largest non-Socialist party, a position they held until 1991. Second, the (previous) 1919 Finnish constitution vested the president with the task of nominating governments and between 1966 and 1981 the long-serving head of state Urho Kekkonen played an active role in the coalition-building process, prevailing upon his former party, the Centre, to participate in a series of broad left-centre ‘Popular Front’ governments against the wishes of the grassroots. Third, a coalition option to the right was precluded by dint of the Conservatives’ status as a ‘pariah party’. Following the traumatic 1970 ‘earthquake election’, the Centre party in the country, disaffected with Popular Front involvement, wanted the two election ‘winners’, the Conservatives and SMP, to form a coalition, but the Soviet Union and Kekkonen ‘persuaded’ the Centre chair Virolainen to reenter a centre-left coalition. This convinced the Centre that ‘domestic politics had become an arm of foreign policy’ (Isohookana-Asunmaa 2006: 654). The Centre’s strategic dilemma can be summarised as follows: (1) In 1966, the two left-wing parties had won an absolute majority of legislative seats. (2) The radical left (Communists), backed by more than one fifth of the electorate, would not cooperate with the Conservatives. Until the late 1980s the Social Democrats, consistently the largest party, would not do so either. (3) The Soviet Union also would not accept Conservative participation in Finnish coalitions, with the result that the Conservatives became a ‘pariah party’. (4) Kekkonen wanted to build surplus majority governments – qualified legislative majorities were needed for much economic legislation – and the Popular Front format, the two left-wing parties together with the Centre, provided in practice the only viable formula. However, the Centre’s perennial governmental presence with the political left further eroded its support and allowed the SMP to feed off its disaffected voters. The Centre’s response was centrist bloc-building. As early as 1967 the Centre chair Virolainen initiated discussions with the Liberals and Swedish People’s Party and there was trilateral consultation and electoral cooperation throughout the 1970s. At the end of 1972, the Centre, Liberals, Swedish People’s Party, and a number of modernisers

The Norwegian Left and the Finnish Centre 33 (remonttimiehet) in the Conservative Party (who wanted to gain the president’s trust as a potential governing party) planned to strengthen non-Socialist cooperation, the ultimate idea being to create a new party. The project name for this ‘new centre’ was ‘K-80’; it was a highly secretive venture (not known to the Centre chair Virolainen) and it ultimately came to nothing (Isohookana-Asunmaa 2006: 402–405; Kääriänen 2002: 166–170; Borg 2015: 125–138). However, in autumn 1980 the new Centre Party chair Paavo Väyrynen and his Liberal People’s Party counterpart considered the formation of a new party, the Centre Union (Keskustaliitto) (Kääriänen 2002: 226). In May 1981, a Centre Council (Keskustan valtuuskunta – KEVA) was formed, comprising the Centre, Liberals, and Swedish People’s Party, and it was expanded at the end of 1985 by the inclusion of the Finnish Christian League (Suomen Kristillinen Liitto). Significantly, following a poor election in 1979 and a subsequent drop in state party funding, the Liberals, at their party conference in summer 1982, voted to become a member organisation of the Centre Party that it was felt would be able to assist the party financially. Lacking an independent identity, the Liberals lost all their MPs at the general election the following year and many of their young ‘names’ defected to the Greens (Arter 1988). According to the Centre Party secretary Seppo Kääriänen (2002: 170), ‘co-operation between the centrist parties represented one of the few strategic weapons in the Centre Party’s arsenal that allowed the party to hold its own against the Social Democrats and the radical left both in general elections and government policy-making’. He concluded that without it the Centre’s ‘danger years’ would have proved truly fateful. Things looked very different from the Liberals’ perspective, and one of its leading figures referred to the conference vote to become a member organisation of the Centre Party as ‘political suicide’ (Borg 2015: 240). Centrist bloc-building undoubtedly served the Centre Party well. The incorporation of the Liberals enabled the Centre to make some headway in the southern towns – so enhancing the party’s ‘catchall tendencies’ (Kääriäinen 2002: 266) – and in the trade unions (Borg 2015: 243). However, centrist party cooperation lost much of its momentum in 1987 when the Centre was forced into opposition – the Liberals had in any event gone solo again in 1986 – and the Centre Council was wound up. See Table 1.3b at http://partyfamilies.eu/index.php/category/chapters/ chapter-1/

Ideology and policy positions Klaus von Beyme (1985) distinguishes between nine party families defined by their basic ideology, and although the Norwegian Venstre can readily be located within the ‘Liberal and Radical’ family, the Finnish Centre Party originally belonged to the Agrarian family. Described another way, although some parties (Venstre) may boast an historic liberal identity, others

34  David Arter (Finnish Centre) have sought to assume a liberal identity as an element in a party change strategy. It was hardly coincidental that when it became simply the Finnish Centre in 1988, the party, concerned to modernise its image, applied for full membership of the Liberal International. The differential ancestry of the two parties is not diminished when noting that Venstre emerged as more a ‘coalition of interests’ (Leiphart and Svåsand 1988: 305) than an organisation grounded in a single, coherent ideology. It was not so much a party as a movement moulded out of diverse elements – nationalist, liberal, radical, and democratic – albeit with two main streams providing the core. First, there were the farmers who quickly organised Venstre associations across the country. Then there were urban radicals – skilled workers, intellectuals, and office personnel. An amalgam of different ideological influences, Venstre none the less pursued a liberal agenda, campaigning for parliamentarism, extending the franchise to all men and women, ending the Union with Sweden, promoting teetotalism, and supporting the [rural] language movement. Venstre has retained its original name (meaning ‘Left’) and the first express reference to the party pursuing a social liberal programme was not until 1949 (Malmström 2005: 148), two years after becoming a founder-member of the Liberal International. But its liberal credentials can hardly be in question. Over the past two decades, Venstre’s policy profile has revolved around three main areas – the environment, education, and social policy. At the 2014 party conference, a resolution was passed supporting legislation designed to make Norway an effectively zero emission country by 2050. At the same party conference, a resolution backed increasing freedom of choice in Norwegian schools, and at the 2016 party conference the focus was on youth and children, including digitalisation in schools. In social policy Venstre is committed to an active role for the state but has advocated cuts in the [very generous] level of sick pay for short-term illness designed to enhance the long-term sustainability of the Norwegian welfare model. It also favours devolving government jobs out of Oslo (the influence of the Danish Venstre has been evident in this) and increasing labour migration as part of a multicultural societal blueprint. Young Venstre (Unge Venstre) has been consistently more radical than the mother party, especially on the question of European integration. It currently favours a new European constitution predicated on federalist lines, a larger, more democratic EU with less emphasis on individual nation-states and Norwegian membership of the EU. Summing up, Venstre has consistently profiled itself as a social liberal party, viewing the state as a provider of equal opportunities and the welfare state as a necessary condition for enforcing positive liberty (Almeira 2012: 99) In the Chapel Hill Expert Survey in 2010, ten Norwegian party experts ranked the seven legislative parties from 0-10 on (1) an economic left-right scale; (2) a green-alternative-libertarian (GAL) versus a traditionalist-authoritarian-nationalist (TAN) scale; and (3) a general leftright continuum. Venstre was to the right of centre (5.80 compared with an

The Norwegian Left and the Finnish Centre 35 average of 5.07) on the economic left-right scale and a general left-right scale (5.44 compared with an average of 5.10) but well to the left on the GAL-TAN scale (3.70 compared with an average of 5.33). Indeed, only the Socialist Left Party was further to the left. Until the 1920s, the Finnish Centre Party, or more precisely its predecessor, was grounded in a type of peasant agrarianism (talonpoikais agrarismi) that was fundamentally nationalist and populist in character. It was nationalist to the point of cultural isolationism in championing the Finnish language against the largely Swedish-speaking Establishment (factory owners, large landowners, and professional elites) and populist in juxtaposing the ‘honest toil’ of the rural bumpkin (maanmoukka) against the sybaritic and parasitical lifestyle of the urban bigwig (herraviha). Economic aims were secondary (Mylly 1989: 109). Rather, the essence of peasant agrarianism was a Jeffersonian-style vision of an independent small-farming Finland that was to be achieved by means of wholesale social engineering and an extensive land reform programme creating land for the landless. A far from secondary aim of the 1920s land reform legislation was to insulate the newly independent republic against penetration from the radical left and communism in particular. The Agrarian-Centre’s initial peasant radicalism was reflected in its international contacts. In 1928, it joined the Green International, which had its secretariat in Prague, and there are references in the organ of the movement, the Bulletin de Bureau International Agraire, to the Finnish and other fraternal parties. Initially, the Finnish party was an active participant but its involvement had become intermittent by the mid-1930s. By then the commercialisation of agriculture and the expansion of the party’s support into the larger farming districts in the south meant that peasant agrarianism had evolved into a producer agrarianism (tuottaja agrarismi). On this basis, the Finnish Agrarians had developed close cooperation with their Swedish sister party Bondeförbundet by the end of the 1930s (Mylly 1978: 57). In the 1940s, the Agrarian parties in Finland, Norway, and Sweden created a cooperative organ that met twice yearly – usually in connection with a party conference – while thereafter cooperation continued within the framework of the interparliamentary Nordic Council that Finland joined in 1956, four years after its foundation (Hokkanen 2002: 29). Unlike its sister parties in Sweden and Norway, the Finnish Centre has not profited from ‘hothouse’ single-issue growth. The Swedish Centre grew strongly on an anti-nuclear power stance in the early 1970s, and the Norwegian Centre grew on the basis of its unequivocal opposition to EU membership at the 1993 general election. In the Finnish case, strategic considerations and internal division have dictated a different narrative. In the 1980s, the Centre leader Väyrynen’s concern that the opposition of the party majority to nuclear power would bring together the pronuclear Social Democrats and Conservatives prompted his proposal for a ‘new Popular Front’ (uusi kansanrintama) comprising the Social Democrats, Conservatives, and Centre (Kääriäinen 2002: 223). On the EU membership question, the Centre

36  David Arter was deeply divided and in Kääriäinen’s words (Kääriäinen 2002: 318) ‘the party leadership took a mega-sized political risk’ in managing the matter. A twin approach was adopted. First, a questionnaire was sent to the party’s local branches. On a 48% response rate, 51.7% were against applying for EU membership (European Economic Area membership was deemed sufficient) and 45.3% in favour, albeit with conditions. Second, the new party leader Esko Aho and former leader Väyrynen undertook a pro-application speaking tour of the party in the country after which they estimated that two thirds were against the EU (the agricultural producers’ organisation MTK was particularly opposed) and one third in favour. At a joint meeting of the party council and parliamentary group in February 1992, the vote was 121–54 in favour of a conditional application. However, matters were severely complicated when in spring 1993 Väyrynen (for unrelated reasons) resigned from the government and began to canvass against Finnish EU membership. With the former leader now in the anti-EU camp, Aho decided before the Centre conference in Jyväskylä in January 1994 to stand down as party leader and prime minister if the vote went against EU membership and he announced his decision beforehand (Aho 1998: 132–134). A conference majority was secured although Eurosceptic elements have persisted in the party, some in 2011 tempted to transfer their support to the anti-bailout True Finns. In multidimensional, multiparty systems locating parties on the basis of their policy positions is no simple task although the three following approaches may go some of the way. The first is to analyse the candidate responses (both those of elected and unelected candidates) to the battery of questions put to them in the Voting Advice Applications (VAAs) during an election campaign. In the Finnish open-list PR electoral system, intra-party competition is fierce and there is an obvious incentive for candidates to promote a clear and distinctive issue-profile. VAAs are legion during Finnish election campaigns, operated by the broadcast media (state and commercial channels), the print media (national and regional) and, increasingly, by various pressure groups. For example, on the basis of a factor analysis of responses to 26 questions in the state broadcasting YLE’s VAA, it was evident that a majority of the 49-strong Centre Party parliamentary group (PPG) elected in April 2015 was to the right of centre on an economic left-right scale and on the right side too on a TAN scale. Equally, on the sensitive question posed in the Helsingin Sanomat VAA of whether to make it easier for persons from outside the EU and European Economic Area to gain work permits in Finland, the Centre Party’s candidates did appear centrist or at least evenly divided: 47% favoured making it easier to obtain work permits whereas 51% were content with the existing ‘conditions attached’ regime. In contrast, 94% of Green candidates supported making the process more conducive to work-based immigration, whereas 89% of True Finn candidates opposed any relaxation of the existing discretionary system.

The Norwegian Left and the Finnish Centre 37 A second approach might be to analyse the roll-call behaviour of party MPs on (unwhipped) ‘conscience questions’ because these should afford insight into the basic attitudinal universe of legislators. On the question of same-sex marriage for instance, the vast majority of Centre parliamentarians have been located on the traditionalist ‘moral right’. When on 29 March 2014 the parliament debated an amendment to the law permitting gender-neutral marriage, opinion was highly polarised. Between 90 and 100% of parliamentarians in the (post-Communist) Left Alliance, Greens, Social Democrats, and Swedish People’s Party were on the libertarian left favouring the legislative change whereas 90–100% of Christian Democrat and True Finn members opposed legalising gay marriage. More than four fifths of the Centre’s PPG were also opposed to the liberalising amendment, confirming the finding from the VAA responses that on ‘value issues’ the party is clearly to the right of centre. This was confirmed in a third approach – using the 2010 Chapel Hill Expert Survey – in which ten Finnish party experts ranked the eight legislative parties. The Centre was close to the centre (5.3) on an economic leftright scale, well to the right on the GAL-TAN scale (6.4) while its ‘general left-right position’ was, with the exception of the True Finns (still then a small party), closest to the centre (5.7).

Structure and organisation Venstre is structured along the lines of a conventional mass party. Members are organised into local branches (lokallag) and county branches (fylkeslag) and convene biennially as a party conference. Conference delegates decision-making authority upwards to the highest body, the central executive committee (Venstres centralstyre), which comprises the party chair, two deputy chairs, and four members (and their deputies) elected by party conference. The executive committee also includes a representative of Young Venstre (Unge Venstre) – founded in 1909 and formally independent of the party – and Venstre’s Women’s Organisation (Norges Venstrekvinnelag). The Liberal Students Organisation is not represented on the central executive committee. Venstre’s membership has not exceeded 10,000 since 1981 and has fallen by virtually two thirds since the late 1950s (Table 1.4a). Although in November 2013 the party ‘celebrated’ ten years of membership growth, this was in practice part of a recruitment drive in the run-up to the 2015 local government election. Candidate recruitment is decentralised and local branches identify potential candidates although in the Norwegian closed-list PR system the party concentrates its efforts on the so-called ‘top candidates’. They are amply briefed on Venstre’s main election themes – climate change, green taxes, child poverty, schools and education, as well as the party’s ‘alternative budget’. See Table 1.4a at http://partyfamilies.eu/index.php/category/chapters/ chapter-1/

38  David Arter For a small party, the leader (chair) can make all the difference although leading a small party inevitably brings with it frustrations. Venstre has elected 13 postwar chairs – compared with eight for the Finnish AgrarianCentre Party – and not all have been members of parliament (Table 1.4b). Håvard Alstadsheim (1990–1992), for example, tried three times without success to be returned to the Storting. Several Venstre chairs brought a record of war-time experience to the post. After the German invasion in April 1940 the Nazis sought to ban the lectures of the first postwar Venstre chair Jacob S. Worm-Müller (1945–1952) – particularly those dealing with nationalism and the separation from Sweden in 1905 – and, as a consequence, he left the country, joined the Norwegian government-in-exile in London and between 1942 and 1945 edited the magazine The Norseman. Helge Rognlien (1972–1974) had been active in the Young Venstre organisation, fell foul of the Nazis and spent years in captivity in the Buchenwald concentration camp. See Table 1.4b at http://partyfamilies.eu/index.php/category/chapters/ chapter-1/ The succession of postwar Venstre chairs is testimony to the fact that small parties are not necessarily more cohesive than larger parties. Gunnar Garbo (1964–1970) belonged to the radical urban wing of the party and for many years met with considerable suspicion from the more conservative, rural elements, especially in the south-west. Garbo took part in the Borten non-Socialist coalition in 1965 and spoke out strongly against the siting of North Atlantic Treaty Organization (NATO) nuclear weapons on Norwegian soil, the war in Vietnam, and Norwegian membership of the EEC. Equally, two Venstre chairs, Bent Røiseland (1952–1964) and Helge Seip (1970–1972), were vocal supporters of the EEC and defected to the Liberal People’s Party over the matter. Whether at times of high or low cohesion, small parties are heavily dependent on the profiling activity of their leader. Proclaiming the cause of gender equality, Venstre’s Eva Kolstad in 1974 became the first female leader of a Norwegian political party. Lars Sponheim (1996–2010) led Venstre into two minority centre-based governments and above all gave the party a ‘media face’. After Venstre’s poor electoral showing at the 2009 general election, he was replaced by the present party chair Trine Skei Grande, who represents the Oslo constituency and has promoted three main themes – the environment, education, and social responsibility. She is presently minister of culture. Like Venstre, which in Duverger’s terms was ‘externally created’, the Agrarian-Centre emerged as a classic class-mass party, boasting consistently the highest membership of any of the Finnish parties and an impressive member-voter ratio – five times higher than the Social Democrats in 1960, for example, (Hokkanen 2002: 157). Indeed, by the late 1950s half the Agrarian voters were party members and this figure peaked at two thirds in 1969, although in 2016 it had fallen to just over one fifth (see Table 1.5a). The member-electorate ratio has ranged from 10% in 1995 to 3.1% in 2016. In its

The Norwegian Left and the Finnish Centre 39 heyday the Agrarian Party was the centre of the local community; meetings were held in a member’s living room (tupailta) and would often be attended by one of the party’s MPs who would undertake the equivalent of Britishstyle constituency service by assisting members with individual problems. This was before the surge of welfare-state related institutions (pensions boards, health boards, etc.) in the 1970s and the later email connections to them. It was also before the negative impact of television on associational life in general and party membership in particular. Members today belong to local branch organisations (paikallisyhdistykset – 2,783 in 2016), which in turn operate under the umbrella of metropolitan branches (kunnallisjärjestöt – 302 in 2016) and district organisation (piirijärjestöt – 21 in 2016). The local branch organisations rather than individual members (as in the True Finns) are represented at the biennial party conference. See Table 1.5a at http://partyfamilies.eu/index.php/category/chapters/ chapter-1/ Candidate recruitment is decentralised. The local branch organisations propose candidates – who may be selected following a postal ballot of members – and a slate of candidates is finalised at a selection meeting of the district party. Nominated candidates will ultimately be issued with an official candidate number and appear alphabetically on the party list. Unlike Norway, the Finnish open-list voting system creates incentives for candidates on the same list to compete with one another and this co-partisan rivalry may be intense, particularly in the party’s stronghold districts. Intra-partisan defeats are relatively commonplace as strong challengers successfully target one of the more marginal incumbents. Although candidate recruitment is managed at the party district level, the central party has on occasions intervened and sought strategically to maximise intra-party candidate competition (and by extension the aggregate list vote) by inserting a celebrity name from outside the world of politics on to a district list. Former skiers, curlers, racing drivers, soap ‘stars’, singers, and ‘Miss Finland’ winners and finalists have also been deployed to ‘paper over’ the electoral cracks – not least in the Centre’s weaker districts in the ‘deep south’ (Arter 2014). Although the party conference is formally the sovereign decisionmaking body in the party, authority is delegated to a 135-member party council (puoluevaltuusto) and a 31-strong party executive (puoluehallitus) and onwards to a small ‘inner cabinet’ of the executive, the so-called ‘work committee’ (työvaliokunta) led by the party chair. The party chair is elected by party conference, and both the workload and the responsibility bound up with the post have been increased by the fact that seven of the eight postwar Agrarian-Centre chairs have also served as prime minister during their period in office. They have led the country and the party. Three of the eight (Virolainen, Väyrynen, and Vanhanen) have also run (unsuccessfully) for the presidency (Table 1.5b).

40  David Arter See Table 1.5b at http://partyfamilies.eu/index.php/category/chapters/ chapter-1/ The first Agrarian Party chair after World War II, V.J. Sukselainen (1945–1964), led the party at the time of the massive Karelian resettlement programme and when, in its strongholds in the north and east of the country, it faced a stiff electoral challenge from communism. When Sukselainen defeated Veikko Vennamo (who headed the Karelian resettlement programme) for the leadership at the 1956 party conference, Vennamo left to found a Smallholders’ Party, later renamed the Finnish Rural Party, which also took votes from the Agrarian-Centre. Sukselainen was narrowly defeated by Johannes Virolainen at the 1964 party conference with more or less discreet support from the (former Agrarian) president Urho Kekkonen. Virolainen’s goal in promoting a change of party name was to extend its support into the towns but in this he also had to face competition from Vennamo-ism (Vennamolaisuus). Virolainen lost out in 1980 to a challenge from Paavo Väyrynen (again with presidential involvement) who set about centrist bloc-building as a means of challenging the Social Democrats’ power hegemony. As part of a further party modernisation package, the name was amended in 1988 and an application lodged for full membership of the Liberal International. When Väyrynen, with his eyes firmly set on higher national office, decided to stand down in 1990, he was succeeded as Centre chair by Esko Aho who, the following year, became the youngest prime minister in Finnish history. Aho contrived to gain majority support in the Centre Party for Finland’s EU application in 1992 but his Centre-Conservative-led government was obliged to manage Finland through the worst recession in its history and he failed to achieve his desired social contract. Indeed, his coalition was faced by a series of general strike threats. Having lost two general elections – although in 1999 the Centre came very close to displacing the Social Democrats as the largest party – and the 2000 presidential election, Aho retired from politics and was succeeded by the first female chair and subsequently Prime Minister Anneli Jäätteenmäki. Under Jäätteenmäki the Centre became the largest party at the 2003 general election. However, her premiership proved short-lived when in a dramatic sequence of events it transpired that she had been frugal with the truth regarding the source of a document she used at the end of the campaign, on the basis of which Jäätteenmäki claimed the (then) Social Democratic prime minister Paavo Lipponen, in a conversation with the U.S. president George W. Bush, had appeared to tie Finland to the Allied coalition favouring an Iraq invasion. Jäätteenmäki was replaced as prime minister and Centre Party chair by her defence minister Matti Vanhanen. In 2007, the Centre Party became the largest party in consecutive elections for the first time in its history but Vanhanen stood down as prime minister and party chair three years later, following revelations about the

The Norwegian Left and the Finnish Centre 41 source of funding for party campaigns – including his own – which had cast an aura of sleaze over the party from which the True Finns were to exploit. Mari Kiviniemi became the second female Centre chair and prime minister after Vanhanen but resigned after the Centre’s disastrous showing at the 2011 general election. She was replaced by the engineer, businessperson, and first-time MP Juha Sipilä under whom in 2015 the Centre became the largest party for the third time in the four general elections since the turn of the new millennium. Against the backdrop of a struggling economy, with exports sputtering, and labour costs too high, Sipilä sought the same type of social contract that Aho had failed to achieve more than two decades earlier. Although space precludes a detailed discussion, it is pertinent to note the way an historic mass party has sought to adapt to the digital era. The Centre has its own television channel where, among other things, recorded showings of the virtual tupailta may be viewed. No longer confined to a member’s living room, the virtual tupailta involves discussion on principal themes led by senior party figures and with social media feedback. For example, a new programme of basic principles (periaateohjelma) was prepared for approval at the 2018 party conference built around the notion of ‘responsible freedom’ (vastuullinen vapaus) and in this connection one virtual tupailta was devoted to discussion and feedback of the core values of the party.

Future challenges This chapter has focussed on Venstre in Norway and the Finnish Centre, both members of ALDE and the Liberal International although neither has been a Liberal Party with a capital ‘L’. Venstre originated as a diffuse anti-establishment opposition movement; the Centre as a class-based peasant and subsequently agrarian producers’ party. Venstre has retained its 19th-century name whereas the Finnish Centre has evolved through two changes of party name. Venstre’s history has been one of factionalism and fragmentation whereas the Finnish Centre has largely withstood challenges from communism and more recently right-wing populism. Venstre is a social liberal party well to the left on the GAL-TAN scale; the Centre hijacked liberalism and the Liberal People’s Party – as an element in a party change strategy – and has been well to the right on the GAL-TAN scale. For the two parties, the way ahead looks very different. Relatively few parties recover their parliamentary status – as Venstre did in 1993 – having fallen below the threshold of representation. The premium will none the less continue to be on survival as a legislative actor. Already dependent on tactical voting, Venstre must remain a credible option for an urban salariat tempted to cast a ‘principled’ liberal vote and, although in 2013 it polled its best result of any district in Oslo, its environmental profile did not prevent the election of a first Green MP for the capital city. The evidence from Sweden in 1885 and 2002 indicates that in a volatile climate

42  David Arter Liberal parties able to ‘capture the mood’ can perform strongly. Obviously too, effective leadership can be an important resource for a minor party. However, a best-case scenario would appear to be ephemeral surges in Venstre support, albeit within a likely range of 5–10% of the electorate. A worst-case scenario would see the oldest Norwegian party condemned to the extra-parliamentary wilderness. For the Finnish Centre, a party that has emerged as the largest party in three of the last four general elections and presently holds the post of prime minister, the challenge is to remain the largest party while recognising that its best results (1991, 2003, and 2015) have come following spells in opposition. The Finnish Centre’s centrism has involved the ability to combine in government with either the Social Democrats or Conservatives, the party’s rank-and-file split in its preferences between an urban liberal wing favouring right-wing cooperation and others inclining towards traditional ‘red mud’ cooperation with the Social Democrats. Either way the challenge is to avoid a resumption of the overarching left-right Social DemocratConservative collaboration that has marginalised the Centre in the past. As a former farmers’ party the Centre’s resilience has been remarkable, if not sui generis. It has withstood the challenge from communism and populism and negotiated the transition from an agrarian to a postindustrial economy while maintaining a mass membership and consolidating its support base. Unlike Venstre and the threat of de-institutionalisation, the Centre occupies highly institutionalised position within the Finnish party system.

References Aardal, B., Bergh, J., Haugsgjerd, A.T. (2014), Velgerandringer og valgdeltakelse ved Stortingsvalget 2013. Oslo: Institutt for samfunnsforskning. Aho, E. (1998), Pääministeri. Keuruu, Finland: Otava. Allern, E.H. (2010), Political parties and interest groups in Norway. Colchester: ECPR Press. Allern, E.H., Karlsen, R. (2014), ‘A Turn to the right: the Norwegian parliamentary election of September 2013’, West European Politics 37(3), pp. 653–663. Almeira, D. (2012), The impact of European integration on political parties. London: Routledge. Arter, D. (1988), ‘Liberal parties in Finland: From perennial coalition actors to an extra-parliamentary role’. In: Kirchner, E.J., Liberal Parties in Western Europe. Cambridge: Cambridge University Press, pp. 326–356. Arter, D. (2001), ‘The Finnish Centre Party: A case of successful transformation?’ In: Arter, D. (ed.), From farmyard to city square?: The electoral adaptation of the Nordic Agrarian Parties. Aldershot, UK: Ashgate. Arter, D. (2013), ‘The ‘hows’, not the ‘whys’ or the ‘wherefores’: The role of intra-party competition in the 2011 breakthrough of the True Finns’, Scandinavian Political Studies 36(2), 99–120. Arter, D. (2014), ‘Clowns, ‘alluring ducks’ and ‘Miss Finland 2009’: The value of ‘celebrity candidates’ in an open-list PR voting system’, Representation 50(4), 453–470.

The Norwegian Left and the Finnish Centre 43 Arter, D. (2016), ‘Neglected and unloved: Does the hinge party deserve that?’, Scandinavian Political Studies Borg, O. (2015), Valtio-oppinut vallan kamareissa. Tallinna, Finland: TallinnaKustanus OY. Hokkanen, K. (2002), Kekkosen maalaisliitto 1950–1962. Helsinki: Otava. Isohookana-Asunmaa, T. (2006), Virolaisen aika – maalaisliitosta keskustapuolue 1963–1981. Porvoo, Finland: WSOY. Jenssen, A.T. (2016), ‘Tactical voting in Norway: Context, motives and occurrence’, Scandinavian Political Studies 39(1), pp. 22–51. Johansen, P.O. (2013), ‘The Norwegian alcohol problem: A failure’, Scandinavian Studies in Criminology and Crime Prevention 14(1), pp. 46–63. Kinsey, B.S. (2006), ‘Cleavage formation in Norway: The contextual dimension’, Scandinavian Political Studies 29(3), pp. 261–283. Kleven, Øyvin, Aardal B., Bergh J., Hesstvedt S., Hindenes Å. (2015), Valundersøkelsen 2013. Oslo: Statitisk sentralbyrå. Knutsen, O. (2003), Generations, age groups and voting behaviour in the Scandinavian countries – A comparative study. Oslo: University of Oslo. Kääriäinen, S. (2002), Sitä niittää, mitä kylvää. Jyväskylä, Finland: Gummerus. Leiphart, J.Y., Svåsand, L. (1988), ‘The Norwegian Liberal Party: From political pioneer to political footnote’. In: Kirchner, E.J. (ed.), Liberal Parties in Western Europe. Cambridge: Cambridge University Press, pp. 304–325. Malmström, C. (2005), ‘Folkpartiet och Venstre – liberale partier i Sverige och Norge’. In: Dempker, M., Svåsand, L. (eds.), Partiernas århundrade. Stockholm: Santérus, pp. 133–158. Mylly, J. (1978), Maalaisliitto ja turvallisuuspolitiikka. Turku, Finland: Tammer. Mylly, J. (1989), Maalaisliitto-Keskustapuolueen historia 2. Maalaisliitto 1918–1939. Helsinki: Kirjayhtymä. Sitter, N. (2006), ‘Norway’s Storting election of September 2005: Back to the left?’ West European Politics 29(3), pp. 573–580. Von Beyme, K. (1985), Political parties in Western democracies. New York: St Martin’s Press. Westinen, J. (2016), ‘Puoluevalinta Suomessa 2000–luvulla’. In: Grönlund, K., Wass, H. (eds.), Poliittisen osallistumisen eriytyminen – Eduskuntavaalitutkimus 2015. Helsinki: Oikeusministeriö, pp. 249–272.

2

The Danish Liberal Parties Karina Kosiara-Pedersen

Introduction The earthquake, or landslide, election of 1973 was devastating to the established parties, reducing their support drastically and doubling the number of parties represented in parliament from five to ten. Several parties have since then gained (and lost) parliamentary representation. Hence, the Danish multiparty system has since 1973 contained between eight and ten parties. Among these parties, three belong to the liberal party family, and are included in this chapter, namely Venstre (The Liberal Party – V), Radikale Venstre (Social Liberals – RV), and Liberal Alliance (Liberal Alliance – LA). These three parties are self-identifying as (social-) liberal parties and they are all either sitting in the liberal group within the European parliament or expected to do so if they get elected. The full name of V includes the appendix ‘Denmark’s liberal party’, the party magazine is called ‘Liberal oversight’, and the party is a member of both the ALDE and Liberal International. The full name of RV includes the appendix ‘Denmark’s Social-Liberal party’, hence the party self-identifies as social-liberal. In addition, RV is a member of the ALDE and of the Liberal International (since 1948). LA self-identifies as a liberal party through its name, however, the party is not represented in the European Parliament, and is not a member neither of the ALDE nor of the Liberal International. Furthermore, as this chapter will show in detail, these three parties are based on the liberal ideology and pursue liberal policies to different extents. Still, they belong to the liberal party family on the relevant dimensions identified in the literature, namely origin, name, transnational affiliation, as well as policies and ideology (Mair and Mudde 1998). This chapter first presents the origins and development of these parties; second, it discusses their ideology and policies; third, it describes their organisation; and, fourth and finally, the chapter exposes the future challenges of all three parties. The chapter highlights differences and similarities across these three liberal parties, though it does not pretend to be thoroughly comparative.

The Danish Liberal Parties 45

Origins and development This section aims to describe the developments of V, RV, and LA since their foundation. This is done, first, through an overall presentation of the parties’ evolution within the party system and of the main intraparty events. Then, the section deals in more detail with the parties’ electoral performances, and with their government participation at national, regional, and local levels of government. Evolution of the liberal parties in the party system and main intra-party events Following Lipset and Rokkan (1967), the two founding cleavages shaping the Danish party system were the rural-urban and employer-employee cleavages (Elklit 1986). Along the rural-urban cleavage first emerged V and the Det Konservative Folkeparti (Conservatives – K, named Højre until 1915), with V representing the rural interests and K the urban businesses. Following the industrial revolution, the employer-employee cleavage emerged together with an urban working class, represented by the Social Democrats on the urban side. When RV split from V, it was basically a split between the employers and the employees in the countryside; the larger farmers represented by V, and the smallholders, or peasants, represented by RV. V is the oldest party in Denmark. Although V’s roots go back further in time, the party was officially constituted in 1870, when three liberal groups in the Danish Parliament united into Det Forenede Venstre (United V). Members of parliament (MPs) from the various groupings composing what would become V tried to mobilise their electorate in 1883 and succeeded in creating local constituency organisations. They even managed to hold what could be termed the first annual meeting of the party in 1888 (Bille 1997: 53). However, disputes within parliament among various fractions halted the organisational development. After the introduction of parliamentary democracy in 1901 and the split of RV (see below), the effort continued and turned out to be successful, as illustrated by the establishment of the party headquarters in 1920 and of a national organisation in 1929 (Bille 1997: 54). Some aspects of the history of V are particularly interesting to point out, and are still relevant today. First of all, V and K have since World War II been electorally complementing each other; if one gained, the other would lose electoral support. Given this trend and the history of governmental cooperation between the two parties, V leader Erik Eriksen attempted to merge the two parties during the 1950s and first half of the 1960s. Due to lack of support and high resistance from within the two parties, Eriksen ended up stepping down as party chair. A second relevant aspect lies in the change in the party’s electoral strategy. Created as an agrarian party aggregating the interests of farmers, the decreasing number of farmers has led V to change its focus and image.

46  Karina Kosiara-Pedersen RV is a splinter party from V. It was created in 1905 as a reaction to the policies of the then-current V government, in regard particularly to the foreign/military issues, but also as a reaction to the divisions between farmers and smallholders and to V’s perceived rightwards turn. The split within the parliamentary group created an incentive to organise locally for those belonging to what at that time was called ‘the Radicals’. The party organisation was created first, and then MPs joined it. Formally, the party was created as an extra-parliamentary organisation, but with strong motivations from current MPs. Hence, the party emerged from a mixed process. Ideologically, RV was built specifically on an antimilitary stance, which was reflected in the first party programme (see below). The creation of New Alliance, the predecessor of LA, was announced at a press conference on 7 May 2007 (threshold of declaration). It was created by MP Naser Khader and Member of European Parliament (MEP) Anders Samuelsen from RV, and by MP Gitte Seeberg from K, though it could be argued that New Alliance at least to some extent was a splinter party from RV. New Alliance was created as a reaction to the government’s rightwards turn on the new politics dimension. ‘Enough is enough’ was their motto in reaction to the (too large) influence of Dansk Folkeparti (The Danish People’s Party – DF) on integration and immigration policies under the Liberal-Conservative government (2001–2011). New Alliance at first experienced a drastic increase in its support and reached the threshold of authorisation already in June 2007, when it collected the necessary signatures to be able to stand for election. Although opinion polls indicated an increasing support, the party almost lost it all in the election campaign, mainly (but not only) because of the poor performance of the party chair Naser Khader in the campaign. However, the party did manage to surpass the 2% threshold of representation with 2.9%. In January 2009, both Naser Khader and Gitte Seeberg left the party, and Anders Samuelsen, who was joined by another MP from RV, took over the leadership of the party. The party was renamed Liberal Alliance, and its political programme took a liberal economic rightwards turn placing the party on the right side of K and V on the redistributive left-right dimension. In 2009, LA gained representation at the local level for the first time, in the municipal council of Horsens. In 2013, the party entered 30 out of 98 municipal councils. In sum, all three liberal parties are internally created parties (Duverger 1964), that is, created by MPs. RV was created as a splinter from V, and LA was created as a splinter mainly from RV. They all share the same origin when looking at the big picture. Table 2.1 sums up the three parties’ successive steps of development, namely declaration, authorisation, representation, and relevance (cf. Pedersen 1982). See Table 2.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-2/

The Danish Liberal Parties 47 Electoral performances Table 2.2 shows the electoral performances of the three liberal parties in the general legislative elections (for Folketinget, which until 1953 was the Lower Chamber, and after 1953 the only chamber in the Danish parliament.) Following World War II, V had the support of almost a quarter of the voters. After decreasing down to 12–15% at the beginning of the 1970s, the party was rewarded in 1975 for its government responsibility after the earthquake election. But after that and until the 1990s, V’s electoral support remained below 15%. After low performances in the 1980s, the ‘heyday’ of K, V quickly peaked up, attracting around 23–24% of the votes in 1994 and 1998. The central explanation for this success takes its roots in Tamilsagen (Tamil Gate), when the K Minister of Justice, Ninn-Hansen, illegally halted the family reunification of Sri Lankan Tamils, causing mistrust of K. K Prime Minister, Poul Schlüter, stepped down in 1993 without an election, which hindered V from capitalising on the case immediately. However, at the following elections, V increased its support, reaching 31.2% of the vote in the 2001 election. Recently (2015), V has gained voters from K but has lost some to both LA and DF. RV was represented in parliament upon their creation in 1905. RV experienced several electoral defeats through the 1970s. After regaining some strength in 1979, the party experienced smaller gains or decreases for a long period. Recently, the party has been on more of a rollercoaster ride with a gain of eight MPs in 2005, loss of eight MPs in 2007, gain of eight MPs in 2011, and loss of nine MPs in 2015. In 2011, RV gained mainly from its upcoming governing partners, the Socialistisk Folkeparti (Socialist People’s Party – SF) and the Socialdemokratiet (Social Democrats – S); but RV also attracted votes from across ‘the aisle’, from V. In 2015, RV delivered votes back to S, but also lost some support to the new party Alternativet (the Alternative – Alt). In sum, RV has now been reduced to a third of its prior1973 strength. The predecessor of LA, New Alliance, passed the electoral threshold in the first election for which it stood. Many changes, in party leadership, ideology, and political programme, took place during this first election period. Even though opinion polls were quite pessimistic, and commentators ruled them out, LA increased its support at the 2011 election (5%) and again in 2015 (7.5%). See Table 2.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-2/ At the local level, V is represented across all 98 municipalities, RV in most, and LA was represented in one of them in 2009, and in 30 in 2013–2017.

48  Karina Kosiara-Pedersen Government participation This section examines the extent to which V, RV, and LA have taken part in governments at the national, subnational, and local levels, and in that case, in which type of government, as summed up in Table 2.3. After the introduction of parliamentary democracy in 1901, V held the keys to the Prime Minister’s office for several years. Until 1908, the party had a majority in parliament. After that period, they started alternating with RV and S. Although most of V’s governments were single-party minority governments, several governments can be especially mentioned. First, V entered into a RV led government with the K in 1968–1971; the first of its kind. Second, V was ready to take on governmental responsibility and formed a one-party minority government after the earthquake election of 1973. S went back into office in 1975; however, S had difficulties governing during these economic hard times. Therefore, frequent elections were held. This explains the third governmental coalition worth noting: the untraditional S-V coalition in 1978–1979. It was an interesting experiment but failed to work because of a lack of trust among the (shadow) ministers of the two parties. The fourth (group of) governments that are of special interest are the K-led governments in 1982–1993, in which V took part without holding the prime-ministership – an unusual feature for V. The 1980s saw an upsurge in the support for K, while V was struggling to transform itself from an agrarian party to a business, economically liberal party. Fifth, in 2001, after more than eight years in opposition, V went back not only in office but also in the Prime Minister’s office. Thanks to the parliamentarian support of DF, V and K could pursue right-of-centre policies during 2001–2011. For the first time since World War II, the government did not include parties placed towards the centre on the traditional, redistributive left-right political dimension. Generally, V can be considered, together with S, as the main party of government in Denmark. Even though the June 2015 general election provided a majority to the four right-of-centre parties, DF, V, LA, and K, the largest party in terms of vote, DF, was not ready to take on government responsibility. Hence, a V minority government formed, with the parliamentary support of the other three parties. However, by fall 2016, it became clear that the government was struggling because of too much public disagreement between the three parties. Therefore, PM Lars Løkke Rasmussen invited LA and K into government; and the government took office by late November. RV has collaborated with both sides of the political spectrum to form governments. RV first participated in government in 1909, when RV chair Zahle led a government supported by S. Between 1929 and 1940, these two parties ruled together. After World War II, RV did not govern until 1968, when the party formed and led a right-of-centre coalition with V and K. During the 1982–1988 period, RV supported the four-party right-of-centre K-led government but at the same time formed an alternative majority with

The Danish Liberal Parties 49 the opposition on issues such as foreign policy, law and order, and environmental protection. RV’s strategy of maximising its policy impact resulted in the party joining the K-V government in 1988. Due to electoral losses, RV exited the government in 1990 but continued supporting it in parliament. In 1993, the party returned to its former government partner, S, and both remained ally since then. RV’s role as pivotal centre party and decisive party in government formation ended in 2001 with the electoral success of DF. Note that, by contrast to both V and LA, RV has currently the strongest tradition of cooperation with the parties towards the left of the traditional left-right spectrum. See Table 2.3 at http://partyfamilies.eu/index.php/category/chapters/ chapter-2/ Similar governments are not formed at the regional and local elections. At these levels, governments are to a large extent characterised by broad coalitions and by coalitions that are not experienced at the national level. A majority within the regional or municipal assembly is needed to support the election of the chief of the executive. At the regional level, Denmark used to have 14 counties, which were amalgamated into five regions in 2006. During the 2009–2013 period, V held the regional mayor position in one out of these, and since 2013 the party has held this position in two of them. The remaining positions are held by S, which indicates that currently V is only the second party at this level of power. At the municipality level, V, again with S, has also attained a strong position. Since 1970, the share of V mayors has been around 40–50%, except in 2005 and 2009 (37% and 32%, respectively) (Kjær and Opstrup 2017). In the 2013 municipal elections, when the party was in opposition at the national level, V won 48 out of the 98 mayorships. The traditional local strength of V is explained by the party’s origins as an agrarian party, with large support from farmers and the countryside population. This is much different for RV. While the party also had certain local strongholds, these are much fewer and have been drastically reduced. In the 1970s, RV used to win 3–4% of the mayorships, but this share has since then decreased. Since 2001, RV has held between 0 and 2 out of the 98 mayor offices (Kjær and Opstrup 2017). LA has gained representation in 30 municipalities, but like other smaller parties, the party has not won any mayor position yet. In sum, V and RV have been parties of government for more than 100 years, at different levels of power and with different strength, whereas LA only entered national government in November 2016.

Ideology and policy positions The main cleavage in the Danish political system today lies in the employer-employee dimension, underlying the redistributive, socioeconomic left-right scale. However, since the mid-1990s, a second dimension

50  Karina Kosiara-Pedersen has been identified among the electorate, namely the New Politics dimension created by policy areas such as immigration and integration, environment/ green, and law and order (Borre and Goul Andersen 1997). Parties may be placed within this two-dimensional space on the basis of the position of their voters. Data collected during the 2015 general election (Hjorth 2017) reveal how close V and LA’s voters are to each other on both dimensions, together with those of K. These three parties are all located very closely, with DF more to the left on the economic dimension and less libertarian on the New Politics dimension. In the opposite, left-of-centre block, which includes RV, the parties’ voters have more distinct preferences. RV’s voters appear to hold quite leftist positions on the New Politics dimension (based on environmental attitudes, law and order, and stance towards immigration and refugees) but a rather centrist position on the economic, redistributive dimension; actually, RV is very close to DF on the economic dimension. In sum, at the aggregate level, RV’s electorate is very distinct from those of V and LA, which are more alike. The European dimension is cross-cutting these two dimensions, as Eurosceptic parties are found at both ends of the other scales, and as all parties comprise both EU positive and sceptics views. This section further describes the positions of V, RV, and LA on these main issues, on the basis of party programs and ministerial portfolios. Starting with the ideological background of the parties, according to its historical roots, V has been commonly categorised as an agrarian party. As argued above, this relates to the party’s origins. But in ideological terms, V has, in particular since the 1980s, while also fulfilling agrarian interests, increasingly emphasised its concerns with private businesses, hence developing a liberal ideology. This switch can be described as one from a focus on rural side ‘ordinary people’s schools’ (højskole-venstre, or ‘folk high school’) to a focus on business schools (handelshøjskole-venstre). This resulted in the party becoming the main bourgeois party; a position they lost to DF at the 2015 election. V has a long list of party principles dating from 2006 (Venstre 2016). These are gathered in a 46-page document that allows the party to address all ministerial areas. First, the liberal view on human society is presented, namely that ‘human beings thrive in freedom, under responsibility’. ‘Freedom’, ‘community’, and ‘broadmindedness’ are key concepts. The right to live one’s own life is important, as is the right to private property. Second, democracy needs to be alive; interestingly, democracy is still emphasised more than 100 years after V – together with S – fought for parliamentary democracy that was de facto introduced in 1901 and formalised constitutionally in 1953. Third, for V, the Danish community is important, and culture breeds community. V is pro-market and pro-free market. V wants a strong and free business sector, with special emphasis on developing a flourishing milieu for entrepreneurs, open competition, time-efficient transport opportunities,

The Danish Liberal Parties 51 and initiatives on agriculture, fisheries, and food. Party members want the labour market to be characterised by a good working environment, state pensions for all, and the freedom to organise and make individual agreements. This translates into to a strong stance against traditional unions, which are mostly associated (at least informally) with S. V wants a modern public sector, and in particular, a world-class healthcare sector. This implies providing social safety, as well as housing, education, and knowledge for all. In research and education, the opportunities granted by the technological development should be exploited. V is not libertarian when it comes to New Politics issues. The party is rather tough on crime, hence pledging for tougher law and order policies. The party program also includes a section on nature and environmental protection where the emphasis is put firstly on solutions brought by the market, and only secondly on solutions brought by state intervention. Family life should be characterised by the freedom to choose, and the acceptance of other forms of family structure in addition to the traditional type with a mother, father, and children. As for foreign policies, V values a binding international community, and the defence of freedom appears as a duty. V also supports the positions that Europe needs to be outward-oriented, and that global trade is important and a way out of poverty. Hence, V puts a strong emphasis on international cooperation. Another way to assess policy priorities is to look into the ministerial portfolios the party holds in coalition governments. In the minority coalition formed with K between 2001–2011 with the parliamentary support of DF, V was responsible for finance; environmental/green issues; schools; internal matters; employment; tax; defence; health; research and higher education; church and clerical affairs; food, fisheries, and agriculture; and integration. In sum, this indicates V’s emphasis on traditional economic liberalism, not cultural or social liberalism. RV has been, on occasion, referred to as ‘Radical Left’ in English language texts, which is the direct translation of ‘RV.’ But this characterisation is quite misleading for a social-liberal party, which is placed just right-of-centre on the economic left-right scale, towards the left on the New Politics dimension, and which has participated in governments headed by S and K Prime Ministers. Hence, the ‘Social Liberals’ is a more appropriate term, one that better reflects the party’s ideology. The first program of RV in 1905 focussed on neutrality, voting rights for all above 21 years of age, anonymous voting in public elections, the introduction of referendums, progressive individual taxing, unemployment insurance, shorter working hours, as well as better care for the elderly, children, and the most vulnerable people in society (Radikale Venstre 2016a). It is quite clear that the antimilitary stance, the development of the political system and democracy in Denmark and basic social rights were at the core of the party’s ideological starting point.

52  Karina Kosiara-Pedersen In its current party principles, created in 1997, the five major headings of the document indicate the party’s priorities, namely ‘human beings and the community’, ‘sustainability’, ‘democracy’, ‘society, the public sector and the individual’, and ‘international obligations’ (Radikale Venstre 2016b). RV believes in free and responsible human beings and is pro-welfare, but wants the social net of security for those who needs help. The party cultivates an image of being economically responsible. It has a green image, and has, together with SF, been the main reason why no traditional green party ever succeeded in Denmark. RV emphasises civil liberties such as same-sex marriage, same-sex adoption, and gender equality politics. It is the most pro-EU integration party within the Danish system (which, of course, does not mean much), and takes a pro-immigration stance along with the leftwing parties. RV’s current key issues, as presented on their website, are (1) more people employed and less on welfare; (2) the social background of children must not determine their future; (3) a green future is worth the cost; and (4) society must take responsibility for refugees (Radikale Venstre 2016c). Again, a mix of social and liberal stances, with a strong emphasis on the New Politics dimension, including refugees and environmental protection. In sum, on the first dimension, RV is just right of centre, and on the second dimension (law and order, immigration, and environment), RV is slightly on the left of the Social Democrats, and located close to the two left-wing parties. When part of the four-party majority coalition led by S in 1993–1994, RV held the ministerial portfolios of the economy, foreign affairs, and education. In 1994, when the Kristeligt Folkeparti (later renamed Kristendemokraterne, Christian Democrats – KD) did not gain parliamentary representation at the general election and were therefore left out of the government, RV also secured the ministry of Nordic cooperation; and when the Centrumsdemokraterne (Centre Democrats – CD) left the coalition in December 1996, RV further received the keys to the ministries of culture and church. In the S-led minority coalition government in 2011–2015, which included the SF until January 2014, RV had similar ministerial responsibilities: the economy, climate and energy, Nordic cooperation, internal matters, development, culture, research and higher education, church and clerical affairs, and equality; supplemented by foreign affairs, tax, social policies, and integration from 2014, when the party left the areas of Nordic cooperation and development. LA occupies the most right-wing position on the redistributive, economic left-right scale and pursues economically liberal policies such as lower tax and more individual freedom. The main agenda of LA is quite straightforward, and clearly economically liberal: ‘more freedom, larger growth, less bureaucracy and lower tax’ (Liberal Alliance 2016a). These are the four political themes the party has promoted throughout its short lifespan. The party manifesto does not expand much upon that. The five principles that its members believe in are (1) personal freedom, (2) broadmindedness,

The Danish Liberal Parties 53 (3) less state and more private, (4) freedom under responsibility, and (5) freedom to dispose of one’s own resources (Liberal Alliance 2016b). LA is ideologically based on liberalism in a pure form, with an emphasis on the economic dimension, but its members also adopt liberal values in the other dimensions. Hence, the three parties have very different positions within the party system. V has for most of the period been the mainstream right-of-centre party, with an emphasis on liberal economic priorities but not on libertarian values, whereas RV is economically at the centre with an emphasis on (pro-) libertarian issues, and LA is a right-wing party especially on the traditional, economic, redistributive left-right scale.

Structure and organisation The structure and organisation of the three liberal parties are analysed in four sections. The first section describes the main party structures with an emphasis on candidate selection and policy formulation. The second section explores the when, how, and why of party leadership changes. The third section looks into party membership figures and characteristics. The fourth and final section focusses on the three parties’ formal and informal links with social movements. Main party structures The purpose of this section is to describe the main party bodies and their competences with a specific focus on candidate selection and policy formulation. V is formally organised as a classic mass party with a hierarchical structure. Members are represented by delegates at meetings at the various levels of organisation, and also at the annual congress, which is the highest authority of the party. The party organisation is characterised by a high level of local autonomy for historical reasons; and the strong support for local autonomy also partly explains why it took a while to build a national organisation. Candidate selection takes place at the local level, that is, within the nomination districts, which currently number 92. Party policies may be discussed and decided at the annual meeting according to the party statutes, and party programs may be developed with the input from members; however, in practice, the parliamentary group and, in particular, the party leadership decide on what to promote and how to frame it. This is, of course, even stronger when the party is in government. The emergence of RV as a party with a membership organisation created independently from the parliamentary group has implications for how the party organises. Most importantly, the parliamentary group is not expected to be obliged to follow the political decisions of the annual congress, even though the highest authority rests within the party congress. Hence, the elected representatives formally have a higher degree of freedom.

54  Karina Kosiara-Pedersen The candidate nomination process is decided at the level of the ten electoral districts, hence, also here there is a high degree of decentralisation. LA seems to have a high degree of top-steering in practice, which may be expected from a new party. In the first election, the top ten candidates are picked by the party leadership. But formally, candidate selection is carried out at the level of the ten electoral districts, as in the case of RV. There are also tendencies towards a higher degree of intra-party democracy, since all members are able to attend and vote at the annual meeting. This gives them a greater direct say in the decision-making than rank-and-file members in the other parties, who only have an indirect say at parties’ annual meetings through their delegates. Evolution of party leadership Whereas the party leader of V is both chair of the membership organisations and the leading figure in the party, RV and LA have untied these two positions – they have a chair of the membership organisation who is distinct from the party leader. Party chairs in V have a central and powerful position; ‘chief’ has been an appropriate designation. Formally, V party chairs are elected by the delegates at the annual meeting. Hence, party leaders are also chairs of the membership organisations. However, leadership selection seems to be a fait accompli in most cases. V has had only one contested election, namely the 1984 contest, opposing the leader of the parliamentarian party and the foreign minister, Uffe Ellemann-Jensen. The latter won a fairly close election at the annual meeting. Fogh Rasmussen was deputy chair before he took over from Ellemann-Jensen, who resigned after the very close electoral defeat in 1998. Lars Løkke Rasmussen was also the deputy chair before taking over, when Fogh Rasmussen became the General Secretary at the North Atlantic Treaty Organization (NATO). Hence, it appears that the deputy chairmanship is a direct highway to become the party chair. However, it has been questioned whether this is currently the case. In summer 2014, Løkke Rasmussen found himself in a media storm over his personal expenses (clothing and family holiday) being paid for by the party (Kosiara-Pedersen 2015b). He was able to convince the parliamentary group and then the national executive that he should stay in the position. However, he also was forced to defend himself at an executive committee meeting of the party organisation because committee members constituted the highest authority within the party between the annual party congresses (according to the party statutes). The deputy chair had a good chance of taking over and sided to some extent with Løkke Rasmussen’s critics. However, the potential coup was deterred when Løkke Rasmussen threatened to field another candidate for the party chair, a candidate who was more popular than the deputy chair, and this seems to have been the major reason why Løkke Rasmussen and the deputy chair reached an agreement that allowed him to stay in office. Hence, the

The Danish Liberal Parties 55 natural succession seems to have been questioned, which may lead to more contested elections in the future between the deputy leader and a challenger. This case also shows that no matter how the formal rules are adhered to in peace time, when there is substantial disagreement, formal rules matter: since the executive committee formally is the highest authority between party congresses, Løkke Rasmussen needed to convince the members of this committee, not just the leadership, the parliamentary party, and the smaller national executive on the good reasons for his retaining his position. RV has separated the positions of party leader and chair of the membership organisation. Party leaders are selected by the parliamentary group, whereas delegate members at the annual meeting elect the chair of the membership organisation. The latter is not an elected politician and is, in general, not well known by the public. Thus the focus in this discussion is placed on the party leader. RV has had only ten party leaders in 110 years (see Table 2.4). The first three leaders of RV led for particularly long periods. The extent to which party leader changes are contested are hard to assess when decided within a parliamentary group. For instance, Marianne Jelved left after the electoral loss in 2007 and was replaced by Margrethe Vestager from the ‘next generation’. On the one hand, this was an obvious choice, which was electorally rewarded in 2011. However, at the time of her selection, in 2007, a minority group of MPs, who were more inclined towards cooperation towards the right of the political spectrum, expressed their preferences for another party leader and backed a candidate, also from the next generation who afterwards chose to leave politics. Hence, in this case the contest was not publicly expressed, but there was some disagreement within the group. Such disagreement did not seem to occur when Vestager left to become EU Commissioner, and Morten Østergaard was entrusted with filling her ‘big shoes’. LA is organised similar to RV with a party leader and a chair of the party membership organisation. As for the formalities of election, all dues-paying rank-and-file LA members elect the chair of the membership organisation at the annual meeting. As in RV, it is the parliamentary group, which decides on the party leader. LA has had only one leadership change, when internal disputes over what policies and strategy to pursue led both party leader Naser Khader and co-founder Gitte Seeberg to leave the party in 2009. Without any contest, the party elite, which had been supplemented with MP Simon Emil Ammitzbøll (formerly RV), decided that Anders Samuelsen, the only remaining party founder, should take over the party leadership. See Table 2.4 at http://partyfamilies.eu/index.php/category/chapters/ chapter-2/ Party membership This section aims to depict the evolution of party membership figures and to examine the profile of members in V, RV, and LA.

56  Karina Kosiara-Pedersen First, Table 2.5 shows the development in the parties’ membership figures. As in other Western democracies, parties with traditional party membership organisations experienced a clear decline. However, it is not all about decline. V, one of the two major membership parties in Denmark, peaked in 1950 with a little more than 200,000 members, representing 46% of the party’s electorate. Since then, figures have progressively and constantly declined. Currently the party enrolls 37,000 members (4% of the party voters). RV has never enrolled as many members as the three other old Danish parties. Membership peaked around 1960, when the party registered about 35,000 members, representing a quarter of its electorate. Its membership has since then also declined steadily. However, at the turn of the century, the V-K government, which was governing with the support of DF, led to an influx of members into RV, raising the party’s membership to a higher level: 9,000 members, a number that they managed to maintain well into the 2000s. However, currently RV seems to be facing declining figures again, reaching 7,000 members in 2015. At first, New Alliance mobilised huge support and enrolled a massive number of members. However, the figures for LA are more modest at around 6,000, which equals 3.4% of their electorate; just below the average of all Danish parties (4%). See Table 2.5 at http://partyfamilies.eu/index.php/category/chapters/ chapter-2/ A party member survey conducted in 2012 allows examining the profile of the members of the three parties (Kosiara-Pedersen 2015a). V is, as are all parties right of S on the traditional left-right scale, characterised by a larger share of men than women among its members; possibly up to two thirds being men. The average age (58 years) is among the oldest of all parties. V’s members are mainly employed in the private sector, or self-employed. RV’s members are somewhat more male than female, but to a lesser extent than V and LA, and are rather young (51 years on average). They are significantly more likely to hold an academic degree, to work in the public sector, and to have an above-average income. Men outnumber women with a ratio of 4:1 in LA. LA’s members are also younger (46 years on average), with a larger tendency to have accomplished higher education, to be employed within the private sector or to have their own company, and to have a higher income. Relation to social movements The establishment of V is linked to the agrarian organisations from late 19th century (Bille and Christiansen 2001: 34). However, the organisational ties have not been formalised in the party statutes; in other words, the agrarian organisations are not granted representation within the party organs (Bille and Christiansen 2001: 49). Since the 1960s, there has been limited

The Danish Liberal Parties 57 personal overlap between personnel in the party and agrarian organisations, and agrarian organisations have lost most of their role as recruitment channels since the 1980s (Bille and Christiansen 2001: 57). In its transformation from a liberal agrarian to a liberal business party, V has developed linkages with employers’ organisations and business organisations in its ‘Liberal business club’, mainly through fundraising (Kosiara-Pedersen 2014). The countryside population also mattered for RV, however, while V was linked to the larger farmers, RV had close ties with the smallholders and their organisations. Currently the party has no formal ties to social movements. However, RV has (had) overlaps with like-minded social movements such as the peace movement and environmental organisations among their supporters. LA does not seem to deliberately relate to social movements. In sum, neither V, RV, nor LA have formal ties with social movements. But the former two have developed some links through personal overlap and funding.

Main challenges for the future V, RV, and LA are all liberal parties according to the four dimensions posited by Mair and Mudde (1998). However, the three parties vary along origin, policies, and ideology, hence illustrating the possible variations within the liberal party family when defined on these criteria. Due to their variation in, among other things, age, party leadership, electoral size, and ideological position within the party system, the main challenges in the future for these three liberal parties also differ substantially both in the electoral and intra-party arenas. V needs to regain its position as the major party in the right-of-centre bloc of parties to retain its grip on the Prime Ministership or its role as leader of the opposition. Electorally, V is challenged by DF, who has attracted voters from both S and V. Part of its challenge is not only the success of DF but also the poor ratings of its current leader, Løkke Rasmussen. His trustworthiness is low after several episodes of misuse of public funds publicly exposed in the media (the more recent was in summer 2014) (KosiaraPedersen 2015b). He does not seem to be able to survive another one of these, given high intra-party dissatisfaction and the readiness of formal and potential lieutenants to take over the party leader position, which have been aired publicly. Currently, it seems likely that there will be a contested party leader election the next time around, an intra-party event that V has experienced only once in recent times (Bille 1997: 102). However, a long time may pass between now and then whereby the succession may be settled within the party elite. In the electoral arena, RV is severely challenged by Alt, the new environmentally friendly party that gained representation at the 2015 general election, apparently by appealing to the urban, eco-friendly, academic, café latte-drinking segment of the electorate, which has usually constituted the

58  Karina Kosiara-Pedersen RV’s bastion. At first, it seems as if Alt is more left wing than RV, whereby the latter may be able to sustain its electoral support by emphasising its combination of ‘social’ and ‘liberal’ ideology. LA seems to be doing well and has not been seriously challenged. Electorally, the party has gained ground. It has changed from being the rebellious party on the side to a government-participating party with six ministerial offices. The main challenge of LA is to strike a balance between their ideological goals and the day-to-day politics not only of ‘their’ ministries but also of government policies on all other fields. The party elite argues that they have given up on certain pure ideological based goals to gain influence on a broader set of policies; it remains to be seen whether their voters appreciate this trade-off. In sum, the Danish party system contains three liberal parties, which vary along origin, electoral success, governmental role, organisation, and challenges, but which all belong to the liberal party family. Hence, the Danish case(s) reflect(s) the diversity of the liberal party family.

References Bille L. (1997), Partier i forandring. En analyse af otte danske partiorganisationers udvikling 1960-1995. Odense, Denmark: Odense Universitetsforlag. Bille L., Christiansen F.J. (2001), ‘Partier og interesseorganisationer i Danmark’. In: Sundberg J. (ed.), Partier och intresseorganisationer i Norden. Copenhagen: Nordisk Ministerråd, pp. 29–77. Borre O., Goul Andersen J. (1997), Voting and political attitudes in Denmark. Aarhus, Denmark: Aarhus University Press. Danmarks Statistik (2016a), Election statistics. http://dst.dk/da Danmarks Statistik (2016b), FVKOM: Folketingsvalg efter valgresultat og kommuner. www.statistikbanken.dk/FVKOM Danmarks Statistik (2016c), VALGK3: Valg til kommunalbestyrelser efter kommune, parti og stemmer/kandidater/køn. www.statistikbanken.dk/VALGK3 Duverger, M. (1964), Political Parties. London: Methuen & Co. Ltd. Elklit J. (1986), Det klassiske danske partisystem bliver til. In: Elklit J., Tonsgaard O. (eds.), Valg og vælgeradfærd. Aarhus, Denmark: Politica. Folketingets Oplysning (2015), Partier opstillet til folketingsvalg siden 1915. http:// www.ft.dk/Folketinget/Oplysningen/Partier/Opstillede_partier.aspx Hjorth F. (2017), ‘Issue voting ved Folketingsvalget 2015’. In Hansen K. H, Stubager R. (eds.), Oprør fra udkanten – Folketingsvalget 2015. Copenhagen: DJØF-Forlag. Kjær U., Opstrup N. (2017), ‘Borgmesterpartierne og borgmestrene”. In: Elklit J., Elmelund-Præstekær C., Kjær U. (eds.), KV13 – Analyser af kommunalvalget 2013. Odense, Denmark: Syddansk Universitetsforlag, pp. 323–341. Kosiara-Pedersen K. (2014), Partiernes kampagneudgifter. In: Hansen K.M., KosiaraPedersen K. (eds.), Folketingsvalgenkampen 2011 i perspektiv. Copenhagen: Djøf Publishing, pp. 57–66. Kosiara-Pedersen K. (2015a), ‘Party membership in Denmark: fluctuating membership figures and organizational stability’. In: van Haute E., Gauja A. (eds.), Party members and activist. London: Routledge, pp. 66–83.

The Danish Liberal Parties 59 Kosiara-Pedersen K. (2015b) ‘Denmark’, European Journal of Political Research. Special Issue: Political Data Yearbook 54, pp. 86–93. Kosiara-Pedersen K. (2017), Demokratiets ildsjæle – partimedlemmer i Danmark. Copenhagen: DJØF Forlag. Liberal Alliance (2016a), Liberal Alliance web site. https://www.liberalalliance.dk/ Liberal Alliance (2016b), Principprogram. https://www.liberalalliance.dk/wp-content/ uploads/2016/01/Liberal-Alliance-Principprogram.pdf Lipset S. M., Rokkan S. (1967), Party systems and voter alignments: Cross-national perspectives. New York: Free Press. Mair P., Mudde C. (1998) ‘The Party Family and its Study’, Annual Review of Political Science 1, pp. 211–229. Pedersen M.N. (1982), ‘Towards a new typology of party lifespans and minor parties’, Scandinavian Political Studies 5(1), pp. 1–16. Radikale Venstre (2016a), Radikale Venstres historie. https://www.radikale.dk/content/ radikale-Vs-historie Radikale Venstre (2016b), Radikale Venstres principprogram. https://www.radikale.dk/ content/principprogram Radikale Venstre (2016c), Det handler om mennesker. https://www.radikale.dk/content/ det-handler-om-mennesker-5 Ritzau (2009), Fakta: LAs op- og nedture. Politiko. http://www.politiko.dk/nyheder/ fakta-liberal-alliances-op-og-nedture Statsministeriet (2015), Regeringer fra 1953 til i dag. http://www.statsministeriet. dk/_p_7812.html Venstre (2016), Venstres principprogram. Fremtid i frihed og fællesskab. Vedtaget på Venstres Landsmøde den 18. november 2006 i Odense. http://v-shop.dk/PDF/ Venstres-Principprogram.pdf

3

The Centre Party and the Liberals The Swedish members of the liberal party family? Niklas Bolin

Introduction ‘The Liberals are the Swedish Liberal party’ is the first line of the party programme of Liberalerna (Liberals – L) (Folkpartiet 2013: 1). Although it comes as no surprise that the self-image of the Liberals is liberal, the statement could also be interpreted as a territorial marker in light of increasing competition over the ownership of partisan liberalism. During most of the 20th century, few would have argued that there were more than one Swedish liberal party. However, during the past decades the originally agrarian Centerpartiet (Centre Party – C) increasingly came to be regarded as a second liberal force in Swedish politics. This perception is also enhanced by the transnational affiliations established since Sweden joined the EU in 1995. Not only did the single Member of European Parliament (MEP) from the Liberals become a member of the European Liberal Democrat and Reform Party (ELDR) but also the two MEPs from the Centre Party were included in the group. The organisational ties with liberalism was soon strengthened after the Centre Party formally joined the ELDR in 2000 and subsequently the Liberal International in 2006.1 While the affiliation of the Centre Party with liberal transnational organisations met opposition internally, as it was claimed that the party in fact was not liberal and risked being perceived as ‘a bad copy of the Liberals’ (Andersson et al. 2000), other voices within the party considered this affiliation to be uncontroversial. Against this background this chapter aims to evaluate to what extent it is reasonable to include both of these parties in the liberal party family in light of three other criteria identified by Mair and Mudde (1998): the parties’ origins, ideology, and organisation. The chapter will proceed as follows. The first section describes the origins and developments of the Centre Party and the Liberals, in terms of party formation, electoral performance, and government participation. The second section examines the ideology and policy positions of the two parties. The third section scrutinises the organisational characteristics of the parties, focussing on the main party bodies, party leadership, and membership

The Centre Party and the Liberals 61 development. The concluding section assesses to what extent the parties actually belong to the liberal party family and discusses the main future challenges for these parties. Note that while the first section takes as its starting point the parties’ formation in the early 20th century, the other sections focus on the postwar period, with a specific emphasis on the past 20 years.

Origins and development While liberal organisations were present in Swedish politics as early as in the middle of the 19th century (Lindström and Wörlund 1988; Särlvik 2002), the origin of the Liberals could be traced back to the formation of the Freeminded National Association (Frisinnade Landsföreningen) in 1902. In 1923, the party split over the issue of alcohol prohibition. However, the two splinters merged a decade later and formed the People’s Party (Folkpartiet) in 1934 (Bäck and Möller 2003: 68). Since then, the party has changed its name twice. In 1990, the suffix ‘the Liberals’ was added to underline the affiliation with the liberal party family. This was further emphasised when the party congress in 2015 decided to delete the first part of the name, which simply became ‘The Liberals’ (Liberalerna). This change occurred in a context where liberal ideas had been adopted by many parties, but where only the Liberals were originally founded upon liberal principles. This change was further motivated by the fact that the reference to the ‘people’ no longer had liberal connotations (Folkpartiet 2015: 62–63), and was often associated with other party families, such as the conservative (e.g., the Austrian People’s Party and the Spanish Popular Party) or the radical right (e.g., the Swiss People’s Party and the Danish People’s Party). The Centre Party is originally a merger of two agrarian parties formed in the beginning of the 20th century. In 1921, the Farmers’ League (Bondeförbundet, formed in 1913) and the National Farmers’ Union (Jordbrukarnas Riksförbund, formed in 1915) merged into one party under the label of the former (see Table 3.1). The party was built upon dissatisfaction with agricultural politics and initially aimed exclusively at defending the interests of farmers. However, structural rationalisation and a diminishing peasantry resulted in a decrease of this specific electorate. As a consequence, the party transformed to broaden its appeal: it broadened its programme to areas such as decentralisation and environmental issues, and changed its name to the Centre Party (Centerpartiet) in 1957 (Christensen 1997).2 See Table 3.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-3/ Electorally, both parties have had their ups and downs (see Table 3.2). In the early 20th century, before the formation of the People’s Party, the Liberals constituted one of three major party groups. In 1911, the party won a record

62  Niklas Bolin high 40.2% of the votes, but its support decreased markedly after universal suffrage was established in 1921. The party was often also represented in the government. Its advantageous position between the two other major parties, the Social Democrats on the left and the Conservatives on the right, offered the Liberals a pivotal role, which enabled them to hold the Prime Ministerial post on six different occasions. During the early postwar years, the party gained support. Under the leadership of the future Nobel Prize laureate in economics, Bertil Ohlin (1944–1967), the Liberals were the main opposition party under Social Democratic governments. At three consecutive elections between 1948–1956, the party polled well above 20% of the votes with a record high of 24.4% in 1952. However, by the end of the 1960s, the party lost its position as the biggest centre-right party in Sweden. Since the abolishment of the upper chamber and the introduction of a unicameral parliament in 1970, the average vote share has been 9.6%. Both forerunners of the Centre Party made small inroads in the parliament during their early elections. In 1921, the first election after the merge, the party won 11.1% and became the fourth biggest party. While slightly losing support over time, the party maintained this level of support during most of the bicameral period. In 1936, the party formed a single-party minority government after the Social Democrats stepped down for not being able to get a parliamentary majority on their defence policy. The government led by Axel PehrssonBramstorp held power only for the remaining three months before the next scheduled election. After the election, the party joined the Social Democrats as a junior partner to form a majority coalition. As part of a war-time national unity government, the Centre Party remained in government until 1945. The Centre Party again coalesced with the Social Democrats in 1951–1957. The cooperation between these two parties then came to an end. As part of the reformation of the party, which led to changing the name and broadening the programme, the party also officially declared to be part of the centre-right bloc. During this ‘phase of successful adaptation’ (Widfeldt 2001) the party increased its electoral support. During the late 1960s and especially during the leadership of Torbjörn Fälldin (1971–1985), the Centre Party became the leading centre-right party, reaching a historical score of 25.1% of the votes in 1973. During this period, discussions took place about a potential merging between the Centre Party and the Liberals. Though the fusion did not materialise, it undoubtedly confirmed the Centre Party’s right-wing turn. During the past four decades, electoral support has decreased constantly, except in 2002 and 2006 where the party managed to slightly gain votes. See Table 3.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-3/ With the exception of the Centre Party minority government in 1936, the Social Democrats held office continuously from 1932 to 1976 (Aylott 2016).

The Centre Party and the Liberals 63 This dominance partly resulted from the continuous dominance of the traditional left-right cleavage (Bergman and Bolin 2011: 254). Given the unidimensional structure of the Swedish party system, the Social Democrats could count on the tacit support from the smaller Left Party (up to 1990 with the suffix ‘the Communists’). While the Social Democrats managed to win a majority of the seats only twice (1940 and 1968), support from the Left Party has kept the centre-right parties out of government. The dominance of the Social Democrats was further reinforced by the inability of the centre-right parties to cooperate. On several occasions, both the Centre Party and the Liberals hesitated about cooperating with the Moderates, which were seen as too far to the right. However, after the 1976 election, the centre-right parties finally succeeded to win a majority (see Table 3.3). Under the lead of the Centre Party Prime Minister, Torbjörn Fälldin, a three-party majority government including also the Moderates and the Liberals was formed. However, the years in government were highly problematic. After internal disagreement over nuclear energy, the government split in 1978. With only 11% of the parliamentary seats, but with the indirect support of the Social Democrats, a Liberal single party government was formed headed by the party leader Ola Ullsten. Despite the three centre-right parties’ problems of sticking together, they managed to win a second election in 1979, but again the government did not last the whole parliamentary term. After an agreement regarding tax policy between the Centre Party, the Liberals, and the Social Democrats that deviated from the position of the Moderates, a new minority government was formed, this time including the Centre Party and the Liberals (Möller 2015). Although hard economic conditions can partly account for this government instability and failures, the fact that these parties did not possess as strong governing competence as the Social Democrats can also be evoked. Although they all shared the objective of getting into government, their political platforms had important differences that hampered cooperation. As a result of these policy divergences, the Social Democrats, once again, came back to office for the coming three parliamentary terms. However, in the late 1980s, a government crisis arose after the Left Party refused to back an austerity package proposed by the Social Democratic government. Cooperation was then initiated with the Liberals. Yet, although talks about joining in a formal coalition took place, internal unease stood in the way. Instead, the Liberals and the Moderates were able to agree on a common economic programme, ‘New start for Sweden’, before the 1991 election. While not formally part of the deal, the Centre Party and the previously not represented Christian Democrats were openly invited to join after the election (Bolin 2012: 184). These four parties later joined in a minority government coalition, which again proved to be problematic. The economic problems, already evident during the former government,

64  Niklas Bolin demanded far-reaching measures. Because the new right-wing populist party New Democracy was not considered as a serious and reliant partner, the minority government instead made two deals with the Social Democrats. Once again, the centre-right had failed to show a level of government competence similar to that of the Social Democrats. After an unsuccessful 2002 election, the Moderates selected a new leader, Fredrik Reinfeldt, and profoundly changed the party’s political stances towards the centre. The change was essentially motivated by office-seeking ambitions (Aylott and Bolin 2007: 625). These positional and strategic changes paved the way for the formation of a pre-electoral alliance, ‘Alliance for Sweden’. Under the leadership of Reinfeldt, the four centre-right parties agreed to work together to oust the Social Democratic government in the 2006 election. Although this objective was nothing new, the comprehensiveness of the deal was quite novel. Together the four parties developed the ‘deepest and broadest’ pre-electoral alliance ‘in Swedish history’ (Allern and Aylott 2009: 277). This was embodied in a common election manifesto, in which many prior differences where settled down, and which would become the common political platform when in government. See Table 3.3 at http://partyfamilies.eu/index.php/category/chapters/ chapter-3/ Their first parliamentary term in office has largely been regarded to be successful. The well prepared programme was implemented to a great extent. The ‘Alliance for Sweden’ was able to stay in government after the 2010 election. However, due to the entry of the radical right party, the Sweden Democrats, they lost their majority. This second parliamentary term has also been regarded to be less successful then the first one. No large reforms alike were implemented, and the coalition was widely seen as a tired government (Aylott and Bolin 2015: 732). Internal problems also arose. Arguably, the ‘Alliance for Sweden’ project prioritised office-seeking goals over other objectives. This also entailed some costs, especially for the fates of the three smaller parties in the coalition. This was obvious for the Liberals. The party was increasingly marginalised as it gradually became associated with only a few issues, education being the most important. The smaller parties also suffered electorally. The 2014 election was the third consecutive election in which the Liberals suffered losses. With only 5.4% of the votes, they endured their second worse election result ever (Dahl 2015), only slightly better than the 4.7% the party secured in 1998. This poor result made the Liberals only the seventh biggest party in the Swedish parliament. In fact, during the past seven elections, only in 2002 did the party secured double digit support. The situation for the Centre Party is not much better. The party has lost ground in two consecutive elections and the 6.1% it won in 2014 is the second worse result since the introduction of universal suffrage in 1921.

The Centre Party and the Liberals 65

Ideology and policy positions Liberalism is a broad ideology with many different currents. Some even regard it as including contradictory elements (Heywood 2012: 26). Unsurprisingly, the liberal party family also seems to be less coherent than other party families (Ennser 2012; Smith 1988: 16). In the Swedish context this has perhaps best been described by Gunnar Helén, leader of the Liberals between 1969 and 1975, who argued that ‘to be liberal is to be split’ between socialism and conservatism (Dahl 2015: 232).3 The relative vagueness of the ideology, accordingly, makes it hard to determine to what extent different parties are liberal. The approach here, therefore, is to examine both how the Centre Party and the Liberals explicitly refers to liberalism in official documents and to use expert survey data and manifesto data in order to assess to what degree these parties display similarities in their political profiles. Today liberalism is embraced to a different extent by various Swedish parties. The biggest centre-right party, the Moderates, are according to their programme ‘inspired by liberalism and conservatism’ (Moderaterna 2013: 8); and the male spokesperson of the Greens have at times made claims about his party being liberal (Strömberg 2016: 226–228). However, traditionally, the Liberals is undoubtedly the Swedish party most strongly associated with the liberal ideology. Unlike other Swedish parties, the Liberals states in its programme that it is ‘the liberal party of Sweden’ (Folkpartiet 2013). The programme also includes a separate section where the concept of liberalism is defined, and refers to traditional liberal values. Amongst other things, it states that liberalism takes the individual as its point of departure, that market economy is a necessity for democracy and that the political objective is to secure individual freedom and her right to pursue happiness. The programme also has plenty of explicit references to social liberalism. Amongst other things, the party emphasises the need to improve the situation for those who are worse off and to increase the freedom for those who have the least freedom (Folkpartiet 2013). The Liberals’ development has also been described as the result of a ‘latent tension between classic and social liberalism’ (Bäck and Möller 2003: 69–70). Although the party initially fought for ideas inspired by classic liberalism, social liberalism was brought early on into the programme. With the selection of new leader Bertil Ohlin, in 1944, the party begun to promote Keynesian ideas. For instance, the party actively encouraged state intervention to stimulate job growth (Bäck and Möller 2003: 68). While existing side by side, each current of liberalism has dominated during different periods. The tensions still exist today. Under the aegis of the former leader, Lars Leijonborg (1997–2007), and the current leader, Jan Björklund (2007–), many commentators have argued that social liberalism has declined in favour of what has been coined as ‘demanding liberalism’ (kravliberalism). Concretely, this tendency is obvious in, first and foremost, the way the party emphasises that it is important not only to grant individuals (e.g., immigrants, pupils,

66  Niklas Bolin or unemployed) rights and benefits but also that it is important that these rights come along with obligations. A case in point is the party’s recurrent proposal of introducing language tests for immigrants to qualify for citizenship, a proposal that has been considered as one of the reasons for the party’s relative success in the 2002 election. The party leadership, however, has been reluctant to agree with the description of demanding liberalism. Jan Björklund has rather argued that there is no contradiction in being demanding and promoting social welfare. He has also explicitly underlined that ‘the Liberals are the social liberal party of Sweden’ (Barrling Hermansson 2010: 209). It is sometimes argued that the Centre Party initially did not represent a specific ideology. Although the party originally embraced some conservative stances, its policy orientation was mainly a collection of proposals on how to improve the conditions of the peasantry (Widfeldt 2001:16). As this specific group constituted a steadily declining part of the electorate, the party soon targeted the whole rural population. Specifically, the party adopted a sceptical stance towards the centralising tendencies that were pursued by the Social Democrats. In addition to promoting decentralisation as a way to protect the interests of the rural population, the party increasingly became oriented towards market economy. Later on the Centre Party also turned out to be the first Swedish environmentalist party. Most importantly, the party was the leading antinuclear party during the 1970s, when energy policy was on the top of the political agenda. The antinuclear stance of the Centre Party was even considered as a major factor of the centre-right parties’ success in ousting the Social Democrats from government in 1976. However, once in government, the party was not able to live up to its electoral pledge about not establishing new nuclear power plants. This failure was one important reason for the formation of the Green Party in 1981 and, in the long run, of the Centre Party’s loss of ownership of environmental policy (Bolin 2016: 159). After continuously losing electoral support, the Centre Party reached its worse result ever with 5.1% of the votes in 1998. During this period, another reform process was initiated. After a far-reaching parliamentary cooperation with the Social Democrats in the middle of the 1990s, the newly selected leader Maud Olofsson (2001–2011) again positioned the party firmly into the centre-right bloc and sought new profile issues. Perhaps most significantly the former rurality profile was complemented with efforts to also attract urban voters. The newly oriented party targeted business owners, while still keeping farmers as its main priority. To overcome a potential conflict of priorities, the party emphasised the entrepreneurial kinship between small business owners and farmers rather than the urban-rural divide (Rosén Sundström and Sundström 2010: 196). Lately, the party has increasingly come to be regarded as a liberal party, though pinpointing a precise point in time is difficult. During the past decade, the party has associated itself with liberalism in its official documents.

The Centre Party and the Liberals 67 This recognition of liberalism as an official part of the ideology can be traced back to the election manifesto of 2006, where the party’s vision was declared to be ‘uniting liberal values with good environment, social responsibility and personal integrity’. In its latest programme the party defines its liberalism as ‘social, decentralised and green’ (Centerpartiet 2013: 3).4 Core liberal ideas are further included, such as the endorsement of market economy and the emphasis on individual freedom. Social liberal aspects are also present, stating for instance that economic freedom should be combined with obligations to help the needy. However, compared with the Liberals, far fewer explicit references to liberalism are made. Let us now compare the two parties using both expert surveys and election manifestos. In Figure 3.1, Chapel Hill Expert Survey data (Bakker et al. 2015a; Bakker et al. 2015b) is presented for the Swedish parties on two dimensions for five time points – 1999, 2002, 2006, 2010, and 2014. On 10

SD

KD

C99

GAL/TAN

C02 C06 M

5 SAP

C10 L10

L06 L02 L99

V

C14 L14

MP

0

0

5 Economic left-right

10

Figure 3.1  Party positions on two dimensions, 1999–2014 Note: Years indicate the position of the Centre Party (C) and the Liberals (L) on two dimensions according to the Chapel Hill Expert Survey. The remaining parties are depicted with dashed ellipses in which all years are included. Figure inspired by Bukow (2016: 127). Source: Bakker et al. (2015a; 2015b)

68  Niklas Bolin the horizontal economic left-right dimension, there is a rather clear pattern of polarisation between the centre-left bloc of the Social Democrats (SAP), the Green Party (MP), and the Left Party (V) on the one hand and the centre-right bloc on the other hand. Generally, the Liberals (L) has positioned itself slightly to the right of the Centre Party (C) but to the left of the Moderates (M) and Christian Democrats (KD). Over time, however, the Centre Party has gone rightward: the party now occupies more or less an identical position as the Liberals on the economic dimension. The long-term trend is that the Centre Party has moved away from an economic centre position to a position slightly to the right. The analysis of the so-called GAL/TAN (Green Alternative Libertarian vs. Traditional Authoritarian Nationalistic) dimension reveals interesting findings. On the lower part of the vertical dimension, that is closer to the GAL end, the Liberals find itself on a similar position as the left-of-centre parties whereas the Centre Party, in the first three rounds of the expert survey (1999, 2002, and 2006), is positioned more to the TAN end. Over time, however, the Centre Party has moved towards a more libertarian orientation. Again we see that the party has moved to a position virtually identical to the position of the Liberals. The magnitude of the move on this dimension, however, is more striking. Whereas Figure 3.1 suggests that the Centre Party and the Liberals occupy a similar position in the Swedish party space, the content analysis of election manifestos (Table 3.4) points out important differences in what the respective parties emphasise. Although economic issues are in general those that are the most salient, the parties differ in regards to the other areas. The Centre Party has put a great emphasis on the environment. In 2014, the party actually emphasised it slightly more than the Greens. Both in the 2010 and the 2014 elections, the party devoted about a fourth of its manifestos to environmental concerns. This seems to be a rather effective strategy, as the issue, like in the 1970s, has become an important factor of Centre Party vote. While the Greens is still the party deemed to have the best environmental policies, the share of voters that thinks that the Centre Party is the most environmentally trustworthy party has increased over the past elections (see, for example, Oscarsson and Holmberg 2016: 232). The party also scores relatively high on the democracy category. A closer examination reveals that this relates almost exclusively to favourable stances on decentralisation. Although this fits well with the traditional rural profile of the party, interestingly the share devoted to this issue decreases over time. The category that is the most important for the Liberals relatively to other parties is (culture and) education. The party and its leader Jan Björklund (2007– ) has during a long period of time called for improvements and more resources to primary and secondary school. The party has been the issue owner up until the 2014 election, and has become so intimately linked with education that some have regarded it to be a single-issue party (Dahl 2015).

The Centre Party and the Liberals 69 Table 3.4  Salience of major policy areas, 1998–2014

1998 2002 2006 2010 2014 AVG

L C Others L C Others L C Others L C Others L C Others L C Others

ECO WEL EUR INT

CUL

21.4 28.9 28.0 18.2 24.0 19.2 24.6 25.6 23.7 19.0 32.5 26.8 24.2 37.0 29.8 21.5 29.6 25.5

6.0 7.9 10.2 14.9 5.7 12.6 11.7 3.0 5.9 12.5 5.0 11.3 13.9 2.6 8.6 11.8 4.8 9.7

10.7 14.9 19.1 7.9 12.2 15.8 16.5 14.3 32.2 15.4 7.8 19.4 19.5 10.8 22.9 14.0 12.0 21.9

3.6 4.1 5.4 6.1 1.9 2.1 0.4 2.6 1.9 1.9 1.1 1.0 2.6 0.0 0.6 2.9 1.9 2.2

7.1 2.1 5.6 7.2 3.4 6.9 1.8 9.8 6.9 6.2 3.9 5.9 4.8 2.4 5.5 5.4 4.3 6.2

DEMO ENV 9.5 15.3 4.7 6.5 12.2 7.1 1.2 11.9 2.0 8.0 8.9 5.0 10.4 7.1 5.1 21.5 29.6 25.5

6.0 11.6 6.8 8.4 9.9 7.1 5.4 10.3 9.6 8.3 24.7 8.1 5.0 26.5 7.2 6.6 16.6 7.8

LST

OTH

9.5 10.3 9.3 16.8 21.7 20.6 25.0 12.1 12.5 21.4 9.7 12.9 13.4 3.1 12.8 17.2 11.4 13.6

0.0 0.8 0.7 1.6 2.3 1.0 2.0 1.9 1.0 1.5 0.0 0.9 2.6 3.7 2.1 1.5 1.7 1.1

Sources: Volkens et al. (2012), Bergman et al. (2015) Notes: Figures represent share of election manifestos devoted to different policy areas. Share based on weighted (by vote share) salience of all other parties represented in parliament the given year. Aggregation of categories according to Dolezal (2016: 25): CUL = culture and education, DEMO = democracy, ECO = economy, ENV = environment, EUR = Europe, INT = international, LST = lifestyle, OTH = others, WEL = welfare. Calculations done by the author.

However, after holding the educational minister portfolio for eight straight years without being able to solve the hotly debated trend of decreasing result in Swedish schools, the party lost its ownership. The party therefore finds itself in a rather serious situation, and seems to be aware of these problems. Indeed, the 2014 election analysis carried out by the party was entitled ‘Time to broaden the Liberal Party of Sweden’ (Folkpartiet 2014a), in sharp contrast to the election manifesto title ‘Vote for the school’ (Folkpartiet 2014b). The analysis concluded that while the issue of education is still important to the party, other issues also need to be emphasised. One of the party’s historical issues is that of gender equality. While this priority can be identified in the manifesto analysis, it is here included in the lifestyle category, which also includes other issues. Next to gender equality, the party manifesto primarily includes mentions about stricter law and order – a surprising finding given the party’s position towards the libertarian end of the GAL/TAN dimension according to expert judgments. Somewhat surprising the Liberals have not emphasised international and European issues to any significant degree. Although widely seen as the most

70  Niklas Bolin EU-friendly party in Sweden, this finding is probably best understood by the fact that these kinds of issues are not a priority among voters. In sum, the expert survey data analysis reveals how the Centre Party has moved from being a slightly authoritarian centre party to become a slightly rightist libertarian party. In a general two-dimensional issue space, the Centre Party, hence, occupies today a very similar position as the Liberals. However, the analysis of manifesto data reveals important differences in issue saliency between the two parties. Whereas the Centre Party, next to the Greens, is the party most inclined to put the environment on the agenda, the Liberals give relatively much attention on educational issues.

Structure and organisation Although Swedish political parties are sparsely regulated by law, parties are organised in a very similar way, resembling the mass party model. They are structured in three territorial levels – national, regional, and local. On each level there is a highest decision-making body, the congress, elected by delegates from lower territorial levels. In general delegates come from the regional level. The Centre Party, however, grants both the local and the regional levels representation at the congress. In addition to these delegates, both the Centre Party and the Liberals grant collateral organisations, such as women, student, and youth organisations, a smaller number of delegates with voting rights. Both these parties, unlike the left-of-centre parties, also grant party board members voting rights at the congress (Bolin 2015a: 115). The congresses in these parties meet every second year. In between these meetings authority is delegated to the party board and the executive committee (Bolin 2015a; Aylott et al. 2013). Formally, party congresses select a number of higher offices, the party leader being the most prominent. Candidate selection, however, is a decentralised procedure in Swedish parties. Generally, any member can nominate candidates. However, the formal decision is made at the regional level. A nomination committee proposes a rank-ordered list, while the formal decision is made by a regional congress (Aylott et al. 2013: 172–173).5 In line with international trends (Sandri and Seddone 2015), most parties, including the Centre Party and the Liberals, arrange primaries where party members have an opportunity to have a say (Aylott et al. 2013). Importantly though, the primaries are not binding and the nomination committee can, and indeed occasionally does, present a list to the regional congress that deviates from the result of the primary. While the leader is formally selected by the congress, the decision is almost always a guided decision resulting in the selectorates having only one candidate to choose from. Like other Swedish parties, the Centre Party and the Liberals have a formal party body appointed to oversee the selection process. Generally, these selection committees nominate one single candidate. Moreover, the nominated candidate is almost never challenged by other

The Centre Party and the Liberals 71 candidates. In practice, therefore, leadership selection in Swedish parties almost always ends with a coronation (Aylott and Bolin 2016). Lately, however, this Swedish practice has been hotly debated.6 The very opaque selection process of the Social Democratic leader Håkan Juholt in 2011 seems to have increased awareness of the top-down approach of these processes. Although closed processes followed by coronations at the congress constitute a general rule, there have been a few exceptions during the past decades. In 1995, the Liberals experienced a selection process where three candidates campaigned openly against each other before Maria Leissner was selected by the congress. More recently, the Centre Party had an open race when the party replaced Maud Olofsson in 2011. In what has been regarded as one of the most open leadership selection processes of modern Swedish politics, Annie Lööf was appointed as the new leader. The process involved more open campaigning and debate between the three different candidates than what Swedish parties and voters are used to. Still, while the selections of Leissner and Lööf took place after open processes, the congress decisions ended in coronations as the competing candidates all withdrew their candidacies before the congress (Madestam 2015: 127–128). Once selected, Swedish party leaders obtain a rather secure position. There are very few examples of leadership challenges, although party leaders are formally (re-)selected on every congress. Leadership changes also occur relatively seldom. Whereas the Centre Party has replaced its leader seven times during the postwar period there has been eight leader shifts in the Liberals. Although it is hard to imagine that 21st-century leaders would stay on for more than 20 years – like Centre Party leader Gunnar Hedlund (1949–1971) or the Liberal Bertil Ohlin (1944–1967) – it is not unusual to hold the leadership for more than 10 years (see Table 3.5). See Table 3.5 at http://partyfamilies.eu/index.php/category/chapters/ chapter-3/ So far, it is hard to find any significant differences in the organisations of the Centre Party and the Liberals. However, a closer examination of the evolution of both organisations reveal relevant aspects. Whereas the Liberals was once formed as a cadre party, the Centre Party resembles a mass party, as it was founded to empower a group in society, the peasantry, that previously lacked formal influence. These differences impacted the organisational development of these parties. Whereas the Centre Party developed into a relatively large organisation in terms of membership, the Liberals remained significantly smaller. The Centre Party early on built widespread local branches in rural areas. From the 1970s, the party also made organisational inroads in urban areas (Widfeldt 2001). For a long period of time, the Centre Party was, in terms of members, the second-largest party after the Social Democrats. In the early 1980s, the party gathered more than 200,000 members. However, during the past decades, membership has decreased

72  Niklas Bolin remarkably (see Figure 3.2). Although the Liberals have also lost members, the decline is not as drastic. Today, both parties’ membership represents only a fraction of the membership they used to have. The Centre Party has less than 30,000 members, and the Liberals have only about 15,000. This has had serious repercussions; such as increasing difficulties to find candidates to run for local office in some smaller municipalities. See Figure 3.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-3/ The identity as a mass party is deeply rooted within the Centre Party. While the membership has drastically declined since the 1980s, the party still publicly cherishes internally democratic processes. In its party programme, for example, it is stated that the party should be the most open Swedish party (Centerpartiet 2013). Although it is hard to assess to what extent the Centre Party objectively is a more open party than the Liberals, it is safe to say the former puts a lot more emphasis on it in official documents. That the Centre Party traditionally prides itself of being a very decentralised party with much influence given to both regional and local branches, of course, is in line with its emphasis on decentralisation. Recent empirical research gives some preliminary support to this. In the Swedish context, the Centre Party seems, next to the Green Party, to be the most internally democratic party (Bolin 2015a; Bolin et al. 2017). However, while these findings are in line with what could be expected based on previous assessments, they should be treated with caution. First, the differences seem to be small. Second, the findings are primarily based on what Katz and Mair (1992: 6–8) referred to as the ‘official story’, that is, what the parties state in their formal documents, and less on the ‘real story’, the less formal aspects that are harder to compare systematically.

Conclusions and discussion: the parties’ main challenges for the future Up until very recently few would argue there were more than one Liberal party in Sweden. Increasingly though there are reasons to question whether this is still the case. Although the origin and early history of the Centre Party would disqualify the party from the liberal party family, on other criteria identified by Mair and Mudde (1998), the case is not as easily rejected. Transnational party affiliation certainly would speak in favour of including the party in the liberal party family. The Centre Party and the Liberals are both members of the ALDE and the Liberal International. Ideologically it is harder to make a final judgment. Both parties, albeit to different extents, refer to liberalism in their official documents. Upon closer examination, however, liberalism appears to be more salient in the Liberals’ programme. Although liberalism might be seen as a rather vague

The Centre Party and the Liberals 73 and sometimes internally incoherent ideology, with different ideological strands identified in previous literature (see, for example, Smith 1988), some aspects such as the emphasis on the individual and the embracement of market economy can be viewed as a common ground of these liberal ideologies. Based on such a parsimonious definition, the Centre Party, the Liberals but certainly also other Swedish parties can qualify as Liberal parties. Interestingly, these aspects have been significantly reinforced in the Centre Party during the past decade. When party positions are scattered on a two-dimensional issue space, the picture reveals that the two parties had distinct positions less than two decades ago, but have today almost identical positions. To be sure, there are important differences in both issue positions and issue emphasis, but on a general level, it can at least be concluded that the parties’ policy orientations are more similar today than they have been in the past. Based on data presented in this chapter, hence, it seems hard not to include the Centre Party within the same party family as the Liberals. Finally, some words about the future of the Swedish liberal parties. The formation of the ‘Alliance for Sweden’ in the first decade of the 21st century was in many ways a success. After having been kept outside of government for more than a decade, the four centre-right parties were able once again to secure executive office. Moreover, this time they managed to stay together and to govern effectively. This success, however, came with a price. And it was primarily the three junior partners of the government who had to pay. When the Alliance was formed, in 2004, the three smaller parties together gathered about 25% of the votes compared with the Moderates’ 23%. This could be compared with the situation in the 2010 election (19.5–30.1%) and 2014 election (16.1–23.3%). Despite that the Moderates lost almost 7 percentage points in the 2014 election, the party was still a lot bigger than the three junior partners combined. In short, the office seeking ambitions of the smaller centre-right parties has had repercussions on their support and affected their political profiles. Although the Centre Party has suffered from being a party that relatively weakly associates with a specific issue (Oscarsson and Holmberg 2016: 172), the Liberals faces the opposite problem. Few parties are considered to be as single-issued as the Liberals. As long as the party was regarded as the best ‘school party’, this was less of a problem. But since the party lost the ownership, this now represents a serious challenge. There are at least two other aspects that makes a continuation of the ‘Alliance for Sweden’ a problematic strategy. First, the parliamentary entry of the Sweden Democrats means that it is unlikely they will gather a parliamentary majority. This implies that they need to rely on some kind of passive support from other parties, either from the left or from the radical right. So far, both the Centre Party and the Liberals have fiercely opposed any sort of cooperation with the Sweden Democrats. Second, the once very politically united alliance has recently started to show disagreement over a number of issues, such as immigration policy. Arguably political unity has

74  Niklas Bolin been one of the main successful components of the ‘Alliance for Sweden’. Without unity, governing effectively seems difficult, and this might in the long run affect how voters perceive these parties’ governing competence. However it is not obvious that cooperation over the bloc divide is a much better alternative. Both parties have invested a lot in the ‘Alliance for Sweden’ and it has in many respects been a success. The problems resulting from the emergence of a big radical right party will certainly not disappear. Moreover, while the parties in the Alliance are not as politically united as they used to be, disagreement would hardly be a smaller problem if cooperation was sought with centre-left parties.

Notes 1. Since 2004, the parties have been members of the successor of ELDR, the Alliance of Liberals and Democrats for Europe (ALDE). 2. To overcome some of the internal skepticism towards a name change the party initially kept its former name in addition to the new. Already in 1958, however, the party congress decided to scrap the original name and has from this point been named the Centre Party (Widfeldt 2001: 6). 3. The author’s translation of the original quote: ‘att vara liberal är att vara kluven’. 4. There was, however, proponents for being more explicit about the party’s commitment to the liberal ideology. In the first draft of the new programme (Centerpartiet 2012), for example, it was stated that ‘the Centre Party is a liberal and decentralised party’ (see also Ankersjö 2012). 5. The Sweden Democrats, however, unlike the other parties run the same candidate list in all constituencies. The composition of this list is decided at the national party level. 6. A similar practice seems to be in place also in some Norwegian parties (Allern and Karlsen 2014: 51–53; Aylott and Bolin 2016).

References Allern E. H., Aylott N. (2009), ‘Overcoming the fear of commitment: Pre-electoral coalitions in Norway and Sweden’, Acta Politica 44(3), pp. 259–285. Allern E. H., Karlsen R. (2014), ‘Unanimous, by acclamation? Party leadership selection in Norway’. In: Pilet J.-B., Cross W. P. (eds.), The selection of political party leaders in contemporary parliamentary democracies. A comparative study. London: Routledge, pp. 47–61. Andersson Y., Carlsson L., Ekengard I., Kollin B.-O., Stensson B. (2000), ‘Lennart Daléus har misslyckats’, Dagens Nyheter, 23 March 2000. Ankersjö P. (2012), ‘Centerpartiet måste bli tydligare’, Svenska Dagbladet, 18 December 2012. Aylott N. (2016), ‘The Party System’. In: Pierre J. (ed.), The Oxford handbook of Swedish politics. Oxford: Oxford University Press, pp. 152–168. Aylott N., Blomgren M., Bergman T. (2013), Political parties in multi-level polities. The Nordic countries compared. Basingstoke, UK: Palgrave Macmillan. Aylott N., Bolin N. (2007), ‘Towards a Two-Party System? The Swedish Parliamentary Election of September 2006’, West European Politics 30(3), pp. 621–633.

The Centre Party and the Liberals 75 Aylott N., Bolin N. (2015), ‘Polarising pluralism: The Swedish parliamentary election of September 2014’, West European Politics 38(3), pp. 730–740. Aylott N., Bolin N. (2017), ‘Managed intra-party democracy: Precursory delegation and party leader selection’, Party Politics 23(1), pp. 55–65. Bakker R., de Vries C., Edwards E., Hooghe L., Jolly S., et al. (2015a), ‘Measuring party positions in Europe: The Chapel Hill Expert Survey trend file, 1999–2010’, Party Politics 21(1), pp. 143–152. Bakker R., Edwards E., Hooghe L., Jolly S., Koedam J., et al. (2015b), 1999–2014 Chapel Hill Expert Survey trend file. Version 1.1. Chapel Hill, NC: University of North Carolina, Chapel Hill. http://chesdata.eu Barrling Hermansson K. (2010), ‘Bland blåklint och totempålar. Om det kulturella tillståndet i Folkpartiet Liberalerna’, Statsvetenskaplig tidskrift 112(2), pp. 203–214. Bergman T., Backlund A., Bolin N., Sandström C. (2015), Umeå Election manifesto database, Umeå, Sweden. Bergman T., Bolin N. (2011), ‘Swedish Democracy: Crumbling Political Parties, a Feeble Riksdag and Technocratic Power Holders?’ In: Bergman T., Strøm K. (eds.), The Madisonian turn. Political parties and parliamentary democracy in Nordic Europe. Ann Arbor: University of Michigan Press, pp. 250–293. Bolin N. (2012), Målsättning riksdagen. Ett aktörsperspektiv på nya partiers inträde i det nationella parlamentet. Umeå, Sweden: Umeå universitet. Bolin N. (2015a), ‘Partikongresser och interndemokrati’. In: Hagevi M. (ed.), Partier och partisystem. Lund, Sweden: Studentlitteratur, pp. 105–119. Bolin, N. (2015b), Party Membership Figures. Sweden 1945–2014. MAPP Project Data Archive. www.projectmapp.eu Bolin N. (2016), ‘Green Parties in Finland and Sweden. Successful cases of the north?’ In: van Haute E. (ed.), Green Parties in Europe. London: Routledge, pp. 158–176. Bolin N., Aylott N., von dem Berge B., Poguntke T. (2017), ‘Patterns of intra-party democracy around the world’. In: Pogunktke T., Scarrow S., Webb P. (eds.), Organizing representation: Political parties, participation, and power. Oxford: Oxford University Press. Bäck M., Möller T. (2003), Partier och organisationer. Stockholm: Norstedts Juridik. Bukow S. (2016), ‘The Green Party in Germany’. In: van Haute E. (ed.), Green Parties in Europe. London: Routledge, pp. 112–139. Centerpartiet (2012), En hållbar framtid – ett förslag till nytt idéprogram. Centerpartiet (2013), Centerpartiets idéprogram. En hållbar framtid. Christensen D. A. (1997), ‘Adaptation of Agrarian Parties in Norway and Sweden’, Party Politics 3(3), pp. 391–406. Dahl S. (2015), ‘Folkpartiet och den förlorade mitten’, Statsvetenskaplig tidskrift 117(2), pp. 231–244. Dolezal M. (2016), ‘The Greens in Austria and Switzerland. Two successful opposition parties’. In: van Haute E. (ed.), Green Parties in Europe. London: Routledge, pp. 15–41. Ennser L. (2012), ‘The homogeneity of West European party families: The radical right in comparative perspective’, Party Politics 18(2), pp. 151–171. Folkpartiet (2013), Partiprogram. Frihet i globaliseringens tid. Folkpartiet (2014a), Dags att bredda Sveriges liberala parti. Rapport från Folkpartiet liberalernas valanalysgrupp. Folkpartiet (2014b), Rösta för skolan. Folkpartiet liberalernas valmanifest 2014. Folkpartiet (2015), Liberal kraftsamling. Landsmöte 2015.

76  Niklas Bolin Heywood A. (2012), Political ideologies. An introduction. Basingstoke: Palgrave Macmillan. Katz R. S., Mair P. (1992), ‘Introduction’. In: Katz R. S., Mair P. (eds.), Party organizations: A data handbook. London: Sage, pp. 1–20. Lindström U., Wörlund I. (1988), ‘The Swedish Liberal Party: The politics of unholy alliances’. In: Kirchner E. J. (ed.), Liberal Parties in Western Europe. Cambridge: Cambridge University Press, pp. 252–278. Madestam J. (2015), ‘Partiledare’. In: Hagevi M. (ed.), Partier och partisystem. Lund, Sweden: Studentlitteratur, pp. 121–132. Mair P., Mudde C. (1998), ‘The party family and its study’, Annual Review of Political Science 1, pp. 211–229. Moderaterna (2013), Moderaternas idéprogram. Ansvar för Sverige. Möller T. (2015), Svensk politisk historia. Strid och samverkan under tvåhundra år. Lund: Studentlitteratur. Oscarsson H., Holmberg S. (2016), Svenska väljare. Stockholm: Wolters Kluwer. Rosén Sundström M., Sundström M. (2010), ‘Ett smalare men vassare Centerparti?’ Statsvetenskaplig tidskrift 112(2), pp. 189–202. Sandri G., Seddone A. (2015), ‘Introduction: primary elections across the world’. In: Sandri G., Seddone A., Venturino F. (eds.), Party primaries in comparative perspective. London: Routledge. pp. 1–20. Seki K., Williams L. K. (2014), ‘Updating the party government data set’, Electoral Studies 34, pp. 270–279. Smith G. (1988), ‘Between left and right: The ambivalence of European liberalism’. In: Kirchner E. J. (ed.), Liberal Parties in Western Europe. Cambridge: Cambridge University Press, pp. 16–28. Strömberg M. (2016), Vi blev som dom andra. Miljöpartiets väg till makten. Stockholm: Atlas. Särlvik B. (2002), ‘Party and electoral system in Sweden’. In: Grofman B., Lijphart A. (eds.), The evolution of electoral and party systems in the Nordic countries. New York: Agathon Press. Widfeldt A. (2001), ‘The Swedish Centre Party: The poor relation of the family?’ In: Arter D. (ed.), From farmyard to city square? The electoral adaptation of the Nordic Agrarian Parties. Aldershot, UK: Ashgate, pp. 1–30. Volkens A., Lacewell O., Lehmann P., Regel S., Schultze H., Werner A. (2012), The Manifesto Data Collection. Manifesto Project (MRG/CMP/MARPOR). Berlin: Wissenschaftszentrum Berlin für Sozialforschung (WZB).

4

Liberalism in the Netherlands The VVD and D66 Gerrit Voerman

Introduction ‘The Dutch liberal tradition is strong. In fact so strong, that we can afford the luxury of having two liberal parties in the Netherlands: D66 and the VVD. But in Brussels we are both “Liberal and Democratic Reformists”. Sometimes it takes Europe, not just to unite people from different countries, but also to unite like-minded people in the same country.’ At the congress of the European Liberal Democrat and Reform Party (ELDR) in November 2003, Jozias van Aartsen, the parliamentary leader of the Conservative Liberal People’s Party for Freedom and Democracy (Volkspartij voor Vrijheid en Democratie – VVD), flirted with the idea of a merger with the Social Liberal Democrats 66 (Democraten ’66 – D66). It was only in the years before that this rapprochement had become visible in the top echelons of the VVD. For a long time this party did not want to have anything to do with the Democrats, and vice versa. One could almost say that this animosity was genetically determined because, although D66 cannot strictly speaking be regarded as an offshoot of the VVD, many Liberals attended its birth. Of the 47 founding members of D66, 16 had been VVD members (Godschalk 1970: 72).1 However, the (not very concrete) merger proposals did not really come as a surprise. They fitted in with the steady ideological, programmatic, and organisational process of convergence between the two parties, which started in the early 1990s. In retrospect, it can be established that this process did not become reality. The VVD and D66 did not merge, and the word ‘merger’ is no longer mentioned. Not only did the programmatic convergence not continue, but both parties have also been successful again in the last few national elections. The VVD has been doing very well, becoming the largest party in Dutch parliament in the elections of 2010 and 2012 and delivering the first Liberal Prime Minister since 1918. D66 also recovered from its deep decline, appearing as the largest party in the European elections of 2014. With upwards electoral trends on both sides, the perspective of a merger is, not surprisingly, further away. The parties included in this analysis are selected on the basis of their membership of the ALDE party caucus and/or of the Liberal International. The VVD as well as D66 belong to both transnational liberal federations. The VVD was

78  Gerrit Voerman one of the founders of the ELDR in 1976 and has been a member since then. After the European elections of 1989, D66 MEP Jan-Willem Bertens joined the ranks of the ELDR caucus in the European Parliament, to which the VVD MEPs have belonged since 1979. In 1994, D66 joined the ELDR party organisation. In 1960, the VVD became member of the Liberal International. D66 was admitted as an observer in1986; in 1999, the party became full member. The common membership of the VVD and D66 to these transnational liberal federations reflects increasing similarities in their ideological beliefs, as demonstrated below. The subsequent section describes how both parties also started to resemble each other in organisational respect. First, however, the origins and historical evolution of the VVD and D66 are analysed.2

Origins and development Like most European polities, the Dutch political system has been shaped by the class cleavage but also by a religious cleavage. In 1879, the first modern – orthodox protestant – mass party in the Netherlands was founded, demanding state subsidies for confessional schools. The Catholics were slower in building a modern party. Both parties were embedded in a growing network of voluntary associations, private schools, and daily newspapers, followed later by universities, trade unions, and radio and television broadcasting associations, called zuilen in Dutch (‘pillars’ in English). Pillarisation isolated Catholics and Calvinists from each other and from the secular minority in the population, thus restricting the influence of the propaganda of the Liberals and the Socialists. The latter established their first party in 1881, and also developed their own subculture. The more individualistic Liberals were always opposed to creating a pillar. Not favouring strongly organized parties, they reluctantly set up a loose federation, the Liberal Union (Liberale Unie) in 1885. The organisation of the Unie tended to be weak and its internal differences strong, resulting in several schisms. In the interwar period there were two Liberal parties: the Social Liberal League of Free Democrats (VrijzinnigDemocratische Bond – VDB), which electorally fluctuated around 5.5%, and the Conservative Liberal State Party (Liberale Staatspartij – LSP), which declined from 9.3% to 3.9% of the vote during this period. After the World War II, the LSP changed its name into Freedom Party (Partij van de Vrijheid – PvdV). At the same time the VDB merged with the Social Democrats into the Labour Party (Partij van de Arbeid – PvdA). Soon after, former VDB-members led by Pieter Oud left the party, which in their view was too dominated by Social Democrats. In January 1948, this group and the PvdV founded the VVD. At the elections of July 1948 the new party received 8.0% of the vote. One month later it entered government – crossing within about half a year the threshold of declaration, authorisation, representation, and relevance (Pedersen 1982). See Table 4.1a at http://partyfamilies.eu/index.php/category/chapters/ chapter-4/

Liberalism in the Netherlands 79 In the history of the VVD (for party chronology, see Table 4.1a), periods of electoral prosperity and of adversity have alternated. In the 1950s, 1970s, 1990s, and 2010s the party was relatively successful, which has to do with firm leadership and a clear strategy directed against Labour (Te Velde 2008). In the intervening periods party leaders were less decisive and less able to develop a strong electoral profile. During the 1950s, the VVD was dominated by Oud, a competent but authoritarian leader who vigorously presided over both the parliamentary caucus and the party organisation (Koole et al. 1988). During these years he waged an increasingly vehement ideological campaign against Labour, which strove (in his eyes) for ‘state socialism’. In the elections of 1959, the VVD obtained 12% of the vote, a growth of almost one-third – a considerable gain in these days of pillarisation and political stability. For the first time since 1952, the party entered government in a centre-right coalition with the Catholic and Protestant parties, which lasted with a short intermission until 1973. In 1963, Oud, at the age of 76, stepped back under the pressure of younger regional party leaders. An interregnum followed, with two competent but not so dominant party leaders (Edzo Toxopeus and Molly Geertsema), who did not attack Labour the way Oud had done. Thus, during the 1960s, the VVD lacked both strong leadership and an aggressive campaign strategy. The party stabilized at some 10% of the vote. The first significant electoral breakthrough of the VVD came in the 1970s, when Hans Wiegel, only 29 years old, became party leader (Vermeulen 2013). In the parliamentary elections of 1972, his party won 14.5%, and in 1977 18.0%, both postwar records. Wiegel combined debating skills, flair, and populist rhetoric, often appealing to ‘the people in the country’, without betraying Liberal principles. Like Oud, he waged an ideological campaign against Labour, which had meanwhile radicalized. Wiegel was able to take advantage of the secularisation, depillarisation, and individualisation of society. He remained political leader even after he and his party joined the cabinet in 1977 with the Christian Democrats (CDA), into which the Catholic party and two Protestant parties had merged. Four years later he again led the party in the elections, although this time to (a modest) defeat. Wiegel soon left the party leadership to Ed Nijpels, a relatively progressive young Liberal, who did well in the early elections of 1982 (23.1%, a new record). His popularity, however, waned rapidly and he was forced to step back in 1986. The party’s top leadership decided to introduce a dual leadership: Vice-Prime Minister Rudolf de Korte and Joris Voorhoeve, chair of the parliamentary caucus. Gradually the latter came to the fore, but he was not a strong leader (Hoedeman 1993). Polarisation against Labour did not seem a suitable strategy because the Social Democrats were in the opposition and weakened by internal problems. When the VVD ended the coalition with the CDA in 1989, the party, internally divided, only received 14.5% of the vote. In 1990 Voorhoeve was not so gently replaced by Frits Bolkestein as party leader. The latter contributed to electoral recovery and unprecedented growth

80  Gerrit Voerman in the 1990s (24.7% in 1998), aided by favourable international ideological and economic circumstances for his party (the rise of monetarism and supply-side economics, the collapse of communism in the eastern bloc, etc.). Like Oud and Wiegel, Bolkestein emphasized the ideological struggle against socialism. This did not prevent him from joining a coalition with D66 and Labour in 1994 – although the former enemy was by then hardly Socialist any more. At the same time, Bolkestein, who had remained leader of the parliamentary caucus, opposed the coalition government on sensitive issues such as immigration policy and European integration, in this way covering the right side of the electoral market. Hans Dijkstal, who succeeded Bolkestein in 1998 when the latter voluntary stepped back, did not follow his example. By joining the cabinet, he could no longer criticise it, and he did not give priority to immigration problems. In this way Dijkstal gave room to the rise of the right-wing populist Pim Fortuyn, who in 2002 could make an inroad in the Liberal electorate. Again the VVD appeared to lack a strong and uncontested political leader. After the heavy electoral defeat in 2002, Dijkstal stepped back and was replaced by Gerrit Zalm. One year later he led the party in the elections. When Zalm became Minister, Van Aartsen, the leader of the parliamentary caucus, claimed the position of party leader, but the party congress did not regard him as such. In this period for the first time in the history of the VVD, a member of parliament (MP) broke away from the parliamentary caucus. In September 2004, Geert Wilders left because he did not want Turkey to join the European Union. Two years later he founded the Freedom Party (Partij voor de Vrijheid – PVV) and gained 5.9% of the vote in 2006, developing the party into a radical right-wing populist rival of the VVD. Until then, the VVD had only rarely been confronted to serious competition from its right. In the 1980s and early 1990s, small xenophobic and nationalist parties competed for voter support from this position, but they were hardly successful. In the run-up to the elections of 2006, Zalm and Van Aartsen stepped back, paving the way for Mark Rutte, a former State Secretary. He became party leader, for the first time directly elected by the party members. His leadership was initially hesitant (Sitalsing 2016). In the membership ballot he gained a narrow majority over his rival Rita Verdonk, who obtained the second position. At the elections she got more votes than Rutte, which was unprecedented and weakened his position as party leader. Verdonk claimed that she was more entitled to lead the VVD than Rutte. In September 2007, he expelled her from the parliamentary caucus. Subsequently, as opposition leader sharply criticising the coalition of CDA, Labour, and a minor party, Rutte gradually gained credibility. At the elections of 2010, the VVD became the largest party with 20.5% of the vote, also due to the fact that other established parties performed poorly. For the first time since 1918 a Liberal became Prime Minister. Rutte led a minority coalition of VVD and CDA, supported by Wilders’ Freedom Party. In 2012, the Freedom Party withdrew its support, causing the fall of the cabinet. At the early elections in September, the VVD scored 26.6% of the vote, an all-time record. In the election campaign Rutte sharply opposed Labour, but

Liberalism in the Netherlands 81 after the elections both parties formed a coalition in no time. In the opinion polls in autumn 2016, the VVD lost nearly one third of its electorate. In the years since its foundation in 1948, the VVD was the only Liberal party represented in Dutch parliament. In 1966, a rival appeared with Democrats 1966, although this new party initially did not label itself ‘Liberal’ – notwithstanding the fact that about one third of the party founders had been member of the VVD. D66 took the threshold of declaration in a remarkable way (Pedersen 1982). On 15 September 1966, an ‘initiative committee’ published a short manifesto (priced one guilder) in which they asked ‘fellow citizens’ to help to ‘re-democratize’ the Dutch political system by supporting the establishment of a new party. They received 2,500 expressions of allegiance (a response rate of more than 20%), after which on 22 October 1966 they decided to found the party Democrats 1966 (D’66, but since 1985 it has been D66 without the apostrophe) (for party chronology, see Table 4.1b). The newcomer declared its intention to take part in the parliamentary elections of 1967, subsequently meeting the conditions of authorisation. See Table 4.1b at http://partyfamilies.eu/index.php/category/chapters/ chapter-4/ In the elections it received 4.5% of the vote (Table 4.2), the best result for a debutant at the national elections until then, crossing the barrier of representation. In 1973, it participated in government for the first time, clearly demonstrating its relevance. See Table 4.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-4/ D66 needed strong leadership perhaps even more than the VVD did. In electoral respect the party was – to different degrees – doing well under the leadership of party founder Hans van Mierlo (1966–1973, 1986–1998), Jan Terlouw (1973–1982), and Alexander Pechtold (2006– ), while it lost support under other leaders. The charismatic van Mierlo regarded D66 as dynamite that had to explode the old political system, creating a new structure with a Conservative and a Progressive political bloc. Although he was successful again at the elections of 1971, tensions within the party were increasing between on the one side, ‘exploders’ like him who preferred close cooperation and even merger with Labour and another progressive party, and on the other side those who preferred an autonomous D66. After the party lost the early elections of 1972, the latter wing gained the upper hand. After van Mierlo led D66 into a government coalition with Labour, he lost the support of his MPs who found him too self-serving and too close to the Social Democrats. Van Mierlo resigned in September 1973 and was succeeded by Jan Terlouw. The party entered a deep crisis: In September 1974 its congress decided to

82  Gerrit Voerman dissolve D66, but the statutory required two-thirds majority was lacking. Party unity was tested as two of the six MPs left the caucus. Unlike van Mierlo, Terlouw considered D66 an ‘ordinary’ party, which had to establish a position between the VVD and Labour, ending its more or less exclusive orientation on the Social Democrats. Profiling his party as the ‘reasonable alternative’ and the ‘fourth political movement’, Terlouw revitalized D66, getting 11.1% of the vote in 1981. Together with the CDA and Labour, D66 formed a coalition that collapsed within a year. D66 stayed in the cabinet with the CDA to prepare for early elections, with disastrous results: it lost nearly two thirds of its voters. Terlouw was heavily criticized for having continued to govern without Labour. He resigned and was succeeded by the largely unknown Maarten Engwirda, who after a few years gave way to party founder van Mierlo who made his return as party leader, this time keeping more distance from Labour. In the opposition, D66 expanded electorally under van Mierlo’s still charismatic leadership to 15.5% of the vote in 1994, the highest score in the history of the party. That year D66 more or less engineered the ‘purple’ cabinet with Labour and the VVD. For the first time since 1918 the Christian Democrats did not participate in the coalition – a long-cherished wish of van Mierlo, who became Vice-Prime Minister. The dreamed coalition was electorally not a success for D66. In 1997, van Mierlo announced his resignation as party leader. He was succeeded by Els Borst who became the first female leader of one of the larger Dutch parties. She was second choice, however, and regarded as ‘interim’ (Van der Land 2003). In the elections of 1998, she lost a third of the party’s voters. It marked the beginning of a period of electoral decline and weak leadership, which would last till 2006. After the defeat in 1998, D66 continued its presence in the purple coalition, although numerically the party was no longer needed. Borst transferred the leadership of D66 to Thom de Graaf. After defeats at the elections of 2002 and 2003, he left and was succeeded by Boris Dittrich. Under the leadership of the latter, D66 formed a centre-right coalition with CDA and VVD. It was the first time that the party opted for governing without Labour (apart from the interim cabinet in 1982), which was highly controversial within D66. Dittrich proved to be an uncertain leader. In February 2006, he resigned after a painful defeat in parliament. He was replaced by Lousewies van der Laan, who a few months later brought down the government in which her party also participated. This implied the end of her leadership: After a polarising campaign she lost the membership ballot to Alexander Pechtold. It was the first time that all members could participate in the party leadership elections. Under the leadership of Pechtold, D66 gained 2.0% of the vote in the elections of November 2006 – a nadir in the party’s electoral history. Under his leadership, however, D66 would recover electorally to 8.0% of the vote in 2012. In the European elections of 2014, the party even scored 15.5%, becoming the largest party. D66 certainly benefited from the fact that the party was in opposition all the time, but its success has also to be attributed to Pechtold, who attracted a lot of attention as an eloquent speaker,

Liberalism in the Netherlands 83 especially in the way he countered the anti-Islam rhetoric of Wilders. In the opinion polls of autumn 2016, D66 scored about 10% of the vote. Electoral performances In 1948, the VVD was the fifth largest party of the Netherlands, with 8% of the vote. In 2012, it was the largest party, with 26.6% – more than three times as large. In the 1950s, the socioeconomic make-up of its electorate differed strongly from other parties: 22% was well-to-do and 48% better-off (middle class) (Staatscommissie J. Donner 1956: 86). Shopkeepers, businesspeople, farmers, and civil servants voted for the VVD. From the 1970s onwards, the VVD, led by Wiegel, would profit from secularisation, which reduced the loyal electorate of the Christian parties. At the same time, class-consciousness and collectivism made way for individualism, eroding the dominance of the collectivist parties, not only the confessionals but also Labour (Van der Kolk 2000). The VVD made inroads on their electorate, in particular among depillarised Christians and individualised workers, while also mobilising its traditional electorate. D66 also managed to capture the depillarised voters. The party’s core electorate was small but its potential electorate was fairly large though rather volatile, which explains its strong electoral fluctuations. In sociodemographic terms, in 2012 the electorates of D66 and VVD resembled each other in terms of gender (respectively 50% and 48% female) and (subjective) social class (93% versus 87% middle and upper class). However, the proportion of voters with higher education degree is nearly twice higher among D66 voters than VVD voters (67% versus 38%, respectively). D66 voters were also much younger: 55% below 45 years, in comparison with 41% of the VVD voters.3 Governmental participation The VVD occupies a somewhat eccentric position in the Dutch party system – more so than its left-wing Liberal rival D66. Nevertheless, the VVD has managed to exercise considerable influence on Dutch politics and has taken part in more postwar governments than any other secular party (Table 4.3). See Table 4.3 at http://partyfamilies.eu/index.php/category/chapters/ chapter-4/ Coalition theory would lead us to expect that the VVD would spend most of its time in the opposition. CDA and Labour are closer to the centre and, in addition, they have usually been larger. Moreover, even if CDA and Labour did not hold a majority in parliament, they could recruit D66 to their coalition. Despite this not too promising point of departure, however, the VVD has participated in governments two thirds of the time from its founding in January 1946 until the end of 2016, which represents 548 out of 827 months (66%). In relative terms D66 has participated less in government than the

84  Gerrit Voerman VVD (201 out of 602 months since its founding in October 1966 – 33% of the time), although the party’s position was closer to the centre. Its exclusive orientation to Labour, which lasted till 2003, might have played a role. From 1967 onwards, the positions of the VVD and D66 have always been ‘antithetical’: whenever one of them participates in government, the other sits in opposition. It is only from 1989 to 1994 that both parties were together in opposition. In 1994, there was a second and more remarkable novelty when D66 and VVD both joined the government together with Labour. Since its foundation D66 had campaigned for a cabinet excluding Christian Democrats but including Labour and the Liberals. The party had always been wary of participating in a centre-right cabinet without Labour, but chose to do so in 2003. In this way D66 terminated its strategic orientation towards Labour, expanding its coalition options and getting closer to the VVD.

Ideology and policy positions The VVD as well as D66 consider themselves to be Liberal, united in their pursuit of the greatest possible freedom for the individual in as many areas as possible. D66 started to label itself ‘Liberal’ nearly 35 years after its founding (Van der Land 2003). The VVD has done so since its founding in 1948, although reluctantly. In the manifesto that was then adopted, the concept ‘liberalism’, tainted since the economic crisis of the 1930s, did not appear (VVD 1948). It was only in the revised manifesto of 1966 that the VVD proclaimed itself to be a ‘Liberal party’ (VVD 1966). From the beginning the VVD accepted that there are limits to individual freedom, which therefore had to be constrained by principles such as responsibility, tolerance, and (social) justice (Voerman and Lucardie 1992). Collectivist values like family and community had to curb individualism. During the process of social change in the late 1960s and 1970s, however, individualisation was welcomed without due consideration. In the Liberaal Manifest of 1981, the VVD demonstrated an optimist perspective on humankind, which was combined with the notion of an active state, which had to stimulate the emancipation and personal fulfilment of the individual (VVD 1982). In the course of the 1980s, this Liberal vision became more pessimistic; the party acknowledged that individualism also had drawbacks. As depillarisation resulted in decreasing social control, the VVD again emphasized law and order. The new version of the Liberaal Manifest, in 2005, not only underlined the protection of individual freedom but also national identity, against the background of the problems caused by immigration and European integration – issues that earlier were put on the agenda by Bolkestein (VVD 2005). Surprisingly, the then-party leader Rutte did not refer to these issues in the manifesto he wrote and which was accepted in 2008, but he did explicitly mention that citizens had to take into account the norms of Dutch society, which was characterized by the Jewish-Christian tradition, humanism, and

Liberalism in the Netherlands 85 Enlightenment (VVD 2008). Partly under the pressure of Wilders’ rightwing populist Freedom Party the VVD moved to the right in this respect. With regard to the socioeconomic dimension the VVD occupies a position to the right of all other established parties in the Netherlands, basically in defence of freedom from state intervention and of inequality – as long as the latter is based on achievement and merit rather than noble birth and family connections. Since the late 1950s, however, the VVD as a governing party has cooperated with the Christian parties in the expansion of the welfare state. In the 1970s, under Wiegel, the party propagated cuts in the social arrangements and other government expenditure, as well as the reduction of bureaucracy and the role of the government to make citizens less dependent. The VVD wanted more market and less state, but did not head towards a complete free market; in the Liberaal Manifest of 1981, for instance, it embraced the ‘social market economy’ (VVD 1982: 27). However, in the version of 2005 it preferred the ‘free market economy’ (VVD 2005). This was also the term Rutte used in his manifesto, in which he wanted to represent the ‘hard-working Dutchman’ (VVD 2008: 2), demanding lower taxes and an activating welfare state with social security as temporary as possible. Initially D66 regarded ideologies as out-dated; problems had to be solved in an undogmatic and ‘pragmatic’ way (Voerman and Lucardie 2001). Party leader van Mierlo was opposed to drafting a manifesto. His successor Terlouw, however, used the term ‘progressive-liberal’ to indicate what he considered to be the task of D66: to protect the individual against the oppressive forces of collectivism and to expand democracy. After both crises of D66 in the mid-1970s and the early 1980s, the call for an ideological positioning increased, but van Mierlo, who became party leader again in 1986, remained reluctant. In the third electoral crisis, starting in the mid-1990s, a group of young party members calling themselves Opschudding (Upheaval) claimed that D66 had to strengthen its link with the voters by presenting itself as ‘social liberal’. In 1998, the party congress accepted their proposal, resulting in the party’s first manifesto, called Uitgangspunten van Democraten 66 (Starting points of Democrats 66). In this document, which asserts the party’s aversion of ‘blueprints or dogmas’, D66 displays an optimistic view on humankind. Individuals are in principle free, equal, and emancipated, but the state should provide conditions for them to fulfil themselves. In its preference for individual freedom, self-determination, and a ‘market without dominance’ (D66 2000: 2) D66 comes close to classic liberalism; yet the pursuit of equality and involvement – a contemporary translation of ‘solidarity’ – and social and ecological preconditions could be considered characteristic of ‘social’ liberalism. D66 wants to promote empowerment of the individual and direct democracy. The call for a democratisation of the state is the reason why the party was founded and distinguishes D66 from all other parties. The party demands more power for the voter by having the Prime Minister and the

86  Gerrit Voerman mayors elected by the people, and by adopting a majoritarian electoral system, corrective referendums as well as people’s initiatives. In the history of the party, the emphasis on constitutional renewal varied. When van Mierlo and De Graaf were party leaders, democratisation was given priority; but Terlouw attached greater importance to environmental pollution and energy scarcity, and Pechtold to education. In programmatic respect, the VVD and D66 show similarities in some policy areas and substantial differences in others.4 In moral and cultural affairs, both parties advocate individual autonomy with respect to abortion and euthanasia. Regarding drugs, fighting crime, and immigration and the integration of immigrants, D66 is usually more liberal than the VVD. The former wants to legalize and regulate the supply of soft drugs, stresses prevention of crime (by education and integrating young people in the labour market), and takes an understanding position regarding migrants, whereas the latter generally takes more conservative or repressive positions (heavier penalties, immigration controls). In relation to the multicultural society, the VVD tries to woe voters from the Freedom Party by advocating tougher positions, whereas D66 under Pechtold turned into the Freedom Party’s most outspoken opponent. On socioeconomic issues, there are currently few differences left between the VVD and D66. Both parties want to strengthen the market economy, even though D66 emphasizes that it has to be a social market economy. They have also accepted recent major decisions such as the rise of the retirement age to 67 years and the further liberalisation of the labour market. Regarding environmental policies, D66 goes further than the VVD, advocating sustainable development and rejecting nuclear power, whereas the latter wants to retain this form of energy. D66 wants to stimulate electric driving, the VVD cherishes the car. As far as the latter party is concerned, economic interests often weigh more heavily in the balance than environmental concerns. The VVD has never appreciated the constitutional reforms of D66, such as the referendum and the direct election of mayors and the Prime Minister. The party regards the referendum as an infringement of the principles of representative democracy. Surprisingly the three reforms were included in the VVD manifesto of 2005, but they have remained disputed within the party. They were not included in the platforms, apart from the election of mayors, which was part of the programmes of 2006 and 2010 and then disappeared in 2012. In its platform of 2017, D66 still promotes these three reforms, although the party does not place the same emphasis on them as in the times of van Mierlo and De Graaf (D66 2016). With respect to European integration, both parties hold divergent views despite their common membership of the pro-European ELDR/ALDE (Vollaard and Voerman 2015). The VVD was for a long time pro-European but became more reserved in the 1990s on the initiative of Bolkestein. He wanted the European Union to limit itself to core tasks in cross-border policy areas, such as the internal market, international trade, migration, terrorism, and energy and climate. For 2017 the VVD advocated economic integration:

Liberalism in the Netherlands 87 the expansion of the internal market and a strict compliance with the Stability Pact. It rejects a political European union and claims that national feelings and interests should be taken into account (VVD 2016a). D66, on the other hand, still favours a ‘federal Europe’, though respecting national differences. The party is willing to give up national sovereignty: it wants fewer vetoes, to strengthen the position of the European Parliament and the Commission (by the direct election of its chair), and a European approach to the financial crisis. Not surprisingly, the programmatic positions of the two parties are reflected in the opinions of their voters.5 On ethical questions such as euthanasia, the electorate of the two parties stand united; most of the voters in 2012 were willing to allow euthanasia at the patient’s request. Besides, with regard to constitutional reform there is surprisingly significant agreement, 66% of D66 voters and 61% of VVD voters are in favour of the elected mayors. With regard to European integration, however, opinions are sharply divided: 60% of D66 voters are pro EU, versus 38% of VVD voters. On the socioeconomic front too, the positions of the two groups differ: 47% of D66 voters and 33% of VVD voters favour more equal income distribution. The most sensitive political issue nowadays is the question of integration. Here, too, the differences between the supporters of the VVD and D66 are substantial. In 2012, 68% of VVD voters believed that asylum seekers who were already in the Netherlands should be sent back to their countries of origin; 34% of D66 voters agreed. In addition, about 69% of VVD supporters felt that immigrants and ethnic minorities should conform to Dutch cultural norms, and only 12% thought that minorities should be allowed to maintain the habits of their own culture. A large minority of D66 voters (40%) also preferred assimilation, and about one third felt that maintaining the original culture was acceptable. Finally, in terms of left and right, D66 voters considered themselves different from VVD voters, but not so strong as in 2003. On a scale of 1 (left) to 10 (right), in 2012 40% of D66 voters (vs. 67% in 2003) placed themselves on the left of the scale; 34% (vs. 5% in 2003) were outspokenly right wing (positions 8, 9, and 10) (data for 2003: van Holsteyn, Lucardie, and Voerman 2004). The vast majority of VVD voters, on the other hand, positioned themselves on the right of the scale: 80% (vs. 42% in 2003) chose a clear right-wing position (8-10). Only 7% of VVD voters considered themselves left wing, and then predominantly moderately left wing (vs. 15% in 2003). In the past 10 years, the voters of the VVD as well as of D66 have clearly moved to the right, but there is still a considerable gap between both parties. It is obvious that this distance and programmatic differences stand in the way of a possible merger of the VVD and D66.

Structure and organisation In many ways the organisations of the VVD and D66 are similar. Like most parties, both have a pyramidal structure: they are based on local branches with on top the party executive and regional organisations in between.

88  Gerrit Voerman In both cases the party congress has a pivotal position. It makes all the important decisions: it elects the party executive and adopts the party manifesto, platform and statutes. Moreover, it also nominates the first candidate (VVD and D66) and the other candidates (VVD only) for parliament, although since the turn of the century this procedure has been profoundly changed in both parties. VVD as well as D66 had a party council (in the period 1948–2015 and 1978–1998, respectively), dominated by delegates of the regional party organisations.6 Originally, the VVD and D66 represented two different models of party organisation (Daalder and Koole 1988). In accordance with its program of radical democratisation, the party structure of D66 was based on the principle of direct democracy (Koole 1992). Since its founding, every D66 member has the right to speak and to vote at the party congress. In the candidate nomination process, members can directly determine (by mail) the ranking of the candidates. The first candidate, however, had to be elected at the party congress, until in 2006 de facto and one year later also formally internal primaries were introduced, in which all members could participate. In 2000, the referendum was introduced, and in 2010, for the first time, the party chair was elected by all members, and no longer by the membership congress (Voerman 2014). The influence of members within D66 has been constrained by centralising tendencies. In 1978, a ‘voting recommendation commission’ elected by the membership congress was introduced in the nomination process. Its purpose was to foster a balanced composition of the caucus, and it has often had a substantial influence on the outcome of the primaries (Voerman 2001). In 2007, the commission was reduced to two persons: the party chair and the first candidate. The latter was in fact the party leader, who – depending on their charisma and electoral success – could exercise considerable informal influence within the party. This was true in particular for party founder van Mierlo. He stepped back in 1973 but returned in 1986, his comeback being well-planned in party backrooms (van der Land 2003). Van Mierlo, who was highly respected within the party, not only proposed twice the person who would succeed him (see below), but he also played a crucial role at the party congress in 2005 that had to decide on the continuation of the participation of D66 in a centre-right coalition (van der Land 2012). In the same vein, Pechtold gained substantial influence, surrounding himself with a few loyal supporters and securing a stronger position for himself in the nomination process (van der Land 2012). Unlike D66, the VVD organisation was based on the principle of representation. The party congress consisted of delegates appointed by the local branches, although formally every member had the right to speak at the congress. Much more than in D66, the regional party organisations, from the early 1960s onwards, were influential within the party. Until 1988 they were represented in the party executive; two years later they lost their dominant formal position in the candidate nomination procedure (Koole 1994).

Liberalism in the Netherlands 89 To compensate for this, the party executive regularly consulted the regional chairs. Regional chairs were also at least several times unofficially involved in the leadership selection process (Schulte and Soetenhorst 2007). In 2002, after a severe electoral defeat, the VVD abolished its structure of representative democracy. All members could directly elect the party chair and its first candidate at the national and European elections, rank the other candidates on the list, and vote on the party congress (Voerman 2014). Fearing to become marginalized, the regional chairs countered the internal democratisation (Schulte and Soetenhorst 2007). They were able to keep their influence by being consulted by the party executive, formally only regarding organisational matters. In 2015, the VVD reorganized again, transforming its hierarchical structure into a so-called ‘network party’, opening up the party to nonmembers without granting them voting rights (Commissie Toekomst van de VVD-structuur 2015). The party council was abolished; membership rights were further expanded, and the number of regional party organisations was reduced from 17 to 7. The position of the regional chairs was upgraded: together with the downsized party executive they formed the National Board of Consultation (Landelijk bestuursoverleg). This new body has important formal competences in the candidate nomination process: it contributes to the list of recommended candidates, and can adopt the definite list in case the turnout at the membership referendum is below the threshold (VVD 2016b). As in the case of D66, successful and charismatic leaders could dominate the VVD. A fine example is Wiegel, who in critical moments was able to impose his will on the party. He left as party leader in 1982, but until the early 2000s a possible comeback hung over the VVD (Schulte and Soetenhorst 2007). Regional party organisations were crucial within the VVD and could not be ignored in times of crisis, although the regional chairs have not been able to prevent the introduction of the system of ‘one person, one vote’. By adopting the principle of direct democracy in the early 2000s, in organisational respect the VVD has become more similar to D66. Evolution of party leadership The VVD and D66, as most other Dutch parties, do not formally recognize a party leader. In daily politics the party’s first candidate at the national elections, who usually becomes the chair of its parliamentary caucus or its most important representative in a government coalition, is considered to be the party leader. All 12 party leaders of the VVD since 1948 had parliamentary and/or ministerial experience before they were appointed. In at least four cases, as far as we know, the incoming party leader (Toxopeus, Wiegel, Nijpels, and Dijkstal) was proposed by the outgoing one, after which he – never a she – was elected by the MPs and/or the congress delegates (Table 4.4a). In 2002, former party leader Bolkestein was involved in the replacement of his

90  Gerrit Voerman successor Dijkstal by Zalm (Schulte and Soetenhorst 2007). In three other cases, the party in public office was also decisive: when the parliamentary caucus had elected its new leader, the party congress would later follow suit. See Table 4.4a at http://partyfamilies.eu/index.php/category/chapters/ chapter-4/ The leadership selection procedure generally added to the continuity of the political line of the VVD. It also demonstrated the hegemony of the party in public office, as well as the docility of the delegates at the party congress. Only after electoral defeat and/or during internal conflict, the party in central office and on the ground were able to counter the dominance of the party in public office. Of the 12 VVD leaders, five were forced to step down (Oud, Nijpels, Voorhoeve, Dijkstal and – more informally – De Korte), partly under pressure of the party chair and/or regional party leaders. In 2006, the party leader was elected for the first time by all members: as the unofficial candidate of the party establishment, Rutte, became party leader by only a short majority. As in the VVD, in D66 the nomination process is dominated by the party in public office. All eight party leaders (van Mierlo serving twice) had been MPs or Minister before – except van Mierlo in his first term, who was a journalist. At least three times the outgoing party leader successfully recommended his successor (Terlouw, Borst, De Graaf; see Table 4.4b). See Table 4.4b at http://partyfamilies.eu/index.php/category/chapters/ chapter-4/ In three other cases, the party leader was appointed by the parliamentary caucus, after which in one case the election of the party leader by the membership congress followed suit (in the two other cases Engwirda and Dittrich became party leader after parliamentary elections but had already stepped down before the next elections). Pechtold was the first party leader to be elected by all members in 2006. Of the nine party leaders (now counting van Mierlo twice), four were more or less forced to resign (van Mierlo in his first term; Engwirda, Dittrich, and Van der Laan after she lost the primaries). D66 had two female party leaders, Borst and Van der Laan, who both only remained in office for about half a year. Party membership Traditional liberal reservations with regard to organisation reflect in the relatively low membership figures. In the 1950s and 1960s membership of the VVD nearly doubled (see Table 4.5).7 In the 1970s, due to depillarisation and fierce opposition of Wiegel against the incumbent centre-left cabinet, membership sharply increased in only six years from 41,000 to 100,000. After a short setback in 1982, the VVD had nearly 103,000 members, an all-time

Liberalism in the Netherlands 91 record. Subsequently membership declined almost without interruption. In 2016, the VVD had fewer members than in the 1950s, while paradoxically it was the largest party of the Netherlands in terms of votes. Correspondingly, the members/voters ratio was 0.3% in 2012, an all-time low. At its peak in 1977 and 1982, the VVD enrolled 1% of its voters. See Table 4.5 at http://partyfamilies.eu/index.php/category/chapters/ chapter-4/ Whereas the membership evolution of the VVD is quite straightforward – an upwards movement until around 1980 and a downfall process until the present day – the membership of D66 has fluctuated more. After its formation, membership increased to reach about 6,000 in the early 1970s, but then dramatically dropped to 300. Terlouw was able to save D66 from its existential crisis, not only electorally but also in terms of members: in 1981 nearly 18,000 members were registered. In the second crisis of the party, membership figures stabilized around 8,000; it took some years in the second period of party leader van Mierlo for it to rise to 15,000 in 1994, the most successful year in the history of D66 in electoral terms. When the party lost again the support of most of its voters, membership also dropped, but not as sharply as its electoral appeal. In its third crisis, the lowest level of membership was higher than in the previous crisis: D66 was able to maintain more than 10,000 members. A few years after Pechtold had become party leader and D66 was electorally doing better, membership started to rise again to reach more than 25,000 in 2016, an all-time record. The members/voters ratio of D66 has always been extremely low, fluctuating between 0.05% and nearly 0.2%. In the early 2000s, when the issue of a merger was seriously raised for the first time, the divergent membership figures of the VVD and D66 crystallised as one of the barriers to a possible merger. In 2005, the VVD had about 42,000 members, for only 13,000 for D66, which in case of a merger would mean that the latter would be taken over by the former. Since then, however, both parties have moved closer to each other: in 2016 the difference was only of about 3,000 members. The membership of both parties does not differ much in sociodemographic respects. In 2008, a quarter of the members of the VVD and a third of D66 were female.8 Members of both parties have a higher professional or university education (D66: 83%; VVD: 76%) and no religious belief (D66: 83%; VVD: 76%). The average age of the members of D66, however, is much higher than of the VVD (62 compared with 51).9 This might explain the relatively high share of pensioners in the D66 membership (44%, compared with 22% of the VVD members). VVD members are more often employed than members of D66 (47% versus 33%), and also more frequently self-employed (23% versus 12%) (Den Ridder 2014: 59–61). It should be noted that these data date from 2008, and that since then membership of the VVD declined, whereas the membership of D66 increased.

92  Gerrit Voerman

Conclusion In 2016, the chances of a merger between the two Liberal parties were quite modest despite some ideological convergences, cooperation at the European level, the abandonment of the strategic preference for Labour by D66, their experience of joint government participation (in 1994–2002 and 2003–2006), a number of common political attitudes of voters, and the democratisation of the VVD’s organisation. The main obstacles to a merger may be the programmatic differences between voters of the two parties on the left–right scale, but also on income distribution, minority problems, and European integration (van Holsteyn, Lucardie, and Voerman 2004). In addition, the relative strength of the VVD and D66 are a great barrier to a merger. Although both parties have become more or less equal in terms of members, electorally the VVD is much larger: 26.6% in 2012, against 8.0% for D66. Since then, according to the opinion polls, the VVD is losing votes and D66 winning. These diverging trends do not facilitate a possible merger: the more a party expects electoral gains, the less it will be interested in collaboration with others, let alone a merger. Moreover, a merger between the two Liberal parties could put off faithful voters on both sides and thus lead to losses. The most left-wing D66 voters would perhaps switch to GreenLeft or Labour, while a number of conservative VVD voters might prefer to vote for the CDA or a more right-wing party like the Freedom Party. A merger would mean that the new party would inevitably be closer to the political centre than the VVD at present. This would create more space for the Freedom Party on the right wing of the political spectrum. Even without a merger, both Liberal parties are influential in Dutch politics, and their prospects look good (Lucardie 2007). Taken together, their share of the parliamentary seats has risen to more than a third in 1994, 1998, and 2012 – more than doubling their total in 1967. This increase reflects the depillarisation and growing individualisation of society: people have become more emancipated and prepared to stand up for their own interest. As long as this process of individualisation continues, liberalism will benefit from it. Moreover, liberalism flourishes in affluent societies with relatively few social and cultural tensions. Class conflict has decreased in the Netherlands since the late 20th century. However, a cultural conflict between immigrants and native Dutch people has since then become more manifest. This might pose a threat for the VVD that for the first time in its history is faced with a serious competitor to its right. At the same time, it provides opportunities for D66, as Pechtold has demonstrated. Nonetheless, radical right populism and its critique of the open multicultural society and of globalisation, free trade, and welfare state reform, has become a serious opponent of liberalism, more than Christian Democracy or Social Democracy.

Liberalism in the Netherlands 93

Notes 1. Two of the 47 founding members did not participate in the survey. 2. The author would like to thank Paul Lucardie for his useful comments. This contribution is partly based on Lucardie and Voerman (2006). 3. The author would like to thank Simon Otjes for providing these data, based on the Dutch Parliamentary Electoral Studies. 4. This comparison is based on the election programmes of both parties from the 1990s onwards. 5. Data provided by Simon Otjes (see note 3). 6. In D66 this organ was called the ‘advisory council’ (Adviesraad). 7. Membership figures of the VVD in this period are not fully reliable, as central registration of members was introduced only in 1972. 8. The last time a more comprehensive survey among the members of the Dutch parties was held, was in 2008. 9. It is very likely that since the influx of new members into D66 from 2010 onwards, the average age of the members has dropped. In 2016, the average age of the D66-members from whose age was known, was 44 years. Communication M. Bordewijk, D66-national office, 25 October 2016. In 2015, the average age of the members of the VVD was 55 years (Commissie Toekomst van de VVD-structuur 2015: 38).

References Commissie Toekomst van de VVD-structuur (2015), De VVD in de 21e eeuw: van bolwerk naar netwerk. Den Haag. D66 (2000), De uitgangspunten van Democraten 66, Den Haag. D66 (2016), Samen sterker – kansen voor iedereen. D66 Verkiezingsprogramma 2017–2021. Daalder H., Koole R. (1988), ‘Liberal parties in the Netherlands’. In: Kirchner E.J. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 151–177. Den Ridder J. (2014), Schakels of obstakels? Nederlandse politieke partijen en de eensgezindheid, verdeeldheid en representativiteit van partijleden (proefschrift). Leiden: Leiden Universiteit. Godschalk J.J. (1970), ‘Enige politieke en sociale kenmerken van de oprichters van D’66’, Acta Politica 5(1), pp. 62–74. Hoedeman, J. (1993), Hans Wiegel en het spel om de macht. Utrecht: Coutinho. Koole R. (1992), De opkomst van de moderne kaderpartij. Veranderende partijorganisatie in Nederland 1960–1990. Utrecht: Het Spectrum. Koole A. (1994), ‘The vulnerability of the modern cadre party in the Netherlands’. In: Katz R.S., Mair P. (eds.), How parties organize. London: Sage, pp. 278–303. Koole R., Lucardie P., Voerman G. (1988), 40 jaar vrijheid en verenigd. Geschiedenis van de VVD-partijorganisatie. Houten: De Haan. Lucardie P. (2007), Nederland stromenland. Een geschiedenis van de politieke stromingen. Assen: Van Gorcum. Lucardie P., Voerman G. (2006), ‘Eccentric yet powerful: The position of the liberals in the Dutch party system’. In: Van Schie P., Voerman G. (eds.), The dividing line between success and failure. A comparison of liberalism in the Netherlands and Germany in the 19th and 20th centuries. Berlin: Lit Verlag, pp. 121–141.

94  Gerrit Voerman Pedersen M.N. (1982), ‘Towards a new typology of party life spans and minor parties’, Scandinavian Political Studies 5(1), pp. 1–16. Schulte A., Soetenhorst B. (2007), Daadkracht & Duidelijkheid. 5 jaar crisis in de VVD. Amsterdam: Uitgeverij Van Gennep. Sitalsing S. (2016), Mark. Portret van een premier. Amsterdam: Prometheus. Staatscommissie J. Donner (1956), De Nederlandse kiezer. Een onderzoek naar zijn gedragingen en opvattingen. Den Haag: Staatsdrukkerij- en Uitgeverijbedrijf. Te Velde H. (2008), ‘De partij van Oud en Wiegel. Leiderschap in de VVD en het primaat van het electoraat’. In: Van Schie P., Voerman G. (eds.), Zestig jaar VVD. Amsterdam: Boom, pp. 27–51. Van der Kolk H. (2000), ‘Het afnemend belang van godsdienst en sociale klasse’. In: Thomassen J., Aarts K., Van der Kolk H. (eds.), Politieke veranderingen in Nederland 1971–1998. Den Haag: SDU, pp. 121–138. Van der Land M. (2003), Tussen ideaal en illusie. De geschiedenis van D66, 1966–2003. Den Haag: Sdu Uitgevers. Van der Land M. (2012), Langs de afgrond. Tien turbulente jaren in de geschiedenis van D66. Delft: Eburon. Van Holsteyn J., Lucardie P., Voerman G. (2004), ‘VVD’66’, Openbaar Bestuur 14(10), pp. 28–31. Vermeulen A. (2013), De liberale opmars. 65 jaar VVD in de Tweede Kamer. Amsterdam: Boom. Voerman G. (2001), ‘De geleide interne democratie van D66’, Democraat 34(3). Voerman G., Lucardie P. (1992), ‘Ideologie en individualisering; De grondslagendiscussie bij CDA, PvdA en VVD’, Beleid & Maatschappij, 19(1), pp. 31–41. Voerman G., Lucardie P. (2001), ‘Liberalisme met een rode rand?’. In: Veldhuizen J. (ed.), D66: een blijvend appèl. 35 jaar werken aan vernieuwing. Den Haag: D66, pp. 108–111. Voerman G. (2014), ‘Kandidaatstelling op landelijk niveau’. In: De Lange, S.L., Leyenaar M., De Jong P. (eds.), Politieke partijen: overbodig of nodig? Den Haag: Raad voor het Openbaar Bestuur, pp. 45–62. Vollaard H., Voerman G. (2015), ‘De Europese opstelling van Nederlandse politieke partijen’. In: Vollaard H., Van der Harst J., Voerman G. (eds.), Van Aanvallen! Naar verdedigen? The Hague, the Netherlands: Boom Bestuurskunde, pp. 99–182. VVD (1948), Beginselprogram van de Volkspartij voor Vrijheid en Democratie. Den Haag. VVD (1966), Beginselprogram van de Volkspartij voor Vrijheid en Democratie. Den Haag. VVD (1982), Liberaal manifest 1981. Den Haag. VVD (2005), Om de vrijheid. Liberaal manifest. Den Haag. VVD (2008), Beginselverklaring. Vastgesteld door de 125e algemene vergadering op 15 november 2008 te Rotterdam. Den Haag. VVD (2016a), Zeker Nederland. VVD verkiezingsprogramma 2017–2021. Den Haag. VVD (2016b), Statuten en Huishoudelijk Reglement. Den Haag.

5

Belgian Liberals Living apart together… Vivien Sierens

Introduction From its inception, the Belgian political system has been deeply marked by the pervasiveness of liberal ideas. The first political movement to organise as a party in continental Europe (Seiler 1993), the Belgian liberals have endured several internal mutations and ideological realignments through their history. From Belgian ‘unionism’ to anticlericalism, from laissez-faire to social liberalism, Belgian liberals have constantly reinvented themselves. Yet, compared with other party families, detailed historiographies of the Belgian Liberal family have been scarce (Noiret 1994). While initially the history of Belgian liberals largely coincided with the history of the Belgian Liberal Party, it now coincides with the vicissitudes of two separate parties evolving in different electoral environments, the Flemish Open Vlaamse Liberalen en Democraten (Open Flemish Liberals and Democrats – OpenVLD) and the French-speaking Mouvement Réformateur (Reformist Mouvement – MR). Dragged into the Belgian federalisation vortex in the 1970s, the unitary party imploded in 1971. Since then, each party has followed separate paths, in terms of electoral performances, party manifestos, and internal organisation. Yet, they have accessed power together at the federal government, remained faithful to a common ideological core, and worked together at the European level. As in some modern families, where partners decide to continue their relationship while living in separate houses, Belgian liberals have decided to continue their relationship by living apart together. The evolution of the Flemish- and French-speaking liberals across the linguistic border has usually been covered by in-depth case studies (Billiet 2002; Delwit 2002). Comparative analyses of Belgian liberals’ evolution since the party split have been scant and covered only a limited time span. Rudd (1988) is a notable exception. He brilliantly covered the history of the Belgian Liberal family until the early 1980s, and paved the way for a comparative analysis of each party after the party split. This chapter aims to follow this path by complementing previous analyses with more recent data. The main objective is to highlight the main commonalities and divergences

96  Vivien Sierens between Flemish and French-speaking Liberals after the party split. The next sections examine the origin and development of both parties, their ideological evolution, and organisational restructuring. The combination of these three dimensions should contribute to a better understanding of the evolution of the Belgian Liberal parties in the past four decades.

Origins and development There are several ways to conceptualise the origins of political parties and to classify them. This section proposes to combine Lipset and Rokkan’s genetic approach of political parties and Pedersen’s lifespan approach to discern respectively the origin and development of Belgian Liberal parties. On the one hand, according to Lipset and Rokkan (1967), the political cleavage structure in which parties emerge weighs heavily on their subsequent ideological stances and organisational culture. On the other hand, according to Pedersen (1982), as mortal organisations, all parties pass through different thresholds (declaration, authorisation, representation, and relevance) during their lifecycles. The crossing of these thresholds represents qualitative breaking points in parties’ organisational developments. Although Lipset and Rokkan’s analytical framework allows them to focus on the stickiness of core organisational elements, Pedersen’s framework allows him to emphasise the dynamic of organisational change. In combining these two frameworks, the objective is to sketch a dialectic account of the commonalities and divergences in the Belgian Liberal parties. Evolution of Liberal parties in the party system and main intra-party events According to Lipset and Rokkan, political parties generally emerged in Western Europe in a revolutionary context (either a national or an industrial one) and alongside two conflictive axes (a territorial and a functional one). Chronologically, the territorial cleavages are assumed to precede the functional ones (Lipset and Rokkan 1967: 13). However, in the case of Belgium, the functional cleavage stemming from the national revolution (the church/state cleavage) preceded the appearance of the territorial cleavage (centre/periphery). As a result, while the unitary Belgian Liberal Party appeared in 1846 as a prototypical anticlerical party, its successor parties stemming from the 1972 party split, the French-speaking Parti de la Liberté et du Progrès (Party for Freedom and Progress – PLP) and Dutch-speaking Partij voor Vrijheid en Vooruitgang (Party for Freedom and Progress – PVV) emerged as a reaction to the increasing saliency of the territorial cleavage. As one of the first political parties to appear in continental Europe, the Belgian Liberal Party achieved very early and almost simultaneously the thresholds of declaration and authorisation. Its creation, in 1846, marks

Belgian Liberals 97 the end of Belgian ‘unionism’, a special alliance between Liberals and Catholics that had dominated Belgian political life since the national revolution and creation of the state in 1830. Closely linked to freemasonry, the Liberal Party found its origins in secular demands for an expansion of public education (Roussellier 1991), constitutional rights (Steed 1988), and abolition of the privileges of the aristocracy and clergy (Bartier 1968). Organised as a network of local Liberal champions (e.g., Eugène Defacqz, Charles Rogier, Pierre-Théodore Verhaegen, or Henri de Brouckère), it established its electoral strongholds in an industrial and urban bourgeoisie (Vandermotten and Médina Lockhardt 2002). Its internal structure, characterised by a low membership and a few charismatic personalities, is the archetype of Duverger’s cadre party (Duverger 1951). In a bipartisan majoritarian system (before the introduction of a proportional electoral system in 1899), this flexible organisation allowed the Liberal Party to become a central player. As early as 1847, the party achieved the thresholds of representation and relevance simultaneously. It won the national election by gaining 52% of the votes (Delwit 2012). From 1847 until 1884, the Liberal Party even managed to govern alone for an aggregated period of 25 years (interrupted by short opposition terms). After this glorious period, the growing salience of the owner/worker cleavage and the emergence of the Socialist Party by the end of the 19th century (Parti des Ouvriers Belges – POB) soon relegated the Liberal Party to the role of third political formation, between the Socialist and the Catholic parties. From 1899 until 1960, the Liberal Party played a pivotal role as kingmaker in governmental coalition negotiations. It took part in 27 of a total of 41 governments, allying either with the Catholics against the POB, or with the POB against the Catholics. Until the 1960s, the unitary Liberal Party barely altered its internal structures and preserved the same name. In 1960, after the pacification of the church/state cleavage that culminated in the School Pact (a political agreement concerning the funding of public and Catholic schools), the Liberal Party operated a rebranding by changing its name into Parti de la Liberté et du Progrès - Partij voor Vrijheid en Vooruitgang (PLP-PVV) (CRISP 1961). At the same time, the party decided to merge with small right-wing parties (Wynants 2011). This rebranding corresponded to the abandonment of the party’s traditional anticlerical stance and to an opening towards the Christian voters. The rebranding of the party was followed by good electoral results in 1965, but very quickly the rise of the centre/periphery cleavage altered these good electoral performances. In 1970, the Belgian political system entered a centrifugal federalisation process by creating three linguistic communities (the Flemish-, French-, and German-speaking communities) (Swenden et al. 2006). As most other national parties, challenged by regionalist parties, in 1972, the Liberal Party split up into two regional subnational parties, PVV and PLP, alongside the linguistic divide between Flemish and French speakers.

98  Vivien Sierens From 1971 to 2014, the development of the two regional successor parties of the unitary Liberal Party followed diverging paths of electoral victory and defeat. Table 5.1 provides a schematic summary of the main developments of the OpenVLD and MR. Three main observations may be formulated to summarise the diverging evolution of the Flemish- and French-speaking Liberals. First, the internal party organisation of the French-speaking Liberal Party followed a much bumpier ride than its Flemish counterpart. Second, in the two parties, internal party dynamics were highly dependent on the succession of leadership figures. Third, both parties opened up at different time towards different groups by forming electoral cartels. See Table 5.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-5/ The structuration of subnational Liberal parties was a complex phenomenon that partly reflected the diverging federalisation aspirations between Flanders and Wallonia. While the Flemish movement had aimed for long at the creation of a federal state based on linguistic communities, the minority Walloon regionalists had strived for the creation of a federal state based on territorial communities. After the party split alongside the linguistic divide, Flemish Liberals, who had previously been actively requesting such a division, smoothly started to coordinate their positions in Flanders and Brussels (note that both Flemish- and French-speaking communities coexist in Brussels). On the contrary, French-speaking Liberals, who had resisted the party split by defending a unitary stance, were much more disorganised in their way to structure French-speaking liberalism across regions. While in Wallonia, the regional branch of the Liberal Party quickly took over the flexible structures of the old unitary party, in Brussels, the party exploded into different local factions. In 1976, the division between Walloon and Brussels Liberals became even more patent when the Walloon branch of the PLP merged with the Rassemblement Wallon (Walloon Rally – RW), a left-oriented regionalist party aiming at the creation of three separate economic regions (Dohet et al. 2014); while the Brussels Liberals were torn apart between three electoral lists: Parti de la Liberté et du Progrès Delforge (Party for Freedom and Progress Delforge – PLP Delforge), the Parti Libéral Indépendant BelgeLiberaaal Onafhankelijk Belgische Partij (Independent Belgian Liberal Party – LIB-LOB), and the Parti de la Liberté et du Progrès de la région de Bruxelles (Party for Freedom and Progress from the Brussels Region – PLP Brussels) (D’Hoore 1997). Actually, it took the French-speaking Liberals almost a decade to build a common party across Brussels and Wallonia, which came into being in 1979. Interestingly, one of the main driver behind the alliance between Walloon and Brussels Liberals at the end of the 1970s was the increasing competition from a right-wing regionalist party, the Front Démocratique des Francophones (Democratic Front of Francophones – FDF)1 that was conquering a great deal of Brussels’ Liberals’ electoral market share.

Belgian Liberals 99 Besides this diverging structuration of liberalism across the linguistic border, both Liberal parties were strongly influenced by their leadership figures. Whereas French-speaking Liberals experienced a series of strong leadership figures (Gol, Michel, Reynders) who regularly attempted to reconfigure party structures and ideology, Flemish Liberals were mainly influenced by one prominent leadership figure (Verhofstadt) who opened up the party structures and rebranded the party as Vlaams Liberaal en Democraat (Flemish Liberal and Democrat – VLD) in 1992. By opening the party, Verhofstadt also attracted defectors from different right-wing parties, among which prominent members of the Flemish regionalist party Volksunie (People’s Union – VU) (e.g., Jaak Gabrieels and André Geens) (Billiet 2002: 201). Finally, both Liberal parties chose the strategy to form electoral alliances with smaller parties and to finally merge with them. Both parties sought to broaden their electoral market share by integrating regionalist parties, but while French-speaking Liberals got closer to Christian voters, Flemish Liberals opened up to conservative and neoliberal voters. Indeed, after the creation of the Parti Réformateur Libéral (Liberal Reformist Party – PRL) in 1979, French-speaking Liberals decided to form an electoral alliance with the Brussels-based FDF in 1993. Six years later, they decided to enlarge this electoral cartel to dissidents from the Christian-democratic party who had formed a small party, the Mouvement des Citoyens pour le Changement (Citizens Movement for Change – MCC). In 2002, this electoral cartel opened up to the German-speaking Partei für Freiheit und Fortschritt (Party for Freedom and Progress – PFF) and transformed into a loosely structured political party, the Mouvement Réformateur (Reformist Movement – MR). The different partners of this movement-party had different weights. While the MCC and PFF were quickly absorbed by the PRL, the regionalist FDF remained more autonomous. This greater autonomy led to its defection from MR in 2011. For their part, Flemish Liberals decided to form an electoral cartel with VIVANT (the acronym standing for Voor Individuele Vrijheid en Arbeid in een Nieuwe Toekomst – for individual freedom and labour in a new future), a small neoliberal party created by a Flemish entrepreneur that had obtained good electoral performances in different Flemish municipalities (Dohet et al. 2014). In 2007, they opened up this electoral cartel to the conservative Liberaal Appèl (Liberal Appeal – LA) and transformed into a rebranded political party, the OpenVLD. Electoral performances Since the party split in 1971, electoral performances of Flemish- and French-speaking Liberals at federal elections have followed different cycles of successes and defeats. These unparalleled electoral cycles are attributable to differences in the structure of the electoral competition across the

100  Vivien Sierens linguistic border. While Flemish Liberals have had to assert themselves in contrast to a strong Christian-democratic party and to a regionalist party, the Nieuw-Vlaamse Alliantie (New Flemish Alliance – NV-A), Frenchspeaking Liberals had to assert themselves mainly in contrast to a strong social-democratic party, the Parti Socialiste (Socialist Party – PS). Electoral fragmentation has also been much higher in Flanders than in Wallonia (De Winter et al. 2006). Figure 5.1.a displays the electoral performances of the French-speaking parties, and Figure 5.1.b displays the electoral performances of Flemishspeaking political parties, both at the federal level. Liberal parties are represented by the dashed bold lines in both graphs. As shown by these graphs, the two Liberal parties have evolved in different competitive environment. See Figure 5.1a at http://partyfamilies.eu/index.php/category/chapters/ chapter-5/ See Figure 5.1b at http://partyfamilies.eu/index.php/category/chapters/ chapter-5/ From the 1970s onwards, the French-speaking Liberal Party has progressively increased its vote share and presence on the political scene. From 1978 until 1985, the party increased its vote share by 4.3 percentage points, from 5.9% to 10.2%. Illustrating the diverging structuration degree between the Walloon and Brussels fractions of PRL, the party’s electoral anchorage diverged between each region. In 1981, the PRL gained 21.7% of the Walloon electorate but only 15.8% in Brussels (D’Hoore 1997), where the party faced fierce competition from the FDF. These good electoral performances were confirmed in the 1985 elections. In 1987, the party faced a severe electoral defeat and stayed in opposition until 1995. In 1995, after its electoral alliance with the FDF in Brussels, it regained strength and enlarged its cartel to the MCC in 1999. This coincided with the proposition from the Flemish Liberals to form a government coalition with the Socialists and the Greens. In 2003 and 2007, the party received its best electoral outcomes with 11.4% and 12.5%, respectively, of the total vote share. This vote share dropped sharply in 2010 to 9.2% and stabilised at 9.6% in the 2014 elections. As far as the OpenVLD is concerned, after some initial turbulences characterised by a succession of rapid electoral successes (1974 and 1981) and defeats (1977 and 1985), the party began a period of constant electoral progression from 1985 until 2003. It rose from 10.7% of the vote share at the national level in 1985 to 15.4% in 2003. These good electoral performances resulted from the modernisation efforts launched under the leadership of Guy Verhofstadt (Govaert 1995). In 1999, the party took the lead in the formation of a governmental coalition, with Verhofstadt becoming Prime Minister from 1999 until 2008. Simultaneous to these good electoral results, the party also faced the growing competition of regionalist and radical right parties in

Belgian Liberals 101 Flanders. From 2007 until 2010, the VLD’s national vote share fell abruptly from almost 16% to 8.6%, and then stabilised at about 9% in the 2014 elections. Table 5.2 shows the evolution of Liberal parties’ seat share in the federal parliament (Lower Chamber). From 1974 to 2003, Flemish-speaking Liberals held a higher vote share than French-speaking Liberals (with the exception of 1985 when the Parti des réformes et de la liberté de Wallonie (Party of Reforms and Freedom of Wallonia – PRLW) scored a big electoral success while the VLD faced a modest electoral defeat). From 2007 to 2014, the trend started to reverse and in the 2014-2019 Parliament, Frenchspeaking Liberals held more seats than their Flemish counterparts. On an aggregated level, seat share for the Liberal family in the lower chamber has oscillated around 20%. It passed from 14.6% in 1971 to 22% in 2014, with a peak at about 30% in the early 2000s. See Table 5.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-5/ Participation to power at various levels A characteristic feature of Belgian Liberals is their recurrent governmental participation and their symmetric access to power. From 1973 to 2014, Liberals have been present in 63% of the governments, that is, 17 out of the 27 governments that formed during that period. This regular governmental participation is explained by their key pivotal role to coalition formation. Interestingly, there is no great difference concerning the participation pattern of Liberals in government before and after the national party split. Of course, the 1970s and 1990s were characterised by a greater governmental instability than previous decades, but this instability affected all parties in a similar way. After the party system split (1972), the participation of Liberals in federal governments occurred through large coalitions involving (in average) at least five political parties. Despite this large coalition aspect, governments including Liberal parties were not significantly more instable than the others. Since 1970, the average length of government without the Liberals has been of 16.9 months, while the average length of government including the Liberals has been of 16.4 months. Besides coalition composition and duration of Liberal governments, one might ask if there are some specialisation patterns regarding the types of ministerial portfolios attributed to the Liberals. Portfolio distribution is the result of closed-door negotiations between the different parties willing to form a governmental coalition. The distribution therefore corresponds to parties’ policy preferences and expertise but also to their bargaining leverage in the governmental agreement. Table 5.3 lists the governments in which the Liberals took part since the party split, and provides details regarding their portfolios. See Table 5.3 at http://partyfamilies.eu/index.php/category/chapters/ chapter-5/

102  Vivien Sierens Concerning the portfolios distribution, one can observe a specialisation trend across time. Since 1971, Belgian Liberals have generally occupied ministries dedicated to Economic Affairs, Foreign Affairs, and ‘law and order’2. In contrast to what Rudd (1988) observed, Belgian Liberals are not ‘always considered [as] the junior member in whichever coalition they have participated’ (Rudd 1988: 200). Thanks to their good electoral performances, they even managed to attain the Prime Minister portfolio four times in the first decades of the 2000s (three times in a row with OpenVLD and one time with MR). Behind these cumulative portfolio specialisation patterns also lies a personal dimension. In several instances, it’s only one individual that occupied the same portfolio during several legislatures. For example, when taking into consideration the different cabinet reshufflings across time, Willy De Clercq (PVV) occupied the position of Finance Minister six years from 1971 to 1977; Didier Reynders (PRL then MR) occupied this same position ten years from 1999 to 2009; Herman Vanderpoorten (PVV) occupied the position of Justice Minister five years from 1973 to 1977, and in 1980; and Sabine Laruelle (MR) occupied the position of Agriculture Minister six times from 2003 to 2014. Besides this personalised dimension of portfolio attribution, one could also observe some specialisation patterns between Flemish- and French-speaking Liberals. Whereas in the past decade, Flemish Liberals have tended to focus on law and order ministries such as Justice, Migration, and Asylum; Frenchspeaking Liberals have tended to focus on Agriculture, Middle Classes, and Development Cooperation. As Rudd (1988) noted, the absence of Liberals from certain departments is as revealing as their specialisation on some issues. Although, they have not remained completely absent from social departments, as Flemish Liberals repeated choice for the Health Department or French-speaking choice for Retirement Department may testify, they have never been present in Labour Ministries or Social Affairs Ministries.

Ideology and policy positions Initially considered as paragons of anticlericalism and champions of industrial capitalism (Bartier 1968), Belgian Liberals have attempted to redefine their ideological identity since the 1960s. Rudd (1988) considered them as ‘economic radicals and social conservative’. The evolution of liberal ideology in the past four decades brings some nuances to this categorisation. Although they have remained ‘economic radicals’ attached to liberal ideals of increased economic freedom, they have taken some progressive stances on a range of societal issues (such as euthanasia, abortion, gay marriage, or environmental issues) which distances them from ‘social conservatism’ and created some internal tensions. To assess ideological evolution within Liberal parties since the 1970s, this section relies on a content analysis of party manifestos (Budge 1987). Manifestos data offer a fertile ground, and content analysis allows dealing with questions of issue saliency and ownership.

Belgian Liberals 103 Before comparing the content of Flemish- and French-speaking party manifestos, one can first notice the diverging evolution of their length. The length of French-speaking Liberals’ manifestos has increased exponentially in the last 40 years. Whereas the PLP’s manifesto was 17 pages long in 1974, the party manifesto of its successor party, MR, was 565 pages long in 2014. Note that the 2014 program was a joint program for regional, European, and federal elections, which might explain this exceptional length. Nonetheless, as exemplified by Dandoy and Pilet (2016), this inflationary trend was particularly remarkable for French-speaking Liberals in the 1990s. This trend is a common feature among the Belgian French-speaking parties, but contrasts with the Flemish-speaking Liberals who have tended to present short and concise manifestos. As a matter of comparison with their Frenchspeaking counterpart, in 2014 the OpenVLD’s manifesto was 57 pages long. In their longitudinal study on the ideological evolution of Frenchspeaking Liberals, Dandoy and Pilet (2016) identified five major policy issues as recurrent in their party manifestos: free market policies, administrative and institutional reform, education, foreign policy (with a special Europhile emphasis), and law and order. Bouteca and Devos (2014) identified similar issues as recurrent for the Flemish Liberals, with a little bit more emphasis on entrepreneurship and pension rights reform. For a first glance at the evolution of liberal policy positions on these different issues across time, data from the Comparative Manifesto Project are useful. As evidenced in Table 5.4, the saliency of each of these five major policy issues within party manifestos has varied across time but in a rather close range of values. The positioning of each party on these issues is also quite similar. Thus, ideologically, not only have both parties been rather similar in the issues they have emphasised, but also in the way they have positioned on them. Both parties have advocated for more free market reforms, especially in the 1970s and 1980s. They have placed special emphasis in their programs on tax reforms, public finance discipline, and more efficient administration. Both parties have Table 5.4  CMP Scores on five dimensions: MR versus OpenVLD, 1974–2010 1974 1977 1978 1981 1985 1987 1991 1995 1999 2003 2007 2010 Free market economy Government admin. efficiency Education expansion EU (+)

MR VLD MR VLD

MR VLD MR VLD Law MR & order (+) VLD

6.8 11.2 6.7 9.2 6.0 5.6 7.5 10.5 11.6 9.4 0.0 2.6 3.6 8.8 3.8 0.9 1.7 1.8 6.1 3.4

2.8 3.8 6.9 0.0

0.0 1.7 1.0 0.0 1.1 3.5 2.1 3.8 1.1 0.0 4.9 2.7 8.6 12.1 11.3 17.1 8.8 16.3 9.6 9.6 9.0 9.0 6.5 12.2

0.0 2.3 1.9 0.9 0.0 1.9

6.9 6.2 1.3 2.3 4.6 0.8

3.6 8.1 8.5 7.2 1.4 1.7 1.6 1.0 1.0 5.2 5.9 11.1

0.0 3.5 4.3 3.3 0.0 0.4

3.6 3.3 4.1 4.7 1.0 0.7

1.3 3.5 0.0 0.0 1.8 1.5

2.9 2.8 1.6 2.7 4.0 4.2

6.5 8.0 17.9 8.9 6.7 12.6

Source: Volkens, Lehmann, Matthieß, Merz, Regel, and Werner (2015)

13.8 0.0 1.2 3.0 4.2 22.0

3.9 0.5 1.0 0.1 4.9 1.5 2.6 1.1 4.0 6.4 6.5 10.4

104  Vivien Sierens also taken rather positive stances towards the EU, although this topic has not been very salient in their respective manifestos with the exception of 1999 (when federal elections coincided with European ones). Both Liberal parties have also regularly emphasised the role of efficient education systems in their federal manifestos. This is remarkable as education is not a federal competence but a community competence (Dandoy and Pilet 2016). Both parties generally linked education issues with topics such as social inclusion of migrants or active labour market policies. The only notable difference between the two parties on these five dimensions appears on the law and order dimension. While Frenchspeaking Liberals have generally supported stricter law and order reforms since the 1980s, this issue was initially much less salient for their Flemish counterpart. However, since the 1990s, with the outburst of child abuses judicial scandals, Flemish Liberals have put much more emphasis on the law and order issue. Finally, Figure 5.2 displays Belgian Liberals’ positioning within the broader political landscape, using expert data collected from 1999 to 2014 in the frame 10.0

GAL-TAN Index

7.5

5.0

2.5

0.0

2.5

5.0

7.5

Left-Right on Economic Issues

Figure 5.2  P  ositioning of Belgian parties in a two-dimensional political space (1999–2014) Note: Each point indicates the position of Belgian political parties on different years (federal manifesto) (1999–2014). Economic dimension (x-axis): all categories related to economic policy, economic ideas, redistribution, and economic groups (0: market regulation to 10: market freedom). Noneconomic dimension (y-axis): GAL-TAN scale (0: green/alternative/libertarian to 10: traditionalism/authority/nationalism). Source: Chapell Hill Expert Survey

Belgian Liberals 105 of the Chapel Hill Expert Survey. The graph differentiates party positioning on two dimensions, an economic dimension (on a left/right scale) and a new politics dimension (ranging from Green/alternative/libertarian (GAL) to traditional/ authoritarian/nationalist (TAN)). The libertarian pole of the GAL-TAN scale typically depicts parties in favour of expanded individual freedoms on ethical issues such as abortion, euthanasia, and gay marriage whereas the traditional pole refer to parties opposing this expansion of individual freedoms and valuing a socially conservative order (Bakker et al. 2015). According to expert surveys, in the past 25 years, Belgian Liberals have consistently remained oriented towards a free market and on the right of the economic agenda, but they have not fallen into conservatism. Interestingly, both Flemish- and French-speaking Liberals seem to occupy a niche in the Belgian political landscape, being the only parties represented in this quadrant of the graph. Since the 1990s, Frenchspeaking Liberals have regularly advanced (at least rhetorically) the idea of social liberalism. Behind this expression, they have advocated, among other things, for better market regulation, fiscal harmonisation, or financial transactions taxes (Dedecker 2011). In a similar vein, the initial Thatcherite ideology of OpenVLD slowly evolved towards a Blairist third way in the 1990s. This mixture of neoliberal ideas and ethic libertarianism is consistent with previous electoral studies on Belgian Liberal voters (Billiet 2002). Nonetheless, beyond the formal ideological stances expressed by both parties within their respective manifestos, they have also been increasingly internally divided on a number of ethical issues. This is particularly the case for MR, which was built as an agglomeration of different rightist parties with different philosophical and religious backgrounds. As a result, MR has deliberately maintained ambiguity over its positioning on ethical issues (Close 2016). On these issues, the party statuses explicitly authorise parliamentarians to vote according to their own religious and philosophical beliefs (Close 2016). Party voters are also divided between conservatives and progressives on these ethical issues. These divergent parliamentary members’ views on societal issues have sometimes fomented internal frictions between MR’s different components. VLD is less concerned by internal divisions on ethical issues but their conservative-progressive divisions have fluctuated across time and according to issues (Wauters et al. 2013).

Structure and organisation Main party structures The evolution of Belgian Liberals’ internal structures is particularly marked by their flexibility. Unlike other partisan organisations characterised by rigid structures, Liberals have demonstrated their ability to pragmatically adapt their organisational structures. This characteristic finds its origin in the historical unitary cadre party. Besides this lability of formal internal structures, informal social practices were also an important characteristic of liberal

106  Vivien Sierens organisational culture. Despite the centrality of national Congresses within formal structures, in practice, large deliberative assemblies have often been marginalised in the party management (Sierens and Van Haute 2016). Since the creation of the Liberal Party, the internal organisation of Belgian Liberals has oscillated between centralisation and disintegration. Their development may be pictured into five sequences: the aggregation of local liberal movements in the 19th century, the creation of coordination structures in the first half of the 20th century, the centralisation process in the 1960s, the split-up and slow reconstruction in the 1970s–1980s, and the federalisation of local and regional movement in the 1990s–2000s (Sierens and Van Haute 2016). In 1972, both Flemish- and French-speaking Liberals inherited from the structures of the unitary Liberal Party. From 1979 until 1987, Frenchspeaking Liberals adopted a simplified structure organised around three central bodies: the Congress (deliberative assembly of the party), the Standing Committee (decision-making body) and the Bureau (in charge of the daily management). From 1979 until 1992, Flemish Liberals adopted a similar structure (Congress, Party Council, and Political Bureau). In both cases, satellite liberal organisations (health insurances, trade unions, youth associations, etc.) were included in formal structures. To adapt to the federalisation process, party statuses were amended regularly. Notable divergences between Flemishand French-speaking Liberals started to show at the beginning of the 1990s. In 1990, French-speaking Liberals reformed their internal structures in an attempt to reinforce the role of regional and local groups in their formal structures. This incremental addition of ad hoc structures stiffened up the internal decision-making process. Besides, at the same time, Frenchspeaking Liberals merged with small rightist parties such as FDF and MCC. Almost at the same time, Flemish Liberals took an opposite step. They simplified radically their organisational structure by focussing on two main bodies: the Congress and the party presidency. They also decided to break formal ties with satellite organisations. Finally, they opened up the party structures to nonmembers by introducing the category of registered voters. In the last decade, French-speaking Liberals did not fundamentally alter their internal formal structure, despite the transformation of their electoral cartel as a party. Working as an asymmetric federation of different parties, the MR did not change the previous structures of its integrating parts. For their part, Flemish Liberals reintroduced more centralisation within the internal party management. Party leadership A common characteristic of Belgian Liberals is the varying importance of party leadership according to their governmental participation. Generally, when party leaders took possession of a ministerial portfolio in the federal government, they delegated party leadership to a lower prominent personality (Dedecker 2011). This meant that in some instances, the personality recognised

Belgian Liberals 107 as the real party leader may have differed from the formal one. Differences have arisen between the two parties with respect to the competitiveness of internal party leadership contests. While both parties have democratised their leadership selection procedures by opening it up to all their members, they have had different tolerances towards open internal confrontation. Although French-speaking Liberals opened up their leadership selection procedure in 1987, they have consistently attempted to limit the competitiveness of internal contests. As exemplified in Table 5.5, of the 14 internal leadership contests organised since 1971, only three included more than one candidate. Among these three cases, two left the party deeply divided and the last one introduced an inefficient leadership. In 1973, before the introduction of members’ universal suffrage as a selection procedure, the leadership contest pitting André Damseaux against Jean Gol torn up the party apart between its two political components, the PLP and RW. In 1990, given the intensity of internal competition, contenders finally decided to run on a single ticket. They teamed up to form a dual leadership. Experience was disappointing and punctuated by internal conflicts. Finally, in 2011, the opposition between Charles Michel and Didier Reynders exposed publicly deep factional divisions within the party. See Table 5.5 at http://partyfamilies.eu/index.php/category/chapters/ chapter-5/ The Flemish-speaking Liberals, on the contrary, have been much more open to competitiveness in leadership contests. Pioneers in the introduction and implementation of closed party primaries (or membership ballot) for their leadership selection in 1991, they have been consistent with their ideological plea for internal party democracy. The number of candidates, their electoral results, and voting dispersion testify to the degree of competition in their internal selection processes. In 1997, up to 15 candidates were contending for party leadership. This competitive setting, in sharp contrast with French-speaking parties, has allowed the party to neutralise the power of internal middle range elites (Wauters 2014) and to open the contest to political newcomers. For example, in 2009, Alexander de Croo (son of Herman de Croo, VLD leader in 1995) managed to win the leadership contest in a tight two-round elections. In contrast with the French-speaking Liberals, due to the high number of contenders, internal elections have not divided the party between two antagonising factions. Another distinctive aspect of the Flemish-speaking Liberal Party leadership has been its greater openness towards female leadership, with two women leading the party in 1985 and 2012. Party membership Differences in organisational culture may also explain the contrasted evolution of membership trends of the two parties the 1970s. In terms of absolute membership figures, whereas MR presents a longitudinal

108  Vivien Sierens declining trend over the period, VLD shows a growing membership trend (Table 5.6). As absolute figures do not take into account longitudinal growth of the electorate, researchers often use the ratio of party membership to the electorate (M/E) for comparative purposes. When M/E ratios are considered, the same trends as the ones observed with absolute figures are confirmed. See Table 5.6 at http://partyfamilies.eu/index.php/category/chapters/ chapter-5/ A closer look at these membership figures points to significant membership drop and increase at the end of the 1980s. After the electoral defeat of 1987, French-speaking Liberals lost about 30 000 members. After that, they never managed to reverse the declining trend and continued losing members until 2012. Flemish Liberals, on the contrary, have seen an upsurge of their membership figures since the early 1990s. Interestingly, this increasing membership pattern contrasts with the general trend in Europe and coincides with the reform of the party’s internal structures. It coincides more particularly with the introduction of competitive leadership contests and the recognition of a new category of sympathisers within the party structures, the ‘registered voters’. Nonetheless, after the glorious early 2000s, party membership started to decline once again, maybe because of the increasing rivalry from other parties such as the Flemish ethno-regionalist NV-A. In both Liberal parties, becoming a member is a relatively easy procedure. Both require a low membership fee compared with other Belgian parties. But rights granted to the members differ. Whereas French-speaking Liberals do not involve their members in their candidate selection procedures or in their manifesto approval (Paulis and Van Haute 2016), Flemish Liberals do involve them extensively in these procedures. Relations to social movements The Belgian political system has long been characterised by a pillarised structure, that is, a political system where each party is linked to a series of satellite organisations such as health insurance and trade unions. Nonetheless, Belgian Liberals have never been a mass-membership movement. Compared with other traditional Belgian party families (the Socialists and Christian-democrats essentially), the Liberals have always had a weak pillar structure (Deschouwer 2012). This shallow dependence on satellite organisations also explains why the OpenVLD was able to break the ties with these organisations in the 1990s. Electoral surveys of liberal voters and surveys of party members have confirmed the success of the depillarisation processes undertaken in both Liberal parties (van Haute et al. 2013).

Belgian Liberals 109

Conclusion Classifying political parties within certain party families is generally not without ambiguity. By combining the genetic party approach and the ideological approach as suggested by Mair and Mudde (1998), this chapter has attempted to solve the classification puzzle for Belgian Liberals. It has been shown that both Belgian Liberal parties may safely be classified as members of the Liberal political family. First, as far as the origins are concerned, Belgian Liberals are a typical case of parties that have emerged on the church/state divide. As their Luxemburgish and Dutch counterparts, they have taken their roots in a democratic anticlerical tradition. Nonetheless, since the 1960s, they have abandoned their anticlerical stances and have managed to position themselves as champions of the free market reforms on the owner/worker cleavage. In contrast with their Scandinavian counterparts, the Belgian Liberals have never positioned themselves on the urban/rural cleavage although they have usually had their electoral strongholds in urban areas. After the party split in 1972, both Liberal regional parties have managed to position themselves on the centre/periphery cleavage by attracting or merging with regionalist movements. In the future, one challenge that will affect both parties will be their positioning on regionalist issues. Both are indeed challenged by regionalist parties (OpenVLD by NV-A; and MR by FDF, although only in Brussels). Should they both stand on a ‘unitary’ position to distinguish themselves from these challengers; try to steal the regionalist issue or adopt diverging positions? Second, ideologically, both parties have consistently promoted individual economic freedom and limitation of state intervention. Although both parties tended to adopt conservative stances on societal and cultural values from the 1960s until late 1980s (Rudd 1988), they have started to adopt more progressive stances on these issues since the 1990s. Nonetheless, the adoption of these more progressive stances has not occurred without ambivalences and internal frictions. These ambivalences may convert into one of their main challenge for the future. So far both parties have occupied a niche market. They have successfully provided voters with a pragmatic policy mix that has combined economic laissez-faire with some progressive stances on societal values. Adopting a more conservative positioning would have different consequences for each party. Given the electoral market in Flanders, this would probably not be a good option for the OpenVLD. Indeed, Flemish Liberals have to face the competition of two major conservative parties (the Christen Democratisch en Vlaams – Christian Democrat and Flemish (CD&V), and NV-A) that have embraced a liberal economic agenda (especially the NV-A) but have remained rather traditional on societal values. However, adopting a more conservative positioning might well be an option for the MR, which has no real challenger. In some way, there is a niche for a more conservative right-wing party on the Francophone side. Nonetheless, if the MR ‘moves’

110  Vivien Sierens from its current policy positioning, there is a high risk of internal divisions if not fission. Besides, if it moves, it will move away from its Flemish counterparts, which might affect their ability to govern together (which until now was facilitated by their ideological closeness). Third, organisationally, both parties have been characterised by their great lability across time. They have adopted flexible and pragmatic approaches and have attempted to open their structures and make them more efficient. Since 1972, both Liberal parties have lowered party membership fees. They have enhanced members’ participation in a series of decisions such as leadership selection. Nonetheless, Flemish Liberals have generally been more consistent with their plea for internal party democracy than their French-speaking counterpart. Once again, both parties seem to face different challenges at the organisational level. Although MR has never been a mass-party, its membership level has declined rapidly in the past decade. How can the party remain attractive for members, and react to this downwards trend? Following its Flemish counterpart, the party may be tempted to change its internal structures and attract new members by enhancing and enlarging party members’ rights. Such reforms risk exacerbating internal frictions between the different existing factions, and be opposed by these factions or middle-level elites who would lose their power and weight in the organisation if such reform was to be implemented. For the OpenVLD, the organisational reforms seem to have been a success so far. Nonetheless, the success has lasted for a relatively short period of time (essentially during the 1990 decade). Since the beginning of 2000s, party membership has also started to decline. On the one hand, if the OpenVLD does not react, it may soon be facing a membership decline similar to the one affecting its French-speaking counterpart. On the other hand, if the party further opens its structures, it risks complicating its internal decision making and ‘diluting’ the organisation.

Notes 1. The party was renamed ‘DéFI’ (Démocrate, Fédéraliste, Indépendant – Democratic, Federalist, Independent) in November 2015. The word défi in French means ‘challenge’. 2. If one would compare the portfolio allocation before and after the party split, one could notice a change in the specialisation pattern and in the cleavage structure. In the postwar period, in the context of heated debates around the financing of public and Catholic school, the unitary Liberal Party had specialised in portfolios dealing with Education, Defence, and Economic Affairs. After the pacification of the church/state cleavage and the party split, Liberal parties have occupied predominantly the Finance portfolios, Justice ministry, and External Trade.

References Bakker R., De Vries C., Edwards E., et al. (2015), ‘Measuring party positions in Europe: The Chapel Hill expert survey trend file, 1999–2010’, Party Politics 21(1), pp. 143–152.

Belgian Liberals 111 Bartier J. (1968), ‘Partis politiques et classes sociales en Belgique’, Res Publica 10(2), pp. 33–106. Billiet J. (2002), ‘Les transformations du libéralisme en Flandre. Les Vlaams Liberalen en Democraten (VLD)’. In: Delwit P. (ed.), Libéralismes et partis libéraux en Europe. Bruxelles: Éditions de l’Université de Bruxelles, pp. 199–225. Bouteca N., Devos C. (2014), ‘Een breuklijn op de terugweg?: Sociaaleconomische kwesties in de partijprogramma’s van liberalen en socialisten tussen 1961 en 2010’, Journal of Belgian History 44(1), pp. 118–156. Budge I. (1987), ‘The internal analysis of election programmes’. In: Budge I., Robertson D., Hearl D. (eds.), Ideology, strategy and party change: Spatial analysis of postwar election programmes in 19 democracies. Cambridge, UK: Cambridge University Press, pp. 15–38. Close C. (2016), ‘La sociologie électorale des libéraux francophones’. In: Delwit P. (ed.), Le Mouvement réformateur. Histoire et actualité du libéralisme en Belgique Francophone. Bruxelles: Éditions de l’Université libre de Bruxelles. CRISP. (1961), ‘La réforme du parti libéral: naissance du parti de la liberté et du progrès PLP’, Courrier Hebdomadaire du CRISP 1961/35 (no. 125), pp. 1–16. D’Hoore M. (1997), ‘Du PLP-PVV au PRL-FDF I. 1961–1992’, Courrier Hebdomadaire du CRISP 1997/9 (no. 1554–1555), pp. 1–50. Dandoy R., Pilet J-B. (2016), ‘L’évolution des programmes électoraux du Mouvement Réformateur’. In: Delwit P. (ed.), Le Mouvement Réformateur. Histoire et actualité du libéralisme en Belgique Francophone. Bruxelles: Éditions de l’Université de Bruxelles. De Winter L., Swyngedouw M., Dumont P. (2006), ‘Party system(s) and electoral behaviour in Belgium: From stability to balkanisation’, West European Politics 29(5), pp. 933–956. Dedecker N. (2011), ‘Le mouvement réformateur’. In: Delwit P., van Haute E., Pilet J.B. (eds.), Les partis politiques en Belgique. Bruxelles: Éditions de l’Université de Bruxelles, pp. 145–62. Delwit P. (2002), ‘Du parti libéral à la fédération PRL-FDF-MCC’. In: Delwit P. (ed.), Libéralismes et partis libéraux en Europe. Bruxelles: Éditions de l’Université de Bruxelles, pp. 179–197. Delwit P. (2012), La vie politique en Belgique de 1830 à nos jours. Bruxelles: Éditions de l’Université de Bruxelles. Deschouwer K. (2012), The politics of Belgium: Governing a divided society. London: Palgrave Macmillan. Dohet J., Faniel J., Govaert S. et al. (2014), ‘Les partis sans représentation parlementaire fédérale’, Courrier Hebdomadaire du CRISP 2014/1(no. 2206–2207), pp. 5–110. Duverger M. (1951), Les partis politiques. Paris: A. Colin. Govaert S. (1995), ‘Du PLP-PVV au VLD. II. 1992–1995’, Courrier Hebdomadaire du CRISP 1995/38 (no. 1503–1504), pp. 1–43. Lipset S.M., Rokkan S. (1967), Party systems and voter alignments: Cross-national perspectives. New York: Free Press. Mair P., Mudde C. (1998), ‘The party family and its study’, Annual Review of Political Science 1, pp. 211–229. Noiret S. (1994), ‘Political parties and the political system in Belgium before federalism, 1830–1980’, European History Quarterly 24, pp. 85–122. Paulis E., van Haute E. (2016), ‘Composition et profil sociologique du parti’. In: Delwit P. (ed.), Le Mouvement Réformateur. Histoire et actualité du libéralisme en Belgique Francophone. Bruxelles: Éditions de l’Université Libre de Bruxelles.

112  Vivien Sierens Pedersen M.N. (1982), ‘Towards a new typology of party lifespans and minor parties’, Scandinavian Political Studies 5(1), pp. 1–16. Pilet J.-B., Cross W.P. (eds.) (2015), The selection of political party leaders in contemporary parliamentary democracies. A comparative study. London: Routledge. Roussellier N. (1991), L’Europe des libéraux. Bruxelles: Éditions Complexe. Rudd C. (1988), ‘The Belgian Liberal parties: Economic radicals and social conservatives’. In: Kirchner E.J. (ed.), Liberal parties in Western Europe. Cambridge, UK: Cambridge University Press, pp. 178–211. Seiler D.L. (1993), Les partis politiques. Paris: A. Colin. Sierens V., van Haute E. (2016), ‘Les structures du Mouvement Réformateur’. In: Delwit P. (ed.), Le Mouvement Reformateur. Histoire et actualité du libéralisme en Belgique Francophone. Bruxelles: Éditions de l’Université Libre de Bruxelles. Steed H. (1988), ‘Identifying Liberal parties’. In: Kirchner E. (ed.), Liberal parties in Western Europe. Cambridge, UK: Cambridge University Press, pp. 396–435. Swenden W., Brans M., De Winter L. (2006), ‘The politics of Belgium: Institutions and policy under bipolar and centrifugal federalism’, West European Politics 29(5), pp. 863–873. van Haute E., Amjahad A., Borriello A. et al. (2013), ‘Party members in a pillarised partitocracy. An empirical overview of party membership figures and profiles in Belgium’, Acta Politica 48(1), pp. 68–91. van Haute E. et al. (2015), ‘MAPP - party membership figures in 32 countries. 1945–2014’. MAPP Project Data Archive. www.projectmapp.eu. Vandermotten C., Médina Lockhardt P. (2002), ‘La géographie du libéralisme Belge et Européen’. In: Delwit P. (ed.), Libéralismes et partis libéraux en Europe. Bruxelles: Éditions de l’Université Libre de Bruxelles, pp. 57–75. Volkens A., Lehmann P., Matthieß T. et al. (2015), The Manifesto Data Collection. Manifesto Project (MRG/CMP/MARPOR). Version 2015a. Berlin: Wissenschaftszentrum Berlin für Sozialforschung (WZB). Wauters B. (2014), ‘Democratising party leadership selection in Belgium: Motivations and decision makers’, Political Studies 62(1), pp. 61–80. Wauters B., Bouteca N., Devos C. (2013), ‘Ideology and party members: the correspondence between election manifestos and members’ opinions in the Flemish liberal-democratic party (1999–2010)’, Politicologenetmaal, Ghent, 30–31 May. Wynants P. (2011), ‘Le libéralisme Francophone du PLP au MR. I. 1961–1999’, Courrier Hebdomadaire du CRISP 7/2011 (no. 2092–2093), pp. 5–77.

6

Diversity, unity, and beyond The Swiss liberals Oscar Mazzoleni

Introduction The family of liberal parties in Switzerland has a rich past. Its more than 180-year history has been characterised by a long series of divisions, reconciliations, and periods of competition. In the past decades, the FreisinnigDemokratische Partei/Parti radical/Partito radicale (Free Democratic Party or Radical Democratic Party – FDP-PLR), the heir of parties formed in the second half of the 19th century, became the foremost Swiss party with a liberal program. After its merge with the Liberale Partei der Schweiz (Liberal Party – PLS), the second largest party of the traditional liberal family, in 2009, the FDP.Die Liberalen (FDP-PLR) became the only Swiss party affiliated to the ALDE group. The FDP also co-founded in 1947 the Liberal International Association, in which it now holds the status of observer. Above all, the FDP (then FPD-PLR) has for long been one of the most influential parties in Swiss politics, and one that has heavily counted in federal government since the 19th century. Thus, the Swiss liberal parties experienced the most permanent national government participation among liberals in Europe and this participation is also determinant for understanding the so-called Swiss model based on power-sharing or consensus (e.g., Lijphart 1999). Long considered as a ‘hegemonic’ Swiss party, and later as ‘predominant’ (Gruner 1977; Labrot 1999), from the 1920s onwards, the development of the FDP was however marked by a slow but continuous decline in electoral support, and by increasing levels of competition in the Swiss political system. Even so, in many ways, the institutional role of the FDP-PLR is still pivotal. This chapter highlights the most important milestones in the development of the liberal family, and especially that of its leading party, also labelled as Radicals, which maintains a relevant participation in government, at national and regional level. It points out changes and continuity in its ideology and policies, examines the organisational model that has prevailed, and discusses the extent to which this model has been able to cope with recent changes in the world of Swiss politics. The chapter concentrates in the first part on the evolution of the Swiss liberal parties, and in particular

114  Oscar Mazzoleni on that of the FDP-PLR, before looking specifically at the FDP-PLR’s ideological stance and organisation. Finally, a concluding section tries to understand the present and future challenges that the party is facing.

From the 19th to the 21st century: from Staatspartei to competition In terms of sociological origins and ideology, the FDP-PLR undoubtedly belongs to the family of European liberal parties (Mair and Mudde 1998). On the other hand, Switzerland is also one of the few countries in Europe in which the history of the liberal party family coincides with that of its state and government. The liberal parties founded in the 19th century maintained their institutional importance and a continuous presence in the party system and in federal government, as part of a democratic system untouched by interruptions or authoritarian takeovers throughout the 20th century. This is particularly obvious when looking at the strongest party of this family, the FDP-PLR, which perfectly embodies the model of the ‘government party’ as described by Angelo Panebianco (1988), or that of the ‘coalition party’ as encapsulated by Von Beyme (1985). Erich Gruner, the leading historian of Swiss political parties, used the expression ‘Staatspartei’ in the 1970s, pointing out how no liberal party in Europe, including the French Radicals or British Liberals, had played a role comparable with that of the Swiss Liberals in terms of nation-building and state-building (Gruner 1977: 73). In the meantime, however, the traditional family of liberal parties suffered a gradual electoral decline. Even though it is still a pivotal party in the government system, the FDP-PLR has struggled to cope with the increasing political competition of recent decades. The origins of a government party To understand the crucial role of the Swiss liberal parties, we need to go back to the 19th century. Their origins reflect that of a whole range of cultural and economic associations and movements that arose in the 1830s. They were the main protagonists of the Swiss Constitution, which was introduced in 1848 and remains largely preserved until now, and which was based on liberal-democratic principles. As the winner of the short but decisive Civil War of 1847, the heterogeneous coalition of subnational parties belonging to the Liberal family (from left-wing to right-wing components) held the reins of political power from the first national elections held in 1848. At this time, Liberal parties, working through electoral committees based on the power of local and cantonal notables, enjoyed almost total dominance in the federal parliament and government. Switzerland is thus an example of how the four thresholds of political party development, as described by Pedersen (1982) – declaration, authorisation, representation, and relevance – all coincided (see Table 6.1). At least until the early 1880s, the dominance of parties

Diversity, unity, and beyond 115 of the liberal family remained virtually unchallenged. Those who lost the brief civil war and were left out of the new constitutional order that was set up in 1848 – which is to say the Catholics – long remained on the fringes of the political system, for they were excluded from the government and had only a few deputies in parliament. Moreover, it was only in 1890 that Socialists began to enter the federal parliament. See Table 6.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-6/ Even though they ruled together for several decades, the liberal parties had to cope with rifts, diversions, and scissions that reflected the profound changes that took place between the late- 19th and the early 21st centuries. Divisions increased between three wings: the ‘Radicals’ (die Freisinnige), which is the moderate centre-leftist component; the left-wing ‘Democratic’ component, which was attracted to the rising Socialist movement; and the right-wing market-oriented component that later formed in some cantons the Liberal Party (PLS). Towards the end of the 19th century and under the pressure of the rising competition in federal elections, these increasing divisions led to the formation of distinct national party organisations. The development of an open competition for federal elections came with the new Constitution of 1874 (Caramani 2003), when male suffrage was extended to all, leading to the formation of the first parliamentary groups. This development was also assisted by the consolidation of the referendum system, which until recently was the only way by which the parties competed nationally. The new Constitution introduced the right to call for facultative referendums, which meant a referendum could be held for every law enacted by parliament, and this was repeatedly used by Conservative Catholics against the Liberal government. The year 1878 saw the creation of the first parliamentary group of the Radicals. This was followed in 1882 by the creation of a group of Conservative Catholics and, in 1883, of a group of ‘Liberals’, thus introducing, for the first time, a formal separation within the liberal family at the national level. The ‘socialist threat’ grew in the 1870s and 1880s, with new and disruptive forms of social mobilisation, including a spread of strikes, which heralded the founding of the Swiss Socialist Party in 1888. This founding was mainly promoted by supporters of the left-wing component of the liberal family. The Socialist Party was the first national party in Swiss history. In this context, after many doubts and uncertainties, the Radicals founded their own national party, the FDP, in 1894. That same year, the Conservatives also formed their own national organisation, the Catholic People’s Party (Altermatt 1991: 173; Meuwly 2007: 64). Unlike several other party systems in Western Europe, which were shaped by the development of so-called ‘mass parties’ (as described by Duverger (1963), among others), the Swiss party system took a different path, partly as a result of the victorious liberalism of the mid-19th century. As stated by

116  Oscar Mazzoleni Martin Schefter (1994: 6), ‘elites who occupy positions of authority within a regime, like outsiders who are seeking to gain entry into that regime, will construct a strong, broadly based party organisation only if it is necessary for them to do so in order to gain, retain, or exercise power’. In the case of Switzerland, the way forward was for the Liberal elites to open up their power to their former opponents (Burgos et al. 2011). This strategy occurred gradually, as a result of two main challenges: the consequences of the development of referendum rights, and the emergence of a ‘capital-worker’ cleavage. Despite the three components of the liberal family (Radicals, Liberals, Democrats) still counted in 1880s more than 65% of votes in elections of the lower chamber, the FDP chose to adopt a general strategy of co-opting their (closer) opponents, that is the Catholic Conservatives who detained only around 28% (see Table 6.2). See Table 6.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-6/ In 1891, when a new constitutional reform introduced the right to launch popular initiatives for partial change of the federal constitution, for the first time, the main component of the Liberal Family, the Radicals (FDP), granted one seat out of seven in the federal government to Catholic Conservatives (Altermatt 1993: 48, 254–255). This rapprochement was brought about partly by the pressure of the referendums against the government, which was applied by the Catholics themselves, and partly by fears of the advance of socialism (Meuwly 2007). This strategy adopted by the Liberals can also be explained by their political culture moulded by historical experience. In 1848, the liberal family had won the civil war against the Conservatives, but the construction of a unitary state implied a strategy of national legitimisation in which cantons, also where the Conservatives were strongest, were granted precise guarantees in terms of institutional autonomy. Adopting a power-sharing strategy, the liberal family, in particular the Radicals, managed to keep a hegemonic role over the Swiss political system at least until the end of the 19th century. From hegemony to power sharing In the 20th century, the hegemony had to make way for a more modest predominance of the FDP (Labrot 1999). Over the decades, it played a role that in many ways continued to be pivotal, but less and less assured in terms of electoral performances (see Tables 6.2 and 6.3). Until recently, this path can be divided into three phases. The first phase came with the end of the absolute dominance of parties belonging to the liberal family in the federal parliament, and especially of the FDP, in the 1910s and 1920s. The decline of the power of the liberal family is also reflected in its internal divisions, as illustrated by the foundation

Diversity, unity, and beyond 117 of the Democratic Party and the Swiss Liberal Party respectively in 1905 and 1913. However, the main introduction of the proportional system constitutes the main turning point. The Radicals defended the first-past-the-post system, which was based on constituencies that corresponded to the cantonal borders. Since 1848, the two chambers of the federal parliament had been elected using this system, which greatly reduced the room for manoeuvre by the opposition. In 1918, the various minority parties, and in particular the Conservative Catholics and the Socialists, managed to amend the federal Constitution by means of a popular initiative. As a result, in 1919, a proportional system was used for the first time to elect the lower house of the federal parliament. This vote marked the end of the absolute majority of the FDP in parliament, whose number of seats was reduced from 106 to 60 (and less than 29% of the votes, see Table 6.3.), revealing also the increasing inability of the liberal family to attract industrial and white-collar workers, albeit some regional variations (Gruner 1977: 248–252; Seiler 1988: 372). Among the supporters of this electoral reform also figured representatives of the newly founded Farmers’ Party. In the early decades of the 20th century, the agrarian wing of the FDP had broken away from the party. Unsatisfied with the level of support that the FDP gave to the rural world, the agrarian wing founded a party of farmers and traders in some cantons. As a response to the increasing pressure of socialism and communism, a representative of the agrarian party was co-opted within federal government in 1929. Many years later, in 1971, this and other parties formed the Swiss People’s Party (SVP), thus bringing together proponents of a strong liberal conservatism (Skenderovic 2013). See Table 6.3 at http://partyfamilies.eu/index.php/category/chapters/ chapter-6/ The second phase lasted from the 1930s to the 1980s. During this period, a process of national political integration took shape in Switzerland under the name of neutrality and independence. During World War II, the Socialist Party was brought into government and a model of broad-based coalition governments gradually established, in which pragmatism became the ideological basis for the work of the parties in the executive body. Under the Cold War and the great postwar economic boom, Switzerland enjoyed a period of exceptional political stability. This was symbolised by the so-called ‘magic formula’ in the late 1950s, which consisted in the emergence and consolidation, for a period of three decades, of an entente between the main parties – the FDP, the Catholic Party (CVP), and the Socialist Party (PS). Thus, in 1959, for the first time, the FDP became a party that counted in government as much as the Catholics and Socialists. Although it was able to insure its uninterrupted participation in government, the FDP was to be – together with the CVP) – the party that was worst affected by the effects of de-politicisation and increasing electoral

118  Oscar Mazzoleni abstention. In the end, this led to a voter turnout in the national elections of the 1970s and 1980s of about 40%. Between the 1940s and 1950s, the FDP also entered a long period in which it lost ground in the elections. In this period, traditional parties of the liberal family faced also the birth of a new opponent party, the Alliance of Independents, a business-oriented party linked to the interests of the leading cooperative of Swiss consumers, which was able to appeal to some Liberal dissenters by the 1980s (Seiler 1988). The third phase started with the end of the Cold War and the collapse of communism, at the end of the 20th century. During this period, the FDP faced its greatest challenges that corresponded to an accelerated electoral decline, bringing the party down to just 20% of the votes in the 1990s elections for the lower house. While the old conservative Liberal Party became increasingly marginal, the FDP had to deal not so much with its old rivals (the Alliance of Independents lost its importance in the 1980s and ultimately disappeared), as with the emergence of new anti-establishment parties. These formed what we might call the new generation of liberal-oriented parties. In particular, they included the Freedom Party, an antitax party set up in 1994 in the wake of a pro-automobile, anti-environmentalist mobilisation, which, in some cantons and especially those with an urban and Protestant tradition, constituted quite a serious challenge for the Radicals. On the opposite side of the political spectrum, the FDP had to deal with another ‘new politics’ party, the Green Liberal Party. Throughout the 1990s and 2000s, the traditional liberal parties were also increasingly under pressure from the SVP, which originally arose as a splinter group from the FDP. In the late 1980s and early 1990s, the old farmers’ party entered into a period of profound change. As well as extending into urban areas, it radicalised towards nationalist and populist positions, but also towards neoliberal economic stances (Mazzoleni 2008). The main response to these challenges (ever-increasing competition, fewer votes, and the loss of several seats in the lower house), was for the FDP to merge with its former competitor and increasingly closer ally, the Swiss Liberal Party. This party was still active in the cantons of Geneva, Vaud, and Neuchâtel, and in the city of Basel. The merger led to a change of name, from Freisinnig-Demokratische Partei (FDP) to FDP-Die Liberalen (FDPPLR). Although the merger failed to stop the decline in votes in the national elections, in particular within the low middle-class voters (Lutz 2012, 2016), it did help maintaining a pivotal role for the FDP-PLR ​​in the federal institutions. The party indeed retained two seats out of seven in the election of the government and kept its position as the largest party in the upper chamber of the parliament in the 2015 elections. In addition, the party also managed to maintain leading positions in cantonal parliaments and governments from 2000 to 2010. In December 2006, the FDP held 42 seats and the Liberal Party 4 seats out of a total of 156 government seats. In April 2016, the FDP-PLR, together with the LP, had a total of 42 seats, again from a total of 156 seats. In the cantonal parliaments, the FDP-PLR ​​arrived in the second position, only surpassed by the SVP (with 533 seats out of 2,609).

Diversity, unity, and beyond 119 Considering the crucial institutional importance of the cantons in the Swiss federal system, the fact that the FDP-PLR has ​​ maintained this strength indicates its enduring pivotal role.

Ideology and policy positions The lasting nature of the key role played by the liberal family in political institutions, and particularly in Swiss government at both national and cantonal levels, is also reflected in its ideological evolution and in its policy positions. At the same time, some of the values and ​​ principles that inspired the creation of this family have remained unchanged through to the present day. The ‘original’ cleavages From an ideological point of view, the liberal family, whose spirit takes root in the Enlightenment and French Revolution, advocated greater freedom and broader political rights for all citizens, together with freedom of opinion, freedom of the press, and free trade and industry. However, the way these values and ​​ principles have been understood has changed considerably. In a rather schematic interpretation, it could be said that in the second half of the 19th century, the main currents of thought were those of the Liberals, of the Radicals, and of the Democrats (Gruner 1977: 77). The Radicals (later synonymous with the Freisinnige) represented the centre-left, defending the right to vote, the strengthening of direct democracy, equal opportunities for education, the creation of a welfare state, and a centralist conception of power. Representatives of industry and trade mainly went along with the Liberals, who supported the decentralisation of institutional powers. The Liberals – market-oriented – stressed the importance of the individual, of freedom and responsibility, and were often influenced by the Manchester school of economic thought. By contrast, the Radicals put the emphasis on equality and the creation of a strong, independent national state based on the will and rights of citizens. The Radicals were also strongly linked to anticlericalism. The Democratic component of the liberal family was composed by representatives and areas of influence that emphasised the importance of social rights and direct democracy, driven by a desire to represent social demands brought about by the industrial revolution. It is no coincidence that supporters of this democratic component were attracted to the socialist ideas that started spreading in the second half of the 19th century. Despite its internal differences, which at times erupted into conflicts between the various strands, in the first decades of the federal state, the liberal family was able to encapsulate within itself the main lines of conflict in Swiss politics. These include the tensions between centre and periphery, between the economy and the state, and between the urban and rural worlds, as well as language differences, because the various currents of the liberal family had considerable weight in all three of the main language areas of the

120  Oscar Mazzoleni country – German, French, and Italian. The religious divide naturally led to opposition with the Catholics, because Protestant currents dominated in the liberal family, but the Liberal-Radical-Democratic ideology also had a high degree of pragmatism and a desire to limit centrifugal tendencies in a country characterised by its great cultural, sectarian, and linguistic divisions. To establish its hegemony in the 19th century, the Liberal-RadicalDemocratic family positioned itself as the ‘party of the nation’, inspired by a desire to represent Switzerland as a common homeland, while respecting cantonal, religious, and linguistic diversity. Between identity and pragmatism The 1930s–1960s period to some extent favoured the Radicals’ sphere, which embraced the concept of the welfare state and state intervention in the economy, in the footsteps of the British reformer William Beveridge. However, in a system where generations of the Freisinnige had been uninterruptedly in government, few incentives for ideological confrontation persisted. During decades, Swiss politics tended towards great stability – a phenomenon that continued through to the 1980s – discouraging strong competition in federal elections. It is not surprising that the programmes of the Freisinnige tended to be rather generic, including a wide range of proposals, and were based on the idea that ideologies were now relics of the past (Meuwly 2010: 122). By the end of this period, faced with the three greatest challenges of the time – the Reaganian liberal revolution, the emergence of environmentalism, and of anti-establishment nationalism – the liberal family needed time to reshape its identity. The first challenge came from within its own political family, which is to say from a new generation of supporters of a renewed form of Manchester-school liberalism, especially in the Liberal Party (Seiler 1988). By contrast, the second and third challenges were external and led to the emergence of new challengers: the Social Democratic Party and the Green Party took up the left, while the ‘new’ SVP managed to siphon off some disappointed FDP voters (Lutz 2012). In addition, the newly created Green Liberal Party competed from a centre-right position in economics (Seitz 2013). It is striking that, just when liberal ideas were triumphant, especially after the collapse of the Soviet bloc, the two traditional Liberal parties appeared to be in decline. Compared with previous periods, the competition had never been so fierce and there was now very little room for manoeuvre. In recent years, some have pointed out how difficult it is to see any trace of the old ideals in the FDP-PLR (Meuwly 2008). When faced with the challenges cited above, the FDP-PLR responded in policy terms by gradually removing its social-egalitarian wing, which had been prominent until the 1960s, and shifted more and more towards a neoliberal economic position. Without denying the importance of the ‘social market’, the most recent FDP programme increasingly focused on individual freedom, responsibility, and

Diversity, unity, and beyond 121 entrepreneurship, thus facilitating the merger with the Liberal Party. In the economic sphere, it called for limited intervention by the state, with guidelines for the state, the market and competition. In the 1990s, the FDP, and later the FDP-PLR promoted lower taxes ‘for all’ in their national electoral programme and took a stand against the expansion of the welfare state and against excessive bureaucracy. They nevertheless wanted a top-quality education system and efficient infrastructure (FDP 1999, 2003, 2011). The FDP also reshaped its identity by revising the traditional concept of national independence. Traditional Swiss patriotism, which became well established in the 1930s among the governing political formations, had called for absolute independence in foreign policy. But the FDP and the LP, together with the federal government, introduced a new approach in the 1980s and 1990s, following the fall of the Berlin Wall and the acceleration of the process of European integration. Embracing a pragmatic approach to relations with the European Union and, in more general terms, a more participatory approach to foreign policy, the party adopted the idea that Switzerland should make an active contribution to peace in foreign countries rather than just hosting talks between belligerent powers. The party’s membership of the ALDE should be seen from this perspective (Baur 2015). The party’s policy with regard to the EU has, however, changed over the past 25 years, even though it has never entirely accepted the principle of full membership for Switzerland. In the 1990s, the proposal for greater integration was mooted, for it was considered necessary to participate to have decision-making powers (FDP-PRD 1991: 13). In recent years, however, in line with a general trend in Swiss politics, the party has adopted a more sceptical approach, nevertheless firmly supporting the bilateral agreements adopted by the Swiss voters through a referendum in 2000. Although some cultural and ethical issues (as abortion or same-sex marriage) have never been strongly relevant in the FDP agenda, regarding law and order and immigration, its policy appears to have become more restrictive in recent years, although the party does stress the need to safeguard the principles of the rule of law. There was ever-greater firmness with regard to asylum policy in the 1990s and 2000s. In the face of growing international migration, the FDP maintained a differentiated approach to these issues: on the one hand, it called for greater severity and a tightening of the conditions for refugee status, while on the other it allowed free access to the labour market for immigrants who were needed by the Swiss economy. Similarly, while the approach to law and order issues has moved away from an insistence on prevention to one that focuses on personal responsibility and punishment, the party has nevertheless continued to distance itself from the more hard-line stance adopted by the SVP. This was especially true in 2015, when there was a move to reduce the power of judges in deciding on legislation regulating the expulsion of foreign criminals from Switzerland. In any case, the party has continued to adopt a policy-seeking strategy, ready to reach compromises with other parties in the field: with the Catholics and

122  Oscar Mazzoleni Socialists, in particular, on European and international issues, and with the SVP on economic issues, as well as on immigration.

Structure and organisation: the party in public office and in central office Partly as a result of this strategy of alliances and of policymaking in parliament and in government, as well as in linking with interest groups, the FDP-PLR has managed to maintain a pivotal role in recent years, by using the opportunity structures offered by a system with no alternation (Mazzoleni 2017), and by strengthening its parliamentary discipline (Schwarz 2009). However, the FDP and other traditional parties elected under the proportional system have faced a decline, which partly reflects the increasing difficulty of party organisations to cope with what the American political scientist Leon D. Epstein referred to as ‘counter-organisational tendencies’ like the role of mass media and capital-intensive and candidate-centred campaign (Epstein 1980: 233). Whilst these tendencies were weakly developed in Switzerland until the 1980s, the rise of new opponents in the 1990s has been accompanied by new forms of campaigning (Weinmann 2009) and competitive patterns which challenged the legacy of the Freisinnige as a government party. Low electoral competition and salience of the public office As Angelo Panebianco (1988) argued, enduring government parties are shaped by some relevant aspects: limited party competition, high availability of public resources as political incentives, high permeability vis-à-vis environment and strong subnational branches. The main consequence is the enduring primacy of the party in public office. The Freisinnige fit perfectly with this configuration. Because of its enduring and largely undisputed pivotal role in federal government, as well as in most cantonal governments, the balance of power within the FDP favoured its representatives in public office. The model of ‘circumscribed’ competition (Kirchheimer 1966: 188), established with the ‘magic formula’, from the 1940s to the 1980s, helped containing the effects of social, cultural, and economic transformations on the party and government system, reducing the incentives for making major investments in election campaigns and relevance of the party in central office. Moreover, the Swiss literature often underlined the weakness of Swiss political organisations at federal level as consequence of the crucial role of referendum rights, through which committees outside the political parties can also challenge the government on singular issues (e.g., Gruner 1977). For the Freisinnige other aspects are also crucial. The dominant and durable presence in the federal government facilitated access and control of public resources – including the selection of civil servants at least until the 1970s (Urio et al. 1989). As strongly rooted in civil society, the Freisinnige

Diversity, unity, and beyond 123 are also shaped by a thick associative network and links with the business power, well beyond those who consider themselves “members” of the party, even though this title is far from clear in the party’s own statutes (Mazzoleni and Voerman 2016). Whilst the party organisation in the Swiss Liberal family have never really developed a formal grass-root membership, they developed ‘amateur’ cantonal organisations under a weak national coordination, as a reflection of the federal state in which liberals identified. The foundation of the ‘national’ organisation of the FDP in 1894 did not lead to a centralised party. In its early decades, the FDP had a very relaxed coordination, with the management of a cantonal branch rotating to act as the national party coordinator, which received its own independent secretariat from 1929. Later, a ‘light’ umbrella structure based on sub-national branches continued to characterise the party throughout the 20th century. In other terms, the main party in the Swiss liberal family is still a kind of network, a sort of a web of cantonal parties where local and cantonal representatives provide strong associative and business linkages. The sustainable federalist legacy under stress However, because of its history as a permanent government party and primacy of the party at subnational level and in public office, the FDP-PLR found it hard to adapt to a context of a growing competition since the 1990s (Mazzoleni 2009). Its organisational features do struggle to keep up with changes. Above all, the changing patterns were fuelled by a shifting strategy of the emerging ‘national-populist’ SVP, which went through a period of exceptional electoral growth in Switzerland’s history. The older agrarian party became the largest party in a few years, not only among the old liberal-inspired parties, but also among Swiss parties as a whole. In the last federal election of 2015, it obtained 29.6% of the vote for the lower chamber of the parliament, and thus almost double the share of the FDP. The electoral success of the SVP was mainly due to a strategy that involved attacking the constituencies of the centre-right parties like the FDP. This was not simply by adopting a neoliberal approach, the defence of national integrity, and an anti-establishment style, but also through significant investment in terms of grassroots mobilisation and funding, while also strengthening the national leadership and centralising the party organisation (Mazzoleni and Rossini 2017). The advantage of the SVP thus also came from the fact that its competitors, among which the FDP, remained faithful, whether willingly or otherwise, to their traditional organisational model shaped by a low-competitive environment. Swiss electoral campaigns profoundly changed, becoming more nationalised and attracted far greater media coverage nationwide. To some extent, as the case of the ads strategy in 2015 federal elections shows, the FDP attempts to adapt (Bühlmann, Zumbach, Gerber 2016). However, its party structure remains very stable. The national party has so far not been based on indirect access through the local and cantonal branches undermining its

124  Oscar Mazzoleni capacity of mobilisation nationwide. The persistent autonomy of the subnational branches also explains why – despite the drafting of a national registry of party affiliates in 2000 – the national party continues to have only rather inaccurate estimates of how many members it has. Rough estimates of the number and the evolution of party membership do indeed appear in the literature, but it should be emphasised that, at least until the 2000s, even the national party itself did not have precise data. This was partly because the ways of becoming a member (only in some cases with recognition of the status of sympathiser) varied not only from one cantonal branch to the next, but also between districts and local branches. By making a stratarchical principle as a pillar of the party’s federalist identity, the selection and appointment of candidates for general elections has remained to this day – as in other Swiss parties – the prerogative mainly of representatives of the cantonal branches and representatives of the party in public office. Similarly, with only a few changes over recent decades, policy formation has been considerably influenced by the role of the delegates of the various cantons and subsidiary associations. In terms of policy orientation, including positions in federal referendums and popular initiatives, the individual cantonal branches maintain a large degree of autonomy as in the previous decades (Seiler 1988). One of the crucial consequences of the persisting weakness of the national organisation is the role of the president of the national party who tends to be a member of federal parliament (MP), but often one in a position that is anything but decisive in the organisation of the parliamentary group. In particular, the national presidency has few privileges and is primarily viewed as a representative and mediator between the various organs of the party. Party legacies fashioned by decentralisation and low professionalisation, but also the dominant political culture that looks with suspicion at any personalist or autocratic leadership in politics, help to explain the low level of competition in the election of national party leadership.

Conclusions In sum, the history of the traditional Swiss liberal parties overlaps with the paths of the federal state built in the 19th century, and its power-sharing model of government arose in the 20th century. Despite its electoral and political decline over the past decades, the FDP-PLR, which is the party that currently represents the legacy of the Liberal family, is also the only one that nowadays officially represents Swiss liberalism at the European level. According to the criteria of origins, ideology, and transnational belonging, the FDP-PLR is undoubtedly a relevant component of the liberal family in Europe. The FDP-PLR is affiliated to the ALDE group, by contrast to other Swiss parties that emerged or transformed in the past decades, such as the Green Liberal Party or the SVP and who share some components of the liberal ideology but don’t belong to a liberal transnational association.

Diversity, unity, and beyond 125 Although the FDP-PLR still encompasses some original elements of its liberal ideology, the party has experienced some significant changes. As we have seen, the ideological history of the Swiss liberal family has its roots in the Enlightenment and in the democratic movements that arose in the early decades of the 19th century. But ever since it came into being, the liberal family had contained ideological currents that were to some extent antagonistic – with Manchester Liberals on the one side and advocates of strong state intervention in the social and economic sectors on the other. Its ideological grounds thus echoed a wide range of positions, from right to left, adopted diversely by other liberal parties in Europe. Regarding the evolution of its ideological profile, two main periods can be identified in the course of the 20th and the 21st centuries. From the 1930s to the 1960s, the FDP, that is the Radicals, clearly supported a state-intervention orientation in the economy, whereas, from the 1980s, it has strengthened its right-wing orientation in socioeconomic issues. However, in time of great new challenges and of gradual electoral decline, particularly for the elections to the lower chamber of the federal parliament, its pragmatic strategy combined with a dominant right-wing orientation in socioeconomic issues does not appear sufficient to secure its future. This adaptive liberal pragmatism – a mix of ideology, strategy and tactics – is part of the factors that contribute to explain their capacity to cope with emerging challenges. Its pragmatism moulded by institutional practices helped the main party of the Swiss liberal family to overcome internal conflict and preserve its influence over the political system. At the turn of the 19th and 20th centuries, with the development of direct democracy and the emergence of political competition, the Radicals launched a strategy to co-opt their former adversaries into the government. This strategy has continued through to the present day, with the aim to safeguard their crucial role in parliament and government. At the same time, its government legacy and the decline of its pragmatism may explain, to some extent, its difficulty to deal with new challenges. Since the 1990s, the FDP-PLR has suffered from the rise of media-oriented national campaigns and from the growth of political competition, with new challengers using liberalism in relation to ecology or successfully mixing neoliberalism and nationalism, at the expense of the traditional liberal formulas. For the FDP, much of the difficulty in adapting to changing patterns in Swiss politics is due to its organisational ‘federalist’ culture expressed through ‘light national party organisation giving considerable power to the cantonal branches, also ties with scepticism towards any predominance of collective action over individual freedom’ (Guner 1977; Meuwly 2007). The uninterrupted participation to power of the Radicals since the 19th century represents also an obstacle for electoral mobilisation. Given its imprinting as a party serving the federal government (which is the persisting meaning embodied in the term ‘Staatspartei’), the FDP-PLR tends to avoid to consider itself as ‘a party like any other’ competing in electoral market (Meuwly 2010: 123). The party legacy shared an imprinting based

126  Oscar Mazzoleni upon a mission of building a Swiss nation-state that had not existed until 1848. As well as being inspired by a liberal, enlightened ideology based on individual responsibility, as founders of the federal state, the Swiss Liberals continued to strive for legitimisation as representatives of the national interest as a whole. They thus resisted the idea of presenting themselves as divisive party in the political contest and organising consequently. Nowadays, some reasons seem legitimising its original imprinting: the FDP-PLR holds two government seats from a total of seven, and it maintains a strong relevance in the cantonal governments and parliaments, as well as in the upper house of the federal parliament. In addition, it still provides close links with the interest groups and the business community that played a key role in its creation and development in the 19th and 20th centuries. Nevertheless, as its pivotal role is certainly not something that can be taken for granted in the future, the question will be whether the FDP-PLR, which represents the traditional liberal family, have the capacity to adjust their strategy and organisational means while at the same time, managing to hold on to their heritage.

References Altermatt U. (1991), Der weg der Schweizer Katholiken ins ghetto: Die entstehungsgeschichte der nationalen volksorganisationen im Schweizer Katholizismus, 1848–1919. Freiburg, Switzerland: Universitätsverlag. Altermatt U. (1993), Conseil fédéral. Dictionnaire biographique des cent premiers conseillers fédéraux. Yens, Switzerland: Cabédita. Baur F. (2015), ‘Assurons une représentation internationale régulière au PLR!’. https:// www.plr.ch/accueil/ Bühlmann M., Zumbach D., Gerber M. (2016), ‘Campaign Strategies in the 2015 Swiss National Elections: Nationalization, Coordination, and Personalization’, Swiss Political Science Review 22(1), pp. 15–28. Burgos E., Mazzoleni O., Rayner H. (2011), La formule magique. Conflits et consensus dans l’élection du Conseil fédéral. Lausanne, Switzerland: Presses Polytechniques et Universitaires Romandes. Caramani D. (2003), ‘The end of silent elections. The birth of electoral competition. 1832–1915’, Party Politics 9(4), pp. 411–443. Duverger M. (1963), Political parties: Their organization and activity in the modern state. New York: Wiley (1st French ed. 1951). Epstein D.L. (1980), Political parties in Western democracies. London: Transaction (1st ed. 1967). FDP (1999), Wahlplattform 1999. Bern, Switzerland: Freisinnige Partei. FDP (2003), FDP. Im Einsatz für Freiheit und Verantwortung. Mehr Chancen für die Schweiz. Wahlplattform 2003 der FDP. Bern: FDP. FDP (2007), Eine Schweiz in bewegung/eine einfolgreiche Schweiz. Bern: FDP. FDP-PLR (2011), Par amour de la Suisse. Avec courage et conviction. Programme de campagne 2011. Bern, Switzerland: PLR. FDP-PRD (1991), Nos convictions. Objectifs et postulats. 1991–1995. Bern, Switzerland: FDP. Gruner E. (1977), Die parteien in der Schweiz. Bern, Switzerland: Francke.

Diversity, unity, and beyond 127 Kirchheimer O. (1966), ‘The transformation of the Western European party system’. In: LaPalombara J., Weiner M. (eds.), Political parties and political development. Princeton, NJ: Princeton University Press, pp. 177–200. Kollman K., Hicken A., Caramani D., Backer D., Lublin D. (2016), ConstituencyLevel Elections Archive. Produced and distributed by Ann Arbor, MI: Center for Political Studies, University of Michigan. Labrot L. (1999), ‘Le Parti radical démocratique suisse: Du parti dominant au parti prédominant?’ Swiss Political Science Review 5(1), pp. 81–104. Lijphart A. (1999), Patterns of democracy. Government forms and performance in thirty-six countries. New Haven, CT/London: Yale University Press. Lutz G. (2012), Eidgenössische wahlen 2011. Wahlteilnahme und wahlentscheid. Lausanne, Switzerland: Selects-FORS. Lutz G. (2016), Eidgenössische wahlen 2015. Wahlteilnahme und wahlentscheid. Lausanne, Switzerland: Selects-FORS. Mair P., Mudde C. (1998), ‘The party family and its study’, Annual Review of Political Science 1, pp. 211–229. Mair P., Müller G. W., Plasser F. (eds.) (2004), Political parties and electoral change party responses to electoral markets. London: Sage. Mazzoleni O. (2008), Nationalisme et populisme. La radicalisation de la ‘nouvelle’ UDC. Lausanne, Switzerland: Presses Polytechniques et Universitaires Romandes. Mazzoleni O. (2009), ‘Des partis gouvernementaux face au déclin électoral. Les cas du parti Radical-Démocratique et du parti Démocrate-Chrétien’. In: Mazzoleni O., Rayner H. (eds.), Les partis politiques en Suisse: traditions et renouvellements. Paris: Houdiard Editeur, pp. 410–442. Mazzoleni O. (2017), ‘Political achievements, party system changes and government participation: the case of the ‘new’ Swiss People’s Party’. In: Wolinetz S., Zaslove A. (eds.), Absorbing the blow. Populist parties and their impact on parties and party systems. Colchester, UK: ECPR Press. Mazzoleni O., Voerman G. (2016), ‘Revitalising grass-root party membership? A comparison among Swiss and Dutch parties’, Paper presented at the ECPR General Conference, Charles University, Prague, 7–10 September. Mazzoleni O., Rossini C. (2017), ‘The Swiss People’s Party. Converting and enhancing organization by a new leadership’. In: Heinisch R., Mazzoleni M. (eds.), Understanding populist organization: The European Radical Right parties. London: Palgrave-MacMillan. Meuwly O. (2007), L’unité impossible. Le parti radical-démocratique à la Belle Epoque. 1891–1914. Hauterive: Attinger. Meuwly O. (2008), La liberté cacophonique. Essai sur la crise des droites suisses. Geneva: Slatkine. Meuwly O. (2010), Les partis politiques, acteurs de l’histoire suisse. Lausanne: Presses Polytechniques et Universitaires Romandes. Panebianco A. (1998), Political parties. Organisation and power. Cambridge: Cambridge University Press. Pedersen M.N. (1982), ‘Towards a new typology of party lifespans and minor parties’, Scandinavian Political Studies 5(1), pp. 1–16. Sartori G. (1976), Parties and party systems: A framework for analysis. Cambridge: Cambridge University Press. Schefter M. (1994), Political parties and the state: The American historical experience. Princeton, NJ: Princeton University Press.

128  Oscar Mazzoleni Schwarz D. (2009), Zwischen fraktionszwang und freiem mandat. Eine untersuchung des fraktionsabweichenden stimmverhaltens im Schweizerischen nationalrat zwischen 1996 und 2005. Norderstedt, Germany: Books on Demand. Seiler D.L. (1988), ‘Liberal parties in Switzerland’. In: Kirchner J.E. (ed.). Liberal parties in Western Europe. Cambridge: Cambridge University Press. Seitz W. (2013), ‘Die Grünliberale Partei: Sind die Grünliberalen eine Rechtsabspaltung der Grünen’. In: Mazzoleni O., Meuwly O. (eds.), Die parteien in bewegung. Nachbarschaft und konflikte. Zurich: NZZ Verlag, pp. 123–156. Skenderovic D. (2013), ‘Bauern, Mittelstand, Nation. Imaginationen and metamorphosen der Schweizerischen volkspartei im 20. Jahrundert’. In: Mazzoleni O., Meuwly O. (eds.), Die parteien in bewegung. Nachbarschaft und konflikte. Zurich: NZZ Verlag, pp. 49–76. Urio P., Baumann E., Arigoni G. et al. (1989), Sociologie politique de la haute administration publique suisse. Paris: Economica. Weinmann B. (2009), Die Amerikanisierung der politischen Kommunikation in der Schweiz. Bestandesaufnahme und Experteninterviews vor dem Hintergrund der Eidgenössischen Parlamentswahlen 2007. Zurich: Rüegger Verlag. Von Beyme K. (1985), Political parties in Western democracies. New York: St. Martin’s Press.

7

Liberal parties in Austria Laurenz Ennser-Jedenastik and Anita Bodlos

Introduction Genuine liberal parties are a recent phenomenon in Austria. In the past, liberal values were incorporated by several parties and especially during the 1970s, the Freiheitliche Partei Österreichs (Freedom Party of Austria – FPÖ) embraced elements of liberalism. When Jörg Haider took over the FPÖ’s party leadership in 1986 and embarked on a more right-wing populist course, five MPs from the party’s liberal wing broke away and, in 1993, founded the Liberales Forum (Liberal Forum – LIF), the first genuine liberal party in Austria after 1945. Led by former FPÖ politician Heide Schmidt, the LIF passed the parliamentary threshold twice (in 1994 and 1995), attracting votes mainly from highly educated urbanites (Liegl 2006: 408–410). Yet the lack of a strong party organisation and competition from Die Grünen – Die Grüne Alternative (The Greens – The Green Alternative) contributed to the loss of parliamentary representation in 1999. The Liberal Forum continued to exist and made comeback attempts in 2002 and 2008, but to no avail. Yet, in 2012, a new liberal party was founded: NEOS – Das Neue Österreich (The New Austria – NEOS). NEOS successfully competed in the general election in 2013 as part of an electoral platform with the LIF and the Junge Liberale Österreich (Young Liberals Austria – JuLis). After a postelection merger with the LIF in early 2014, the party was renamed NEOS – Das Neue Österreich und Liberales Forum (The New Austria and Liberal Forum – NEOS) and also gained seats in five regional assembly elections, and in the European Parliament (Johann et al. 2017). This chapter will cover the LIF, as a historic party, and NEOS, the current liberal party of Austria. These two parties are the only genuine and relevant incarnations of political liberalism in Austrian postwar politics. Although the residual LIF merged into NEOS, we consider NEOS to be a new party. As our survey of NEOS members and activists shows, only 11% of respondents are former LIF members. Also, NEOS has a much more grassroots-driven organisation than the LIF, which originated from the political establishment.

BK-TandF-9780815372387_TEXT_CLOSE-181001-Chp07.indd 129

14/01/19 2:17 PM

130  Laurenz Ennser-Jedenastik and Anita Bodlos Both parties are or were members of transnational liberal party organisations (ELDR/ALDE and/or the Liberal International) (Liegl 2006: 408; NEOS 2016: 4). Thus, both parties meet the first criterion of Mair and Mudde’s (1998) classification for defining party families, which constitutes the selection criterion for this volume on liberal parties. This chapter proceeds as follows. The next section discusses the origins and development of genuine liberal parties in Austria and thus covers LIF and NEOS. Then, the following sections focus on the currently active liberal party, NEOS. More precisely, we describe the policy profile of NEOS and the party’s structure and organisation.

Origins and development For most of Austrian history, liberalism has been a marginal political force. During a brief period around 1870, a number of liberal politicians and administrators exerted great influence in the government of the Dual Monarchy, promoting constitutionalism, secularism, centralism, and economic freedom. Yet they were soon marginalised, and many liberal politicians aligned themselves with German nationalism (Luther 1988). During the First Republic (1919 to 1934), German liberals were therefore organised in parties such as the Großdeutsche Volkspartei (Greater German People’s Party – GDVP) and the Landbund (Rural Federation – LBd), where liberalism typically took a backseat to German nationalism. Besides, a natural constituency for liberal politics, the Jewish citizenry in Austria, was either expelled or annihilated during the Nazi era (Pelinka 1996). After 1945, liberalism was a political current that existed, to varying degrees, in the three main parties, yet none of them had it as its core ideology. The Sozialdemokratische Partei Österreichs (Social Democratic Party of Austria – SPÖ) embraced elements of political liberalism (such as the separation of church and state), and the Christian Democratic Österreichische Volkspartei (Austrian People’s Party – ÖVP) promoted some features of economic liberalism. The FPÖ was founded in 1956 by a group of former Nazis, yet later tried to shift into more liberal territory under enduring leader Friedrich Peter. The party even joined the Liberal International in 1979, yet the transition towards liberalism met strong resistance from the party’s nationalist wing and therefore never succeeded (Luther 1988). The Liberal Forum was established in February 1993 by five Freedom Party MPs (Table 7.1) who found themselves in strong disagreement with the FPÖ leadership over the direction the party had taken during the early 1990s. Under Jörg Haider’s control, the FPÖ had become a populist radical right party, with a strong anti-immigrant platform and Eurosceptic tendencies – both of which were anathema to the (few) liberals that had remained in the party. The FPÖ’s ‘Austria first’ popular initiative, conducted in early 1993 and signed by 7.4% of the electorate, constituted the catalyst for the party split (Liegl 2006).

Liberal Parties in Austria 131 The five LIF MPs were granted the status of a parliamentary group by the president of the lower chamber. As a consequence, they enjoyed not only representation on parliamentary committees, but also access to public funds for parliamentary parties, and greater media presence than they would have had as an extra-parliamentary party (Liegl 2006: 407; Gerstl 1998: 156–159). See Table 7.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-7/ The Liberal Forum held its founding congress in November 1993. This congress produced a first party program, and, by an overwhelming majority of 96%, the delegates elected Heide Schmidt as the party leader (Gerstl 1998: 245–255). Schmidt had been the FPÖ candidate in the 1992 presidential election (earning 16% of the vote), held the position of deputy party leader, and served as the third president of the Nationalrat between 1990 and 1994 (Parlament 1999). Having been the LIF’s parliamentary caucus leader since the party’s inception, Heide Schmidt became the public face of the party during the 1990s. She was unopposed in her three reelections for the party leadership in 1995 (93%), 1997 (96%), and 1999 (90%) (Ennser-Jedenastik and Müller 2014). By applying a broad definition of Sartori’s (1976: 119–125) criteria for the ‘relevance’ of a party, the LIF can be classified as relevant throughout its representation in the Nationalrat. The defection from the FPÖ resulted in a lack of majority for a potential coalition between the ÖVP and the FPÖ and the LIF was always able to influence party competition. Still, it lacked the electoral strength for forming a majority coalition with any other party (Müller 2006b: 293). Throughout these years, the policy profile of the LIF included both liberal positions on socioeconomic issues and on cultural issues (Liegl 2006: 408). Especially in the 1990s, the party emphasised issues of market liberalisation in its manifestos. Examples are the abolition of state monopolies, the privatisation of state-owned companies, and the liberalisation of opening hours for businesses (LIF 1994: 2; LIF 1995: 12, 28). The LIF also always endorsed liberal stances on cultural issues in its manifestos such as same-sex marriage, liberal family laws (LIF 2008: 21), and gender equality (LIF 1994: 4), and a rejection of right-wing extremist agitation (LIF 1995: 35). A third prominent issue for the LIF was European integration. The LIF was the only opposition party that advocated for EU membership in the 1994 referendum campaign, as both the FPÖ and the Greens campaigned against Austria’s EU-accession (Fitzmaurice 1995). In line with its parliamentary origins, the LIF adopted a relatively centralised organisational structure. While party members could participate in policy formulation through committees and working groups, the most important decisions about policy and personnel remained in the hands of the party elites. This was mirrored in the fact that the LIF never developed into a membership-based party. It counted 3,000 party members during the

132  Laurenz Ennser-Jedenastik and Anita Bodlos mid-1990s – thus a little more than 1% of its electoral base in 1994 and 1995 (Liegl 2006: 408). The top-down organisational nature of the party was also reflected in the fact that it failed to take root in subnational political arenas. Between 1993 and 1996 it entered three of the nine regional assemblies (Landtage). Yet neither in Lower Austria, nor in Vienna or Styria did it manage to be reelected. By late 2001, the LIF had lost all parliamentary representation at the European, national, and regional level. After losing all seats in the Nationalrat in 1999, Heide Schmidt resigned her leadership position in 2000 (Liegl 2006: 409–410). See Table 7.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-7/ Notwithstanding these setbacks, the party stood in the 2002 and 2008 parliamentary elections, receiving a vote share of 1% and 2.1%, respectively (Table 7.2). In 2008, the party could even persuade Heide Schmidt to mount a comeback as the LIF’s lead candidate in the election, yet the poorly resourced campaign did not gain traction in the ever more competitive Austrian party landscape. In 2006, the LIF did not run its own campaign, but its then party leader, Alexander Zach, was elected on the SPÖ’s party list as a result of an electoral pact between the two parties (Müller 2009: 515). While the LIF was represented in legislative bodies at the national and regional level, the party remained excluded from executive office. At the national level, the Grand Coalition of SPÖ and ÖVP governed until early 2000, and at the Land level the LIF failed to clear the hurdle to be represented in the Proporz governments that award executive positions to all parties according to their seat share. With all parliamentary representation gone by 2001, the 2000s witnessed the slow demise of the Liberal Forum. Neither an organisational reform of the party in 2008, nor the adoption of a new party program in 2009 could do much to halt that trend (for instance, the party received only 0.7% of the vote in the 2010 regional election in Vienna). It was only through an electoral pact and later a merger with a new party that the LIF was saved from political insignificance. This new party was initiated in 2012 by Matthias Strolz, an entrepreneur and former political advisor, and Veit Dengler, an international manager, who were disenchanted with the existing party system. Against the background of political stalemate in the SPÖ-ÖVP coalition government and a conservative turn in the ÖVP under party leader Michael Spindelegger, these politically engaged people organised public discussions (called ‘citizen forums’) and workshops on issues such as education, the welfare state, or democracy, thereby broadening its reach. During 2012, they decided to establish a new political party. The founding congress of ‘NEOS – The New Austria’ took place on 27 October 2012 in Vienna, where 250 party members elected Matthias Strolz as party leader (with 96% of the vote).1

Liberal Parties in Austria 133 In March 2013, NEOS announced an electoral pact with the LIF and the JuLis who had abandoned their erstwhile mother party. The pact also incorporated the small Online Partei Österreichs (Austrian Online Party – OPÖ). During much of 2013, NEOS activists used a strategy based on personal networks and social media to prepare for the parliamentary elections in September. Money was raised through crowd-funding efforts (larger donations from former LIF politician Hans Peter Haselsteiner arguably helped) (Strolz 2013), and candidates were selected in a three-tier process (combining at equal weight an open primary, a membership vote, and a party executive vote) (NEOS 2012b). In the run-up to the election, polls saw NEOS close to the 4% threshold (Aichholzer et al. 2014: 24). The final vote tally was 5%, thus providing the party nine out of 183 seats in the lower chamber of the Austrian parliament. It was only the second time since 1950 that an extra-parliamentary party entered the Nationalrat (the other being the Greens in 1986). The party attracted much of its support from the young and highly educated (Johann et al. 2014). In January 2014, a congress of all party members was held to finalise the merger between NEOS and LIF. Matthias Strolz was elected leader of the newly constituted NEOS, obtaining 99% of the membership vote. Later that year, the Young Liberals were incorporated into NEOS as the party’s youth organisation (now called Junge liberale NEOS – JUNOS). NEOS experienced a quick upsurge in the polls after the 2013 election, yet by early 2015 had reverted to slightly above its 2013 strength (not least due to the ÖVP’s leadership change towards a more liberal party leader). In 2014 and 2015, the party ran candidates in all five regional elections, but only entered two Land parliaments (Vienna and Vorarlberg), while failing in Upper Austria, Styria, and Burgenland. Also in 2014, it secured 8.1% of the vote in the European Parliament elections and thus gained one seat (taken up by former LIF leader and NEOS deputy leader Angelika Mlinar) (Johann et al. 2017; Belafi 2014). At the time of writing, NEOS counted about 2,500 members (NEOS, 9 August 2016), of which around 1,000 were in Vienna. Although this seems to be a rather small number in comparison with ÖVP and SPÖ, this figure compares favourably with the Green Party that, 30 years after its establishment, only counted about 6,000 members in 2015 (EnnserJedenastik 2014). At the time of writing, the status of NEOS in terms of ‘relevance’ resembles the LIF in the 1990s: NEOS certainly inf luences party competition in Austria. Hence, by applying a broad definition of Sartori’s (1976: pp. 119–125) ‘blackmail potential’, we consider NEOS as a relevant party of the Austrian party system. However, at the moment, NEOS has not enough parliamentary seats to ensure a majority with any other party (although a coalition of three parties would be arithmetically possible).

BK-TandF-9780815372387_TEXT_CLOSE-181001-Chp07.indd 133

14/01/19 2:17 PM

Figure 7.1  Policy positions and saliency scores for the general election in Austria in 2013 Note: Party abbreviations: G = Greens, TS = Team Stronach. Saliency scores give percentages of the issue area in the manifesto (y-axis); saliency scores do not add to 100 as other

Liberal Parties in Austria 135 Figure 7.1  (Notes Continued) policy areas and the residual category are not shown. For assessing positions, we include only issues that can be assigned to the underlying scale (see list below). Positions are only calculated if at least five of these statements are published. Thus, parties with fewer positional statements are missing in the corresponding graphs. The poles for the position of the issue areas are: (1) socioeconomic issues: state intervention, welfare expansion (–1) vs. free market, welfare retrenchment (+1); (2) cultural issues: liberal social values (–1) vs. conservative social values (+1); (3) law and order: civil liberties (–1) vs. law and order (+1); (4) immigration: restrictive immigration policies (–1) vs. liberal immigration policies (+1); (5) Europe: Euroscepticism (–1) vs. support for European Integration (+1); and (6) education: Inclusive forms of education (–1) vs. selective forms of education (+1) Source: AUTNES manual content analysis of election manifestos (Müller et al. 2017; see also Dolezal et al. 2016)

Ideology and policy positions This section describes the ideology and policy positions of NEOS in comparison with the other competing parties in the Austrian general election in 2013. The government before and after the election, like in the majority of the years after 1945 (Müller 2006a), was formed by the SPÖ and the ÖVP. Furthermore, the Austrian party system is characterised by a strong Green Party (the Greens’ vote share has exceeded 10% since 2002) (Dolezal 2016) and an even stronger radical right populist party (FPÖ) (McGann and Kitschelt 2005). Team Stronach für Österreich (Team Stronach for Austria – Team Stronach), founded by defecting BZÖ-MPs a year before the election, managed to pass the threshold for the first time. The Bündnis Zukunft Österreich (Alliance for the Future of Austria – BZÖ), a splinter party from the FPÖ, lost parliamentary representation (Dolezal and Zeglovits 2014). Besides reports from the literature, a good way of estimating the parties’ policy profile is by analysing quantitative data gathered by manual content analysis under the auspices of the Austrian National Election Study (AUTNES) (Müller et al. 2017). We analysed manifestos from six parties that were published before the 2013 parliamentary elections. NEOS did not publish a manifesto before the general election in 2013, therefore the generic party platform (Grundsatzprogramm) was used as a substitute. The content analysis is based on the AUTNES coding procedure that extracts standardised statements from every grammatical sentence. After this unitising step, every statement is coded into one of more than 700 issue categories. In addition, the position towards the issue is coded as positive, neutral, or negative (for details, see Dolezal et al. 2016). This coding scheme allows for estimating saliency scores and policy positions for a varying number of issue areas. For the analysis at hand, the fine-grained issues are aggregated into six policy areas, plus a residual category. The policy positions and saliency scores are displayed in Figure 7.1.

136  Laurenz Ennser-Jedenastik and Anita Bodlos To calculate the socioeconomic left–right positions, the parties’ statements are assigned along a scale where the negative pole (−1) indicates support for state intervention, welfare expansion, and bigger government, whereas the positive pole (+1) indicates support for free markets, welfare retrenchment, and smaller government. As Figure 7.1 shows, NEOS is positioned roughly in the centre of the socioeconomic scale (+0.1). This position is a result of putting almost equal emphasis on issues on both poles. For example, NEOS supports the implementation of a Tobin tax in the European Union and a tax cut for low incomes (NEOS 2013: 28, 32), yet it also rejects mandatory membership for workers and businesses in corporatist organisations and advocates for a more liberal regulation of working hours (NEOS 2013: 32–33). Still, compared with other Austrian parties (mean: −0.2, standard deviation: 0.4), NEOS takes a relatively liberal position in this policy field that is almost a standard deviation to the right of the mean position of all parties. NEOS dedicates about 30% of its platform to socioeconomic issues, less than any other party (mean: 39.5; standard deviation: 7). Cultural issues constitute a second important policy field and combine views on homosexuality, family norms, the role of religion, gender equality, and political extremism. Positions on cultural issues are calculated on a scale ranging from liberal social values (−1) to conservative social values (+1). With the exception of the extreme outlier FPÖ, all Austrian parties take positions on the liberal side of the cultural scale. NEOS is situated almost perfectly in the middle of these parties on the liberal side. Examples of liberal positions emphasised by NEOS are rights for homosexuals or encouraging fathers to invest in child care (NEOS 2013: 52). Still, NEOS takes more conservative stances than the Greens and the SPÖ. This is because NEOS supports some issues classified as conservative such as more rights for divorced fathers and the recognition of child-rearing periods in the calculation of pensions (NEOS 2013: 50, 52). In sum, NEOS expressed a more conservative view in its 2013 party platform on cultural issues than the LIF in its manifestos in the past (not shown). Cultural issues only played a minor role in NEOS party platform in 2013: the party only dedicated about 4% of its platform to these issues, the lowest share for cultural issues of all parties (mean: 7.6, standard deviation: 2.5). A third important policy field for Liberal parties in Austria, as well as for liberal parties in general (von Beyme 1984: 57), is Europe. The FPÖ’s increasingly negative attitude towards the European integration constituted one of the reasons for the subsequent party split in 1993 (Liegl 2006: 407). Europe was also one of the core issues during the formation process of NEOS. The first party platform published right after the founding congress contained five chapters, one of which was about European issues (NEOS 2012a). The salience of Europe in NEOS’ platform published before the 2013 general election is still slightly above average (4.6%). However, with the exception

Liberal Parties in Austria 137 of the FPÖ, who takes a strong and pronounced Eurosceptic stance, the salience of European integration across parties is low compared with the previously described issue areas (mean: 4, standard deviation: 3). Like the LIF in the past, NEOS takes a clearly pro-European position. For example, NEOS advocates a European constitution and the transformation of the European Union into a federal state (NEOS 2013: 22–23). In sum, NEOS joins three other pro-European parties (Greens, ÖVP, SPÖ) and emphasises this issue slightly more than its competitors. Education is considered as another important issue for liberal parties (von Beyme 1984: 57) and constitutes also a prominent issue in the policy profile of NEOS. NEOS dedicated almost 19% of its platform to education, and thus emphasised this policy field more than any other party (mean for all 7 parties: 12.1, standard deviation: 5.3). In addition, the party leader described education as the ‘core issue’ of the election campaign in 2013 (Strolz 2013: 100). The scale for assessing positions ranges from inclusive forms of education (−1; e.g., comprehensive schools) to a more selective educational system (+1, e.g., implementation of tuition fees). As Figure 7.1 indicates, NEOS takes a position almost perfectly in the middle of the scale. The party promotes both inclusive forms of education (such as equal chances for all children in schools independently from their parents’ education, or more full-time schools) and selective forms of education (for example, more restrictive admission policies at universities) (NEOS 2013: 16, 18). Thus, these statements cancel each other out and position the party around the centre. Still, compared with the mean across parties (mean: −.4, standard deviation 0.5), NEOS takes a position closer to the ‘selective’ pole than most parties. Figure 7.1 also visualises policy positions and saliency scores for the immigration issue. NEOS takes a position that is close to the more liberal pole of the scale. Examples for pro-immigration issues are work permits for immigrants depending on the ‘demands of the labour market’ or encouragement and support for learning the German language (NEOS 2013:11, 27). According to the party platform, immigration also constitutes a more important field for NEOS than for most other parties, although the salience of this issue is limited across most parties (mean: 5.4, standard deviation 5.7). The last graph in Figure 7.1 shows the saliency scores and position for the policy field ‘law and order’. Here, the saliency scores are so low that estimations of policy positions have to be treated with caution. Still, it is safe to assume that this issue plays a minor role in the ideological profile of NEOS. Likewise, the endorsement of the liberal view by NEOS is in line with the position of the LIF in the past. In addition to the policy fields displayed in Figure 7.1, NEOS dedicated larger parts of its platform to the issue area ‘environment’ (8.6%). Moreover, the policy fields ‘democracy’ and ‘corruption’ each constituted about 3% of the platform.

138  Laurenz Ennser-Jedenastik and Anita Bodlos Summarising the policy profile of NEOS, it can be stated that the party does not take any extreme positions compared with its competitors. NEOS endorses a position on the more market-oriented pole of socioeconomic issues (at least compared with other Austrian parties), is pro-European and takes liberal stances on immigration as well as (less clearly) on cultural issues. Still, these party positions are hardly unique in the Austrian party system. Education is central to the policy profile of NEOS, as also stated by the party leader (Strolz 2013: 100).

Structure and organisation Main party structures NEOS has adopted a quite open party structure, yet retained an element of elite control in candidate nominations. The first party statutes were published in 2012. Throughout this document, the party emphasises transparency and the participation of nonmembers in certain areas, such as candidate selection and policy development. Members require either an Austrian citizenship or a permanent residence in Austria (NEOS 2012b: 4). The main party body is the membership assembly (Mitgliederver­ sammlung), where party members have to meet formally at least once a year, but in fact do so more frequently (around four times). The membership assembly is open to the public, still, only members have voting rights. The membership assembly elects the party leader and the executive committee. Both are elected for a period of three years. Moreover, the membership assembly decides on ‘cooperation with other political parties’, thus also upon coalitions (NEOS 2012b: 6). NEOS has adopted the most inclusive candidate selection method among all parliamentary parties in Austria. It consists of a hearing, a nomination by the membership assembly, followed by voting in three tiers. Members and nonmembers alike have the opportunity to present their candidacy at a hearing. Those who are nominated by the membership assembly can then participate in the election process. The first selectorate is the general public who can vote online, a novel procedure in Austrian politics. For the 2013 parliamentary election, 1,355 citizens seized the opportunity to cast a valid online vote (Jenny 2016: 89). Next, the executive committee casts its votes, followed by the third tier in the nomination process, the membership assembly. All three selectorates use a version of the Borda count method. Each selector ranks his or her five most preferred candidates and awards them between one and five ‘confidence points’. The total sum of points for each candidate per tier is then divided by the number of selectors, and summed across tiers, thus producing a rank-ordering of candidates for the election list (NEOS 2012b: 7–8). NEOS (2012b: 9) also offers nonmembers the possibility to participate in the formulation of policies. According to the statutes, the membership

Liberal Parties in Austria 139 assembly and the executive committee can implement ‘citizen forums’. These forums are online platforms or offline groups for which anyone can register to take part in discussions and votes. Although participation is open, the membership assembly has the final say on election manifestos and party platforms. Although not established in its party statutes, NEOS also has a substantial number of policy working groups (called Themengruppen or Themennetzwerke) that are open to members and nonmembers. The first NEOS party platform originated from the policy papers produced by these working groups. After the merger of NEOS and LIF in 2014, new party statutes were adopted (NEOS 2016). The candidate selection process remains largely the same. Likewise, the party maintained the opportunity for implementing ‘citizen forums’ and its openness to the public when it comes to formulating policies (if not otherwise regulated). Extensions with respect to the previous statutes concern regulations of regional and local party organisations and elections as well as the explicit implementation of a youth organisation (JUNOS). NEOS also explicitly offers Austrian citizens living abroad the opportunity to form a party organisation (called NEOS X) that is equivalent to the party organisations in the nine Bundesländer (NEOS 2016: 17). Besides, the ‘extended executive committee’ (Erweiterter Vorstand) is added as an additional, official party body. It consists of the executive committee, the leaders of the regional and youth organisations and ten elected members (as well as high-ranking party officials such as the head of the parliamentary fraction who do not have voting rights). This party body has comprehensive powers: it forms the electorate for the second tier of the candidate selection process, it nominates candidates for executive positions (such as government ministers) and it preselects policy proposals for the membership assembly that retains the right to decide on coalitions and cooperation with other parties. Evolution of party leadership The party leader of NEOS at the time of writing is Matthias Strolz2, who was also one of the driving forces behind the formation of the party. He was officially elected as party leader at the founding congress in October 2012 with 96.2% of the valid votes. After the party merger with the LIF, he was reelected by 98.7% of the participants of the membership assembly. The former LIF-leader Angelika Mlinar was chosen as deputy leader with 78% of the votes. Party members As mentioned above, there is no tradition of mass membership in Austrian Liberal parties. The Liberal Forum never boasted a large number of members. Even during its electorally most successful period, in the middle of the

140  Laurenz Ennser-Jedenastik and Anita Bodlos 1990s, the LIF had only 3,000 members (Liegl 2006: 408). NEOS continuously publishes its membership figures on the party websites. At the time of writing, the party had 2,558 members (NEOS, 9 August 2016). To gain further insights, we conducted an online survey among all NEOS party members and (nonmember) activists. An invitation to take part in the survey was sent by email to all members and activists (which numbered 3,296) by the NEOS party academy (NEOS Lab) together with a link to the online questionnaire. The survey was fielded on 6 April 2016, and respondents had until 23 April to fill out the questionnaire. Altogether, responses from 412 individuals (83% party members, 16% affiliates, 1% undeclared) were collected, which represents a response rate of 12.5%.3 Although this is not an overwhelming number, the sample proportions correspond strongly to the two characteristics of the overall population for which we have information on: region and gender (only Vienna is slightly overrepresented, with a deviation of +7 percentage points; all other deviations are smaller than 3 percentage points). To begin with, the gender distribution is heavily skewed (as in most parties, see Scarrow and Gezgor 2010). 80% of all survey respondents are male (which is exactly the proportion in the total membership). The mean (and median) age is 45, thus about ten years younger than the typical party member in Austria (compared with data from the AUTNES preelection survey, see Kritzinger et al. 2016). About 41% of all members are from Vienna (48% in the survey). Like the Liberal Forum in the past, NEOS attracts highly educated individuals. A whopping 59% of respondents report having a university degree. By contrast, only about 11% of the sample have completed a formal education below the level of a high school degree. The occupation profile is rather unusual, too: 44% work in white-collar jobs, and about the same proportion are self-employed – a proportion that is about four times the national average. About 10% of the respondents are still in education. A secularist attitude was associated with the Liberal Forum in the past. This trend appears to continue within the party members of NEOS: only 47% of the respondents declare to be a member of the Roman Catholic Church, compared with 62% of the total population in 2014 (Katholische Kirche 2014). Moreover, 40%, more than twice the percentage among the entire population, declared not to belong to any religion. Irrespective of the denomination, only 20% of all respondents declare themselves to be fairly or very religious, compared with 47% of eligible voters (Kritzinger et al. 2016). The typical NEOS member is thus a middle-aged, secular male from Vienna with a university degree, working in a white-collar job or running his own business. Beside this demographic information, we can also shed light on the political characteristics of party activists and members, such as activity levels, former party membership, and policy preferences. To begin with, the vast

Liberal Parties in Austria 141 majority (about 70%) has never been a member of any party before. 11% report earlier membership of the LIF, and 9% report earlier membership of the ÖVP. The respondents appear to be quite active in the party. No less than 74% report having attended a membership assembly at the national or regional level during the past two years, and a third of all respondents have been to more than three. The median respondent reports one hour of unpaid work for the party per week (with only 286 valid responses, however). What is more, no less than 45% of respondents have been active in one or several of the party’s policy working groups (Themengruppen). Although we lack comparative evidence for other parties in Austria, it is safe to say that all these indicators point to a very high overall level of party activism. To be sure, it is very likely that more active members and activists have higher participation rates in the survey, so that figures overestimate the level of party activism. Party members and activists perceive their party as largely in line with their own ideological position. On average, respondents position NEOS in the centre of an 11-point left–right scale (mean: 4.9, standard deviation: 1.2). This value is almost identical to the respondents’ average self-placement (5.0, standard deviation: 1.5). Only 16% of all respondents place themselves further than two points away from their party on the left–right scale. Members and activists also support the pro-European position of the party: about 88% of respondents declare Austrian membership in the European Union to be ‘a good thing’. Only 2% take a negative stance, whereas 9% rate EU membership as ‘neither good nor bad’. Participants in the survey also had the possibility to name up to three ‘most important problems’. A frequency analysis shows that socioeconomic issues are ranked first (30% of all responses). This priority for socioeconomic issues is in line with the saliency scores obtained from the 2013 party manifesto. The second most important issue for party members and activists is education (15%), which also mirrors the party’s strong emphasis on the issue (19% in the party platform). Education is closely followed by the policy field of immigration (14%). This value is considerably higher than the saliency score in the party platform (6%), which is most likely due to the European refugee crisis that started in 2015. Corruption was named as a most important issue by 5% of the respondents, followed by Europe (2.4%). All other policy fields of the applied coding scheme were named by less than 2% of the respondents. Relation to social movements Neither LIF nor NEOS have strong anchoring in social movements or interest groups. While the establishment of NEOS was clearly fuelled by loose associations and circles of citizens who were disenchanted with the perceived political deadlock under the SPÖ – ÖVP coalition, there are no formal organisations that serve as a societal backbone for the party.

142  Laurenz Ennser-Jedenastik and Anita Bodlos However, our survey includes a number of questions that shed light on the relationships between party members and interest groups. Only 10% of all respondents report membership in a trade union, which is just a little more than half the share in the electorate as a whole; 18% are members of the Chamber of Labour; and 28% are members of the Chamber of Commerce (both are corporatist interest groups with mandatory membership). One in twenty respondents is a member of another chamber with mandatory membership (e.g., for lawyers, doctors, architects, or veterinarians).

Conclusion The starting point of this chapter was that LIF and NEOS fulfil one criterion for defining party families according to Mair and Mudde (1998): membership of liberal transnational organisations. The operationalisation of the other criteria, origins, and policy profile, is less clear cut. Compared with other liberal parties in Europe, the LIF as first liberal party of Austria emerged late (Smith 1988: 19). Although LIF-activists proclaimed liberalism and perceived themselves in line with a liberal tradition, the origins of the party as a split from the FPÖ in 1993 are rather unusual. Still, the LIF clearly developed a policy profile that highlighted economic and social aspects of liberalism. Thus, the LIF was, in our point of view, without doubt a liberal party. The origins of NEOS are even more unusual for a liberal party. Whereas the activists of the LIF could draw on a liberal tradition of a fraction within the FPÖ, the activists engaged in the formation process of NEOS had no common history: some were party members of the LIF, some of the ÖVP, most were previously unaffiliated to a party. Only the cooperation with the JuLis and the LIF resulted in the participation of activists who were organised in a party that operated under the label ‘liberal’ (Aichholzer et al. 2014: 22). Still, the policy profile of NEOS resembles that of liberal parties and the LIF in many respects. NEOS positions itself to the right of most parties on the socioeconomic scale (although this implies a central position in the Austrian case) and is found among the social-liberal parties. Besides, NEOS supports the European integration and emphasises education, which are common aspects of a liberal policy profile (von Beyme 1984: 57). Hence, we count NEOS among the liberal parties due to its policy profile and the membership in international party organisation, although its origins do not resemble that of established liberal parties in Europe. Finally, a few words on the main challenges for liberal parties in Austria in the future. Keeping in mind the fate of the Liberal Forum, the main challenge for NEOS is to maintain parliamentary representation. At the time of writing, NEOS polls about 6% or 7% of the votes. Although the danger of losing parliamentary representation is not imminent, it may still linger upon the party. Another challenge is the party’s organisational

Liberal Parties in Austria 143 strength and electoral success in the rural areas of Austria. NEOS successfully passed the threshold for the regional assemblies in Vienna and Vorarlberg, where the party was also strongest in the 2013 general election. Still, NEOS failed to enter regional assemblies in three other regions where the party also performed below average in the general election. Thus, the strength of the party varies significantly across the country. A third challenge lies in a process of disillusion due to the reality of everyday political life. NEOS was founded as a grassroots process with highly engaged and optimistic activists who perceived themselves as part of a civil society movement to initiate fundamental changes in the Austrian political system. As the survey among members shows, the activists still invest above average time in unpaid work for the party. The challenge remains to keep up this spirit while still managing an effective day-to-day political functioning.

Notes 1. Information about the formation process of NEOS and demographic data about its members was gathered through interviews with high-ranking party officials. 2. Due to the recent history of NEOS and the accordingly low number of members and leaders, we refrain from using tables and rather describe the developments in the text. 3. To be more precise, out of 3,296 e-mail recipients an estimated 1,262 opened the message (an open rate of 39%); and out of those, 431 clicked on a link contained in the email message (a click rate of 34%). There were 46 messages that could not be delivered to the listed email addresses), leaving a total of 1,988 messages that were delivered but left unopened.

References Aichholzer J., Kritzinger S., Jenny M., Müller W.C., Schönbach K., Vonbun R. (2014), ‘Die Ausgangslage’. In: Kritzinger S., Müller W.C., Schönbach K. (eds.), Die Nationalratswahl 2013: Wie Parteien, Medien und Wählerschaft zusammenwirken. Vienna: Böhlau, pp. 9–38. Belafi M. (2014), ‘Die Europawahl 2014 in Österreich: Personalisierung als Strategie in einem ausdifferenzierten Parteiensystem?’. In: Kaeding M., Switek N. (eds.), Die Europawahl 2014. Wiesbaden, Germany: Springer VS, pp. 197–207. Dolezal M. (2016), ‘The Greens in Austria and Switzerland: Two successful opposition parties’. In: van Haute E. (ed.), Green parties in Europe. Abingdon, UK: Routledge, pp. 15–41. Dolezal M., Ennser-Jedenastik L., Müller W.C., Winkler A.K. (2016), ‘Analyzing manifestos in their electoral context: A new approach applied to Austria, 2002–2008’, Political Science Research and Methods 4(3), pp. 641–650. Dolezal M, Zeglovits E. (2014), ‘Almost an earthquake: The Austrian parliamentary election of 2013’, West European Politics 37(3), pp. 644–652. Ennser-Jedenastik L. (2014), ‘Party membership figures. Austria 1945–2013’. MAPP Project Data Archive [www.projectmapp.eu].

144  Laurenz Ennser-Jedenastik and Anita Bodlos Ennser-Jedenastik L., Müller W.C. (2014), ‘The selection of party leaders in Austria: Channelling ambition effectively’. In: Pilet J.-B., Cross W.B. (eds.), The selection of political party leaders in contemporary parliamentary democracies: A comparative study. London: Routledge, pp. 62–76. Fitzmaurice J. (1995), ‘The 1994 Referenda on EU membership in Austria and Scandinavia: A comparative analysis’, Electoral Studies 14(2), pp. 226–232. Gerstl A. (1998), Die Entstehung und Entwicklung des Liberalen Forums (LiF) 1993/94. Doctoral thesis, University of Vienna. Jenny, M. (2018), ‘Intra-party Democracy and Internet: The Case of NEOS in Austria’. In: Cordero, G., Coller, X. (eds.), Democratizing Candidate Selection. New Methods, Old Receipts? Basingstoke: Palgrave Macmillan, pp. 77–98. Johann D., Glantschnigg C., Glinitzer K., Kritzinger S., Wagner M. (2014), ‘Das Wahlverhalten bei der Nationalratswahl 2013’. In: Kritzinger S., Müller W.C., Schönbach K. (eds.), Die nationalratswahl 2013: Wie Parteien, Medien und Wählerschaft zusammenwirken. Vienna: Böhlau, pp. 191–213. Johann D., Jenny M., Kritzinger S. (2017), ‘Mehr Wettbewerb bei Österreichs Wahlen? Die neue Partei NEOS und ihre engsten Konkurrenten’, Zeitschrift für Parlamentsfragen 47(4), pp. 814–830. Katholische Kirche (2014), ‘Kirchliche Statistik der Diözesen Österreichs (Katholiken, Pastoraldaten) für das Jahr 2010’, http://www.katholisch.at/dl/ qNtoJKJKKknLmJqx4kKJK/Statistik_2014_Katholiken_Pastoraldaten_Version_ Amtsblatt.pdf Kritzinger S., Zeglovits E., Aichholzer J., Glantschnigg C., Glinitzer K., Johann D., Thomas K., Wagner M. (2016), ‘AUTNES Pre- and Post Panel Study 2013’. Dataset, GESIS Data Archive, Cologne, ZA5859 Data file Version 2.0.0. Liegl B. (2006), ‘Kleinparteien’. In: Dachs H., Gerlich P., Gottweis H., Kramer H., Lauber V., Müller W.C., Tálos E. (eds.), Politik in Österreich. Das Handbuch. Vienna: Manz, pp. 402–411. LIF (1994), ‘Die Politik des Liberalen Forums’, Manifesto. LIF (1995), ‘Die offensive Mitte, Persönlichkeiten und Programm für Österreich’, Manifesto. LIF (2008), ‘Fairness’, Manifesto. Luther K.R. (1988), ‘The Freiheitliche Partei Österreichs: Protest party or governing party?’. In: Kirchner E.J. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 213–251. Mair P., Mudde C. (1998), ‘The party family and its study’, Annual Review of Political Science 1, pp. 211–229. McGann A., Kitschelt H. (2005), ‘The Radical Right in the Alps: Evolution of support for the Swiss SVP and Austrian FPÖ’, Party Politics 11(2), pp. 147–171. Müller W.C. (2006a), ‘Regierung und Kabinettsystem’. In: Dachs H., Gerlich P., Gottweis H., Kramer H., Lauber V., Müller W.C., Tálos E. (eds.), Politik in Österreich. Das Handbuch. Vienna: Manz, pp. 168–187. Müller W.C. (2006b), ‘Parteiensystem: Rahmenbedingungen, format und mechanik des parteienwettbewerbs’. In: Dachs H., Gerlich P., Gottweis H., Kramer H., Lauber V., Müller W. C., Tálos E. (eds.), Politik in Österreich. Das Handbuch. Vienna: Manz, pp. 279–304. Müller W.C. (2009), ‘The snap election in Austria, September 2008’, Electoral Studies 28(3), pp. 514–517.

Liberal Parties in Austria 145 Müller W.C., Bodlos A., Dolezal M., Eder N., Ennser-Jedenastik L., Kaltenegger M., Meyer T. M., Praprotnik K., Winkler A.K. (2017), ‘AUTNES Content Analysis of Party Manifestos 2013’. Dataset, GESIS Data Archive, Cologne, Germany, ZA6877 Data file Version 1.0.0. NEOS (2012a), ‘Pläne für ein neues Österreich’, Party platform. NEOS (2012b), ‘Satzung NEOS – Das neue Österreich’, Statutes. NEOS (2013), ‘Pläne für ein neues Österreich’, Party platform. NEOS (2016), ‘Satzung NEOS – Das neue Österreich und Liberales Forum’, Statutes. NEOS (9 August 2016), ‘Wir erneuern Österreich’, https://neos.eu/wer-wir-sind/ Parlament (1999), ‘Biografie von Mag. Dr. Heide Schmidt’, https://www.parlament. gv.at/WWER/PAD_01739/#tab-Ueberblick Pedersen M.N. (1982), ‘Towards a new typology of party lifespans and minor parties’, Scandinavian Political Studies 5(1), pp. 1–16. Pelinka A. (1996), ‘Liberalismus und Judentum’. In: Hauer N. (ed.) Liberalismus und Judentum. Vienna: Liberales Bildungsforum, pp. 125–136. Sartori G. (1976), Parties and party systems: A framework for analysis (Vol.1). Cambridge: Cambridge University Press. Scarrow S.E., Gezgor B. (2010), ‘Declining memberships, changing members? European political party members in a new era’, Party Politics 16(6), pp. 823–843. Smith G. (1988), ‘Between left and right: The ambivalence of European liberalism’. In: Kirchner E.J. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 16–28. Strolz M. (2013), ‘Eine pralle Mischung aus Idealismus und Professionalität’. In: Hofer T., Tóth B. (eds.), Wahl 2013. Macht Medien Milliardäre. Münster, Germany: LIT Verlag, pp. 95–105. von Beyme K. (1984), Parteien in westlichen Demokratien. Munich, Germany: Piper.

8

It’s (not only) the economy, stupid? Past and future of the German Liberal Party Sebastian U. Bukow

Introduction Liberal parties have a long tradition in Germany, dating back to the mid-19th century. After World War II, the Freie Demokratische Partei (Free Democratic Party – FDP)1 was set up. This new party incorporated the different prewar strands of liberalism. It quickly evolved into the most important (economic-) liberal party in Germany. Due to its specific role as kingmaker in the German party system, FDP’s political importance has often been higher than its electoral weight. The party’s path is characterized by alternation and variation in terms of political alliances as well as organisational and electoral success. Most recently, after more than 60 years in parliament, FDP lost all representation in the federal parliament. However, the party managed to come back at several Land level elections and has revitalised its national party in the general elections in 2017. The chapter addresses the versatile development of this most relevant party in the German system. As we show, the story of the FDP is strongly entangled with the development of the German party system.

Origins and development Evolution of the party in the party system and main intra-party events German liberal parties are ideologically rooted in the Aufklärung and have been present since the mid-19th century (Kaack 1971: 21–28). After World War II, the idea to revitalize the liberal tradition is pushed by Theodor Heuss and others. This strategy was favoured by a broader set of circumstances such as the general distrust against the new confessionintegrative Christian-Democratic party (CDU, CSU) and the socialist imprint of the Social Democrats (SPD; Dittberner 2010: 32; Kirchner and Broughton 1988: 66). Overall, the FDP perpetuates the tradition of liberal parties in Germany after World War II but overcomes the liberal prewar schism and incorporates the several strands of liberalism in one organisation (Dittberner 2010: 28–35; Kirchner and Broughton 1988).

It’s (not only) the economy, stupid? 147 The FDP passed the threshold of declaration even before the founding of the Federal Republic. Due to formal regulations, weak party roots after the Nazi regime, and organisational path-dependency, Land level precursors were most important in the founding process. Furthermore, due to the electoral success of Land level precursors, the FDP was represented in the constituent assembly (Parlamentarischer Rat) and played a crucial role for enforcing liberal rights in the constituency. These subnational party groups were already present in several Länder parliaments when the federal party was founded in December 1948.2 Consequently, and due the regional heterogeneity (Dittberner 2010; Gutscher 1984; Kaack 1971: 201–202), the federal party and its central office has remained weak (Bukow 2013). Nevertheless, the federal party in public office has progressively gained importance. From the very start, the development of the party has been entangled with its parliamentary life and governmental participation. The FDP has never been a mass-member party and has suffered when in opposition. It has used incumbency as an ‘opportunity to project itself into the limelight’ (Kirchner and Broughton 1988: 67). Less than one year after the formal foundation of the federal party, the FDP passed the thresholds of authorisation, representation, and relevance in the first federal election of 1949 (Table 8.1). The FDP joined government with the Christian Democrats (CDU/CSU) and other smaller parties (Chancellor Konrad Adenauer, CDU). In addition, the FDP got the first federal President (Theodor Heuss, 1949–1959). This early success and the party’s positioning are most relevant for the later role of the party. The FDP managed to survive the process of party system consolidation in the early 1950s and survived as third party in the consolidated German party system (beside CDU/CSU and SPD, Niedermayer 2006). Although the FDP has remained modest in terms of membership and electorate size, the party has been the pivotal player in the German two-and-a-half party system for almost 40 years, as the party operates as ‘kingmaker’ by providing parliamentary majority (Kirchner and Broughton 1988: 66). Accordingly, the relevance of FDP in the political system is much higher than its electoral results would indicate. See Table 8.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-8/ In the decades after its post–World War II rebirth, the party faced several electoral ups and downs, but overall it was successful in terms of vote-, office-, and policy-seeking. Every now and then, the FDP changed its coalition preferences, e.g., due to party leaders’ deliberate decisions or to intraparty conflicts. These changes are critical junctures for the party; they constitute electoral challenges for the party and come along with a substantial renewal of party members. Initially, the liberals discussed whether to position themselves to the right of the CDU/CSU or at a central position (i.e., left of the CDU/CSU but

148  Sebastian U. Bukow right of the SPD; Gutscher 1984: 120–215; Treibel 2014: 51–55). In 1952/1953, this conflict culminated during the party manifesto process and the ‘Naumann-affair’ in North Rhine-Westphalia (Gutscher 1984: 151–159). The liberals finally positioned themselves between CDU/CSU and SPD, but tensions quickly arose about whether to be closer to one or the other. Because of these tensions, of the electoral loss in the 1953 elections, and of conflicts in federal politics (especially on the issues of Ostpolitik, Saarland, and a reform of the electoral system), the FDP veered away from CDU/CSU (Dittberner 2010: 38–40; Treibel 2014: 54–55). The FDP-chair Thomas Dehler emphasized the party’s autonomy and aimed to position it between CDU/CSU and SPD, as ‘third power’ (Dritte Kraft, Dittberner 2010: 40–42). The FDP left government in 1956 but this strategy did not pay off in the 1957 election. After this episode, the FDP moved back close to CDU/CSU and rejoined the government in 1961. The coalition lasted until 1965/1966 (with an interruption due to the Spiegel affair in 1962, Dittberner 2010: 43–44), when a conflict on tax policies lead the FDP to leave government (Oppelland 2015). Until the mid-1960s, the FDP had been quite successful as a milieubased party. However, the party had to face societal change, dealignment, electoral defeats, and a loss in party members as well as party elites (Oppelland 2015). In response to these threats, the FDP revised its programmatic base and realigned as a policy-based party emphasising the liberal constitutional state and a new Ostpolitik (Vorländer 2013: 270). Intraparty conflicts and further loss of members occurred but the party survived (Dittberner 2010: 46). In 1969, the FDP supported SPD’s candidate for federal President, Gustav Heinemann, (Niedermayer 2006: 117), and after the federal election Willy Brandt (SPD) was elected as chancellor (1969–1974) by SPD and FDP. The FDP reinvented itself as a social-liberal party and was back in government, where it remained until 1998. Walter Scheel (1969–1974 Minister of Foreign Affairs, 1974–1982 Federal President), Hans-Dietrich Genscher (1969–1974 Minister of Home Affairs, 1974–1992 Minister of Foreign Affairs) and Otto Graf Lambsdorff (1977–1984 Minister of Economic Affairs) all gained importance in the party. Despite ongoing intraparty conflicts (left-liberals vs. market-liberals), the coalition survived difficulties and even a change in chancellorship (Helmut Schmidt, 1974–1982). This continuity rested upon FDP’s governmental pragmatism instead of being an ideology-based policy-seeking party (Vorländer 2013: 271). The FDP acted and presented itself as a ‘functional party’, ensuring the parliamentary majority and controlling the coalition partner (either Social Democrats, 1969–1982, or Christian Democrats, 1982–1998). The social-liberal era came to an end in the early 1980s, when the FDP drew nearer towards CDU/CSU (Oppelland 2015) due to ongoing internal disputes and with the SPD regarding economy, welfare, and foreign policies (Dittberner 2010: 48–59). In 1982, the FDP supported Helmut Kohl (CDU) in a constructive vote of no confidence. Kohl became Chancellor, and the major

It’s (not only) the economy, stupid? 149 party of the minimal-winning-coalition was substituted (CDU/CSU instead of SPD), but the FDP remained in government. This was a critical juncture for the party, accompanied once again by extensive membership renewal (Oppelland 2015) and massive electoral defeats. However, the party did not split and it regained electoral support in the mid-1980s (Vorländer 2013: 271). The FDP outlasted the 1980s more or less unharmed and as a governing party (Niedermayer 2006: 117), but change was on the horizon: The electoral success of the Greens transformed the party system and put an end to the FDP’s position of sole kingmaker (Bukow 2016; Patton 2015: 188). The German unification in 1989/1990 deferred this transformation, since Hans-Dietrich Genscher (Minister of Foreign Affairs, FDP) played a leading role in the unification process, which facilitated the FDP’s electoral success in these years (Niedermayer 2006: 124–125). The FDP was the first nationwide party after the unification (Vorländer 2013: 271) and was represented in all 16 Länder parliaments. However, this success could not be sustained. From 1993 until 2000, the FDP raced to the bottom, and only won in two federal and five Land level elections (out of 33 elections at European, federal, and Land levels, Vorländer 2013: 272). In the 1993/1994 elections, the FDP exited 12 Länder parliaments. This electoral decline has several reasons (Niedermayer 2006: 125; Vorländer 2013: 272–273): The party’s core clientele decreased to less than 5% of the electorate, and the FDP depended on votes that were ‘borrowed’ from the designated coalition partner. This strategic voting model worked well a functional party, but it collapsed as the party system fragmentation increased in the 1990s. Furthermore, after the retirement of Hans-Dietrich Genscher, Otto Graf Lambsdorff, and others, no leading FDP politician represented FDP issues prominently. Intraparty conflicts exploded and the party struggled with its image of ‘party for high-earners’, especially in recently unified eastern Germany (Niedermayer 2006: 125). The party suffered in the 1990s and failed at its goal of keeping Kohl as chancellor in the 1998 elections. The CDU/CSU and FDP lost votes and, after almost 30 years in power, the FDP was no longer a governmental party at the federal level. In the 2000s, the FDP reemphasised liberal core issues such as free market policies (Niedermayer 2006: 126). Guido Westerwelle (2001–2011 party leader, 2006–2009 chair of the parliamentary group, 2009–2013 Minister of Foreign Affairs) was the party’s most important politician. Despite some conflicts and scandals,3 the FDP increased its electoral results and proclaimed its ambition to form a CDU/CSU-FDP coalition in the 2005 elections campaign. This borrowed-votes-strategy was successful and the FDP became the third-largest party again. However, the party remained a peripheral party (Debus and Müller 2011: 174), because there was no parliamentary majority for a CDU/CSU-FDP coalition and no political will for a SPD-FDP-Greens coalition (Vorländer 2013: 274). Instead, a Grand Coalition was set up and the FDP remained in opposition (Bukow and Seemann 2010).

150  Sebastian U. Bukow The FDP benefited from its time in opposition and gathered electoral success at the Land level (in 2009, the FDP had members of parliament (MPs) in 15 of the 16 Länder parliaments and participated in half of all Länder governments, Patton 2015: 188). Emphasising liberal core issues (especially, the pledge to reduce taxes), the FDP gained 14.6% of the votes in the 2009 federal elections. This was its best federal results ever, and the long-awaited CDU/CSU-FDP coalition was implemented (Angela Merkel, CDU, remained Chancellor). However, this was no easy process. First, the coalition negotiations were ‘surprisingly tough’ (Saalfeld and Zohlnhöfer 2014: 238) and the FDP failed to enforce their main pledge (lower taxes) in the negotiations. In fact, there was no real common governmental project. Second, Guido Westerwelle became Minister of Foreign Affairs (not Minister of Finances), which weakened the FDP’s position in the coalition bargaining. Third, governmental work started suboptimal (e.g., a very first success for FDP was to push tax cuts for the hospitality sector; but generous party donations by a hotel chain became public). Consequently, the FDP’s ratings in polls collapsed, and it lead to several Land level electoral defeats and parliamentary drop-outs; Philipp Rösler replaced Guido Westerwelle as party chair in 2011 (Patton 2015: 188). The federal governmental participation did not pay off and the 2013 elections turned into a nightmare for the party, because it fell from its highest to its lowest results ever and exited the federal parliament (Table 8.2). There are multiple reasons for this defeat. First and foremost, the FDP’s borrowed-votes-strategy did not work, as the CDU/CSU claimed during the campaign that both votes should be balloted for Merkel’s CDU (Patton 2015: 188). Furthermore, the Alternative für Deutschland (Alternative for Germany – AfD), a populist and partly neoliberal competitor, enticed liberal voters. Finally, FDP’s valence dropped dramatically and the party had no strategy to prevent the electoral loss (Zur 2017). FDP’s exit from the federal parliament is not the first time, but it is a very rare event. FDP lost all its 93 federal MPs and its parliamentary staff (about 90 parliamentary party group and 635 MP employees). Furthermore, the party in central office had to further reduce its number of employees (which was already small – about 35 employees in 2012 – Bukow 2013: 175). Consequently, the FDP experienced extra-parliamentary opposition at the federal level for the first time since 1949. See Table 8.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-8/ Electoral performances of the party From 1949 until 2013, the FDP was continuously present in the German Bundestag. Being the ‘third party’ in the German party system, the FDP is much smaller than the CDU/CSU and the SDP. However, a pluralisation of the German party system takes place with the emergence of the Greens in

It’s (not only) the economy, stupid? 151 the 1980s, and with the former socialist party PDS in the 1990s (now: Die Linke). Nowadays, the FDP competes with the Greens, Die Linke, and the AfD for the position of third party in the German party system (Figure 8.1). The FDP’s vote share at the federal elections has varied between 4.8 and 14.6% (Table 8.2). After being in parliament for more than 60 years, the FDP exited the Bundestag for the first time in 2013, due to falling below the 5% threshold. See Figure 8.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-8/ The FDP cannot rely on a large number of traditional voters. Its electoral success depends much more on its ability to mobilise ‘borrowed votes’, that is, votes based on strategic voting in Germany’s mixed-member electoral system. Even though the party’s survival has never been an issue at the federal level until 2013, its electoral performances vary over time (Figure 8.2). There is more variation at the subnational level, where parliamentary dropouts and comebacks are quite common. The FDP holds an uninterrupted Land level parliamentary representation only in Baden-Württemberg (Patton 2015: 190), that is, in 1 out of 16 Länder. See Figure 8.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-8/ Although the party still lost votes and parliamentary representation in the 2014 Land level elections (Saxony, Brandenburg, Thuringia), the FDP managed to come back in 2015/2016. The party reentered three Land level parliaments (Bremen, Rhineland-Palatinate, Berlin) and increased its electoral scores (Schleswig-Holstein, 11.5%; North Rhine-Westphalia, 12.6%). In the most recent Land level elections, the FDP’s electoral share varied from 1.5% (Brandenburg) to 12.6% (North Rhine – Westphalia). In the summer 2017, FDP was represented in the European parliament and in nine Land level parliaments. At the federal level, the FDP scored about 8% in election polls in summer 2017. It is most likely that the FDP will reenter the German federal parliament after the 2017 elections. The subnational parliamentary groups, their electoral success, and FDP’s young party chair Christian Lindner, are important factors in the federal comeback. Relation of the party towards participation to power at various levels From 1949 until 1998, the FDP has been part of the federal government with very few interruptions; the party ‘spent more time in government than any other party’ (Zur 2017: 1) . Consequently, the FDPs’ history is strongly related to its governmental participation and its role as ‘junior partner’ in minimal-winning coalitions.4 The party has played a crucial role as kingmaker for many decades. However, with the rise of the Greens, the FDP

152  Sebastian U. Bukow lost its pivot position and relevance. Since 1998, the FDP sits in opposition at the federal level mainly (except 2009–2013). At the Land level, the FDP’s governmental participation varies over time. In Germany, minimal winning coalitions with two parties are the preferred coalition format. At the federal level, coalitions are led by the CDU/ CSU or the SPD. Even though the FDP is able to join coalitions with both parties, it usually positions itself closer to CDU/CSU even though there has been a social-liberal era at the federal level (1969–1982; see Table 8.3). This preference for coalitions with the Christian Democrats is also evident at the subnational level, although several multiparty-coalition formats have been tested as well (Switek 2015). See Table 8.3 at http://partyfamilies.eu/index.php/category/chapters/ chapter-8/ Table 8.3 summarises the FDP’s governmental participation at the federal level. In Germany, basic law defines some portfolios whereas others are the result of coalition negotiations (and therefore change over time). The smaller coalition partner is usually appointing the Vice-Chancellor (who has a ministerial portfolio as well, generally Foreign Affairs). Therefore, when in government, the FDP gets the Vice Chancellor and a various number of ministers (mainly Foreign Affairs, Justice, Finances, and Economic Affairs). During the social-liberal coalition, the FDP obtained the Minister of the Home Affairs instead of the Minister of Justice.

Ideology and policy positions Liberal parties often occupy a pivotal position on a right-left scale (Keman 1994; Kirchner 1988) and are said to be ideologically more flexible than Christian Democrats or Social Democrats (for Austria, Denmark, Germany, and the Netherlands, see Franzmann 2006). This applies particularly to the German FDP, often labelled as an excessively flexible party (Franzmann, 2006), being more interested in offices than in policies (Dittberner 2010). One reason for this negative image is the FDP’s former position as a kingmaker and its changing coalition preferences. Overall, the three traditional competitors in the German party system (CDU/CSU, SPD, and FDP) show comparatively high congruence on issue salience (Franzmann 2012), as economics and social affairs are the by far most important issues for all German parties (Pappi and Seher 2009: 412). With regard to policy positions, differences between parties are more evident. Thereby it is helpful to locate parties in a two-dimensional space, whereby one axis covers socioeconomic and the other political-cultural issues (Figure 8.3), since ‘a general left-right-scale does not scope with the position of the liberal FDP whose socioeconomic positions is sometimes to the right of the CDU and whose position

It’s (not only) the economy, stupid? 153 10

GREENS

Linke

8 1987 1983 2013 1980 1990 1994 1998 2002 2005 1976 1972 1965 1957 2009

SPD

1961 1969 FDP 19531949

6

Economic Dimension 2

4

6 4

2

0

CDU/CSU Non-Economic Dimension (Gal-Tan)

0

8

10 AfD 2013 X

Figure 8.3  Parties in the German political space Note: Party positions calculated according Franzmann and Kaiser (2006), ovals embrace parties’ positions over time; diamonds show each position of FDP’s electoral manifestos in the labelled year. Author’s own calculation.

concerning law and order issues is to the left of the CDU’ (Pappi and Seher 2009: 405). In the last years, the relevance of these two dimensions has even increased (Niedermayer 2006: 120). The general position of a party in the two-dimensional space is quite stable, as indicated by the ovals contouring the parties’ positions over time on Figure 8.4). The two larger parties (CDU/CSU and SPD) are geared to the centre, whereas the smaller parties (FDP, Greens, and Die Linke) tend towards more extreme positions. As shown in Figure 8.3, the FDP’s position in the German party system is unique. It is the only party in the upper right sector of the political map, that is, it combines economic liberalism with cultural liberalism. Both dimensions correspond to the party’s preferred ministerial portfolios. There is no doubt that the FDP is a liberal party in all definitions of the term (Kirchner and Broughton 1988: 62). The FDP emphasises individualism and freedom by and through the state (Dittberner 2010) and presents itself as promarket, proconstitutional-state, and procultural liberalism (Vorländer 2013: 504). Thereby, intra-party conflicts and the emergence of factions has been hidden (Vorländer 2013: 504), even though there have been relevant programmatic conflicts (e.g., between the national, market-, radical- or social-liberals in the party). Furthermore, the type of liberalism represented by the FDP has changed over time. In the early years, the FDP’s ideological core was anticollectivism and antisocialism

154  Sebastian U. Bukow Federal Committee 34 Committee members; further ex officio members

Executive Committee (up to 11 members, partly ex officio)

Party Congress (662 delegates)

Members Abroad (EU) (2 Delegates)

16 Land Branches Party Congresses (660 Delegates)

Party Members (Local/Regional Groups)

(Executive) Committees

Collateral Organisations

(Youth Organisation, University-Groups, Medium-sized Businesses Initiative, Women’s Organisation, Seniors’ Organisation, Councillors’ Organisation etc.)

Figure 8.4  Formal party structure Note: Simplified structure of the formal party organisation (e.g., not shown: arbitral courts, party in public office, party central office/staff). Author’s own figure based on party statute regulations.

(Glaeßner 2006: 457), but there is a strong national-liberal faction as well. In the 1970s, the national-liberal wing declines and the FDP transforms into a more radical-liberal party. Finally, in the 1980s the FDP becomes an economic-liberal party (Franzmann 2015: 166; Franzmann 2011). Since the 1980s, the FDP has emphasized economic issues and has been the economic-liberal party in the German party system. These are the issues that the party emphasises most prominently in campaigns, although there are several other salient issues addressed in the party manifestos (e.g., freedom and human rights, governmental efficiency, and technology, see Table 8.4). Emphasising promarket, liberal economic positions (e.g., free trade, deregulation, and simplified taxation) and the idea of a lean state is crucial for the parties’ self-conception and its public image, but it also strengthens the party ownership over economic-liberal issues. It provides the party with a unique position in the two-dimensional political space of the German party system. In the wake of the global finance and debt crisis, the FDP has been the only relevant party arguing for market liberalism, deregulation, and tax reduction. It has helped the party winning swing voters that were not convinced by the CDU/CSU postcrisis left-shift

Table 8.4  Left-right position and issue saliency, FDP 1949–2013 1949 1953 1957 1961 1965 1969 1972 1976 1980 1983 1987 1990 1994 1998 2002 2005 2009 2013 Mean R-L (Laver/ 5.33 4.94 5.00 4.11 Budge) R-L (Franzmann/ 5.12 5.03 4.99 5.27 Kaiser) Economic 6.64 6.31 6.30 6.98 Dimension Noneconomic 6.14 6.13 6.28 5.96 Dimension FDP Voters (L-R) EU: Positive 0.00 2.25 0.00 0.94 EU: Negative 0.00 0.00 0.00 0.94 Freedom and 2.80 7.87 4.35 3.74 Human Rights Democracy 10.28 3.37 4.35 2.80 Free Market 1.87 2.25 2.90 6.54 Economy Market 0.94 4.49 1.45 2.80 Regulation Environmental 0.00 0.00 0.00 2.80 Protection Governmental and 6.54 6.74 2.90 8.41 Administrative Efficiency Technology and 14.02 5.62 1.45 17.76 Infrastructure: Positive Culture: Positive 1.87 2.25 1.45 0.94

5.05

4.07 6.35

4.72

4.64

5.20

5.56

5.09

5.08 5.75

5.23

5.80

5.21 5.70 5.16

4.88

4.91 4.52

4.53

4.01

4.35

4.57

5.00

5.11 5.35

5.54

5.83

5.80 5.68 5.03

7.26

7.28 6.80

6.64

6.53

6.94

6.85

6.92

6.88 7.23

7.59

7.89

8.13 7.56 7.04

6.46

6.24 6.65

6.53

7.23

7.28

7.25

6.89

6.71 6.61

6.62

6.59

6.55 7.36 6.64

-

6.30

6.64

6.69

7.34

6.47 6.32

6.41

6.32

6.39 6.45 6.53

5.66 0.00 5.66

-

-

2.00 3.85 0.00 0.00 4.67 3.85

5.11 0.00 7.23

3.41 5.09 1.95 0.16 0.00 0.49 7.91 10.55 11.71

3.35 0.00 4.80

5.19 1.21 0.11 0.00 1.66 4.44

2.37 0.45 4.54

5.80 1.49 8.70

3.64 1.59 2.97 0.76 0.97 0.30 7.33 9.89 6.20

0.47 10.67 3.85 0.47 4.00 7.69

4.26 1.70

6.05 1.09

1.09 0.73

1.46 2.44

0.15 8.30

1.11 1.33 2.98 9.96

1.67 5.05

1.91 6.72

1.69 3.06 3.31 7.20 3.88 4.21

1.89

7.33 3.85

5.11

3.10

2.55

2.93

1.89

3.65 0.63

3.03

3.89

7.42 4.96 3.44

2.36

0.00 3.85

2.13

8.37 15.64 10.73 12.81 10.83 4.50

1.87

2.83

5.02 3.72 4.86

7.08

6.67 3.85

4.26

3.88

4.73

7.32

9.43

6.00 3.85 10.64 13.95

5.09

9.76 11.94 11.38 6.34 11.91

6.58

7.96 7.37 8.95

1.89

1.33 0.00 10.64

1.09

6.34

2.33

1.78 1.55 2.78 (Continued)

3.41

3.35 10.50 8.62 10.60 10.18 10.58 8.03 6.90

3.64

4.75 3.30

1.57

Table 8.4  Left-right position and issue saliency, FDP 1949–2013 (Continued) 1949 1953 1957 1961 1965 1969 1972 1976 1980 1983 1987 1990 1994 1998 2002 2005 2009 2013 Mean Equality: Positive Welfare State Expansion Welfare State Limitation Education Expansion Education Limitation Law and Order: Positive Multiculturalism: Positive Multiculturalism: Negative

0.94 2.25 2.90 0.94 2.80 4.49 4.35 10.28

1.42 1.89

0.67 3.85 6.00 0.00

3.83 5.53

1.24 7.29

3.27 1.46

3.42 2.44

3.20 6.11

2.87 3.11 1.33 2.66

2.37 3.28

3.11 1.84

4.22 3.68 2.63 2.22 1.67 3.65

0.00 0.00 0.00

0.00

0.67 0.00

1.28

0.31

1.82

1.46

1.46

3.20 3.81

4.04

0.71

1.20 1.90 1.21

0.00 1.12 5.80 13.08 15.57

8.67 3.85

5.96

4.81

1.09

2.93

4.37

4.42 4.31 10.65

3.40

5.82 3.10 5.50

0.00 0.00 0.00

0.00

0.00

0.00 0.00

0.00

0.00

0.00

0.00

0.00

0.00 0.95

0.25

0.14

0.13 0.39 0.10

0.00 1.12 0.00

0.00

0.00

0.00 3.85

1.28

1.40

0.00

0.00

0.29

3.98 2.79

2.83

1.56

1.96 0.81 1.21

0.94 3.37 1.45

0.94

0.00

2.67 0.00

0.00

0.31

0.00

0.00

0.44

0.00 0.76

1.31

0.57

0.67 1.86 0.85

0.94 0.00 0.00

0.00

0.00

0.00 0.00

0.00

0.00

0.00

0.00

0.00

0.00 0.82

0.91

0.21

0.00 0.27 0.18

0.00

Note: Policy positions: L-R (0: left to 10: right) calculated according to Laver and Budge (1992); transformation: (L-R + 100)/20 and Franzmann and Kaiser (2006); economic/noneconomic dimension, see Figure 8.3; FDP voters: Politbarometer (ZA2391; weight by Repräsentativgewicht; cp. GESIS Data Catalogue. https://www.gesis.org); saliency: Volkens et al. (2016).

It’s (not only) the economy, stupid? 157 on economic issues in the 2009 elections, which has been most relevant for the FDP’s outstanding electoral success (Treibel 2014: 76). This is facilitated by the fact that economic-liberal politicians in the CDU/CSU have progressively lost their influence, which even strengthens the exclusive representation of economic liberalism by the FDP. At the same time, this programmatic focus on liberal-economic issues threatens the FDP’s survival (Dittberner 2010: 294). Its disastrous governmental performance on their economic core issues in 2009/2010 lead to the party’s collapse in federal polls and in Land level elections. In reaction to this, the FDP aimed to broaden its programmatic portfolio and launched a programmatic process in 2010, aiming to replace the 1997 programme. FDP’s secretary general Christian Lindner (2009–2011; since 2013 party chair) wanted to revitalise elements of traditional social-liberalism in this new basic programme. The social-liberal and economic-liberal wings of the party reinvigorated during the programmatic process and intraparty discussions. Due to conflicts with the party chair Philipp Rösler, Lindner resigned from his secretary position on the verge of the completion of the final draft. According to Lindners’ aims, the new programme combined elements of the former, 15-year old basic programme with a broader set of liberal core issues. The FDP emphasized its identity as party of freedom and self-determination. The programme points out that individualism is a core aspect of FDP’s liberalism (FDP 2012: 21). At the same time it addresses the necessity of societal solidarity (FDP 2012: 24). This moderate renewal carefully prepared FDP for new coalitions with the SPD and the Greens, although these options were not the party’s preference (Treibel 2014: 78). The new party basic programme was adopted at the party convention in 2012, and the 2013 election manifesto reflects some changes in saliency. However, these changes did not contribute to secure electoral success in the 2013 elections (Vorländer 2013: 504–505). In 2013, the FDP was threatened by the emergence of the AfD. This new party combines economic liberalism, that is, a promarket-economy position, with anti-EU-positions (especially anti-Euro), whereas the FDP represents pro-European, economic-liberal positions. The FDP lost its ownership on economic-liberal issues in the 2013 elections and shared the ownership on cultural-liberal issues (as usual) with the Greens (Bukow 2016). In the 2013 election, AfD and FDP were strong competitors for the small group of promarket-liberal voters, and the FDP lost many voters to the AfD (Schmitt-Beck 2014). Since then, the AfD has shifted its emphasis to antimigration instead of anti-Euro-issues and the FDP has regained ownership of market-liberal issues in Germany. Furthermore, it is the only party in Germany that combines economic-liberal and cultural-liberal positions. Referring to the party programme, the 2017 electoral manifesto outlines a liberal story including ‘world-leading education’ (aiming for fair chances at beginning of life instead of welfare-state redistribution), individual empowerment and achievement, self-determination of life, and aspects of constitutional and cultural liberalism. Economy and free market policies are still

158  Sebastian U. Bukow the most salient issues, but they are framed in a much broader conception of liberal individualism. Overall, the FDP’s position in the two-dimensional political space is still unique. This specific position explains how the FDP can be quite flexible in its coalition preferences. The party is on the right of the CDU/CSU on a general right-left scale (Franzmann 2015) due to its core positions on economic issues, but the distance with the SPD on some noneconomic issues is lower than with the CDU/CSU. Consequently, FDP participates in three different Land level coalitions in summer 2017, including coalitions with Greens, SPD, and CDU (coalitions: CDU/FDP in North Rhine-Westphalia, SPD/Greens/ FDP in Rhineland-Palatinate, and CDU/Greens/FDP in Schleswig-Holstein).

Structure and organisation Main party structures In line with the party law as regulatory framework as well as German federalism (Bukow 2013), the liberal party is a multilevel organisation with three levels of party organisation (local/regional level, Land level, and federal level). In this chapter, the formal story of the federal party and specific aspects of multilevel integration are analysed. The formal story is codified in the party statutes and several additional documents/standing orders, since all German parties must have extensive statutes by law, including binding regulations referring to the party structure, party bodies, decision-making procedures, and financial aspects. The principle of intra-party democracy, understood as a formal bottom-up, delegation-based process of decision making, is most important. Consequently, addressing the formal story, the federal party congress is the highest intra-party body. Decisions of the party congress are binding for the lower party levels and members (§ 11, 2 FDP statutes). It meets once a year, in some years there can be additional meetings as well (for a full list, see Dittberner 2010: 170–172).5 The congress’s main responsibility is to discuss and decide on policies (e.g., basic programmes, party manifestos) as well as organisational matters (e.g., party statutes, party budget, etc.), to elect the party leader, the party executive committee and further intra-party posts. The federal party congress does not select candidates for the federal parliament (Bundestag), as they are (s)elected by Land level congresses (Länder lists) and by constituency assemblies (constituency candidates, although the FDP does not have any constituency MPs in the federal parliament). All members have the right to join the congress, but only 662 delegates are allowed to decide.6 660 of these delegates are elected by the 16 Land branches, and two by the ‘members abroad in Europe’. There are two main differences with other German parties: (1) Delegates are elected by Land level congresses and not by regional assemblies/branches. It increases the

It’s (not only) the economy, stupid? 159 power of the Land level branches within the party. Furthermore, middle-level elites rather than rank-and-file members are usually party delegates (Treibel 2014: 92). Consequently, the party convention is a convention of local and subnational elites (Treibel 2014: 86), which is in line with the liberals’ tradition of party of notables. (2) The organisational strength of local party members is decisive for the distribution of half of the delegates (330) only. The other half (330) is distributed based on the electoral success of each Land party in the last federal election (§13, 3 FDP statutes). According to party law, up to 50% of delegates can be distributed on this basis. The FDP’s use of this maximum emphasises its self-understanding as electoral party rather than membership-based organisation (Bukow 2013).7 Controversial discussions within the congress are rare. Conflicts are usually negotiated informally before the congress takes place. In addition, nonpublic meetings of the Land level delegates take place during the congress session – another evidence for the power of the Land branches within the federal party. The party youth organisation organizes a delegates meeting as well, since approximately 200 of the Land level delegates are also member of the youth organisation. Therefore, the youth organisation is the most powerful collateral organisation and acts as a 17th Land branch during party congresses (Treibel 2014: 95). However, addressing the real story, it is the federal committee, that is the most important body of the federal party (Dittberner 2010: 173), since the 2013 extinction of the (traditionally most important) party in public office. Within the federal committee, the federal executive committee meets more frequently. It decides on day-to-day politics and aims to link the party in central office and the party in public office (§ 17-19 FDP statutes). In addition to this executive body, the broader federal committee links the federal party with the multilevel organisation. Therefore, 16 (out of 32) committee members are not elected individually by the congress but in a linked election. Each Land branch has the right to propose one person for this package deal (§5, 5 FDP standing orders (BGO)). This procedure guarantees that each Land branch is represented in the federal committee. The most important individual player at federal level is the party leader (currently, Christian Lindner, who is not only the federal party leader but also the leader of the North Rhine-Westphalian Land party branch and parliamentary party). In fact, since 2013, the party leader is the most visible party politician. He and the party committee are supported by a small party bureaucracy that has been reduced after the 2013 defeat (Bukow 2013).8 Party leadership Party chairs are usually central figures in Germany (Detterbeck and Rohlfing 2014). They preside over the multilevel party organisation and have relevant intra-party power, since they are – normally – not only party chairs, but powerful party leaders (Detterbeck 2013: 273). German party leaders are usually

160  Sebastian U. Bukow front-row members of the party in public office as well, and serve in relevant public (governmental) positions. Due to party law, the formal leadership selection process does not differ between German parties. Only the right to nominate candidates differs in nuances (e.g., the FDP is unique to that extent that some party actors have the explicit right to propose a candidate, see Detterbeck and Rohlfing 2014: 83). The federal party congresses have by law the exclusive right to elect the federal party leadership (§9, 4 Party Law). Intra-party primaries for leadership selection are not allowed. Even consultative intra-party primaries for leadership-decisions are highly disputed (in contrast to primaries electing the electoral frontrunner, which is not a formal position). Due to the lack of intra-party primaries and due to the widespread practice to solve conflicts before and not during a party congress, federal leadership elections are seldom competitive. In Germany, party chair elections are much more a coronation than a competitive process, since there has been only one candidate in approximately 9 out of 10 intra-party elections (1965–2012, cp. Detterbeck and Rohlfing 2014: 84). However, FDP leadership elections are more often contested, even though a real choice is rare (Table 8.5; for further details, see Detterbeck and Rohlfing 2014: 85). Once elected, FDP chairs have an above-average incumbency length (6.2 years, Detterbeck and Rohlfing 2014: 88). According to party law, they have to be reelected after two years, but re-election is not limited. The FDP’s statutes do not even mention any procedure of early deselection (however, early withdrawal took place in some cases, e.g., Rösler most recently). Notably, there has not yet been a female party chair in the federal FDP. See Table 8.5 at http://partyfamilies.eu/index.php/category/chapters/ chapter-8/ Party membership The FDP has never been a mass-member party. Only in the wake of the German unification has the party exceeded the 100,000 members mark (1990–1992; see Table 8.6). In 2015, the party counted approximately 53,000 members (December 2015, whereas rounded numbers for other parties are as follows: Left 59,000; Greens 59,500; CSU 144,500; SPD 443,000; CDU 444,500 members; Niedermayer 2016). Since then, the FDP numbers have slightly increased (spring 2017: 57,000). See Table 8.6 at http://partyfamilies.eu/index.php/category/chapters/ chapter-8/ Three membership conditions need to be stressed: The minimum age is 16, the applicants must have lived in Germany for at least two years (but German citizenship is not required), and the membership is exclusive, that is, no other party membership is allowed (§ 2 FDP statutes). Further formal membership

It’s (not only) the economy, stupid? 161 rules can be set by the Land branches, since direct affiliation to the federal party are exceptional. Usually, members are affiliated to the nearest local branch (based on their place of residence) and to the according Land branch. Party members in Germany are not representative of society (Klein 2011). In all parties, the typical party member is highly educated and male. This is also the case for the FDP (approximately 56% of them have a higher education degree). However, women are even more underrepresented than in other German parties (female FDP members represent less than 23% of total members, see Niedermayer 2016). One reason for this might be that the FDP is the only party that does not have any kind of women quotas for party positions and candidate lists (Bukow and Poguntke 2013: 182). Compared with other German parties (except the Greens) the FDP has a larger share of young members (approximately 11% of those 16–29 years of age, Niedermayer 2016). In terms of professional background, the share of self-employed and entrepreneurs has been significantly higher than in other parties (FDP 41%, all parties 26%, society 9%, see Klein 2011: 50). Civil servants, still overrepresented in contrast to society, are less represented in the FDP compared with other parties (FDP 27%, all parties 35%, society 7%, Klein 2011: 50). Rank-and-file-members have never been the core of the party organisation. In fact, their intra-party power is limited. They can participate in the societal aspect of party life at the local level (Treibel 2014: 86), but party conventions are dominated by middle-level elites. However, in some cases party members have a relevant impact on party politics. In all German parties, elements of direct democracy are increasingly popular for policy and coalition decisions. At the federal level, intra-party ballots are limited to policy issues, and even then they are a rather rare event. The federal party has established three different types of intra-party ballots: binding ballots, consultative ballots, and members’ requests. The federal party convention, the federal committee, at least five committees/conventions of the Land branches, at least 100 regional branches, or 5% of the total membership, have the right to initiate a binding ballot. Besides, not all issues may be addressed, for example, changes of the statutes, intra-party elections and candidate selections, and financial aspects are not accepted (§20 FDP statutes). The threshold for consultative ballots is lower, and changes of the statutes might be questioned (§ 21a FDP statutes; not allowed are intra-party elections and candidate selections, and financial aspects). Finally, members’ requests can be launched by 250 members (§21b FDP statutes; not allowed are intra-party elections and candidate selections, and financial aspects). If initiated, the federal committee has to discuss the requested topic. Despite this broad range of instruments, intra-party ballots are exceptional at the federal level. Only two ballots were cast in the past 25 years (Bukow and Poguntke 2013: 181–182). The first ballot on wiretapping almost split the party in 1995, as the FDP in government was overruled by the FDP members to accept wiretapping. The second ballot (2011) on FDP’s pro-Euro position was highly controversial as well, but the anti-Euro-position missed

162  Sebastian U. Bukow the quorum in the final decision.9 Even though there are additional elements of direct democracy at the Land level (Detterbeck 2013), all in all, the FDP is like all German parties mainly based on the model of representative intraparty democracy. Middle-level elites and the party in public office, rather than rank-and-file members, are central for party policies and decisions. Collateral Organisations Like all German parties, the FDP has a number of collateral organisations, linking party and society (Figure 8.4), but most of these organisations are less important than in other parties, since they are seldom involved in intra-party decisions (Dittberner 2010: 170; Treibel 2014: 106). The most important collateral organisation is the party youth organisation (Young Liberals) and the party-related but independent political foundation (Friedrich-Naumann-Foundation). There are also other smaller and less important collateral organisations, for example, for higher education students, women, small and medium-sized businesses, seniors, councillors, lawyers or the LGBT community.

Main challenges for the future The German FDP has a discontinuous history. Starting after World War II as a party for all strands of liberalism in Germany, the party quickly played a central role as a kingmaker in the consolidated German party system. Therefore, the FDP has always emphasized its public office, its self-understanding as an electoral party, and its unique role as a kingmaker. However, the liberals lost this position due to the pluralisation of the German society and party system in the 1980/1990s. After more than 60 years of continuous presence, the FDP even exited the federal parliament in 2013. This was the party’s most fundamental crisis ever. The survival of the party was questioned in politics, media, and academia. However, in the past two years, the FDP seems to have made a comeback. Based on an ideological renewal towards a more comprehensive ‘liberal story’ beyond the focus on free market/economic issues, and on a personalized communication, the liberals have again been electorally successful in several Land level elections. Reentering the federal parliament in autumn 2017 is a likely possibility. However, this recovery remains fragile. Two main problems must be tackled after the 2017 federal elections: The FDP has to organize a sustainable evolution in terms of policies and personnel. In terms of policies, the party is challenged – like all German parties – by the increasing fragmentation of the party system and the necessary flexibility in coalition formation. This is especially a challenge for the FDP, because the party was often accused to prefer offices over policies. However, the liberals have already addressed this problem in their new basic programme and manifesto that emphasize a more comprehensive story telling of individualism and self-determination.

It’s (not only) the economy, stupid? 163 Future coalitions will show to which extent this story can work in practice. In terms of personnel, the FDP is challenged by the threat of being a one-man show. Recent electoral successes were strictly linked to individualities, namely party chair Christian Lindner (federal party; North RhineWestphalia) and Wolfgang Kubicki (Schleswig-Holstein). So far, it has not been the manifesto, but the personnel that has been most relevant for the FDP’s comeback. However, reentering the federal parliament will offer the opportunity to increase the pool of political talents (the lack of leading women is the most evident) and will come along with a larger number of parliamentary party staff. Overall, the FDP undoubtedly occupies the position of liberal party in the German party system, as it uniquely combines economic and cultural liberalism. It survived its pivotal crisis of 2013 and will most likely be back in the federal parliament (and maybe even in government) after the 2017 elections and remain a relevant player in the German party system in the coming years.

Notes 1. The FDP (Freie Demokratische Partei – Free Democratic Party) used the abbreviation F.D.P. from 1968 until 2001, since then again FDP. The party-own short term is Die Liberalen (The Liberals) until 2015, since then Freie Demokraten (Free Democrats). 2. In the British, French, and US-American occupation zone (see Gutscher 1984). In the Soviet occupation zone, the Liberal-Democratic Party of Germany (LDPD) was founded, but due to the early division of east/ west-Germany the LDPD did not become the German Liberal party but was included in the Eastern German bloc party system (Glaeßner 2006: 187). The LDPD and the party system in Eastern Germany (GDR) are not addressed in this chapter. 3. Most important was the so-called Spaßpartei – concept (‘fun party’), i.e., a media-focussed campaign pattern and the emphasis of party campaigning in the context of entertainment. Furthermore, Jürgen Möllemann’s anti-Semitic populism caused serious conflicts. Both conflicts boiled down with Möllemann’s death in 2003 (Vorländer 2013: 273). 4. Since the largest party normally claims the Chancellor/Prime Minister position, the FDP in government has appointed ministers only. However, in Baden-Württemberg, Reinhold Maier (FDP/DVP) was the first prime minister (1952–1953), but since there has been no FDP prime minister in Germany. In addition to this, the FDP did manage to appoint two federal presidents (Theodor Heuss, 1949–1959, and Walter Scheel, 1974–1979). 5. A ‘small congress’ (Bundesausschuss), allowing congress meetings between the full congress, was repealed (Bukow 2013: 132). A ‘Europe congress’ is held by the federal party for the election of candidates for the European parliament. 6. All delegates and further ex officio members (e.g., committee members, representatives of collateral organisations, etc.) have the right to give a speech, and in addition to this, guests might join discussions as well. 7. Die Linke, Greens, SPD, and CSU refer to membership strength only, but CDU refers to electoral success as well (whereby only 200 of 1,000 delegates are distributed by electoral success).

164  Sebastian U. Bukow

8. In addition to this, the FDP owns some smaller companies for specific organisational tasks. Due to these companies and centralized party services (e.g., member- and contact-management, campaigning tools, etc.), a gradually centralization of the party took place recently (Bukow 2013; Treibel 2014). 9. A third attempt to cast a ballot addressing an army reform could be mentioned, but it even failed to reach the quota for getting initiated (Detterbeck 2013).

References Bukow S. (2013), Die professionalisierte Mitgliederpartei. Politische parteien zwischen institutionellen Erwartungen und organisationaler Wirklichkeit. Wiesbaden, Germany: VS Verlag für Sozialwissenschaften. Bukow S. (2016), ‘The Green Party in Germany’. In: van Haute E. (ed.), Green parties in Europe. London: Routledge, pp. 112–139. Bukow S., Seemann W. (eds.). (2010), Die Große Koalition. Regierung – politik – parteien 2005–2009. Wiesbaden, Germany: VS Verlag für Sozialwissenschaften. Bukow S., Poguntke T. (2013), ‘Innerparteiliche organisation und willensbildung’. In: Niedermayer O. (ed.), Handbuch parteienforschung. Wiesbaden, Germany: Springer, pp. 179–209. Debus M., Müller J. (2011), ‘Government formation after the 2009 federal election: The remake of the Christian–Liberal coalition under new patterns of party competition’, German Politics 20(1), pp. 164–185. Detterbeck, K. (2013), ‘The rare event of choice: Party primaries in German Land parties’, German Politics 22(3), pp. 270–287. Detterbeck K., Rohlfing I. (2014), ‘Party leader selection in Germany’. In: Cross, W.P., Pilet J.-B. (eds.), The selection of political party leaders in contemporary parliamentary democracies. A comparative study. London, New York: Routledge pp. 77–92. Dittberner J. (2010), Die FDP. Geschichte, personen, organisation, perspektiven. Eine einführung (2nd ed.). Wiesbaden, Germany: VS Verlag für Sozialwissenschaften. FDP (2012), Verantwortung für die freiheit. Karlsruher freiheitsthesen der FDP für eine offene Bürgergesellschaft. Retrieved from https://www.fdp.de/sites/default/files/ uploads/2016/01/28/karlsruherfreiheitsthesen.pdf Franzmann S. (2006), ‘Parteistrategien auf oligopolistischen Issue-Märkten. Eine empirische Analyse der wahlprogrammatik in Deutschland, Dänemark, Österreich und den Niederlanden mit Hilfe des Gutenberg-Modells’, Politische Vierteljahresschrift 47(4), pp. 571–594. Franzmann S. (2012), ‘Die liberale parteienfamilie’. In: Jun U., Höhne B. (eds.), Parteienfamilien. Identitätsbestimmend oder nur noch Etikett?, Opladen, Germany: Barbara Budrich, pp. 157–186. Franzmann S., Kaiser A. (2006), ‘Locating parties in policy space. A reanalysis of party manifesto data’, Party Politics 12(2), pp. 163–188. Franzmann S. (2015), ‘The failed struggle for office instead of votes: The Greens, Die Linke and the FDP’. In: D’Ottavio G., Saalfeld T. (eds.), Germany after the 2013 elections. Breaking the mould of post-unification politics?, Farnham, UK: Ashgate, pp. 155–179. Glaeßner G.-J. (2006), Politik in Deutschland (2nd ed.). Wiesbaden, Germany: VS Verlag für Sozialwissenschaften.

It’s (not only) the economy, stupid? 165 Gutscher J.M. (1984), Die Entwicklung der FDP von ihren Anfängen bis 1961 (2nd ed.). Königstein/Taunus, Germany: Hain. Heisterkamp U. (2014), Think tanks der parteien? Eine vergleichende analyse der deutschen politischen Stiftungen. Wiesbaden, Germany: Springer VS. Kaack H. (1971), Geschichte und struktur des deutschen parteiensystems (1st ed.). Opladen, Germany: Westdeutscher Verlag. Keman H. (1994), ‘The search for the centre: Pivot parties in West European party systems’, West European Politics 17(4), pp. 124–148. Kirchner E.J. (ed.) (1988), Liberal parties in Western Europe. Cambridge: Cambridge University Press. Kirchner E.J., Broughton D. (1988), ‘The FDP in the Federal Republic of Germany: The requirements of survival and success’. In Kirchner E. J. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 62–92. Klein M. (2011), ‘Wie sind die parteien gesellschaftlich verwurzelt?’. In: Spier T., Klein M., Alemann U.V., Hoffmann H., Laux A., Nonnemacher A., Rohrbach K. (eds.), Parteimitglieder in Deutschland. Wiesbaden, Germany: VS Verlag für Sozialwissenschaften, pp. 39–59. Niedermayer O. (2006), ‘Das parteiensystem Deutschlands’. In: Niedermayer O., Stöss R., Haas M. (eds.), Die parteiensysteme Westeuropas. Wiesbaden, Germany: Springer VS, pp. 109–133. Niedermayer O. (2016), Parteimitglieder in Deutschland: Version 2016-NEU. Retrieved from www.polsoz.fu-berlin.de/polwiss/forschung/systeme/empsoz/schriften/ Arbeitshefte/index.html Oppelland T. (2015), Freie Demokratische Partei. Retrieved from www.bpb.de/politik/ grundfragen/parteien-in-deutschland/42106/fdp Pappi F.U., Seher N.M. (2009), ‘Party election programmes, signalling policies and salience of specific policy domains: The German parties from 1990 to 2005’, German Politics 18(3), pp. 403–425. Patton D.F. (2015), ‘The prospects of the FDP in comparative perspective: Rest in peace or Totgesagte leben länger?’, German Politics 24(2), pp. 179–194. Saalfeld T., Zohlnhöfer R. (2014), ‘From ‘dream team’ to ‘marriage of convenience’? An introduction’, German Politics 23(4), pp. 237–250. Schmitt-Beck R. (2014), ‘Euro-Kritik, wirtschaftspessimismus und einwanderungsskepsis: hintergründe des Beinah-Wahlerfolges der alternative für Deutschland (AfD) bei der Bundestagswahl 2013’, Zeitschrift für Parlamentsfragen 45(1), pp. 94–112. Switek N. (2015), Bündnis 90/Die Grünen. Koalitionsentscheidungen in den Ländern, Baden-Baden: Nomos. Treibel J. (2014), Die FDP. Prozesse innerparteilicher führung 2000–2012 (1st ed.). Baden-Baden, Baden-Baden: Nomos. Vorländer H. (2013), ‘Die Freie Demokratische Partei (FDP)’. In: Niedermayer O. (ed.), handbuch parteienforschung (1st ed.). Wiesbaden, Germany: Springer VS, pp. 497–507. Zur R. (2017), ‘When valence crushes: Explaining the electoral failure of the German FDP in the 2013 election’, German Politics, DOI: 10.1080/09644008.2016.1257612

9

The UK Liberal Democrats Liberalism at a crossroads Alan Wager and Tim Bale

Introduction The recent history of the Liberal Democrats is defined by their decision to enter coalition with the Conservative Party in May 2010. The formation of the Conservative-Liberal Democrat government marked the first time the party, existing in its current form since 1988, had entered national office. It was a decision widely seen as counterintuitive and led to a dramatic fall in the party’s electoral fortunes at the election of 2015, prompting profound questions about its ideological position, as well as the robustness of its organisation and support. Even if their heritage had always been more complex, the Liberal Democrats had long been seen as a party of the centre-left: was this still the case? They also liked to see themselves as organised from the bottom up, yet they accepted a top-down coalition agreement as a fait accompli and subsequently became part of the structures of government (Evans and Sanderson-Nash 2011: 459). Tensions around what a British liberal party should represent, and its electoral niche within the party system had long been elided by the ambiguity that is only really possible in opposition: In government how could hard choices be avoided and grassroots influence maintained? For Nick Clegg, Liberal Democrat leader and Deputy Prime Minister throughout the party’s time in government, the reasons for its crushing defeat in May 2015 went beyond the parochial. His resignation speech the day after the general election placed those losses in the context of a cross-national decline for liberal parties: arguing that ‘liberalism here, as well as across Europe, is not faring well against the politics of fear’ (Clegg 2015). Two months later his successor, Tim Farron, described a blueprint for recovery that both echoed the organisational tenet of ‘community politics’, while also pointing to the example of the resurgence of the Dutch D66 party as cause for optimism (Farron 2016). This chapter will provide an understanding of the Liberal Democrats’ place within this liberal party family. It will do so, firstly, by tracing the development of the party (Table 9.1), sketching out the evolution of liberalism in Britain. It will then present an overview of contemporary developments in the

The UK Liberal Democrats 167 Liberal Democrats’ policy platform. Finally, it will map out the party’s organisational structure and constitutional mechanisms. See Table 9.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-9/

Origins and development From majority government to minor party status, 1906–1945 The general election of 1906 was, on the face of it, an overwhelming Liberal triumph: the party won 400 seats to the Conservatives’ 133, an electoral landslide. But the election proved to be the Liberals’ swansong. Their success was exaggerated by a plurality electoral system that would go on to be hugely disadvantageous for the Liberal Party once it began to lose support. It was also enabled, in part, by an electoral alliance with a burgeoning Labour movement. Labour gained a parliamentary foothold in 1906 and in less than ten years’ time went on to usurp the Liberals as the default alternative to the Conservative Party in Britain. The Liberal interwar leader Herbert Samuel famously argued that ‘the abiding problem of Liberal statesmanship is to raise the enthusiasm of the working class without frightening the middle class’. The arrival of a rival party, predicated on the representation of working-class interests through the mechanism of parliamentary democracy, made this task much more difficult. Whether a liberal party could ever succeed as one of the UK’s two main parties when the dominant political cleavage was one of class is a moot point (McKibbin 1974). Internal party schisms during the World War I were driven in large part by the antagonism between two rival liberal leaders in Asquith and Lloyd George, the former much less comfortable about state action than the latter. This led to a period – as Table 9.1 shows – of fractures, fissions, and electoral decline. In 1918, factions in the Liberal parties splintered and campaigned separately: some under Lloyd George in an electoral coalition with the Conservatives, others led by Asquith seeking election independently. A continuing state of flux briefly halted when Liberal factions united over free trade in 1923, producing a brief resurgence. But, by 1931, Liberals were again hopelessly divided and the Liberal National party parted from what they saw as an increasingly left-leaning party to join a Conservative-led government. Although different accounts stress the primacy of structure or agency in explaining these changes, what is certain is that the interwar period saw a political realignment in British politics that fundamentally weakened the Liberal Party’s institutional foothold in Britain. The Liberal Party was the largest party in the House of Commons at the outset of World War I in 1914, leading a reformist and radical minority government. By the 1945 election, the party took just 9% of votes cast, and just 12 members of Parliament (MPs), less than 2% of the total, were Liberals.

168  Alan Wager and Tim Bale Liberal revival, 1945–1981 The British Party system retained this electoral shape – Labour on the left, the Conservatives on the right – for the next 30 years. So squeezed were the Liberals that their continued survival as a political force became a textbook puzzle for exponents of rational choice approaches to electoral behaviour (Cox 1997: 69–98; Duverger 1972: 23; Ware 2009). But historical legacies matter: anyone looking at the geographical spread of votes for the Liberal Democrats in 2015 is immediately struck by the similarities with the 1950s (Steed 2015). By 1955, the party’s representation had halved to six seats, and 2.5% of the popular vote. The Liberal National splinter party began to cooperate, then merge, with the Conservative Party led by Winston Churchill, himself a former Liberal politician. There was pressure for the rest of the Liberals in parliament to do the same but, because there was little attempt at strategic thinking in adapting to the party’s new circumstances (Wager 2015), that pressure was resisted by the party’s immediate postwar leader, Clement Davies. Under the leadership of Davies’s successor, Jo Grimond, who led the party from 1956 to 1967, the party’s fortunes improved markedly. Grimond halted the party’s steady movement towards political irrelevance, reversing a glacial movement to the right and, after the 1959 election, pursuing a clear commitment to the ‘realignment of the left’ (Dutton 2006). In policy terms, greater emphasis was placed on higher public investment and social equality, and a strong support for British membership of the European Economic Community (EEC), commonly known in the UK at the time as the Common Market. The party’s geographical areas of strength in rural areas of Wales and Scotland, the ‘Celtic Fringes’, was boosted by middle-class ‘progressive’ voters in English suburbs and disaffected former Conservative voters who could not bring themselves to vote Labour. However, any significant parliamentary breakthrough in a general election remained elusive. Over time, success in local government became both the Liberal Party’s raison d’être and the source of political momentum and morale. Advancing a localised or ‘pavement’ politics had electoral benefits, but detached the party from any attempt to actively define liberalism as a national creed. As Table 9.2 demonstrates, under the leadership of Jeremy Thorpe from 1967 to 1976 the party surged electorally and profited hugely from an unpopular Conservative government. A general election in February 1974 led to a hung parliament, and post-election negotiations on the prospect of a Conservative-Liberal government. While unsuccessful – the Conservative Party could not, and would not, countenance a form of proportional representation as the price of any deal – the Liberal Party’s share of the vote had shot up from 7.5% to 19.3%, though the UK’s plurality electoral system afforded it just 14 MPs in return. The nation’s ‘third party’ could reasonably claim to be closer to the other two than it had been for decades. See Table 9.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-9/

The UK Liberal Democrats 169 David Steel’s rise to the leadership in 1976, following Thorpe’s embroilment in scandal and resignation, revivified the links between the Liberal Party and Labour. Steel explicitly pronounced achieving office as a priority in his leadership election, and within a year was operating a form of ‘contract parliamentarism’ (Bale and Bergman 2006) through an explicit deal with a Labour Party operating a minority government. Though the ‘Lib-Lab Pact’ achieved little in policy terms, and appeared to have negative electoral consequences, Liberal members of parliament were once more orientated towards the benefits of office. The founding of the Social Democratic Party (SDP), the Alliance, and the creation of the Liberal Democrats, 1981–1988 The success of Liberal Party politics, since interwar realignment, had always been contingent on the fortunes of their larger competitors. In early 1981, Labour’s social democratic, ‘moderate’ wing party splintered and a new party, the Social Democrat Party (SDP), was formed. Steel had been hugely encouraged by this development, and actively encouraged the creation of a new political force that could appeal beyond the Liberal Party’s reach, whilst acting closely with it to ‘break the mould’ of British party politics and ensure electoral reform (Crewe and King 1995; Brack et al. 2015). Indeed, the Liberal Party and the SDP quickly created a comprehensive preelectoral coalition. The two parties in tandem quickly surged in opinion polls, with one survey putting what became known as ‘The Alliance’ at more than 50%. However, an economic recovery and a rapid rise in support for Margaret Thatcher’s Conservative Party meant the SDP, and the Alliance, ‘went up like a rocket [but] came down like a stick’ (Crewe and King 1995: vii). In the 1983 election the Alliance secured 25.4% of the vote, just 2% shy of overhauling the Labour Party. But this was little consolation given the Alliance had to be content with just 23 seats: 17 for the Liberals and 6 for the SDP. Steel held a strong working relationship with Roy Jenkins, the SDP’s leader in 1983. Jenkins’s replacement, David Owen, held a different conception of how that relationship should operate: at a local level cooperation was largely accepted, but tensions became increasingly apparent at the top of both parties. The 1987 general election saw no improvement on the 1983 result for either part of the Alliance, and discord at a leadership level – not least, contrasting views on the possibility of a Conservative-Alliance coalition – meant the sustainability of the Alliance was increasingly questioned. Amid the postelection disappointment, Steel instantly called for a merger of the two parties. Owen was resolutely opposed and, despite a narrow vote by the SDP membership in favour of fusion, formed a short-lived splinter party, the ‘Continuing SDP’. As a result, merger negotiations were messy – defined by division around the name, and the preliminary policies, of the new party. But in April 1988 the Social and Liberal Democrats (SLD) were created and a year later, following a name change, became the Liberal Democrats.

170  Alan Wager and Tim Bale Paddy Ashdown, ‘The Project’ and ending equidistance, 1988–1999 The task of establishing the new party, following these debilitating negotiations, fell to a new leader, Paddy Ashdown. The Liberal Democrats’ prospects looked bleak. The two parties lost almost half their members in the act of fusion. In the 1987 general election, the Alliance had secured 25.4% of the vote. From May 1988 to August 1990 the Liberal Democrats failed to reach 10% in monthly polling averages, and finished fourth in the 1989 elections to the European Parliament with 6% compared with the Greens, in third, with 15%. The most immediate existential threat was the possibility of financial bankruptcy (Ashdown 2001). Yet, not for the first time, a series of by-election victories rejuvenated the party and, in the election of 1992, the Liberal Democrats won 20 seats – not a remarkable success but a result that ensured the party’s continuation. Ashdown wanted to lead a party that believed in market forces, but considered the Conservatives too zealous in pursuing them. In a speech a month after the 1992 election, Ashdown called on his party to abandon the premise that their political position was equidistant between Labour and the Conservatives, and called on the Liberal Democrats ‘to work with others to assemble the ideas around which a non-socialist alternative to the Conservatives can be constructed’. Tony Blair, leader of the Labour Party from 1994, was receptive to the idea (pressed upon him by Jenkins) that the two party’s policies and intellectual histories were intertwined, and their electoral fates symbiotic and interdependent. But ‘The Project’, an attempt to forge close links between the two parties, was at best a partial success. It led to a joint Labour-Liberal Democrat cabinet committee, and a constitutional reform agenda that Liberal Democrat pressure played a part in achieving. Yet the possibility of a coalition agreement between the two parties – and, with it, electoral reform – receded following New Labour’s landslide victory in 1997. Ashdown, feeling let down and disappointed, resigned as Liberal Democrat leader. From Lib-Lab to Lib-Con, 1999–2010 Ashdown’s replacement, Charles Kennedy, was a politician of the centre-left whose roots were in the SDP rather than the Liberals. He was endorsed by Ashdown, but clearly had a different conception of the Liberal Democrats’ strategic role as one not of ‘constructive opposition’, but increasing differentiation and distance from Labour. This meant the carving out of a liberal identity on specific issues such as environmentalism and civil liberties. The 2001 election was not particularly fruitful for the Liberal Democrats electorally but led to a small increase in parliamentary representation. However, Kennedy’s firm and consistent opposition to the Iraq War in 2003 established him in the public consciousness and led to a groundswell of support for the party – while the Liberal Democrats

The UK Liberal Democrats 171 expected a greater return in 2005, their total of 62 seats was still the highest achieved by the party since its formation. The party’s problem was similar, though not identical, to that which had been identified by its interwar leadership: the party’s voters had the social profile of Conservative voters and the political outlook of Labour voters; it therefore had no natural base of support, at least relative to its two bigger rivals. After losing a long battle against alcoholism, Kennedy was replaced as leader by Ming Campbell, who made little impact during a period of Conservative revitalisation under the fresh leadership of David Cameron. Nick Clegg was chosen to replace Campbell in 2007, at least in part due to an overwhelming perception that he would most appeal to the electorate. This judgement appeared to be vindicated. The opportunity of leadership debates prior to the 2010 general election was one that Liberal Democrats had longed for, and gave Clegg and the party a platform on an equal footing with the other two main parties. The result was an undoubted success for the party and resulted in ‘Cleggmania’: the party’s opinion poll rating shot up above the Labour Party’s during the election campaign, although by polling day it had fallen back again and the promise of a breakthrough result went unfulfilled. The party found itself, however, a kingmaker in a hung parliament. After five days of talks with both the Conservative and Labour parties, Liberal Democrat negotiators ultimately agreed a coalition – overwhelmingly ratified by a special meeting of party members – with the Conservative Party (Adonis 2011; Laws 2012). As Table 9.3 shows, for the first time in almost a century a liberal party was holding national office in the UK. See Table 9.3 at http://partyfamilies.eu/index.php/category/chapters/ chapter-9/ The Conservative–Liberal Democrat coalition, 2010–2016 The coalition agreement with the Conservative Party was comprehensive, and five Liberal Democrats were in cabinet for the next five years. The agreement appeared to set out some key policy benefits for the party (discussed below), though the thrust of the government’s economic framework – particularly the priority, rate, and style of deficit reduction – represented a big win for the Conservatives (Bale 2012). Key to Liberal Democrat acquiescence to a coalition was a referendum on electoral reform, albeit on the introduction of the nonproportional Alternative Vote (AV) rather than the proportional Single Transferrable Vote (STV) the party had long-advocated. The proposal’s defeat by 68% to 32% was put down, at least in part, to Clegg’s growing unpopularity with voters. Coalition government had meant the abandonment of key pledges – most emblematically the trebling, rather than the promised abolition, of university tuition fees. But in truth the party’s popularity dived as soon as it entered a coalition of the centre-right, given their voters broadly perceived themselves as centre-left. It did not show any serious sign of recovery (Cowley and Kavanagh 2015).

172  Alan Wager and Tim Bale As the 2015 general election drew nearer, Liberal Democrats were torn between two strategic choices: radical differentiation from the Conservatives or proud association with the actions of their coalition. Rather than opting for one or the other, they perhaps understandably tried a bit of both – with disastrous results. Liberal Democrat representation in the House of Commons dropped by 86% – from 57 to just 8 MPs. Their vote share fell by 66%, from 23.0% to 7.9%. Apart from when the Liberal Party had fractured and splintered in 1918 and 1931, this was the largest fall in vote share for any political party, in any British general election, ever. The Liberal Democrats had previously profited from their lack of electoral and sociological definition (Cutts 2011). The ‘streak of ambiguity’ (Smith 1988:16) that runs through the European liberal family was particularly evident in a party that had relied on concentrations of regional strength, and the support of those looking to vote for ‘none of the above’ (Curtice et al. 2010: 404). After 2015, the party clearly needed to think afresh.

Ideology and policy positions The election of Tim Farron as Liberal Democrat leader was a clear shift in ideological direction for a party reeling from the consequences of coalition. Farron, as President of the Party from 2011 to 2015 but not a minister, had stood aside from and even opposed some of the biggest symbolic compromises of the coalition government. His claim that the party had been ‘muzzled’ in coalition suggested that the party had not made the best of the arrangement (Farron 2015) – a tacit acceptance, and a pointed criticism, that the Liberal Democrats had not been muscular or assertive enough when promoting liberal policy positions in inter-party negotiations. The scene seemed to be set for a swing back to the left, although whether voters could be convinced was another matter. Yet defining the party afresh means doing more than renouncing elements of the coalition’s programme in government. Any attempt to seriously review the Liberal Democrats’ policy platform was overshadowed, at least initially, by a desire to reinvigorate the membership of the party and to focus once again on upping its presence in local government – something that had been badly eroded by its years of unpopularity in coalition between 2010 and 2015. Sharp changes in the political wind have fundamentally redefined the Liberal Democrats’ task and room for manoeuvre. A tumultuous period following the 2015 general election saw the rise of Jeremy Corbyn in the Labour Party, and a referendum in which the UK voted to leave the European Union. As a result, the electoral market for the Liberal Democrats as a radical, insurgent party rapidly returning to its left-of-centre roots is limited. This meant some row-back on Farron’s rhetoric during his path to the leadership. The referendum increased the saliency and urgency of a long-standing liberal cause in the UK, internationalism and Britain’s relationship with the EU. But it also

The UK Liberal Democrats 173 demonstrated an electorate ill-at-ease with liberal solutions to the broad economic questions of the 21st century. Socioeconomic policy While the path to a Conservative-Liberal coalition in 2010 took many commentators by surprise it was soon retroactively traced to 2004, and the publication of The Orange Book: Reclaiming Liberalism (Marshall and Laws 2004). A faction, predominantly but not exclusively on the centre-right, produced an edited volume with the stated aim of disavowing ‘soggy socialism and corporatism’ (Marshall and Laws 2004: 29). For some Liberal Democrats this was a reclaiming of a classical liberalism of free markets, free trade, and limited government abandoned by the party in 1900 in favour of a ‘social liberalism’ of economic interventionism. The core thesis was that a decade of Labour government demanded a rebalancing of the economy towards a smaller state, and market-centred solutions. Yet the extent to which there is a clear arc from the Orange Book to the 2010 Coalition Agreement has been exaggerated. The ideological equilibrium of the Liberal Democrats had clearly moved. But its activist base, as well as what passed for its core, still thought of itself as broadly centre-left. A lukewarm reception for the book from Kennedy, leader at the time, was matched by an angry reaction from social liberals. A counterpublication, Reinventing the State, explicitly critiqued New Labour’s tax policies, and record on income inequality, from the left (Brack et al. 2007). It also argued sacrificing equality ‘on the altar of purist economic liberalism and the market’ would be an abandonment of the Liberal Democrat’s historical identity. This tempered a movement away from an ideological stance that had, as Kennedy was keen to point out, been fairly electorally successful for the party. As a result, the 2010 manifesto was muddled in its macro-economic thinking – splitting the difference between the two main parties, and between an increasingly stark intra-party divide. Deficit reduction had to be ‘confronted’ in a ‘fair’ way, and the Liberal Democrats’ forecast plans were both the most detailed, and the least reliant on public expenditure cuts, of the three main parties. By 2015, analysis showed their plans were more sure-footedly ‘centrist’ while retaining specific tax and spend pledges designed for coalition negotiations. Notably, a ‘Mansion Tax’ on properties worth more than £2 million; increases in capital gain tax; and, a call for the raising of the income tax threshold. The Mansion Tax was quickly abandoned in negotiations with the Conservatives in 2010, but was seen as a crucial point of agreement with Labour in 2015; increases in capital gains had been one of the party’s few clear wins in government. The raising of the income tax threshold to £10,000 quickly became (with limited success, given voters primarily saw it as a Conservative achievement) one of the party’s ‘flagship’ policies. It was emblematic of the party’s economic balancing act

174  Alan Wager and Tim Bale between social and economic liberals: ostensibly economically redistributive for the lowest paid, while also beneficial to higher-rate taxpayers, and sellable as a simplification of the tax system. The policy owed its survival in 2015 – a further incremental increase to £10,400 – and its place on the front-page of the manifesto, to these political considerations as well as to the fact that the document was fully intended to steer negotiations with the Conservative Party for the widely predicted renewal of their coalition. The act of entering coalition in 2010 had arguably cemented the position of economic liberals within the Liberal Democrats; however, the party’s catastrophic election defeat in 2015 meant their predominance was only temporary. True, in 2005 only a dozen MPs (roughly 20% of the parliamentary party) could be classified as economic liberals, while by 2010, most senior positions were occupied by that wing of the party (Quinn and Clements 2011: 70). True, too, that the willingness of the Liberal Democrats in 2010 to accept the idea that, as one of them put it, a deal with the Conservatives ‘could be capable of delivering the “tough but tender” economics which I had long believed in’ (Laws 2010: 270) had shocked Labour’s underprepared coalition negotiators (Adonis 2013). But Liberal Democrat adherents of free-market solutions were principally younger, up-and-coming MPs in 2005, and as a result of the party’s collapse in 2015 many of them disappeared from parliament. This does not mean, however, that market liberalism will now simply be swallowed by social liberalism. Farron’s first major speech on the economy as leader, in November 2015, was mindful of both the left and right of the party in its emphasis both on measures of redistribution and on ‘the entrepreneurial spirit’ of liberalism (Perraudin 2015). Religious, moral, or cultural issues Rifts over the broad shape of economic policy can be neatly contrasted to near-universal agreement within the party on issues of social liberalisation. The Liberal Democrat Party has been a consistent advocate of LGBT rights. In government this manifested itself in the passing of legislation on samesex marriage, a policy resisted by swathes of Conservative activists and the majority of the Conservative party’s MPs. Shared parental leave was also a policy principally owned by the Liberal Democrats, and sold as promoting the breaking down of traditional notions of gender roles. Driven by the party’s attempts at policy differentiation, and the freedom of opposition, there has been a hardening of support for liberal measures on moral and cultural issues. Liberal Democrats have even begun, for example, once gain to call for the legalisation of cannabis. The party’s Health Spokesperson Norman Lamb, in his speech to conference advocating the policy, argued: ‘we have all committed to the fight back. If we are to succeed, if we are to stand for something, we have to articulate liberal values’ (Lamb 2016). The promotion of liberalism has been matched by positive support for multiculturalism. While, in 2011, David Cameron spoke of an ‘active, muscular

The UK Liberal Democrats 175 liberalism’ and against a ‘passively tolerant society’, Clegg responded with a call for an ‘open, liberal society, (in which) individuals are free to live in the manner of their choosing, so long as they do not harm others’ (Cameron 2011; Clegg 2011). Within Britain, as across Western Europe, immigration has increasingly been defined not just as an economic issue but a sociocultural one, too (see, for example, Hainmueller and Hopkins 2014). The party has repeatedly made the social and economic case for migration, against the imposition of targets, and against the inclusion of students in them. Indeed, its position has increasingly become more bullish: in a book released in the wake of the EU referendum on the prospects and merits of cross-party cooperation on the left and centre-left, Farron (2016) wrote the chapter on immigration calling for parties to coalesce around ‘positive and clear language about the overall benefits we get from immigration’. Europe and ‘Brexit’ The Liberal Democrats’ support for continued membership of the European Union could be a defiantly positive statement of its internationalism, during a period where Britain’s role in Europe remains wholly uncertain. Equally, it could be the advocacy of a policy at best politically unfeasible, at worst a direct reversal of the democratic will. Following the referendum, Farron pledged to fight any future election on a pledge to reverse the process of ‘Brexit’. The party’s supporters were, unsurprisingly, the most likely to vote ‘remain’ in the referendum. Yet paradoxically key areas of geographical strength (and potential strength) for the Liberal Democrats also voted strongly for ‘leave’, while the party struggled for relevance and coverage during the referendum campaign. Certainly, the postreferendum flux in party politics – and the anger felt by some of Labour’s well-heeled and well-educated supporters towards their own party leader’s less than enthusiastic effort to keep the UK in the EU – did not lead to any significant increase in national support for the Liberal Democrats, although membership of the party rose. The election of a new leader, the veteran Vince Cable, in the wake of the party’s unimpressive performance in the post-referendum ‘snap’ election of 2017 made very little impact on voters either. Law and order/civil liberties The Liberal Democrats’ focus on civil liberties has been a consistent point of policy differentiation. During New Labour’s period in office between 1997 and 2010, the party marked out its credentials through uncompromising opposition to increases in detention without trial, and the introduction of ID cards. Parliamentary cooperation on these issues appeared to mark a growing alignment with the Conservative Party that, in part, contributed to the idea that the coalition agreement was not as incongruous as it first appeared to many (Fox 2010: 613). The Coalition agreement argued that

176  Alan Wager and Tim Bale ‘the British state has become too authoritarian, and over the past decade it has abused and eroded fundamental human freedoms and historic civil liberties’ (Coalition Agreement 2010). Civil liberty campaigners, who had long held a constructive relationship with the Liberal Democrats, became disenchanted by coalition legislation on the introduction of ‘secret courts’. Some of Clegg’s strongest attempts at differentiation from the Conservatives, as the parliament drew to a close, came in this area – notably in the party’s strong opposition to the so-called ‘Snooper’s Charter’.1 Centre-periphery issues Support for electoral reform, and the devolution and localisation of political power, has been a long-running political priority for the Liberal Democrats. But regionalism may have reached the limit of its political effectiveness in Scotland. Support for independence has become the key cleavage issue for voters in Scotland and, resultantly but reluctantly, for the political parties. Federal solutions to demands for greater regional and national control are, however, the only solution for a party that maintains opposition to an independent Scotland. The party opposes calls for an English Parliament on similar grounds, arguing it would prove destabilising for the UK. Liberal Democratic policy of regional ‘devolution on demand’ attempts, ultimately unconvincingly, to bypass calls for greater powers for national assemblies and parliaments, and the creation of a new legislature in England. It is striking that a party set firmly in favour of a federalist approach has not provided a convincing federalist answer to these problems within the British constitutional settlement (Adams 2014). Meanwhile, Tim Farron stated that a ‘massive, massive red line’ in future coalition negotiations is the automatic implementation of electoral reform for elections to Westminster. After the failed referendum on the Alternative Vote in 2011, however, persuading other parties to accede to that demand, even in the unlikely event that the Liberal Democrats are in any position to make it any time soon, will be difficult if not impossible.

Structure and organisation The Liberal Democrats are proud of their federal structure and decentralised organisation. This focus on the dispersal of decision-making is at odds with British political tradition and culture and its ‘power hoarding’ approach to democracy. In comparison to their counterparts in other British political parties, members feel that they are listened to, and have more power over policy and decision-making (Johnson 2014). It is seen as emblematic of their commitment to the distribution of power and ‘community politics’, and belief in democratic participation more generally (Russell and Fieldhouse 2004: 52). But this feeling of control is not shared to the same extent when members consider the party’s broader electoral and political strategy

The UK Liberal Democrats 177 (Johnson 2016). Adaptation in light of political developments – such as an increase in regionalisation and devolution in the UK, the party’s fluctuating electoral fortunes, and the conducting of coalition negotiations – can lead to pressure on the structures that institutionalise party members’ organisational role. This brings into question how important the party’s constitution is in dictating and defending the development and distribution of power within the Liberal Democrats. When the Liberal Democrat Party was being created, tensions around how the new constitution would operate drove disagreement and discontent. For the SDP, enshrining fundamental policy commitments was key. For Liberal negotiators, the overarching aim of those drafting the constitution was a commitment to retaining what they saw as the party’s organisational identity and integrity. The party’s structure, as laid out in the constitution (the document, in its present form, runs to more than 280 pages) clearly sets out a framework for how policies should be drawn up, how the leader of the party is elected and removed, and how candidates for parliament and devolved assemblies are chosen. Policy position formulation Party conference The supreme policymaking body of the Liberal Democrats, at least in theory, is the party’s bi-annual Federal Conference. Until 2015, delegates eligible to vote at conference were the party’s representatives and prospective candidates, along with those drawn from local parties. In September 2015, the conference adopted One Member One Vote for all registered conference attendees on matters of policy, and elections to federal committees. This was a change rationalised by attempts to reinvigorate the grassroots following the coalition, and enfranchise what was a significant surge in membership following the 2015 general election. The Federal Conference Committee sets the agenda of the conference. This is a process that has input from leaders and party spokespeople (Sanderson-Nash 2011). Radical, key votes since 2015 have included banning fracking and the legalisation of cannabis at the spring 2016 conference; both votes were preceded by campaigns on these issues by party spokespeople, rather than emanating from the conference floor. Equally, during the period of coalition, conference votes on the ‘bedroom tax’ and the use of secret courts directly contradicted government policy, and caused significant problems for the Liberal Democrat leadership. Federal Policy Committee The role of the Federal Policy Committee (FPC) is to research and develop Liberal Democrat policy in consultation with its members. Meeting monthly, it is considered the most influential and important body within

178  Alan Wager and Tim Bale the party. The committee comprises elements of each strand of the party’s federal structure: the leader; the president; representatives and MPs (as far as possible) each of the state parties of England, Scotland, Wales; elected councillors; and members drawn democratically from the grassroots party. As a result of the FPC’s composition, key policies have continued to make their way onto the party manifesto, despite not being supported by the party’s leadership and Westminster parliamentary elites. A key example was the party’s policy to remove tuition fees, which was included the 2010 manifesto in spite of key members of the leadership being keen to dilute it in light of wider fiscal commitments (Bowers 2012; Cowley and Kavanagh 2015) but was then reneged upon by the coalition. Members of the FPC were then elected, following resurgence in interest in positions on the committee, on the grounds they would act against elements of party ‘modernisation’, including the abandonment of the pledge. Clegg’s opposition as leader ensured abolition did not make it onto the party’s headline policies, but it remained party policy. The party’s constitution is clear that the FPC’s role is to enact, interpret, and present the policies decided by the party’s votes at conference. This mediating role – drilling down on the details, writing up the policy and determining the party’s priorities for general election manifestos – plays a significant role in what the party’s policy platforms look like to voters. Equally crucially, their front-page commitments in 2010 and (according to speculation) in 2015 appeared to be act as an approximate guide for what the key Liberal Democrat areas were likely to be in any interparty negotiations. The committee can downgrade policies to long-term aspirations; equally, they can be elevated to flagship commitments. Although the party’s policy process appears unambiguous, there is scope for flexibility in how the FPC’s requirement of party consultation is realised in practice. Between 1997 and 2010 the increased role for the national parliamentary party, and the gradual creation of a de facto veto over party policy was seen, in large part, to have resulted from the party’s sharp increase in parliamentary representation (Russell et al. 2007: 90–92; Sanderson-Nash 2011). The fall in 2015 should, subsequently, see a diminution of the role of MPs as policymakers. The party has long used a special group to draft the manifesto, which is now appointed principally by the party leader as FPC chair. In 2016, the prospect of an early election following a change of Conservative Party leader and Prime Minister saw the FPC set up the party’s Manifesto Group with the aim of the swift production of a manifesto – through necessity bypassing some, but not all, of the constitutional responsibility of consultation. Leadership selection The Liberal Democrats’ system of leadership selection formally gives significant power to party members. This is not surprising given the party’s history. The Liberal Party, with the election of Steel in 1976, became the

The UK Liberal Democrats 179 first major British party to elect its leader by a popular ballot; likewise, the SDP also adopted a system of One Member One Vote for the selection of its leaders. Each leader has to obtain parliamentary support above a threshold of 10%, rather than just a ‘proposer and seconder’, a change put in place in 2005. However in practice, in 2015, the 10% figure meant the support of just a single MP for a candidate to be nominated. Each candidate also requires the endorsement of 200 party members, drawn from 20 different local parties – in practice a nominal figure, but one that cements the need for initial support beyond the parliamentary party. The election then takes place under the Alternative Vote system. Although the institutional functionality of leadership elections is clear, in practice the election and ejection of leaders is something in which the will of the parliamentary party tends to dominate. As Table 9.4 shows, in three of the six Liberal Democrat leadership elections that have been conducted, the party’s choice has been between two candidates – others have dropped out before a membership vote due to a lack of backing among fellow MPs. The need for parliamentary support and confidence in the leader remains key, both in appointing and rejecting leaders. In 2005, the forced resignation of Kennedy was seen as a step-change in the way Liberal Democratic leaders were selected or removed: the party-at-large remained largely supportive but his authority drained away as close colleagues lost confidence in his ability to do the job. The choice of Campbell as Kennedy’s successor was also driven principally by the idea he would be able to command parliamentary support, and his campaign stressed his ‘continuity’, ‘professionalism’, and ‘rigour’ (Denham and Dorey 2007: 41); but when that support evaporated (owing largely to the perception that he could not compete with the younger, fresher Cameron), he was also effectively (although not with any malice) forced out by his colleagues. Conversely, Clegg’s position as leader of the Liberal Democrats remained secure, even when things looked very dark for the party during the coalition years, not least because of the lack of consistent opposition to him within the parliamentary party. Ultimately, however both parliamentary and extra-parliamentary support matters: a concerted effort to force Clegg’s resignation in 2014 ultimately failed due to a lack of concentrated support from both MPs and members (Wintour and Watts 2015; Pack 2015). See Table 9.4 at http://partyfamilies.eu/index.php/category/chapters/ chapter-9/ Candidate selection The Liberal Democrats have a constitutional clause to fight every parliamentary seat in England, Scotland, and Wales. The party’s federal structure is embedded in the selection of candidates: beyond the requirement that candidates have to pass a vetting process, the choice of candidate

180  Alan Wager and Tim Bale remains wholly in the hands of local constituency parties, provided they are large enough (more than 30 members). All members of the local constituency party have a vote in the selection of candidates. This lack of central control over the process has caused some problems of representation. In the 2010–2015 parliament a mere seven of the party’s 57 MPs were women; in 2015, all eight of the party’s MPs elected were men. Having repeatedly resisted their introduction on ideological grounds, in 2016 the Liberal Democrats conference strongly endorsed ‘All-Women Shortlists’ in any constituency an incumbent MP may will not refight (Evans 2011). Liberal Democrat membership Liberal Democrat Party members were surveyed by the ESRC-funded party membership project just after the general election in 2015: a defeat that led, as Table 9.5 demonstrates, to a relative resurgence in membership levels following a significant decline when in government.2 The average age for members was 51, which was the same as the average for party members of the UK’s six biggest parties. The average length of membership of a Liberal Democrat member was between 17 and 18 years. Some 56% of Liberal Democrats were university graduates – a figure that matches that for both Labour and Green Party members and that is considerably higher than for Conservative (38%) and UKIP (23%) members, partly because of many of the latter are older. Liberal Democrats emerged as the most middle class of all the parties’ members, with 76% of them falling into the ABC1 marketing category now routinely used by pollsters and pundits.3 See Table 9.5 at http://partyfamilies.eu/index.php/category/chapters/ chapter-9/ When asked to place themselves on a left-right scale running from 0 (very left wing) to 10 (very right-wing), Liberal Democrat members placed themselves on average at 4.1, compared with 7.7 for Conservative Party members and 2.4 for Labour Party members. Their overwhelmingly liberal attitude on moral questions was evident in the 79% of members who expressed their support for legislation on equal marriage, whereas their social, as opposed to economic, liberalism was testified to by the similarly overwhelming 75% who agreed that government should redistribute from rich to poor. Interestingly, while most were downcast by the more recent election result, it had not dimmed their faith in coalition per se: perhaps because it is the only way the party is ever likely to make it into national office, some 71% still believed coalitions were better than single-party government. Nor had the experience of some of their dearest hopes being dashed by their own leaders led to cynicism and alienation: some 85% thought the party leadership respected ordinary members and 81% felt that membership of the party had lived up to their expectations. Although about 35% of Liberal

The UK Liberal Democrats 181 Democrats would like to see members given more control over party policy compared with just 5% who though the leadership should have more; about 56% though the balance was about right. In short, the Liberal Democrat grassroots may have been disappointed by the outcome of their party going into government, but there were few if any recriminations. Liberal Democrats and social movements The relatively radical policy positions of certain subsections of the Liberal Democrats has meant that social movements and campaigners advocating environmentalism, nuclear disarmament, civil liberties, and electoral reform have tended to hold a relatively strong association with the party – for example, with their attendance and campaigning, along with the organisation of fringe events, at party conferences. Links to the party have grown stronger through disillusionment with the Labour government (1997–2010), especially on issues such as environmentalism and civil liberties where Liberal Democrats demonstrated policies that have been more attuned to social movements than to the incumbent government. Friends of the Earth, for example, promoted the Liberal Democrats as the greenest of the major parties. Predictably, these relations were strained by the party’s period in national office. Previously supportive organisations, such as the civil liberties campaign group Liberty, have become more critical of the party.

Conclusion Despite its roots lying in both liberalism and social democracy, the Liberal Democrats can be firmly placed within the European liberal party family. In policy terms, the rise of economic liberalism within the parliamentary party took place as part of an attempt to redefine liberalism in the British context. Social liberalism was an alternative conception of liberalism, rather than a clear assertion of the party’s alternative, social democratic DNA. Although this debate had a clear policy dimension it was, as Mair and Mudde (1998: 220) emphasised, about ‘what parties are, rather than … the question of what parties do’. It was an ideological debate about the party’s identity, conducted almost wholly in the language of liberalism. The party’s organisational structure is also justified and defended on the grounds that mass participation, the dispersal of power, and decentralisation are inherently liberal concepts. A tendency sometimes to obsess about organisational procedure as much as national political strategy is something leaders have fought against – a fight they appeared to be winning as the party gained greater parliamentary representation and eventually seats around the Cabinet table. The party’s return to being a party of opposition (winning just 12 parliamentary seats in 2017) and local government may, however, reverse this process of ‘professionalisation’.

182  Alan Wager and Tim Bale Britain’s parliamentary voting system has meant the Liberal Party has consistently defined itself by its centrism. The 2015 election saw the Liberal Democrats emphasise their potential moderating influence on other parties as much as any inherently liberal flavour that their presence would supposedly bring to government. This typified the electoral pragmatism that saw liberalism’s broad centre-left origins in the UK consistently emphasised through much of the Liberal Party’s history, helping to ensure that the fate of liberalism and social democracy were often seen as intertwined, even symbiotic. The logical conclusion for some was a shared party and one that leant towards the Labour Party. But the Conservative-Liberal Democrat coalition redefined perceptions of the party, leading to a shedding of its centre-left support if not its identity. As a result, the Liberal Democrats now appear to be as ‘betwixt and between’ as they ever were, at least in voters’ eyes, if not in their own view. Whether they can once again meld together social and economic liberalism in such a way as to threaten the dominance of the two big parties will, as it has throughout their history, depend as much on the fate of those parties as on their own efforts.

Notes

1. The Draft Communications Data Bill, proposed in 2013 but ultimately vetoed by Liberal Democrats in government, which would require Internet service and mobile phone providers to maintain records of each user’s Internet browsing activity for 12 months. 2. For more details, see http://esrcpartymembersproject.org. 3. ABC1 is a market research category within the NRS demographic classification system used in the UK. The NRS system categorises people as Grade A, B, C1, C2, D, and E, according to occupation. The categories ABC1 are often bracketed to equate to the middle class; C2DE to working class.

References Adams S. (2014), ‘The Lib Dems failings on the English Question’, LSE British Politics and Polity. Retrieved from http://blogs.lse.ac.uk/politicsandpolicy/thelib-dems-failings-on-the-english-question-are-a-damning-indictment-for-a-partywhich-claims-to-believe-in-a-federal-future-for-the-uk/ Adonis A. (2013), 5 days in May. London: Biteback. Ashdown P. (2001), The Ashdown diaries: 1988–1997. London: Penguin. Bale T. (2012), ‘The black widow effect: Why Britain’s Conservative–Liberal Democrat coalition might have an unhappy ending’, Parliamentary Affairs 65(2), pp. 323–337. Bale T., Bergman T. (2006), ‘Captives no longer, but servants still? ‘Contract Parliamentarism and the new minority governance in Sweden and New Zealand’, Government and Opposition 41(3), pp. 422–449. Brack D., Grayson R., Howarth D. (eds.). (2007), Reinventing the state. London: Politico’s. Brack D., Little T., Igham R. (eds.). (2015), British Liberal leaders. London: Biteback. Bowers C. (2012), Nick Clegg: The Biography. London: Biteback.

The UK Liberal Democrats 183 Cameron D. (2011), ‘PM’s speech at Munich Security Conference’, Retrieved from https://www.gov.uk/government/speeches/pms-speech-at-munich-security-conference Clegg N. (2011), ‘Nick Clegg speech: An open, tolerant society’. Retrieved from http://www.libdems.org.uk/nick_clegg_speech_an_open_confident_society Clegg N. (2015), ‘Nick Clegg resigns: The Liberal Democrat leader’s resignation speech text in full’, The Independent. Retrieved from http://www.independent. co.uk/news/uk/politics/generalelection/nick-clegg-resigns-the-liberal-democratleaders-resignation-speech-text-in-full-10235830.html Coalition Agreement (2010), The Coalition: our programme for government. London: Cabinet Office. Cowley P., Kavanagh D. (2015), The British general election of 2010. London: Palgrave Macmillan. Cox G. (1997), Making votes count. Cambridge: Cambridge University Press. Crewe I., King A. (1995), SDP: The birth, life, and death of the Social Democratic Party. Oxford: Oxford University Press. Curtice J.K., Fisher S., Ford R. (2010), ‘The British general election of 2010: The results analysed’. Paper presented at the American Political Studies Association Conference, Retrieved from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1643443 Cutts D. (2011), ‘Yet another false dawn? An examination of the Liberal Democrats’ Performance in the 2010 general election’, British Journal of Politics and International Relations 14(1), pp. 96–114. Denham J., Dorey P. (2007), ‘The ‘Caretaker’ cleans up: The Liberal Democrat leadership election of 2006’, Parliamentary Affairs 60(1), pp. 26–45. Dutton D. (2006), ‘Anticipating the project: Lib-Lab relations in the era of Jo Grimond’, Contemporary British History 20(1), pp. 101–117. Duverger M. (1972), Party politics and pressure groups. London: Nelson. Evans E. (2011), Gender and the Liberal Democrats. Manchester, UK: Manchester University Press. Evans E., Sanderson-Nash E. (2011), ‘From sandals to suits: Professionalisation, coalition and the Liberal Democrats’, British Journal of Politics and International Relations 13(4), pp. 459–473. Farron T. (2015), ‘In full: Tim Farron’s speech’, Lib Dem Voice. Retrieved from http://www.libdemvoice.org/in-full-tim-farrons-speech-transcribed-by-joanneferguson-46820.html Farron T. (2016), ‘Celebrating the movement of people’. In: Lucas C., Nandy L., Bowers C. (eds.), The Alternative: Towards a new progressive politics. London: Biteback, pp. 112–124. Fox R. (2010), ‘Five days in May: A new political order emerges’, Parliamentary Affairs, 63(4), pp. 607–622. Hainmueller J., Hopkins D. (2014), ‘Public attitudes toward immigration’, Annual Review of Political Science 17, pp. 225–249. Johnson C. (2014), ‘The importance of local parties and incumbency to the electoral prospects of the Liberal Democrats’, Politics 34(3), pp. 201–212. Johnson C. (2016), ‘Analysing the electoral incentives and obstacles to co-operation between Labour and the Liberal Democrats’, Paper presented at the Political Studies Association Conference, Brighton, UK, 21–23 March. Lamb N. (2016), ‘Keynote Speech’, 2016 Liberal Democrat Party Conference, Brighton, UK, 17–20 September. Laws D. (2010), 22 Days in May. London: Biteback.

184  Alan Wager and Tim Bale Mair P., Mudde C. (1998), ‘The party family and its study’, Annual Review of Political Science 1, pp. 211–229. Marshall P., Laws D. (2004), The Orange book: Reclaiming liberalism. London: Profile. McKibbin R. (1974), The evolution of the Labour Party, 1910–1924. London: Oxford University Press. Pack M. (2015), Opinion polls, Retrieved from https://www.markpack.org.uk/ opinion-polls/ Pedersen, M. (1982), ‘Towards a new typology of party lifespans and minor parties’, Scandinavian Political Studies, 5(1), pp. 1–16. Perraudin F. (2015), ‘Tim Farron: Entrepreneurs are natural liberals’, The Guardian. Retrieved from http://www.theguardian.com/politics/2015/nov/19/ tim-farron-entrepreneurs-natural-liberals-lib-dems Quinn T., Clements B. (2011), ‘Realignment in the centre: The Liberal Democrats’. In: Allen N., Bartle J. (eds.), Britain at the polls: 2010. London: Sage, pp. 62–88. Rallings C., Thrasher R. (2012), British electoral facts. London: Biteback. Russell A, Fieldhouse E. (2004), Neither left nor right?: The Liberal Democrats and the electorate. Manchester, UK: Manchester University Press. Russel, A., Fieldhouse, E., Cutts, D. (2007), ‘De facto Veto? The Parliamentary Liberal Democrats’, The Political Quarterly, 78(1), pp. 89–98. Sanderson-Nash E. (2011), Obeying the iron law? Changes to the intra-party balance of power in the British Liberal Democrats since 1988. Doctoral thesis, University of Sussex, UK. Sartori, G. (1976), Parties and Party Systems. Cambridge: Cambridge University Press. Smith G. (1988), ‘Between left and right: The ambiguity of European Liberalism’. In: Kirchner E. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 16–28. Steed M. (2015), ‘2015 – disaster, or darkness before dawn?’, Journal of Liberal History 88(3), pp. 69–73. Ware A. (2009), The dynamics of two-party politics. Oxford: Oxford University Press. Wager A. (2015), ‘Friends with benefits: A temporal comparison of electoral pact negotiations in the British context’, British Politics. Wintour P., Watts N. (2015), ‘The Clegg catastrophe’, The Guardian, 24 June. Retrieved from http://www.theguardian.com/politics/2015/jun/24/the-nick-clegg-catastrophe

10 Fianna Fáil In the Liberals but not of the Liberals Conor Little and David M. Farrell

Introduction: Fianna Fáil and the Liberals1 Fianna Fáil’s (FF) membership of the Liberal Europarty2 since 2009 is a recent chapter in the party’s difficult history of finding its place in the European Union’s complex of party families and groups. In the early 1980s, it formed the European Democratic Alliance (EDA) with the French Gaullists; later, it formed the Europe of the Nations (UEN) group, which it shared with several decidedly right-wing parties (O’Brennan 2010). By 2004, contacts between the Liberals and Fianna Fáil through Dick Roche (then Minister for European Affairs), Brian Cowen (Minister for Foreign Affairs and Deputy Leader of Fianna Fáil), and EU Commissioner David Byrne led to the party negotiating membership of the European Liberal Democrat and Reform Party (ELDR). However, Fianna Fáil’s Members of the European Parliament (MEPs) opposed this initiative, citing policy differences on the Common Agriculture Policy, European integration, and ‘moral issues’ (Watson 2010: 97, 109). On that occasion, in the face of party leader Bertie Ahern’s nonconfrontational style of leadership, the MEPs won out. As the UEN continued to admit MEPs with far-right views in the 2004–2009 European Parliament term, party leader Bertie Ahern faced criticism of Fianna Fáíl’s continued membership of the group. In 2006, he publicly expressed a preference for joining the Liberals, while in 2007 Dick Roche assured the Liberals that Fianna Fáil would join (Watson 2010: 173). The prospect of the UEN disintegrating during the 2004–2009 term and the formation of the European Conservatives and Reformists (ECR) group provided further impetus to the prospective move. Meanwhile, the Progressive Democrats (PDs) – a small but influential Liberal party that was a member of the ELDR – disbanded in the aftermath of its defeat at the 2007 general election. In February 2009, with Cowen as party leader, after negotiations led by Roche, and with a motion in favour of membership approved by its party conference, Fianna Fáil’s leadership announced that it would join the ELDR. The decision was again hotly contested, not least by its most electorally successful MEP and leader of the UEN Brian Crowley.

186  Conor Little and David M. Farrell This time, however, the leadership pressed ahead. Since joining, Fianna Fáil has played an active role in the ELDR: in 2012, it hosted the ELDR Congress in Dublin; during the European Parliament election campaign in 2014, the party leader highlighted the importance of Alliance of Liberals and Democrats for Europe (ALDE) membership; and Fianna Fáil has been represented among the Vice Presidents of the ALDE party group. Nonetheless, opposition to the party’s membership of ELDR/ALDE continued. After the European Parliament election in 2014, Crowley was Fianna Fáil’s only MEP, having once again achieved an extraordinary vote tally (27.4% of first-preference votes) in the four-seat constituency of Ireland South. After the election, he decided to sit with the ECR. An assiduous constituency worker without a strong profile on policy, he argued that his membership of the new group would allow him to ‘work more effectively for [his] constituents in Ireland South’ and to ‘protect some jobs of Irish staff in Brussels’ (Kelly and Beesley 2014). The latter was a reference to ALDE’s alleged refusal to give long-term contracts to two of Crowley’s staff, although ALDE claimed that it had offered to keep them on for the 2014–2019 term. More generally, the decision came in the context of very poor relations between Crowley and the party leader, Micheál Martin, who had denied him the opportunity to run in the Irish presidential election in 2011; Crowley’s senior position in the UEN, which afforded him personal and political advantages in parliament; his good personal relations with individuals involved in the ECR; and his discomfort with some Liberal policies. Fianna Fáil responded to Crowley’s decision by expelling him from the Parliamentary Party. The party argued that ALDE ‘mirror[ed] most closely the policy principles that Fianna Fáil want to highlight in the European Parliament on behalf of the Irish people’ (Fianna Fáil 2014), referring to policies on job creation and economic development in particular. It described the ECR as containing elements that are ‘racist, xenophobic… homophobic’ and eurosceptic and as having member parties and policies that ‘directly contradict core principles of the Fianna Fáil organisation’ (ibid.). Behind closed doors, the party whip described them as ‘a crowd of headbangers’ (Hand 2014). Despite their lack of representation within the ALDE group in the European Parliament in the 2014–2019 term, Fianna Fáil remained a member of the ALDE party. ALDE hailed their general election result in early 2016 (ALDE 2016) and the Fianna Fáil party leader referred to Commissioner Vestager as ‘a valued colleague of ours in the ALDE group’ (Fianna Fáil 2016). How valid is the claim that Fianna Fáil and the Liberals are a good fit? More generally, how similar is Fianna Fáil to its liberal partners? Although Lane and Ersson (1987) and Hix (1999) considered Fianna Fáil a member of the liberal party family, this view is not widely shared. One participant in the development of ELDR/ALDE observed that the party was ‘not previously seen as Liberal’, with its membership ‘bringing a more agrarian

Fianna Fáil 187 focus to deliberations and representing a victory for geographical widening over ideological deepening’ (Smith 2014: 112). A study of Fianna Fáil conducted around the time of its accession to the ELDR observed that, ‘Fianna Fáil would struggle to define itself as a liberal party’ (Hayward and Fallon 2009: 505). Nor does Bressanelli’s multidimensional policy-based predictive model of party group membership place the party in the ALDE group in the 2009–2014 term (Bressanelli 2012; correspondence with the authors). At first sight, then, Fianna Fáil presents a hard case for the ‘transnational federation’ approach to party family classification, an approach that, due to its simplicity and its incorporation of choices made by a party, is ‘one of the most widely accepted standards by which to group individual parties into party families, especially at the EU level’ (Mair and Mudde 1998: 216). The case of Fianna Fáil highlights two weaknesses of this approach identified by Mair and Mudde. First, it does not account for changing memberships over time, of which Fianna Fáil’s accession to the ELDR is an example (see also Bressanelli and Piccio 2014). Second, it does not account for inconsistency between levels: unlike the vast majority of its partners in the ALDE party, Fianna Fáil is not a member of the Liberal International. Further, although its membership of the ELDR/ALDE ‘follows the party’s own choices’ (Mair and Mudde 1998: 216), this was clearly a constrained and contested choice. This chapter examines this ‘hard case’ and in doing so interrogates Fianna Fáil’s compatibility with its new party family. It provides a description of recent ideological and organisational developments in the party and where it sits in relation to its liberal partners. Before doing so, it provides a brief overview of the party’s history to place these developments in context.

Fianna Fáil’s origins and development Fianna Fáil emerged from the short but bitter Irish Civil War of 1922–1923, which was fought over whether to accept or reject a treaty negotiated with Britain to end the War of Independence (1919–1921). The new party was formed from the antitreaty faction of Sinn Féin (We Ourselves – SF), which had been the dominant political party in the final years of British rule over Ireland, from 1918–1922. Fianna Fáil’s trajectory across the thresholds of declaration, authorisation, representation, and relevance happened in pretty short order (Table 10.1). The point of its declaration was in early 1926 when Éamon de Valera, who had led the antitreaty side, left Sinn Féin and established Fianna Fáil: at the heart of the split was the issue of whether the party should enter the Dáil (Irish parliament), the principal bone of contention being the requirement that TDs (MPs) had to swear an oath of allegiance to the British crown. Although Fianna Fáil had parliamentarians in its ranks from its foundation, it was not until August 1927 that it ended its ‘abstentionist’ policy and crossed the threshold of representation. In Sartori’s (2005: 108; and therefore Pedersen’s 1982: 7) terms, it arguably crossed the threshold of relevance before crossing the threshold of representation, as it

188  Conor Little and David M. Farrell influenced party competition even as an abstentionist party. Its relevance was confirmed when it entered government for the first time in 1932. See Table 10.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-10/ Mair and Mudde (1998) refer to party origins as one of the features of importance in categorising parties, and therefore it is worth considering the degree to which by its origins Fianna Fáil might be judged as fitting within the liberal family. On the face of it the answer would seem to be ‘No’. In its nationalist and conservative origins, Fianna Fáil shares little with liberal parties, who tended to emerge from and structure a church-state cleavage (Lipset and Rokkan 1967). By the 2009 party conference, where the party’s decision to join the ELDR was announced, Minister Noel Dempsey argued that the party had ‘led the way’ to a liberal society, but there had accumulated considerable evidence to the contrary: if anything, Fianna Fáil was – and largely remains – on the church’s side of the church-state cleavage. This tallies with the conclusions of most comparative analyses of the Irish case. The most comprehensive treatment of this is by Steed and Humphreys (1988) who distinguish two alternative models of a liberal party (more generally, see Kirchner 1988; Margulies 2015): the Anglo-Scandinavian and the Continental. The distinguishing feature of the former is that liberal parties tend to be positioned in the centre, between the Social Democrat and Conservative poles. These tend to be left-liberal in orientation and can be easily distinguished from the conservative right-wing parties. The Continental model, by contrast, tends to see an ideological overlap between Liberals and (statist) Christian Democrats on the right-of-centre of the spectrum facing the Social Democrats on the left. Steed and Humphreys have little difficulty in locating the liberal parties of most of the established European party systems. The singular exception is Ireland: ‘only in Ireland has any political party yet to identify itself as part of the liberal family’ (1988: 420). They review previous efforts by scholars to try to locate either Fianna Fáil or their principal rival, Fine Gael (FG), in the liberal camp, before concluding: ‘no useful purpose is served by a procrustean effort to force either party into the liberal mould. Doing so obscures the significant point about the Irish party system, the complete absence of anything resembling a liberal party in a country where the existence of church-state issues and the 19th-century influence of British liberalism might have been expected to produce one of either the continental or the Anglo-Scandinavian type’ (p. 421). Their explanation for this conundrum is the lack of a church-state cleavage in Irish politics,3 resulting from the overlapping interests of Catholicism as a nonestablished church in the 26 countries of (then) southern Ireland and nationalism ‘which left no place for a secular liberal nationalism’ (ibid.).4 Ironically, their analysis was published just as a genuinely liberal party – the Progressive Democrats, which contested its first general election in 1987

Fianna Fáil 189 and became a member of the ELDR – was in the process of emerging, led by former Fianna Fáil minister Des O’Malley. (As we have noted above this party has since disbanded.) From the 1930s through to very recently, Fianna Fáil was the single most important actor in Irish politics and the Irish party system, its vote collapse in 2011 somewhat reducing its influence ever since (Marsh et al. 2017). Throughout most of the history of the state party politics in Ireland could be characterised as ‘Fianna Fáil versus the rest’. The trends in Figure 10.1 speak for themselves. In the period leading up to the 2011 election, Fianna Fáil could be pretty much guaranteed a vote total of around 40% in an election, on occasions even creeping up to 50%. No other party ever came close to matching them. Fianna Fáil was the ‘party of government’. See Figure 10.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-10/ By 2011 – the point of its electoral meltdown – Fianna Fáil had held office for almost 14 consecutive years and had dominated government for 61 years of the previous 79 (see Table 10.2). Until 1989 its extensive periods in office were always single-party (albeit sometimes minority) governments. It refused to countenance coalition; it didn’t need to. See Table 10.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-10/ The party faced external stimuli for change – though more in the form of trends than any sudden developments. One of the current authors was amongst a bunch of commentators on Irish party politics in the 1980s and 1990s writing of electoral change, party system transformation, and challenges to the hegemony of Fianna Fáil (Farrell 1994, 1999; Mair 1987). The party successfully saw off these earlier challenges, which had witnessed the rise of new parties (most notably the Progressive Democrats in the late 1980s) and the (associated) increase in electoral volatility, as a result of which governments rarely survived elections and coalition government became the norm. By the early 2000s, many of these threats appeared to be waning: new parties had disappeared or were disappearing, volatility was dissipating, the new norm of coalition governments had bedded in. Having passed the Rubicon in consenting to coalesce with other parties in 1989, Fianna Fáil succeeded in maintaining its status as ‘the party of government’. Fianna Fáil’s leadership has been characterised, with a couple of recent exceptions, by long periods of stability (Table 10.2). Factionalism became a feature of intra-party politics from the 1970s, centred on the Parliamentary Party which had the power to select the leader (Farrell 1994: 232). This tendency was quelled under the tenure of Bertie Ahern (1994–2008). Amidst

190  Conor Little and David M. Farrell allegations of impropriety in his financial dealings, he handed the leadership to then-Finance Minister Brian Cowen in 2008 but Cowen’s tenure lasted less than three years, as he was forced to step down after the EU and IMF ‘bailout’ of late 2010, and immediately prior to the party’s worst electoral result ever. The major reason for the party’s long term electoral success was its ability to attract support across all class categories – a tendency that stretches back to soon after the party’s formation in the 1920s (Sinnott 1995), making it one of the first parties in Europe to manifest ‘catch all’ tendencies ‘while the phrase was still a glint in Kirchheimer’s eye’ (Gallagher 1981: 271). A lack of social bases – the key factor behind the various assessments of Irish party politics as ‘sui generis’ (e.g., Carty 1981; Whyte 1974) – has benefited Fianna Fáil most. It also manifests itself in high levels of cross-class support for Fine Gael, though that party’s cross-class appeal has largely disappeared in recent elections (Marsh and McElroy 2016). Although the parties of the left (Labour, and now also Sinn Féin and some micro parties) tended to attract votes from working-class bases, the plain fact is that between them they traditionally attracted significantly less working-class support than Fianna Fáil – a sign of the weakness of the left in Irish party politics and the strength of Fianna Fáil.

Fianna Fáil’s policy positions since 2009 Liberal parties are known to be ideologically heterogeneous (Ennser 2012; Marino 2014: 68). Conservative and Social Liberal subgroups are often distinguished and their overlap with Conservatives and Christian Democrats is frequently noted (Ennser 2012; Smith 1988; Steed and Hearl 1985). This is especially the case on the general left-right dimension (Bressanelli 2012: 744–745). Liberals do have some core characteristics on other dimensions, such as EU integration and civil liberties (Smith 2014: 110; see also Chapter 17, this volume). To describe the positions taken by Fianna Fáil while a member of the Liberals, we focus on three policy areas that are salient both to European liberals and in Ireland’s domestic politics. These are economic policy, EU integration, and policy on individual freedom, including ‘moral policy’ issues. We examine the party’s positions and developments in party policy since 2009 and we compare them with the liberal party family and with the other main party families at two points in time (2010 and 2014) during the party’s membership of ELDR/ALDE using data from the Chapel Hill Expert Survey (CHES) (Bakker, de Vries et al. 2015; Bakker, Edwards et al., 2015).5 This CHES data contain time points both before and after Fianna Fáil’s major electoral shock of 2011, which provided an impetus for a series of organisational initiatives to renew its policies while reengaging its members. It does not depend on the party’s election manifesto and this is arguably an advantage in this instance. In 2011, the manifesto was substantively different from previous and subsequent election manifestos, focusing as it did on the economic crisis that consumed

Fianna Fáil 191 Irish politics at the time; it may reflect more the circumstances in which it was drafted than the ideology of the party that presented it. Further, at the time of writing in 2016, data for the 2016 manifesto had not been published by the Comparative Manifestos Project (Volkens et al. 2016). Economic policy Economic policy has in recent decades been the most salient issue for the Fianna Fáil party (Bakker, Edwards et al. 2015) and for Irish politics, not least during the years of the ‘Celtic Tiger’ economy, until 2008, and in the economic crisis that followed. Fianna Fáil has been seen as a centre-right party on economic and social issues. During its time in government from 1997 until 2011, it was influenced by its economically liberal coalition partners, the Progressive Democrats, until their dissolution in 2009. After its heavy electoral defeat in 2011, Fianna Fáil had the opportunity as a party of opposition and the motivation, as the electorate objected to ‘austerity’, to position itself on the left-of-centre on social and economic issues, consistently emphasising values such as ‘fairness’ and protecting the vulnerable (Little and Farrell 2013: 11). Under the leadership of Micheál Martin, known within the party as a ‘social democrat’, it has had a leader with the inclination to do so. This repositioning is reflected in expert survey estimates from 2010 and 2014 (Figure 10.2). Even taking into account experts’ likely scepticism of the

Fianna Fáil Liberal Conservative Christian Democrat Radical right Agrarian/Confessional Socialist Green Radical left Fianna Fáil Liberal Conservative Christian Democrat Radical right Agrarian/Confessional Socialist Green Radical left

2010

2014

4 6 8 2 Position of the party in terms of its stance on economic issues with 95% Cls. (0 = Extreme left; 5 = Centre; 10 = Extreme right).

Figure 10.2  Positions of Fianna Fáil, the Liberals, and other party families on the economic left-right dimension Source: CHES

192  Conor Little and David M. Farrell party’s shift, this is not surprising, as it went from implementing austerity policy in government in 2010 to opposing some of the same measures in opposition from 2011. With regard to the liberal party family, the expert survey data show a shift from an estimated position to the right of the party family to a position significantly to its left. EU policy Upon joining the ELDR, Fianna Fáil’s public statements and media briefings put a strong emphasis on the party’s compatibility with the Liberals on EU integration and, conversely, on its incompatibility with Eurosceptic members of the UEN. One of the party’s lead negotiators, Dick Roche, argued that, ‘Fianna Fáil is a strongly pro-European party, and should be at the heart of Europe. The primary focus was to get into a group more reflective of Fianna Fáil’s actively pro Europe view, and at the centre of European policy and evolution.’ (Sunday Business Post 2009). Hayward and Fallon (2009) describe Fianna Fáil’s pro-EU position as ‘the primary justification’ given by party leader Brian Cowen for joining the ELDR. There was a practical advantage to being part of a pro-integration Europarty: it could provide support to Fianna Fáil during Ireland’s (relatively frequent) referendums on EU issues. After the defeat of a first referendum on the Lisbon Treaty in 2008, the party was facing a second referendum in late 2009; one source cited the prospect of receiving support from the Liberals as an advantage of joining, although Roche insisted that this was ‘not a major consideration’ (Smyth 2009a; Sunday Business Post 2009). Fianna Fáil’s pro-European position is borne out by expert survey data, which estimate that the party’s position on EU integration was similar to that of the strongly pro-integration Liberal party family (Figure 10.3). Notwithstanding this evidence, a closer examination of the party reveals some divergence from its Liberal partners. Fianna Fáil is not a ‘Euro-federalist’ party; nor does it share ‘the ideal of a united Europe’ (Smith 2014: 110). Indeed, Fianna Fáil’s continued preference for a ‘Europe of the nations’ was displayed in stark terms in October 2009, months after joining the ELDR, when its three MEPs contributed to the narrow defeat of a resolution that condemned constraints on press freedom in Italy on the grounds that the party opposed interfering in member states’ domestic politics. With the exception of one Italian MEP, all other Liberal MEPs voted together (EurActiv 2009); later, the Fianna Fáil leader was ‘harangued’ by the ALDE leadership (O’Brennan 2010). Hayward and Fallon (2009) suggest the party has a history of ‘tenuous Europeanism’, evident in its emphasis on the instrumental economic value of the EU; its difficulty in persuading its own voters to support the Nice Treaty on both occasions that it was put to a referendum in 2001 and 2002; its effective failure to campaign on the Lisbon Treaty referendum in 2008, which was defeated; and its largely domestic and electoral political motivation for activating a campaign for the second referendum on the Lisbon

Fianna Fáil 193 2010

Fianna Fáil Liberal Christian Democrat Socialist Green Regionalist Conservative Agrarian/Confessional Radical left Radical right

2014

Fianna Fáil Liberal Christian Democrat Socialist Green Regionalist Conservative Agrarian/Confessional Radical left Radical right 2

3

4

5

6

Overall orientation of the party leadership towards European integration with 95% Cls. (1 = Strongly opposed; 7 = Strongly in favour).

Figure 10.3  Positions of Fianna Fáil, the liberal party family, and other party families on the EU integration dimension Source: CHES

Treaty. Its membership of the ELDR could also be interpreted as a further episode in what they suggest is a long (pre-economic crisis) history of using EU institutions to showcase its competence as a governing party (Hayward and Fallon 2009: 503). Indeed, a central argument put in favour of joining the ELDR was strategic: the Liberals were a more influential group (at the time, the ELDR had 100 MEPs, four Prime Ministers, and seven Commissioners), while UEN membership was increasingly a domestic embarrassment to the party (see also Hayward and Fallon 2009: 505). One reason for the party’s sometimes-lukewarm Europeanism is the range of opinion within the party. Expert survey data also shows markedly more dissent within Fianna Fáil on EU integration than in the average liberal party, especially when in opposition in 2014 (not shown). This was evident within the party after the ‘No’ vote on the Nice Treaty, when the result was welcomed by senior party figures (Hayward and Fallon 2009: 500). They included one Minister – Eamon O’Cuiv, grandson of the party’s founder – who reported that he voted ‘No’; 11 years later, he was removed from the office of Deputy Leader due to his opposition to the Fiscal Compact, which was also subject to a referendum. Individual rights and ‘moral’ policy issues Despite their differences, then, Fianna Fáil and the Liberals have not been grossly incompatible on the most important dimensions in the politics of the EU: economic policy and EU integration. However, there is one dimension

194  Conor Little and David M. Farrell Fianna Fáil Liberal Radical right Agrarian/Confessional Conservative Christian Democrat Regionalist Socialist Radical left Green Fianna Fáil Liberal Radical right Agrarian/Confessional Conservative Christian Democrat Regionalist Socialist Radical left Green

2010

2014

2 4 6 8 Position of the party in terms of its ideological stance on democratic freedoms and rights with 95% Cls. (0 = Extreme left; 5 = centre; 10 = Extreme right).

Figure 10.4  P  ositions of Fianna Fáil, the liberal party family, and other party families on the Gal-Tan dimension Source: CHES

that is salient both to liberal parties and in Irish politics, on which Fianna Fáil differs considerably from its liberal partners. This relates to individual rights and especially with regard to so-called ‘moral issues’. These differences emerge in a variety of indicators in the CHES data. The most general of these is the Gal-Tan dimension (Figure 10.4), which refers in CHES’s formulation to abortion, euthanasia, and marriage equality, amongst other issues.6 In 2010 and 2014, Fianna Fáil was placed clearly to the right, in a similar position to Conservative and Christian Democratic parties, whereas the liberal party family is placed to the left (Figure 10.4). This pattern is repeated in the data on other items: civil liberties versus law and order; the role of religious principles in politics; and the party’s position on ‘social lifestyle’ issues (not shown). It is borne out by data from other sources: one measure of Fianna Fáil’s voting record on LGBT issues in the 2009–2014 European Parliament shows that its three MEPs had the most conservative record among ALDE member-parties (ScoreEP 2014). In recent years, Fianna Fáil has increasingly come to be seen as divided on these issues, with marriage equality and abortion having been the main ‘moral policy’ issues on the agenda. Before the February 2016 general election, one political journalist suggested that about half of its parliamentarians were conservative, and a minority (about one quarter), including its party leader, were ‘more liberal’, with the remainder somewhere in between (McGee 2016b). On marriage equality, there are indications that the party has attempted to ‘modernise’ its position, while on abortion it has remained more reticent.

Fianna Fáil 195 Marriage equality During the 2007–2011 Fianna Fáil-led government, under pressure from its main coalition partner (the Green Party) and a gathering movement for marriage equality, the party agreed to introduce legislation for civil partnership (e.g., Mullally 2014). Since 2011, under Micheál Martin’s leadership, and in parallel with its main competitor, Fine Gael, the party took a more liberal position. Party members passed motions in favour of marriage equality at their party conference in 2012, while at the party conference in 2013, the party leader signed a pledge supporting marriage equality (Little and Farrell 2013: 11). In 2015, Fianna Fáil imposed a whip on legislation relating to the marriage equality referendum (i.e., dissenters would be expelled from the Parliamentary Party) and several figures associated with the party – former President Mary McAleese, former candidates Noel Whelan and Tiernan Brady, and adviser Peter McDonagh – played central roles in the ‘Yes’ campaign. The campaign exposed divisions within the party. Ultimately, one Senator – Jim Walsh – resigned the whip due to his opposition to the referendum and the associated Children and Family Relationships Bill, albeit he returned to the Parliamentary Party in January 2016. Another Senator, Averil Power, described by an editorial in The Irish Times as ‘the emblem of the attempted rebranding of Fianna Fáil, of its break with the past, of its hoped-for future – liberal, iconoclastic, young, women-friendly… ’ (The Irish Times 2015) – resigned from the party after the referendum campaign for quite different reasons: she complained that the party and its leadership had not mobilised strongly enough in favour of a ‘Yes’ vote and she reported that she was laughed at by colleagues in the Parliamentary Party when she suggested that canvassers working on a rural by-election campaign should also canvass for a ‘Yes’ in the referendum. Other TDs acknowledged that the party was divided. One observed that, ‘The realpolitik of it for Fianna Fáil was that if we moved too quickly we would lose some of our constituency’, another that, ‘We are trying to accommodate a broad church’ (McGee 2015), and the party leader acknowledged that some TDs were more enthusiastic in campaigning than others. However, he also insisted that her departure was caused by the party’s candidate selection strategy in Power’s local constituency and that liberal views were widely held within the party, not just among a small subset of TDs. Abortion Ireland has had an extremely restrictive abortion law, which has been the subject of multiple referendums.7 For many in Fianna Fáil, abortion was a key issue of difference from its Liberal partners, and was frequently cited in the debate over whether to join the Liberals: in 2004, one MEP described the ELDR’s members as being ‘very, very pro-abortion’ (Humphreys 2004).

196  Conor Little and David M. Farrell It became an issue of contention almost immediately upon its accession to the ELDR: in 2009, ahead of the Lisbon Treaty referendum, the party clashed with two Liberal MEPs who signed a petition asking for the EU to make abortion legal in Ireland, Malta, and Poland, with the aim of initiating a Citizens’ Initiative. In the course of this controversy, one Fianna Fáil MEP revealed that the Liberal group had written to the party stating that it would ‘fully respect the position of [Fianna Fáil] on policies that relate to socio-ethical issues’ (Smyth 2009b). The death, in 2012, of a woman while having a miscarriage compelled the Fine Gael–led government to legislate for abortion under extremely limited circumstances (a risk to the woman’s life), with the Protection of Life During Pregnancy Act. Although the Fianna Fáil party leader Micheál Martin expressed support for the legislation, which was opposed by antiabortion campaigners, he was forced to allow a free vote due to the conservative majority in his Parliamentary Party. In the final vote on 12 July 2013, five Fianna Fáil TDs voted in favour of the Bill and 14 voted against it. In the Upper House, three Senators voted for it, ten against, and one abstained. Ahead of the February 2016 general election, Martin said that Fianna Fáil ‘would not initiate moves to repeal the 8th [Amendment to the Constitution]’ (Irish Examiner 2015), which is the constitutional basis for Ireland’s abortion regime. The party supported the status quo despite broad public support, even among the party’s rural core, for broader access to abortion (McEnroe and Baker 2015; RTÉ/B&A 2016). A survey of candidates at the 2016 general election showed some division within Fianna Fáil on whether the status quo should be made more liberal or more restrictive. Of the 22 candidates who disagreed with their parties on the issue, ten were from Fianna Fáil; five had more liberal positions than the party, and five took more conservative positions (Costello 2016). After its relative success at 2016 election, the antiabortion Pro-Life Campaign claimed that the party had benefited from an antiabortion vote and the document underpinning the ‘confidence and supply’ arrangement between Fianna Fáil and Fine Gael that emerged after the 2016 election did not mention abortion, despite Fine Gael’s policy of setting up a Citizens’ Assembly to discuss repealing the 8th Amendment. This indicated that it was too divisive for Fianna Fáil to manage (McGee 2016a).

Fianna Fáil’s structure and organisation since 2009 The Irish electoral system of proportional representation by the single transferable vote (STV) facilitates a strong localist emphasis in Irish party politics, which has been reflected in Fianna Fáil’s party organisation. STV is a candidate-based system operating within multimember constituencies, in which voters are expected to rank the candidates in order of preference. Candidates from the same party are therefore often in open competition with one another, pressing their competing claims by emphasising a focus

Fianna Fáil 197 on constituency work (Farrell 1985). This has important implications for party politics in Ireland, where traditionally the most important criterion determining a voter’s decision is a candidate orientation rather than a programmatic one (Marsh and Cunningham 2011). Fianna Fáil has been most adept at working this localist and candidate-centred system, often supported by its role in government. Assessing the state of Fianna Fáil’s grassroots organisation over time has traditionally not been easy, for want of reliable data. Farrell (1994: 229) observes that Fianna Fáil did not keep membership records and points to a ‘crude’ party estimate of 75,000 (reported by Marsh et al. 1993 in the Irish Political Studies Data Yearbook), warning that ‘these are not regarded as reliable, even by the party itself’ (ibid.). Similar estimates in the Yearbook for 1996 and 1997 are 65,000 (and 3,000 party branches), supplied by the party, with the party claiming to have 30,000 active members and the remainder ‘associate members’ (King and Gillespie 1998; King and Wilford 1997). The estimates included in the Yearbook for around the turn of the century ranged between 50,000 and 65,000 members, based on the number of party branches and the number of youth organisation members (Lyons 1999; MacCárthaigh and Totten 2001). In May 2008, the estimate of 65,000 members appears again (Hayward and Fallon 2009: 493), once again based on a fairly rough estimate. In 2009, Hayward and Fallon (2009: 494) wrote that, ‘While much has changed in the sphere of Irish domestic politics in recent years, Fianna Fáil still relies heavily on an only-slightly modified version of its original organisation structures’. Members joined the party through its cumainn (local branches); the party did not maintain a central register of members. Delegates from the local branches voted on candidate selection and on nonbinding motions at the annual party conference, but neither they nor ordinary members had the opportunity to influence major party decisions, such as entry into government. Power was concentrated in the leader, supported by the General Secretary, and the Parliamentary Party. Indeed, the predominant party model in Ireland has always been one of ‘stratarchy’, characterised in addition by the independence of the Parliamentary Party from the rest of the organisation, with a powerful role for the party leader. The principal abiding feature is that it is the public office face that predominates; there is a weak (and until recently relatively underresourced) central party office. The party’s dramatic election defeat in 2011 provided an impetus to review Fianna Fáil’s organisational structure, as part of the party’s process of ‘renewal’, and organisational changes were discussed at the 2012 and 2013 party conferences (see Little and Farrell 2013). In discussing these and other reforms, the party aimed to engage and retain members and to mend disaffection that had grown between members and the leadership. It also aimed to send out a message to voters that the party was breaking with the past and modernising.

198  Conor Little and David M. Farrell It adopted a new party constitution, which modernised many aspects of the party organisation. In 2012, the party introduced a new individual party membership model. By April 2013 it had 19,600 members (Poguntke et al. 2016). It gave members votes in local candidate selection conventions (one-member one-vote [OMOV]) in place of the system of giving votes to delegates from local cumainn (local branches). A new leadership selection process was put in place after consultation with party members in 2013, with the next party leader to be elected by OMOV. The new party rules also stated that party members would have a vote on any decision to enter government – a stipulation that may have influenced the party’s decision to support (rather than join) a Fine Gael-led government in 2016. The number of registered cumainn, which had collapsed in 2011 (to 1,010, from 1,864 in 2009 and 2,032 in 2010) began to recover, rising to 1,385 in 2012 (Fianna Fáil 2012, 2013). Fianna Fáil’s organisational reforms touched on other dimensions too. Reduced public funding resulting from a drastically reduced parliamentary contingent and reduced private funding due to the crisis-hit economy meant that its financial and human resources were challenged. It obtained revenue from its individual membership model and focused on small donations (Fianna Fáil 2012, 2013). This new model of party organisation, with its focus on individual members, seems more consistent with liberal conceptions of individual rights and democracy. How does the party compare empirically to other liberal parties? Data for the Political Party Database (PPDB) was collected partway through Fianna Fáil’s process of organisational reform and therefore reflects only part of these reforms (e.g., it does not include later reforms such as new rules on leadership selection). However, it allows a preliminary assessment of the party in relation to other liberal parties and in relation to other party families. One key dimension is intra-party democracy. Fianna Fáil seems to be less internally democratic than other member parties in ALDE, according to both measures of intra-party democracy introduced by Poguntke and colleagues (Figure 10.5; Poguntke et al. 2016). Nonetheless, given its subsequent reforms, we can expect that it has moved somewhat towards the Liberal mainstream on both measures.

Main challenges for the future At the time of writing, the main challenges for Fianna Fáil are only loosely related to its fit – or otherwise – with its adoptive party family in the EU. One measure of the task of rebuilding the party is that its February 2016 election was regarded (by it) as a triumph. Certainly, it gave pause for thought to those who had predicted its demise after 2011 (see Whelan 2016), and although this was in a context in which all the established parties lost electoral ground to new parties and other electoral forces, it was its second-worst general election result ever, after the 2011 result (see Farrell and Suiter 2016). In the wake of the 2016 election

Fianna Fáil 199 Fianna Fáil ALDE EGP PES PEL ECR EPP 0.2

0.4

0.6

0.8

Assembly-based intra-party democracy index.

Fianna Fáil ALDE PES ECR EPP EGP PEL 0.2

0.4

0.6

0.8

Plebiscitary intra-party democracy index. Members of main Europarties, 2011–2013; FF 2012. Data: Bolin et al. (2017)

Figure 10.5  Fianna Fáil and party families compared on two measures of intraparty democracy

it has positioned itself as the most important legislative prop to an extremely vulnerable minority coalition government, and – at the time of writing – this seems to be doing it no harm in terms of opinion poll trends. What are the prospects for the relationship between the Liberals and Fianna Fáil? Any episodes of dissonance must be read against the background of general compatibility on the most important policy dimensions. However, on moral issues, there is likely to be limited and uneven convergence. Although Fianna Fáil has become more liberal in recent years – its support for marriage equality would have been unthinkable until recently – it is not moving inexorably towards a socially liberal profile. The failure of a leading social conservative and former government minister, Mary Hanafin,

200  Conor Little and David M. Farrell to be elected in 2016 was significant. However, there is no shortage of conservatives among the new cohort of TDs, even among the Parliamentary Party’s youngest members. Its future leaders are at least as likely to come from among its conservatives, like Michael McGrath, as from its more modernising TDs, like Billy Kelleher. One area to watch for the future is Fianna Fáil’s position on Europe especially in the light of the 2016 Brexit vote in the UK. Given (as we discussed above) the party’s occasional bouts of ‘tenuous Europeanism’, it will be interesting to see how the party positions itself in the discussions over future relations between the EU and the UK. Political space could well open up towards the Eurosceptic end of the spectrum, not least given the grave uncertainties over how Brexit will impact on the island of Ireland. This may be compounded by the party’s differences of opinion with liberal partners on other issues, as illustrated by its current opposition to Commissioner Vestager’s position on state aid to Apple. Fianna Fáil is not the first party to have agonised over its compatibility with the Liberals (Steed and Hearl, 1985). Nor is the partnership between the Liberals and Fianna Fáil the first marriage of convenience in the European Parliament; the UK Conservative Party’s membership of the EPP present one high-profile case of many (Bressanelli 2012). Ultimately, the relationship between Fianna Fáil and the Liberals is likely to be governed more by the realpolitik of internal party politics and the party’s incentives in the European institutions, rather than by ideology alone.

Notes 1. Unreferenced information in the chapter comes from articles published in newspapers (The Irish Times, Irish Independent, Irish Examiner, Sunday Business Post) and by public service broadcaster RTÉ. Parts of the chapter draw directly on Little and Farrell (2013). 2. The European Liberal and Democratic Reform Party (ELDR) until 2012, then renamed as the Alliance of Liberals and Democrats for Europe (ALDE). 3. This fact is accepted by all political scientists who have sought to fit Lipset and Rokkan’s cleavage model to the Irish case, starting with the classic essay by John Whyte (1974). 4. They go on to note that this doesn’t preclude the potential for a secularising nonsocialist party to emerge, and they suggest that they (then emergent) Progressive Democrats could be seen as a potential candidate for this position. 5. The comparisons that follow are with liberal (and other) families as coded in the CHES, which in turn takes its lead from Hix and Lord (1997) (Bakker, de Vries et al. 2015; Bakker, Edwards et al. 2015). 6. The Gal-Tan dimension refers to the ‘position of the party in 2014 in terms of their views on democratic freedoms and rights. “Libertarian” or “postmaterialist” parties favor expanded personal freedoms, for example, access to abortion, active euthanasia, same-sex marriage, or greater democratic participation. “Traditional” or “authoritarian” parties often reject these ideas; they value order, tradition, and stability, and believe that the government should be a firm moral authority on social and cultural issues’ (Bakker, Edwards et al. 2015, codebook). 7. In 2018, the constitutional ban on abortion was removed by referendum.

Fianna Fáil 201

References ALDE (2016), Fianna Fail double their seats in the Irish Parliament, ALDE. Retrieved from https://www.aldeparty.eu/en/news/fianna-fail-double-their-seats-irishparliament Bakker R., Edwards E., Hooghe L. et al. (2015), 2014 Chapel Hill Expert Survey. Version 2015.1. University of North Carolina, Chapel Hill. Retrieved from https:// chesdata.eu Bakker R., de Vries C., Edwards E. et al. (2015), ‘Measuring party positions in Europe The Chapel Hill expert survey trend file, 1999–2010’, Party Politics 21(1), pp. 143–152. Bressanelli E. (2012), ‘National parties and group membership in the European Parliament: ideology or pragmatism?’ Journal of European Public Policy 19(5), pp. 737–754. Bressanelli E., Piccio D. (2014), ‘Fit or misfit? Italian parties in Europe’, Representation 50(2), pp. 231–244. Carty R.K. (1981), Party and parish pump: electoral politics in Ireland. Waterloo, Canada: Wilfrid Laurier University Press. Costello R. (2016, 25 February), ‘Abortion issue divides opinion in parties, survey shows’, The Irish Times,. Retrieved from http://www.irishtimes.com/news/politics/ abortion-issue-divides-opinion-in-parties-survey-shows-1.2547941 Ennser L. (2012), ‘The homogeneity of West European party families. The radical right in comparative perspective’, Party Politics 18(2), pp. 151–171. EurActiv (2009), ‘Parliament votes down EU moves on press freedom’, EurActiv.com. Retrieved from https://www.euractiv.com/section/public-affairs/news/parliamentvotes-down-eu-moves-on-press-freedom/ Farrell B. (1985), ‘Ireland: from friends and neighbours to clients and partisans: some dimensions of parliamentary representation under PR-STV’. In: Bogdanor V. (ed.), Representatives of the people? Parliamentarians and constituents in Western democracies. Aldershot, UK: Gower, pp. 237–264. Farrell D.M. (1994), ‘Ireland: centralization, professionalization and competitive pressures’. In: Katz R.S., Mair P. (eds.), How parties organize: Change and adaptation in party organizations in Western Democracies. London: Sage, pp. 216–241. Farrell D.M. (1999), ‘Ireland: A party system transformed?’ In: Broughton D. (ed.), Changing party systems in Western Europe. London: Continuum International, pp. 30–47. Farrell D.M., Suiter J. (2016), ‘The election in context’ In: Gallagher M., Marsh M. (eds.), How Ireland Voted. London: Palgrave Macmillan, pp. 277–292. Fianna Fáil (2012), ‘Renewal’. Programme for Fianna Fáil’s 73rd Ard Fheis. 2–3 March 2012. Retrieved from http://www.scribd.com/doc/83168294/Ard-Fheis-Clar-2012 Fianna Fáil (2013), ‘A fairer way to recovery’. Programme for Fianna Fáil’s 74th Ard Fheis. 26–27 April 2013. Retrieved from http://www.fiannafail.ie/content/pages/9719/ Fianna Fáil (2014), ‘Joint Statement from FF Chairman Brendan Smith and Whip Seán Ó Fearghaíl’. Fianna Fáil. Retrieved from https://www.fiannafail.ie/ joint-statement-from-ff-chairman-brendan-smith-and-whip-sean-o-fearghail/ Fianna Fáil (2016), ‘Failure to set out a coherent strategy to EU leaves Ireland without voice on crucial debates – FF Leader’. Fianna Fáil. Retrieved from https://www. fiannafail.ie/failure-to-set-out-a-coherent-strategy-to-eu-leaves-ireland-withoutvoice-on-crucial-debates-ff-leader/

202  Conor Little and David M. Farrell Gallagher M. (1981), ‘Societal change and party adaptation in the Republic of Ireland, 1960–1981’, European Journal of Political Research 9(3), pp. 269–285. Hand L. (2014), ‘Keep calm and carry on: It’s brave faces all round as party plays down shock switch’. Independent.ie. Retrieved from http://www.independent.ie/ opinion/columnists/lise-hand/keep-calm-and-carry-on-its-brave-faces-all-round-asparty-plays-down-shock-switch-30381996.html Hayward K., Fallon J. (2009), ‘Fianna Fáil: Tenacious localism, tenuous Europeanism’, Irish Political Studies 24(4), pp. 491–509. Hix S. (1999), ‘Dimensions and alignments in European Union politics: Cognitive constraints and partisan responses’, European Journal of Political Research 35(1), pp. 69–106. Hix S., Lord C. (1997), Political parties in the European Union. London: Palgrave Macmillan. Humphreys J. (2004, 28 June), FF split over joining new EU group. The Irish Times. Retrieved from http://www.irishtimes.com/news/ff-split-over-joining-neweu-group-1.1146701 Irish Examiner. (2015, 22 September), Fianna Fáil will not campaign to repeal the amendment on abortion. Retrieved from http://www.irishexaminer.com/ breakingnews/ireland/fianna-fail-will-not-campaign-to-repeal-the-amendment-onabortion-697203.html Irish Times (2015, 27 May.), The loss of power for Fianna Fáil. Retrieved from http://www.irishtimes.com/opinion/editorial/the-loss-of-power-for-fiannaf%C3%A1il-1.2226763 Kelly F., Beesley A (2014, 24 June), FF disputes MEP Brian Crowley claim over European Parliament staff. The Irish Times. Retrieved from http://www. irishtimes.com/news/politics/ff-disputes-mep-brian-crowley-claim-over-europeanparliament-staff-1.1842840 King S., Gillespie G. (1998), ‘Irish political data 1997’, Irish Political Studies 13(1), pp. 211–279. King S. Wilford R. (1997), ‘Irish political data 1996’, Irish Political Studies 12(1), pp. 148–210. Kirchner E.J. (ed.). (1988), Liberal parties in Western Europe. Cambridge: Cambridge University Press. Lane J.-E., Ersson S.O. (1987), Politics and society in Western Europe. London: Sage. Lipset S.M., Rokkan S. (1967), Party systems and voter alignments, cross-national perspectives. New York: Free Press/International Yearbook of Political Behavior Research, Vol. 7. Little C., Farrell D.M. (2013), ‘Political failure and party change? The case of Fianna Fáil’, Paper presented at the Annual Meeting of the American Political Science Association, Chicago. Retrieved from http://papers.ssrn.com/abstract=2300961 Lyons P. (1999), ‘Irish political data 1999’, Irish Political Studies 14(1), pp. 191–256. MacCárthaigh M., Totten K. (2001), ‘Irish political data 2001’, Irish Political Studies 16(1), pp. 287–351. Mair P. (1987), The changing Irish party system: Organisation, ideology and electoral competition. Recent changes in European party systems, London: Pinter. Mair P., Mudde C. (1998), ‘The party family and its study’, Annual Review of Political Science 1(1), pp. 211–229. Margulies W. (2015), ‘Liberal party performance and polarisation’, Australian Journal of Political Science 50(2), pp. 241–257.

Fianna Fáil 203 Marino B. (2014), ‘The alliance of Liberals and Democrats for Europe Group: Towards an inevitable decline?’ In: De Sio L., Emanuele V., Maggini N. (eds.), The European Parliament elections of 2014, CISE, pp. 67–72. Marsh M., Cunningham K. (2011), ‘A positive choice, or anyone but Fianna Fáil?’. In: Gallagher M., Marsh M. (eds.), How Ireland voted 2011: The full story of Ireland’s earthquake election. London: Palgrave Macmillan, pp. 172–204. Marsh M., Farrell, D.M., McElroy, G. (2017), A conservative revolution? Electoral change in twenty-first century Ireland. Oxford: Oxford University Press. Marsh M., McElroy G. (2016), ‘Voting behaviour: Continued de-alignment’. In: Gallagher M., Marsh M. (eds.), How Ireland voted 2016: The election that nobody won. London: Palgrave Macmillan. McEnroe J., Baker N. (2015), ‘Fianna Fail would not hold abortion law vote.’ Irish Examiner. Retrieved from http://www.irishexaminer.com/ireland/fianna-fail-wouldnot-hold-abortion-law-vote-355316.html McGee H. (2015, 26 May), Power’s resignation takes the wind out of Martin’s sails. The Irish Times. Retrieved from http://www.irishtimes.com/news/politics/ power-s-resignation-takes-the-wind-out-of-martin-s-sails-1.2225770 McGee H. (2016a), ‘Analysis: FF could face blow back from water charge payers’. The Irish Times. Retrieved from http://www.irishtimes.com/news/politics/ analysis-ff-could- face-blow-back-from-water-charge-payers-1.2635249 McGee H. (2016b), ‘Fianna Fáil is changing but not quickly enough’. The Irish Times. Retrieved from http://www.irishtimes.com/opinion/fiannaf%C3%A1il-is-changing-but-not-quickly-enough-1.2498638 Mullally U. (2014), In the name of love: The movement for marriage equality in Ireland. Dublin: History Press. O’Brennan J. (2010), ‘Ireland and the European Union: Mapping domestic modes of adaptation and contestation’. In: Hogan J., Donnelly P.F., O’Rourke B.K. (eds.), Ireland: Business and society: Governing, participating & transforming in the 21st century. Dublin: Gill & Macmillan, pp. 379–397. Pedersen M.N. (1982), ‘Towards a New Typology of Party Lifespans and Minor Parties’, Scandinavian Political Studies 5, pp. 1–16. Poguntke T., Scarrow S.E., Webb P.D. et al. (2016), Party rules, party resources and the politics of parliamentary democracies. How parties organize in the 21st century. Party Politics 22(6): pp. 661–678. RTÉ/B&A. (2016), RTÉ/Behaviour & Attitudes 2016 General Election Exit Poll Report. Sartori G. (2005), Parties and Party Systems: A Framework for Analysis. Colchester: ECPR Press. ScoreEP. (2014), ScoreEP. Retrieved from http://score-ep.org/lgbt.html#bar_country Sinnott R. (1995), Irish voters decide: Voting behaviour in the elections and referendums since 1918. Manchester, UK: Manchester University Press. Smith G. (1988), ‘Between left and right: the ambivalence of European liberalism’. In: Kirchner E.J. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 16–28. Smith J. (2014), ‘Between ideology and pragmatism: Liberal party politics at the European level’, Acta Politica 49(1), pp. 105–121. Smyth J. (2009a), ‘FF expected to apply to join Liberals in EU Parliament’. The Irish Times, 28th February. Retrieved from http://www.irishtimes.com/news/ ff-expected-to-apply-to-join-liberals-in-eu-parliament-1.711518

204  Conor Little and David M. Farrell Smyth J. (2009b), ‘Irish MEPs criticise Liberals’ support for abortion’. The Irish Times, 31st July. Retrieved from: http//www.irishtimes.com/news/irishmeps-criticise-liberals-support-for-abortion-1.709858 Steed M., Hearl D. (1985), Party families. An examination of the evidence gathered by political scientists on the characteristics and kinship patterns of Liberal and related parties in the European Communities. London: Liberal European Action Group. Steed M., Humphreys P. (1988), ‘Identifying Liberal Parties.’ In: Kirchner E.J. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 396–435. Sunday Business Post (2009, 15 March), ‘FF to join liberal Euro group next month’, Sunday Business Post. Retrieved from http://www.lexisnexis.com.ep.fjernadgang.kb.dk/ lnacui2api/api/version1/getDocCui?l ni=7V79-C551-2RR9- G55F&csi=270944,270077, 11059,8411&hl=t&hv=t&hnsd=f&hns=t&hgn=t&oc=002 40&perma=true Volkens A., Lehmann P., Matthieß T., Merz N., Regel S., with Werner A. (2016), The Manifesto Data Collection. Manifesto Project (MRG/CMP/MARPOR). Version 2016a. Berlin: Wissenschaftszentrum Berlin für Sozialforschung (WZB). Watson G. (2010), Building a Liberal Europe: The ALDE Project. London: John Harper. Whelan C., Russel H., Maître B. (2016), ‘Economic Stress and the Great Recession in Ireland: Polarization, Individualization or ‘Middle Class Squeeze’?’, Social Indicators Research: An International and Interdisciplinary Journal for Quality-ofLife Measurement 126(2), pp. 503–526. Whyte J.H. (1974), ‘Ireland: politics without social bases’. In: Rose R. (ed.), Electoral behavior: A comparative handbook. New York/London: Free Press/Collier Macmillan, pp. 619–651.

11 Nuanced liberalism The weakness of liberal parties in Spain Lidia Núñez

Introduction In 1833, the Austrian Prince of Metternich stated that ‘[t]he Spanish character is incapable of nuance, for which it is meaningless to speak of a liberal Spain’ (Marichal 1995; cited in Millán and Romeo 2004: 284). Liberalism is a multifaceted concept. The political development of Western countries is so intertwined with liberalism that their political systems are often referred to as liberal democracies. In these countries, some of the most important political ideologies are rooted in liberal postulates. As such, beyond slogans and party names, identifying which parties represent a liberal ideology remains a difficult task. However, what is beyond doubt is that in some countries liberal parties have had – and sometimes still have – an important role, whereas in other countries there are virtually no liberal parties or they are located at the margins of the political system. This chapter focusses in one of these countries in which Liberal parties have traditionally been very weak or even mostly nonexistent: Spain. Since the fall of the Francoist regime, there has not been a single major Liberal party, with the exception of Unión de Centro Democrático (Union of Democratic Centre – UCD), which, although sometimes referred to as a liberal party, was above all an ideologically ill-defined party of notables from the old regime. Several liberal parties have existed in Spanish politics, but only those that have emerged at the beginning of the 21st century have managed to gain significant parliamentary representation on their own: Ciudadanos (Citizens – Cs) and Unión, Progreso y Democracia (Union, Progress, and Democracy – UPyD). The Liberal parties that existed before tended to merge into larger parties, such as UCD and Alianza Popular (People’s Alliance – AP), later on called Partido Popular (People’s Party – PP). In this chapter, we argue that it is precisely the nuances within the liberal ideology that can explain the weakness of liberal parties in Spain. In this sense, it is important to note that liberalism as an ideology is not alien to Spanish citizens. In a preelectoral survey conducted in 2011, citizens that declared having an ideological preference were then asked which ideology they felt closest to. Among the respondents, liberalism appeared as the third most preferred option (18%), after socialism (24%)

206  Lidia Núñez and conservatism (19.6%). These figures have been stable over the past decade and nothing indicates that they were very different before. If the liberal ideology is so widespread among the population, what can explain the traditional weakness or the nonexistence of liberal parties in Spain? This question can be addressed from two perspectives: from the demand-side (voters) and from the supply-side (parties). From a demand-side perspective, surveys show that there seems to be an electoral potential large enough to guarantee the viability of a liberal party. However, the question to be answered is whether this electorate represents a homogenous group of voters. From a supply-side perspective, liberal parties’ weakness might be explained by the fact that parties of other ideologies might include liberal ideological stances in their positions and manifestos, hence attracting liberal voters. This chapter addresses these questions by focussing on the two new liberal parties, UPyD and Ciudadanos, through a comparison between the positions of liberal voters and the political parties. In doing so, several dimensions are covered. First, a brief historical overview of the evolution of liberal parties in Spain shows how, in spite of the importance of liberal ideas in Spain, these have not contributed to the existence of strong liberal parties. Second, the chapter focusses on the contemporary Spanish political arena. It shows the main features of the liberal electorate and disentangles the extent to which they present a distinct political group. In so doing, the chapter analyses three dimensions that have played a key role in Spanish politics since the 19th century: economic liberalism, political liberalism, and the centre-periphery cleavage (Lipset and Rokkan 1967). Third, the chapter analyses the extent to which liberal voters’ preferences are represented by the current political offer. Spanish political parties’ positions differ on economic and political liberalism and on decentralisation and nationalism. The analysis shows that the 21st-century liberal political parties constitute a real novelty in Spanish politics, by representing a political ideology that contains for the first time a mixture of economic and political liberalism, while opposing further decentralisation. Next, an overview of the internal organisation of the two new liberal parties is presented. Finally, the last section of the chapter concludes by discussing the future challenges for these parties in terms of the development of their organisational structures, their survival in a highly restrictive electoral system, and their high degree of personalisation.

Origins and development: a story of permanent weakness Some works have analysed the weakness of liberalism in Spain from a historical perspective, in terms of the development of liberal political institutions (e.g., Millán and Romeo 2004; Salvadó 2003). Scholars have pointed out that, for most of the 19th and 20th centuries, this weakness can be attributed to the Spanish society’s agrarian social structure during the 19th and most of the 20th century, to the dissociation between political and economic

Nuanced liberalism 207 powers, to the strength of religiosity, or even to the political influence of the Catholic Church. However, the relative weakness of liberal political institutions does not entail that liberalism as an ideology did not have a leading role in Spanish politics since the beginning of the 19th century. Indeed, liberalism has permeated the evolution of Spanish political systems and institutions and has marked the character and nature of political struggle ever since. From the 19th century to democracy: a story of changing landscapes The term ‘liberal’ was used in politics for the first time in Spain in the context of the Peninsular War (also named the Spanish Independence War1) when the Courts of Cadiz promulgated the first Spanish Constitution2. It was the first time sovereignty emanated from the nation, which in practical terms put an end to feudalism. The separation of powers and the respect for individual liberties were also included in the Constitution. Also for the first time, universal male suffrage was introduced and male citizenship was considered to be universal, hence considering in equal terms those born in Spain and those born in the colonies. This Constitution did not last for long: it was restored several times through the 19th century in a context of intermittent restorations of the Ancien Regime. During the reign of Isabel II, the Liberals split in two branches that had a leading role in Spanish politics until the end of that century: the moderate liberals (Partido Moderado – Moderate Party) and the progressive liberals (Partido Progresista – Progressive Party). The former were primarily composed of ‘big landowners, the big traders and the intellectuals’, whereas ‘the low bourgeoisie’ was incorporated into the later with ‘a more egalitarian and exalting vision of the middle classes’ (Marcet Morera 2000: 185). By the end of the century, the two liberal movements merged again in response to the ideologies that were increasingly spreading among the working class: socialism and republicanism. After a short revolutionary period (1868–1874), from the last decades of the 19th century until 1923, the Spanish political system is known as the Bourbon Restauration. It was characterised by the alternation of two political forces. On the one hand, the Partido Liberal (Liberal Party) led by Práxedes Mateo Sagasta constituted the continuation of the former progressive liberals. On the other hand, the Partido Moderado, which was the successor of the former moderates, was led by Cánovas del Castillo. Although elections were regularly organised, these were not competitive. Alternation in power of these parties was based on electoral fraud and agreements between elites in a system that was called el Turno (or ‘The Turn’) whose aim was to ensure political stability. From 1923 to 1930, the dictatorship of Primo de Rivera put an end to the Bourbon Restauration, ‘el Turno’, and the constitutional order. A period known as the Second Republic started just after Primo de Rivera’s

208  Lidia Núñez authoritarian regime. The republican system brought the holding of regular and competitive elections. The most important actor representing liberal ideology at the time was the Derecha Liberal Republicana (Liberal Republican Right) led by Niceto Alcalá Zamora, who was the first Prime Minister of the new Republic for some months and, later, President of the Republic (1931–1933). The Second Republic finished with the coup d’état that occurred in 1936 and that led to the 40-year long Francoist dictatorship. After the death of the General Franco, transition to democracy was characterised by a period of huge political instability in terms of political supply: Many political parties emerged and most of them did not last long (Table 11.1). Among newly emerged liberal parties, the most notable ones were the Partido Demócrata Popular (People’s Democratic Party) founded by Ignacio Camuñas, the Federación de Partidos Demócratas y Liberales (Federation of Democratic and Liberal Parties) of Joaquin Garrigues Walker and Antonio Fontan, and the Partido Liberal (Liberal Party) of Enrique Larroque. These parties, together with other parties from the conservative side of the political spectrum, merged into the Union de Centro Democratico (UCD), which was the largest party in Spain during the first and second democratic elections. The leaders of the liberal parties occupied important positions while UCD was in government (1977–1982). Nonetheless, UCD only managed to have a short-lived success, to the extent that it was soon dismantled. Its leader and former Prime Minister (1976–1982) Adolfo Suarez founded the Centro Democrático y Social (Democratic and Social Centre) in 1982 and became president of the Liberal International in 1989. This new party was never able to reach the shares of votes previously won by UCD. The Centro Democrático y Social obtained its best results in 1986 (9.2% of the votes) and in 1989 elections (7.9%), but in the following elections it never managed to win more than 2%. A key aspect to understand the history of liberal parties in recent decades in Spain is the role played by the Partido Popular in attracting in its ranks the small Liberal parties that had (re-)emerged after the end of the Francoist regime and after the dismantlement of UCD. During the years of transition to democracy, conservative politicians had difficulties in marketing their ideas to a society that was looking for change in politics. The mismatch between conservative elites coming from the backbenches of the authoritarian regime and the Spanish society explains the early success of UCD, a party of local notables without a solid structure and ideologically ill-defined compared with other options such as Alianza Popular or the Partido Socialista Obrero Español (Spanish Socialist Workers’ Party – PSOE). However, as time passed by, and as UCD was in clear decline because of internal conflicts between competitive factions, the leadership of Manuel Fraga Iribarne consolidated in Alianza Popular. Some small liberal parties existing at the time – Union Liberal3 and the Partido Liberal – merged into Alianza Popular first as a federation of parties and then as a real fusion

Nuanced liberalism 209 Table 11.1  Main events related to liberal parties since the transition to democracy Year of birth 1974 1976 1976 1976 1977 (1879) 1977 (1895) 1977 1977 1977

Party birth Partido Demócrata Liberal (Camuñas) (Liberal Democratic Party) Partido Popular (People’s Party) Federación de Partidos Democratas y Liberales (Federation of Democratic and Liberal Parties) Partido Liberal (Enrique Larroque) (Liberal Party) Partido Socialista Obrero Español (Spanish Socialist Workers’ Party) Partido Nacionalista Vasco – Euzko Alderdi Jeltzalea (Basque Nationalist Party) Union de Centro Democrático (Union of the Democratic Centre)

1978

1978 1982 1982 1983 1984 1984

Centro Democrático y Social (Democratic and Social Centre) Partido Demócrata Popular (People’s Democratic Party) Unión Liberal (Schwartz) (Liberal Union) Partido Reformista Democrático (Democratic Reformist Party)

1986

1987 1989 1996 2006 2007

Ciudadanos, partido de la ciudadanía (Citizens, party of the citizenship) Unión, Progreso y Democracia (Union, Progress and Democracy)

Other important events

Elections for the Constituent Assembly (Legal) foundation of the Federación of Partidos Demócratas y Liberales (Garriques Walker) Beginning of the alliance between Convergència Democràtica de Catalunya (CDC) and the Unió Democràtica de Catalunya (UDC) Federación de Partidos Demócratas y Liberales and the Partido Liberal support UCD

The Partido Liberal joins Alianza Popular (former PP) The Union Liberal merges to the Partido Liberal (Larroque) The Partido Liberal runs in the general elections together with Alianza Popular and PDP in the Coalición Popular The Partido Liberal abandons Coalición Popular IX Congress of Alianza Popular: creation of Partido Popular and definitive merging of the liberal parties The Partido Popular wins a general election for the first time

210  Lidia Núñez within the party. Alianza Popular adopted a new structure in 1989 and this was translated into a change of name: Partido Popular. The small liberal parties that had existed until then merged into this new political party at what was call the Congress of Re-foundation4. This ‘Re-foundation’ was characterised by a much more liberal and prodecentralisation positions in sharp contrast with the conservative, centralist and interventionist positions that the party had defended until then (Garcia-Atanze 1990: 229). The aim was to transform the party into a political space where liberal, conservative, and Christian democratic ideas could coexist. The impact of this change and the subsequent evolution of the party on these positions will be presented further on in this chapter, but it suffices to say for the moment that the changes in the PP’s positions on some of the key elements of liberal ideology can help to understand the emergence of the new liberal parties: UPyD and Ciudadanos. Another brief attempt to build a liberal party occurred just before the 1986 elections: the Operación Roca (Operation Roca). The operation was named after Miquel Roca, who was at the time the spokesperson of the Catalan Convergencia I Union (Convergence and Union – CiU). Together with Antonio Garrigues Walker, he launched the Partido Reformista Demócratico (Democratic Reformist Party – PRD). Their aim was to build a strong liberal party that could act as hinge party in Spanish politics. However, the severe defeat in the 1986 elections led to the disappearance of the party. Other parties in Spain have also received the label of liberal, but the degree to which their positions are based in the liberal ideology is, to say the least, unclear. A notable example of this lack of definition is Convergencia Democràtica de Catalunya (Democratic Convergence of Catalinan – CDC; but Democràcia i Llibertat – Democracy and Liberty in the 2015 general elections). This party is a member of the Liberal International and of the Alliance of Liberals and Democrats for Europe (ALDE) Party. Clearly positioned as a centre-right nationalist party (Giordano and Roller, 2002), its political manifesto cannot be considered to be rooted in liberal ideology, nor could the majority of its voters be labelled as liberals (Barberá and López 2006). Another party that is sometimes classified as Liberal is the Partido Nacionalista Vasco – Euzko Alderdi Jeltzalea (Basque Nationalist Party). Although it is a member of ALDE, it is more often classified as a centre-right nationalist party (Pallarés and Keating 2003: 242; Gillespie 2000). The 21st century: a new breath of hope for the liberals? In the first decades of the 21st century, two liberal parties emerged in Spain (for main events, see Table 11.1). They were both born in the context of opposition to peripheral regionalism: first, Ciudadanos in Catalonia and, then, UPyD in the Basque Country. The leaders of these parties justified their creation to better represent nonnationalist voters that were according

Nuanced liberalism 211 to them not being effectively represented by the political offer available at the time. As such, they were born around some of the most salient and long-lasting political issues since transition to democracy in Spain: federalism and the political weight of nationalist parties in national politics. The Spanish decentralisation process has been unequal in the sense that not all regions have the same financial systems, most notably the Basque Country and Navarre, and the devolution of competences to the subnational level has not been done at a similar pace across regions. This situation was meant to be more rebalanced after 2001 with the transfer of education and health systems to all the regions and the approval of a new law on regional finances. However, the Basque Country and Navarra financing systems are different from the rest of the regions. Furthermore, there are still large discrepancies in the degree of self-government of other competences (e.g., police). Political tensions over the state financing system of regions have been a permanent feature of the Spanish political debate. In this context, a group of citizens created in Barcelona in 2005 a civic platform to express their opposition to Catalan nationalism, which was the seed of the creation in 2006 of Ciudadanos as a political party. The political objectives of the party echoed those of the civic platform and were fixed in the party statutes: ‘the party will defend citizens’ common interests beyond territorial differences and the secularism of public authorities in identity issues’5. They ran for the first time in the 2006 Catalan elections, winning three seats. In 2008, the party ran for the general elections, obtaining barely 0.2% of the total vote share. Figure 11.1 displays the evolution of the vote shares of the main parties since 2008 in the general and European elections. The electoral results of Ciudadanos started an upwards trend in the 2009 European elections; the party achieved nearly 14% of the vote in the 2015 elections, which made it the fourth largest statewide political force. See Figure 11.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-11/ In September 2007, Rosa Díez, a former member of the PSOE, founded Unión Progreso y Democracia together with a group of intellectuals, after a major disagreement over the antiterrorist policies of the PSOE that was back then in government. The party was close to antiterrorist civic associations, such as Basta Ya!, as well as to associations of victims of terrorism. The newly created party had a clear social liberal position and was opposed to regional nationalisms. Díez clearly stated that the party’s aim was to attract both right-wing liberals and left-wing voters. The party manifesto was structured around the defence of ‘the equality and liberty of all Spanish citizens within an efficient and viable State, free from the pressure of nationalist parties’. It defended ‘a strong State with a clear division of competences’ and in which regions would have the same competences and the same financial system.6 UPyD ran for general

212  Lidia Núñez elections for the first time in 2008, winning 303,246 votes (1.2%) and one seat in the Congress of Deputies. The party reached its best results in the 2014 European elections (6.5% of the vote and four seats). Surveys have since then showed a declining trend, to the point of not winning any representation in the 2015 general elections. These two parties aimed at attracting a sector of the Spanish electorate characterised by an opposition to decentralisation and a weak defence of regional identities, the defence of an liberal economic agenda and secularist stances. These similarities across party manifestos and positions triggered several proposals from the side of Ciudadanos about the convenience of running jointly to elections since 2008. However, UPYD has so far always refused these proposals. UPYD had developed quicker than Ciudadanos a statewide offer able to compete across the different electoral districts. However, several factors can account for the fall of UPYD in the 2015 general elections. First, the 2008 economic crisis was accompanied by an increase of electoral fragmentation and electoral volatility. Voters severely punished the two largest parties, PP and PSOE, for not being able to redress the economic situation. A part of the PP’s electorate switched their votes to Ciudadanos instead of UPYD. The latter was perceived to be less competitive due to its internal conflicts and the declining strength of Diez’s leadership. Between the 2014 European elections and the 2015 general elections, many of the party’s elected representatives abandoned UPYD and joined Ciudadanos, and so did a part of the party’s grassroots members. Although the context of a severe economic crisis might account for part of the evolution of these parties, the objective of this chapter is also to shed light on the degree of coherence between Liberal parties and liberal voters. Is the growth of Ciudadanos in the last general elections based on the attraction of liberal ideas? Or is it just a temporary electoral consequence of the economic crisis? To address this question, we now analyse the profile of the Spanish liberal voters.

Ideology, policy positions, and voters: who attracts the liberal voters? The ideological preferences of Spanish citizens in terms of their closeness to ideological families have two main features: stability and consolidation. There have not been profound changes in the past decade, even in a context of abrupt changes in the political scene witnessed during the economic crisis that started in 2008. Figure 11.2 presents the evolution of the ideology of Spanish voters. The figure shows that there is a clear stability in the distribution of ideological preferences between three ideological families: liberalism, left-wing ideologies (social democracy, socialism, and communism), and right-wing ideologies (Christian democracy and conservatism). There are more fluctuations

Nuanced liberalism 213 45% 40% 35% 30% 25% 20% 15% 10% 5% Apr-09 Jun-09 Aug-09 Oct-09 Dec-09 Feb-10 Apr-10 Jun-10 Aug-10 Oct-10 Dec-10 Feb-11 Apr-11 Jun-11 Aug-11 Oct-11 Dec-11 Feb-12 Apr-12 Jun-12 Aug-12 Oct-12 Dec-12 Feb-13 Apr-13 Jun-13 Aug-13 Oct-13 Dec-13 Feb-14 Apr-14 Jun-14 Aug-14 Oct-14 Dec-14 Feb-15 Apr-15 Jun-15 Aug-15 Oct-15

0%

Conservatives and Christian-Democrats

Liberals

Left-wing ideologies (socialdemocrats, communists, socialists)

No ideology, don’t know and no answer

Figure 11.2  Evolution of voters’ self-declared ideology in Spain (in %) Note: Citizens were asked which ideology they felt closest to. Answer categories: Conservatism, Christian democracy, Liberalism, Progressivism (included in surveys in 2009 and 2010), Social Democracy, Socialism, Communism, Nationalism, Feminism, Environmentalism, Other, No ideology. Source: Surveys administered by the Centre for Sociological Research, 2009–2015

around those declaring themselves as apolitical or not responding to the question. Ideological preferences are therefore consolidated around the three ideological blocs. This stability leads us to question whether they constitute a homogeneous electorate. Profile of liberal voters The most important feature of this group of voters is its nonspecificity: it shows nonsignificant differences in almost any of the sociodemographic characteristic when compared with the rest of the electorate. This lack of specificity is also translated into a centrist position on the left-right scale, as displayed in Figure 11.3. Nonetheless, this apparent centrism does not entail that this electorate is not significantly different from the rest of the voters in terms of ideological preferences, especially when it comes to three main dimensions: two of them are related to the main facets of liberalism; that is, political and economic liberalism, and the third refers to the one of the most important cleavages in Spanish politics, the centre-periphery cleavage. Survey data shows that liberal voters in Spain distinguish themselves from the rest of the electorate on these three dimensions (see Table 11.2). Liberal voters are more in favour of reducing taxes even to detriment of the provision of public services, which goes in line with traditional liberal economic stances. There are also significant differences between liberal and nonliberal voters in terms of secularism as feature of political liberalism. To a question asking whether homosexual couples should be authorised to adopt children, liberal voters show a statistically significant

214  Lidia Núñez 50% Rest of the voters

40% 30%

25,3% 17,7% 16,5%

20% 10%

11,5% 4,9%

7,4%

6,3%

5,7%

2,5%

2,2%

1,7%

1,0%

9

10

0% 50% Liberal voters

40% 33,3% 30% 20% 12,4% 10% 3,0%

4,1%

1

2

15,4%

14,3%

9,4%

5,3%

0% 3

4

5

6

7

8

Figure 11.3  Left-right position of liberal voters versus rest of the electorate Note: Preelectoral study carried out in 2011. Study 2.915. Centro De Investigaciones Sociológicas

greater support for it compared with the rest of the electorate. This group of voters also shows a statistically significant more negative view on decentralisation. Ideological positions of parties In the light of these data, the question that might be asked is why liberal parties have not managed to achieve a greater electoral success. To look into this question, one needs to broaden the scope of the analysis and look into the positions of the main competitors in Spanish political arena: the PSOE and, especially, the Partido Popular. These two parties have concentrated a very large part of the votes for most of the recent history of Spanish democracy. Between the 1993 and 2011, both parties managed to attract more than 70% of the votes. The evolution of the positions of these two parties can help to explain the traditional weakness of liberal parties as well as the emergence of the new liberal parties, UPyD and Ciudadanos.

Nuanced liberalism 215 Table 11.2  Positions among liberal voters and the rest of the electorate N In favour of reducing taxes even if that entails a reduction in public services In favour of a greater decentralisation

Rest of the 4,885 electorate Liberal voters 771 Rest of the 4,914 electorate Liberal voters 785 Against the adoption of children Rest of the 4,854 by homosexual couples electorate Liberal voters 787

Standard Mean Mean deviation difference 4.2

2.4

0.2*

4.4 2.7

2.4 1.1

−0.1*

2.6 4.0

1.0 3.6

−0,6***

3.5

3.2

Source: Centro De Investigaciones Sociológicas (2011), preelectorial study 2.915 (N = 6,082) Note: * p < 0.05, ** p < 0.01, *** p < 0.001. The answer categories for the items on taxes and on the adoption of children are formulated are Likert scales ranging from 0 to 10, where 10 means complete agreement with the statement. The item on decentralization has five categories in which lower values indicate a more favourable position towards centralization.

In this sense, the evolution of the PP is especially significant. After the Re-foundation of the party in 1989, one of the most significant features of the Partido Popular has been its ability to unite very different views across the right-side of the political spectrum. It could be described as representative of the new right, which in Heywood’s terms ‘is a marriage between two apparently contrasting ideological traditions’ (2012: 93): economic liberalism and tradition-based authoritarianism. Classic economic liberalism, or neoliberalism, is based on a strong defence of free market and small state. Positioned as very liberal on the economic dimension in its beginnings, the PP’s positions have moderated over time (Montero and Clavo 2000). The Partido Popular has included strong arguments in favour of the maintenance of the welfare system in its party manifesto. However, this is not to be mistaken for closeness to social liberal positions, such as the urge to compensate for the failures of the system to provide equal opportunities to every citizen. The most important privatisations in Spain’s recent history were made during the Aznar’s governments (1996–2004).7 With the exception of the small liberal parties that emerged in the 21st century, UPYD and Ciudadanos, no other party has shown a greater commitment to economic liberalism and social conservatism than the PP. The second important dimension of the new-right parties is related to tradition and a certain degree of authoritarianism. Some authors refer to this type of ideology as neoconservatism. Historically close to the positions of the Catholic Church on moral issues, neoconservatism is characterised by a strong defence of ‘the family’ understood in traditional terms as the basis of social cohesion and moral values (Heywood 2012: 98). The Partido Popular can be seen as an example of this kind of party. Not surprisingly, it has traditionally been very successful in attracting religious voters (Montero and Calvo 2000).

216  Lidia Núñez 160 140 120 100 80 60 40 20 1989

1993

1996

2000

2004

2008

2011

Figure 11.4  Evolution of the positions of the Partido Popular on free market economy: Percentage difference with respect to the average number of favourable mentions on the topic in the period 1989–2011 Note: Calculations based of the Markeco variable of the Manifesto Project Dataset: a composite indicator of the number of favourable mentions on free market economy and economic orthodoxy (variables 404 and 414). For more information, see Volkens et al. (2016a).

It has fiercely fought against homosexual marriage, adoption by homosexual couples, and is very conservative regarding abortion. The Partido Popular, in line with the new right ideology as defined by Heywood, puts the emphasis on Law and Order: ‘crime and punishment can only be countered by a fear of punishment’ (Heywood 2012: 99). In terms of the centre-periphery cleavage, the PP, a conservative party with a very centralised and hierarchical structure, was traditionally opposed to decentralisation. However, it has become increasingly comfortable with the development of regional governments. Figures 11.4 and 11.5 show the evolution of the positions of the Partido Popular regarding market economy and decentralisation in party manifestos (Volkens et al. 2016) taking as a reference the period between the year of the Re-foundation (1989) and 2015. Figure 11.4 shows the evolution of favourable mentions towards market economy in the party manifestos over time. In 2004, there was a clear declining trend of PP’s defence of liberal economic stances. The attention that the party pays to this issue increased back in 2008, shortly after UPyD and Ciudadanos emerged. Regarding the centre-periphery cleavage, it is interesting to highlight the evolution of the main parties on decentralisation. Figure 11.5 shows how the PP has evolved on this dimension. The PP was traditionally the party that showed a clear opposition towards decentralisation. However,

Nuanced liberalism 217 140 120 100 80 60 40 20

1989

1993

1996

2000

2004

2008

2011

Figure 11.5  Partido Popular’s positions on decentralisation: Percentage difference with respect to the average number of favourable mentions on the topic on the period 1989–2011 Note: Calculations based of the decentralisation variable (per301) of the Manifesto Project Dataset. For more information, see Volkens et al. (2016a).

Figure 11.5 shows that the party’s positions on this topic have evolved since 2004, towards more moderate positions. As a result of this evolution, the party distanced itself from liberal voters opposed to further regionalisation. This might help explaining the emergence and evolution of UPYD and Ciudadanos. Figure 11.6 shows a simplified view on the configuration of the positions of the main competitors, PP and PSOE, and of the new liberal parties along the three main dimensions highlighted in this chapter: economic and social Centre Ciudadanos

Partido

UPyD

Popular

State

Church Partido Socialista Periphery

Figure 11.6  Party positions of Spanish parties on a two-dimensional political space

218  Lidia Núñez Table 11.3  2015 preelectoral survey: vote recall and vote intention of Liberal voters (in %) Vote recall PP PSOE UPyD Ciudadanos Did not vote/ Would not vote N

31.0 19.0 3.0 16.0

Vote intention 16.4 9.7 0.4 20.6 8.4 1,962

Source: Centro De Investigaciones Sociol of 10 (2015), preelectoral study

liberalism and decentralisation. It shows how the new liberal parties are located in a political space that was until then not represented in the political arena. The two parties hold slightly more liberal positions than the PP on the economic dimension, but they differ with it in terms of their stronger positions in favour of centralism and social liberalism. Voters identifying with liberalism as their preferred ideology might therefore face difficulties in feeling represented by just one political party. Their votes might be contingent upon the weight they attribute to the three dimensions stressed in this chapter. Social liberals who are concerned about economic inequalities and value individual freedom in lifestyle choices would not feel at ease with the positions defended by the Partido Popular. They might feel closer to left-wing parties as the main advocates of social and cultural liberties, favouring the legalisation of homosexual marriage and children adoption and abortion. However, these left-wing parties would not do well when it comes to attracting economic liberals in favour of limiting state intervention and guaranteeing free market. Different nuances within liberalism can produce contradictory voting choices. Table 11.3 shows the vote recall of voters whose preferred ideology is liberalism at the previous elections (2011) and their vote intentions for 2015. It reflects how in 2011 most liberals would have voted for the PP (31%), while four years later, the votes of liberals were more fragmented: 21% of them preferred the new option of Ciudadanos and only 16% preferred the PP.

Structure and organisation The various liberal parties that have emerged in Spain since the breakdown of the Francoist regime have an important feature in common: Most of them have been founded and led by small groups of people with close ties with the worlds of culture and economy. In these parties, leadership is key and party membership is mostly insignificant. This section provides further details on the main structures of the new Liberal parties: UPyD and Ciudadanos, as the most recent and relevant examples of Liberal parties

Nuanced liberalism 219 in Spanish politics in recent decades. The following paragraphs present an overview of the main features of these parties in terms of leadership, structures, and membership. UPyD The political strategies and evolution of UPyD have been very much focussed on and defined by the party leadership. Díez, a former member of parliament (MP) of the PSOE, led the party from 2007 until 2015. Criticism within the party became important from 2014 onwards. The party had obtained its best results ever in the 2014 European elections, winning one million votes. However, in the next 2015 local and regional elections, the party only managed to win 200,000 votes. This loss intensified the criticisms within the party and Díez resigned. The next leader, Andrés Herzog, was elected in an extraordinary party conference with 43.3% of the votes. He did not stay long in the position, resigning just after the bad results obtained in the 2015 general elections when the party did not win representation. Gorka Maneiro, also a former militant of the PSOE, took the baton and has led the party since then. In terms of party organisation, at the national level UPyD relies on three main organs: the Party Conference, the Direction Council, and the Political Council. The most important body is the Party Conference that gathers the members of the party on a regular basis (every four years) or extraordinarily if the Direction Council decides so. The Direction Council is the most important structure between two party conferences. It is composed by the party spokesperson and 20 members elected by the Party Conference through a closed-list system. The Political Council is in charge of controlling the rest of the structures within the party, including the party’s elected representatives. As in the prior liberal parties in Spanish history, UPyD has an extremely weak party membership basis, although with important fluctuations over time. In 2008, a year after the creation of the party, it only counted 3,665 members, but that number had doubled by 2011, the year where the party reached its highest membership numbers so far. Party membership was then rather stable until 2014, when it rapidly declined back to 4,028 in 2015 (El País 2015). Part of this decline is due to the fact that many party members joined Ciudadanos.

Ciudadanos The leader of the party since its inception has been Albert Rivera, a lawyer who was 26 years old when the party was launched. The electoral strategy of the party has been very much focussed on the image of Rivera, to the extent that secondary figures of the party have only gained relevance when it started winning office at the regional level and in the period of government

220  Lidia Núñez formation that followed the 2015 and 2016 general elections, when the party became key for government formation. The most important party structures are the General Assembly, the General Council, and the Executive Committee. The General Assembly represents party members. The General Council is the main organ between two meetings of the General Assembly and is in charge of designing the main political directions. It is composed of 70 members, of which 60 are elected through open-list system by the General Assembly. The rest of the members of this council are five elected representatives, three members that are elected by the Executive Committee, the president of the party, and the Secretary General. The third important body is the Executive Committee, which is composed of between 20 and 40 members elected by the General Assembly through a closed-list system. It is in charge of the administrative and political directions of the party. Members of this Committee are elected by the General Assembly. Regarding party membership, data is scarce. Recent data reported in the press reflects that although the membership is very limited, it is experiencing a quick growth: in 2013 the reported number of party members was a little more than 2,000; by 2015 that number had reached 25,495. In spite of having its origins in Catalonia, most of the party members come from other regions: Madrid (5,091), Valencia (2,511), Alicante (1,512), and Málaga (1,059), whereas in Barcelona the party only counts 2,095 members. Both parties show significant similarities. First of all, in spite of having democratic organisation procedures and offering party members an important say in the election of the main bodies within the organisation, these parties follow very personalised dynamics. Leaders in both parties, especially at the time of their creation, Díez and Rivera, concentrate most of the power in their hands. In the case of UPyD, a large group of elected representatives resigned or were expulsed from the party due to divergences with the party leader or have denounced an excessive control from the party leadership, such as Francisco Sosa Wagner, Fernando García, or Carlos Sevillano. Following Katz and Mair’s (1993) three faces of the party, these parties show a clear ascendancy of party in central office. The party on the ground is characterised by an extreme weakness both in terms of numbers and in terms of loyalty. Indeed, UPyD’s membership has been halved since 2014 due to the transfers of members to Ciudadanos. The party in public office has had a similar dynamic in the last years, with many elected representatives from UPyD joining and running for office as candidates of Ciudadanos.

Conclusion This chapter has shown that liberal parties in Spain have always been rather weak, but that it does not emanate from an absence of liberal voters. The chapter has shown how the PP, as a right-wing catch-all party has been able

Nuanced liberalism 221 to attract the largest share of liberal voters and that only recently (in 2015 and 2016 general elections) this position has been challenged by Ciudadanos. The chapter has shown that liberal voters in Spain can be considered as an electorate with its own characteristics: These voters are typically more liberal on the economic and political dimensions than the rest of the population, and tend to have more negative stances towards decentralisation. However, the new parties have only had a mild success in capitalising on their specific positions on these dimensions, possibly due to the changes in positions operated by the Partido Popular to counterbalance the emergence of these parties. However, what is beyond doubt is that these two parties can be categorised within the liberal family. Following Mair and Mudde’s (1998), UPyD and Ciudadanos can be classified as liberal parties on the basis of their ideological profile. Both recently created, their genetic origin does not offer much basis for a classification of these parties. Other criteria, such as belonging to transnational federations of political parties, offer only a partial support for this categorisation. UPyD and Ciudadanos both seat within the Alliance of Liberals and Democrats for Europe (ALDE) group in the European Parliament although only Ciudadanos is a full member of the party. These parties do not yet belong to any international federation, such as the Liberal International. This absence can at least partially be explained by the centre-periphery debates in the Spanish political arena. Indeed, the only Spanish parties belonging to the ALDE party and the International Liberal were regionalist parties: the Partido Nacionalista Vasco (Basque Nationalist Party) in the former, and Convergencia Democratica de Catalunya (Democratic Convergence of Catalonia) in both. After the transition to democracy, these parties decided to join these federations to be represented on the international scene without having to share the same affiliation as the Alianza Popular/Partido Popular, who belonged to the European People’s Party. Given that UPyD and Ciudadanos were created around opposition to regionalism, regionalist parties, and specially Convergència, have opposed the entry of these parties in these federations. In fact, the only party within ALDE that voted against the incorporation of Ciudadanos was Convergència Democràtica de Catalunya. In spite of sharing comparable origins, UPyD and Ciudadanos have had an unequal evolution; the latter grew exponentially to the expenses of the former. Whether this situation will be maintained in future elections, only time knows. However, at this stage, what is clear is that both parties target the same electorate and that their evolution is contingent upon a series of factors. First, the evolution of the positions held by the Partido Popular is capital. The extent to which this party manages to sustain a clear message in several dimensions will mark the evolution of the new liberal parties, most notably on the extent to which it is able to transmit the image of the economy being well managed, a moderate message on moral issues, and

222  Lidia Núñez strong positions against the independentism in Catalonia and other peripheral nationalisms. Second, these new parties face another significant challenge that can be even more important: the constraints that the Spanish electoral system poses to minor state-wide parties. The electoral system used in Spain is a closedlist proportional system that produces majoritarian outcomes mainly due to the low district magnitude in many districts. The Spanish electoral system hence benefits the larger political parties, and most notably those that get better results in smaller districts. This is the case for the Partido Popular in what Montero and Riera (2008) refer to as the conservative bias of the Spanish electoral system. In the 2015 general elections, the share of seats won by the Partido Popular was 7 percentage points higher than its share of votes, while the difference for Ciudadanos was only 2 percentage points. An optimal distribution of votes across districts – especially in larger ones – is therefore crucial for these parties. In addition, survey data shows that there is still a pool of liberal voters who currently are not represented by any of the existing political parties and who prefer to abstain in elections. Survey data shows that 15% of these voters claim not to have voted in the 2011 general elections (Centro de Investigaciones Sociológicas Study 3317). Future electoral success of these parties might well be contingent upon their capacity to attract this sector of the liberal voters. Finally, the evolution of these parties’ internal dynamics and structures is also important. Ciudadanos is currently the stronger of the two liberal parties. However, as a new party it faces the same kind of internal challenges as UPyD did. First, both parties face a need to develop and strengthen the party on the ground. A solid membership basis can serve as control of the elected representatives at the local and regional levels, something that is even more relevant in newly created political parties. In addition, the development of the party on the ground plays an important role for the future of these parties, as it is one of the major instruments for the selection and recruitment of political personnel. In addition, the recent access to public office at different levels of government – local, regional, national, and European – poses the challenge of developing and maintain consistent and coherent positions and strategies across levels of government. In conclusion, this chapter has shown that there is a liberal electorate in Spain, but that there are different nuances of liberalism within this electorate. The future of the new Liberal parties and of the Spanish political scene will largely depend on the extent to which these new parties succeed in representing the interests of this electorate.

Notes

1. This war started with the Dos de Mayo uprising on 2 May 1808 and ended on 17 April 1814.

Nuanced liberalism 223 2. This Constitution came into effect several times during the 19th century: in 1812, 1820, and 1836. 3. Union Liberal was led by Pedro Schwartz and only lasted for a year before it merged into Alianza Popular. 4. IX Congress, Avanzar en Libertad (Progress in Liberty). 5. The party statutes are available at https://www.ciudadanos-cs.org/estatutos. 6. Resoluciones políticas del 2º Congreso UPyD (2013). 7. During the Aznar government, the total amount of revenues obtained through privatisations reached 40,000 million euros (Vidal-Beneyto 2004).

References Alvárez Junco J., Shubert A. (eds.) (2000), Spanish history since 1808. London: Arnold. Barbera O., López A.B. (2006), ‘Convergència i Unió: From stability to decline?’ In: De Winter L, Gómez-Reino M., Lynch P (eds.), Autonomist parties in Europe: Identity politics and the revival of the territorial cleavage. Barcelona, Spain: Institut de Ciències Politiques I Socials (ICPS), pp. 101–142. El Pais (2015, 8 July), La afiliacion en UPyD, El Pais. http://elpais.com/elpais/2014/08/22/ media/1408739957_952572.html García-Atance García de Mora M.V. (1990), ‘IX Congreso Nacional del Partido Popular’, Revista de Derecho Político (31), pp. 229–234. Gillespie R. (2000), ‘Political polarization in the Basque country’, Regional & Federal Studies 10(1), pp. 112–124. Giordano B., Roller E. (2002), ‘Catalonia and the ‘Idea of Europe’s competing strategies and discourses within Catalan Party politics’, European Urban and Regional Studies 9(2), pp. 99–113. Heywood A. (2012), Political ideologies: An introduction. London: Palgrave Macmillan. Katz R.S., Mair P. (1993), ‘The evolution of party organizations in Europe: The three faces of party organization’, American Review of Politics 14(4), pp. 593–617. Lipset S.M., Rokkan S. (1967), ‘Cleavage structures, party systems, and voter alignments: an introduction’. In: Lipset S.M., Rokkan S., (eds.), Party systems and voter alignments. New York: Free Press, pp. 1–64. Mair P., Mudde C. (1998), ‘The party family and its study’, Annual Review of Political Science 1(1), pp. 211–229. Marcet Morera J. (2000), ‘Convergencia Democrática de Catalunya: The only Spanish Liberal reference at a European level’. In: De Winter L., (ed.), Liberalism and liberal parties in the European Union. Barcelona, Spain: Institut de Ciènces Politiques i Socials, pp. 11–24. Marichal J. (1995), El secreto de España. Madrid: Taurus. Millán J., Romeo M.C. (2004), ‘Was the liberal revolution important to modern Spain? Political cultures and citizenship in Spanish history’, Social History 29(3), pp. 284–300. Montero J.R., Calvo K. (2000), ‘Religiosity and party choice in Spain’. In: Broughton D., ten Napel H.M. (eds.) (2000), Religion and mass electoral behaviour in Europe (Vol. 19). London: Routledge, pp. 118–135. Montero J.R., Riera P. (2008) Informe sobre la reforma del sistema electoral. Madrid: Comisión de Estudios del Consejo de Estado. Pallarés F., Keating M. (2003), ‘Multi-level electoral competition regional elections and party systems in Spain’, European Urban and Regional Studies 10(3), pp. 239–255.

224  Lidia Núñez Salvadó F.J.R. (2003), ‘The great war and the crisis of liberalism in Spain, 1916–1917’, Historical Journal 46(4), pp. 893–914. Vidal-Beneyto J. (2004, April), ‘Balance de la era Aznar’, Le Monde Diplomatique. Edición Cono Sur (58). Volkens A., Lehmann P., Matthieß T., Merz N., Regel S., Werner A. (2016), The Manifesto Data Collection. Manifesto Project (MRG/CMP/MARPOR), Version 2016a. Berlin: Wissenschaftszentrum Berlin für Sozialforschung (WZB).

12 Liberal parties in Central and Eastern Europe Between success and failure Blagovesta Cholova and Jean-Michel De Waele

Introduction After the fall of communist regimes in 1989, analysts and media predicted that the then unchallenged capitalism and liberal ideology would triumph in the countries of the Eastern bloc (Cholova and De Waele 2013; Hanley et al. 2007; Krastev 2007; Szacki 1995). A quarter of a century later, this predicted triumph did not take place, and in some countries (Poland, Hungary, Bulgaria) even the basic principles of democracy are put into question. Why did liberalism, which seemed to have a fertile ground for development in these countries, fail to establish itself as a leading ideology? In this chapter, we discuss the fate of the liberal party family in Central and Eastern European (CEE) countries. We review the reasons of their relative (lack of) success, by examining historical, political, and sociological factors. Our analysis points out that, although regional differences and national particularities can account for the diversity of liberal parties in the region, common features can be identified that help understanding the (mis)fortune of these parties. On this basis, we discuss the possible outcomes for the future. The first section presents the criteria used to distinguish liberal parties from other centre-right political entities in CEE. The second section describes the development of liberal parties in CEE, before and after the communist era, and highlights their electoral instability. The third section provides some explanations for the relative weakness or success of liberal parties in most CEE countries. In our concluding remarks, we address the potential of liberal parties and the future challenges that they may face in this fast-changing environment.

Identifying liberal parties in CEE The classification of political parties by party families requires detailed selection criteria and method to avoid bias (Hanley 2006; Mair and Mudde 1998). The question is especially relevant in cases where the political background is not straightforward and where one party can potentially belong to

226  Blagovesta Cholova and Jean-Michel De Waele different categories, depending on the time frame or on the selection criteria (De Waele and Gueorguieva 2002; Hanley 2006; Vachudova 2008). Most parties in CEE countries do fall in this category. In the following, we review the defining criteria of a party family and we show the extent to which they are problematic to identify liberal parties in CEE. We contend that liberal parties in CEE lack a clear ideological profile and that they use party labels mainly as strategic tools. As a consequence, we use the last criteria (transnational affiliation) to select our cases, in line with the other chapters in this volume. Party label After the fall of the communist regimes, the political system first structured around a cleavage related to the past, dividing anticommunists and postcommunists (Lawson et al. 1999; Rogier and De Waele 2004; Todorov 2004). Parties then gradually adopted classical labels from Western European party families. In a context where party identity emerged almost from scratch, choosing a label constituted a strategic step towards international recognition, to gain competitive advantage or explore an available niche on the political spectrum. Consequently, one label could embrace different parties across national contexts (Hanley and Szczerbiak 2006). We therefore don’t use the party label as a defining criterion. Cleavage formation and ideology Another difficulty in distinguishing liberal parties from their conservative rivals in CEE countries relates to the process of cleavage formation. After the fall of communist regimes, two main blocks often opposed each other: on one side, the successor parties (the ‘left’), which, as transformed heirs of the dominant Communist Party, still controlled much of the state resources and benefited from a clear organisational structure and a huge electoral base (Bozóki and Ishiyama 2002; Grzymala-Busse 2002; De Waele 2004; Sznajder 2011); on the other side, anticommunists parties and groupings (the ‘right’), most of which were created after the fall of the regime, and whose main ideological principles and policy priorities were related to the economic, social, and political transitions (Held 1993; Tavits and Letki 2009). During the period from 1996–1997 to 2005–2007, left- and right-wing parties aligned to facilitate the process of accession to the European Union and proposed relatively similar economic policies (Tavits and Letki 2009; Vachudova 2005; Whitefield and Rohrschneider 2009). This strong European leverage didn’t leave much flexibility to the parties, which agreed to promote the economic transition to liberal market and the privatisation of state-owned industry. This subsequently led to the perceived ‘liberal consensus’ among left- and right-wing parties on economic policy and made the economic cleavage irrelevant (Krastev 2016). This dominant ‘liberal’

Liberal parties in Central and Eastern Europe 227 stand made (economic) liberalism the most popular ideology during the first decade of transition and guaranteed a fertile ground for the development of parties embracing this ideology (Krastev 2007). At the same time, the cleavage dividing communist heirs from other parties persisted, and on this dimension, the right-wing parties remained united in coalition against the left. During this period, given their common support to liberal economic policies and opposition to the successor parties, Liberals and Conservatives had few ideological differences and were therefore hardly distinguishable. Towards the end of the transition period, the ‘liberal consensus’ faded away and new cultural issues became salient (Hanley and Szczerbiak 2006; Szczerbiak and Taggart 2008). This cultural cleavage involved a conflict of values between openness (multicultural) and closeness (nationalistic), and between pro-European and Eurosceptic positioning. In the 2000s, divisions emerged between liberal, conservative, and nationalist-populist parties, but the salience of this cultural cleavage didn’t have the same intensity across countries1. Therefore, the identity of those movements remained quite fluid (Hanley 2006; Hanley and Szczerbiak 2006; Hanley and Dawson 2016). Splits, mergers, and new parties complicated the process of identity formation even more. This fluidity prevents us from using cleavage structure and ideology as a criterion to identify liberal parties in CEE countries. European-level affiliation A last criterion to identify liberal parties in CEE countries is to turn to their own choices and alliances at the European level. After their foundation and especially after accession to the European Union, all parties had to position themselves within the European political landscape and searched for allies in the European Parliament (De Waele and Gueorguieva 2002). The European party families tried to attract the parties deemed close to their ideological stances, especially if these parties were powerful at the national level. By choosing their partners at the European level, political actors acknowledged their ideological proximity with a certain party family (Cholova and De Waele 2013). This criterion allows identifying liberal parties more easily, that is, those parties that joined the ALDE. Besides, it allows considering changes over time, as parties can realign, for instance to the right, by becoming gradually liberal-conservative (or conservative). It is the case for political movements that quit the ALDE and joined the EPP or other conservative European parties, such as Platforma Obywatelska (Civic platform – PO) in Poland and Magyar Polgári Szövetség (Hungarian Civic Platform – Fidesz) in Hungary. This approach is, however, problematic. Although ideological proximity constitutes one of the main criteria in building alliances, European political groups often have a catch-all strategy, such as the European People’s Party (EPP) that gathers a broad range of parties, from more conservative to liberal and centrist progressive parties (Jansen and Van Hecke 2011). This strategy

228  Blagovesta Cholova and Jean-Michel De Waele allows them to attract the strongest parties of the CEE countries, whose ideological profiles are not as clear as those of Western European parties. Furthermore, the individual parties have strong incentives to join the EPP (because of its influence), and can also adjust their own stances to be eligible to adhere (see above, PO and Fidesz). In spite of these limitations, this approach is the most straightforward to identify parties belonging to the liberal party family in CEE countries. Case selection and data collection Keeping in mind these limitations, our case selection includes CEE members of the ALDE from 2001 to 20172. We selected NDSV (Nacionalno dviženie za stabilnost i văzhod – National Movement for Stability and Progress, 2001–2013 and member of ALDE since 2002) and DPS (Dvizhenie za prava i svobodi – Movement for Rights and Freedoms, 1990–present, member of ALDE since 2007) in Bulgaria; ODA (Občanská demokratická alliance – Civic Democratic Alliance, 1990–2007, member of ALDE 2004–2007), US-DEU (Unie Svobody–Demokratická unie – Freedom Union-Democratic Union, 1998–2011, member of ALDE since 2004) and ANO2011 (Yes2011, 2011–present, member of ALDE since 2012) in Czech Republic; UW (Unia Wolności – Freedom Union, 1994–2005, not affiliated to ALDE), PD (Partia Demokratyczna – Democratic Party), RP (Ruch Palikota – Palikot’s Movement, 2011–present, member of ALDE 2011–2013), and .N (Nowoczesna – Modern, 2015–present, member of ALDE since 2015) in Poland; PNL (Partidul Național Liberal – National Liberal Party, 1990–2014, member of ALDE 2007–2014) and ALDE (Alianța Liberalilor și Democraților – Alliance of Liberals and Democrats, 2015– present, member of ALDE since 2015) in Romania; ANO (Aliancia Nového Občana – Alliance of the New Citizen, 2001–2007, member of ALDE 2004–2007), SDKU (Slovenská demokratická a kresťanská únia – Slovak Democratic and Christian Union, 2000–2016, not member of ALDE) and SaS (Sloboda a Solidarita – Freedom and Solidarity, 2009–present, member of ALDE 2009–2014) in Slovakia. Parties that did not directly joined the ALDE (SDKU and UW) were selected as predecessors of the parties that later joined, given their direct links in terms of prominent personalities, structures and ideological profile. The analysis focusses mainly on the second and third decades (2001–2017), although historical developments are also taken into account. We analyse new parties (e.g., Nowoczesna in Poland and ALDE in Romania, created in 2015), parties that have existed at the national level for several years (27 years in some cases, e.g., PNL and DPS) and former parties that are now irrelevant (e.g., NDSV or SDKU). This chapter adopts a broad comparative approach and highlights shared features and evolutions within the specific context of CEE. Although the selected liberal parties differ in their structure, longevity, and political

Liberal parties in Central and Eastern Europe 229 strength, some common trends can be outlined in terms of voters’ profile and ideology. This chapter is based on data gathered within the framework of the project ‘Liberal parties in CEE: weakness and potential’ (Cholova and De Waele 2013). In this project, we collected data on voters’ profile, electoral geography, and historical tradition. We also conducted a series of interviews and focus groups3 with members and voters of liberal parties. Although the project dates back from 2012–2013, this chapter discusses new cases, Nowoczesna and ALDE, which were created in 2015, and provides some update regarding the other cases.

The development of liberal parties in CEE countries This section describes the historical legacies of liberal parties in CEE. It highlights the lack of liberal tradition in the precommunist period in most of the cases and suggests how historical legacies contribute to understand the fate of liberal parties in the postcommunist era. Second, it analyses the electoral performances and the development of liberal parties in CEE countries over the past 26 years and more specifically after 2001. Third, it points out their relative weakness and instability and describes the patterns of electoral volatility of liberal voters. The (lack of) historical legacies The history of liberal parties in CEE before World War II varies across national contexts. In some countries, such as Romania and Bulgaria, liberal parties were strong actors in the precommunist period and liberalism can be considered as one of the mainstream ideologies. In these countries, ‘historical’ parties with liberal orientations were revived after 1989 – e.g., Narodna Liberalna Partia (National Liberal Party – NLP) in Bulgaria, PNL in Romania. Yet the gap between them and their predecessors impeded them to recover their former influence and electoral support. In other countries, such as Poland and Czechoslovakia, liberal movements were much weaker and didn’t manage to play an autonomous role on the political scene. In Romania and Bulgaria, liberal parties were created during the state-building period after gaining independence from the Ottoman Empire in the second half of 19th century (e.g., Tatarescu’s party in Romania and the Democratic Alliance in Bulgaria). They were created as ‘copies’ of their Western counterparts but adapted to the national context. Liberalism was understood as the defence of civic rights and individual freedoms against the conservative parties that wanted to impose some restrictions, for instance on voting rights. In both countries, liberal parties played a prominent role until the interwar period when they gradually lost their influence because they were deemed responsible for the consequences of the war, but also because new political actors emerged after the introduction of universal

230  Blagovesta Cholova and Jean-Michel De Waele suffrage: the Socialist, Social-Democrat, and the agrarian parties, which were ‘class-based’ parties representing the workers (or peasants) and relied on large grassroots bases outside the parliament. After World War I, in the newly created countries of Central Europe (mainly Czechoslovakia and Poland), liberal parties did not succeed in gaining support. This can partly be explained by the fact that they were perceived as ‘foreign’ by the local population. Besides, nationalism and conservatism were much more popular ideologies, given the new quest for national identity and the traditional values of the mostly Catholic rural areas. In fact, during the interwar period, all liberal parties endured some drawbacks: They were mainly urban parties grounded in the bourgeois milieu and supported by smallholders and could not face the outbreak of new rival mass parties (Cholova and De Waele 2013). In addition, liberal parties were considerably weakened by the establishment of authoritarian rule in the region. After World War II, liberal parties in Bulgaria and Romania were deemed responsible for the outcome of the war, along with other traditional parties – both countries were initially Hitler’s allies. This impeded liberal parties to reconstruct after World War II and eventually reinforced their decline after the introduction of communist regimes. During communism, liberal-minded figures often joined the dissidence and expressed their criticism through organised manifestations or strikes. Part of these figures who were active before and during World War II went in exile in Western Europe and remained active, supporting the dissidence. Examples include the Polish government in exile, or prominent figures such as author Milan Kundera (Czechoslovakia). This trend was more significant in Central Europe. In southern part of Eastern Europe, the first real protests appeared shortly before 1989, as illustrated by the Ekoglasnost movement in Bulgaria founded in 1988, or by the dissident movements in Romania from 1989. In general, the liberal tradition in CEE countries did not offer a solid base for legitimising the new parties after 1989. Even in the southern part of Eastern Europe where the tradition was stronger, new parties had few direct links with their predecessors but rather relied on a symbolic legacy. The Romanian case is interesting in that regard: The PNL managed to gather important figures who were active before the communist regime, such as Radu Campeanu, and who could transmit and legitimise the legacy. These figures, especially the ones coming from the dissidence, played an active role in the creation of the new liberal parties, but provoked internal disputes given their individual visions of liberalism, and in some cases their resentment towards parties as such – Vàclav Havel in Poland is a prominent example. In many ways, their presence within the parties constituted a force and a weakness, especially in the first years when these parties were fighting against the well-organised rival communist parties. Even though the liberal tradition was weak in most of the CEE countries, new liberal parties had a great potential after 1989, since at the beginning of

Liberal parties in Central and Eastern Europe 231 the transition the majority of the population widely supported liberal values (Krastev 2007; Zuzowski 1998). The emergence of liberal parties after 1989 was not (only) part of a revival of historical parties and movements but related more to a process of ‘modernisation’ of politics where liberal values symbolised the return to Europe and the Western way of life (DIMAP 2009). The legacies proved to be important where they existed but also had a negative impact on the institutionalisation of the young liberal parties. Electoral performances and development after 1989 Table 12.1 displays the electoral performances of the liberal parties after 1989. At the aggregate-level, a relatively stable proportion of the electorate in the region (10–15%) casts a ballot for liberal parties, although this share varies from one country to another and from one election to the other. In most of the cases examined, the electoral system is proportional, which makes these scores significant in terms of seats. Often, voters cast their preference for new entrants embracing the liberal identity. Liberal parties’ vote share in the region remains quite stable after the initial breakdown of anticommunist coalitions in the early 1990s. However, the liberal vote is highly fragmented, and new parties emerge as ‘old’ ones decline. Two cases deviate from this rule: Romania, where the PNL holds a significant share of the vote; and Bulgaria, where the DPS is exceptionally stable (the party received between 7 and 10% over 20 years), but where the NDSV displays important electoral volatility. Overall, the situation is paradoxical: Individual parties’ scores lack stability but it is counterbalanced (or triggered) by a stable aggregate share of votes for liberal parties in the region. Despite difficulties in periods of turmoil, liberal parties are present in all those countries, although in most cases, they cannot mobilise steady support over time. In terms of development, liberal parties in CEE follow similar patterns. After the initial participation in the anticommunist umbrella movements, most liberal parties tried to run for election independently but their scores as individual parties remained below those of the former coalition. Following this failure, most of them tried to reform. This second phase was marked by numerous splits and mergers. In most countries, a liberal party has participated in a government coalition as a minor partner at least in one legislature but this often proved disastrous for the party at the next election, and only few liberal parties survived this threshold – namely, the PNL and DPS. Electoral volatility occurs mostly within the liberal bloc: The majority of the voters who voted for a liberal party in the previous election tend to vote for a (new) liberal party at the next one. This is especially true in countries such as Bulgaria, Slovakia, and Poland. For instance, in Slovakia, at the 2002 election, half of the voters of ANO were either first-time voters or SDK voters (Slovenská demokratická koalícia – Slovak Democratic Coalition); at the 2010 election, 40% of the SaS voters had voted for SDKU-DS at the previous

Table 12.1  Electoral performance of liberal parties (% of the votes, national legislative election) Bulgaria DPS NDSV

1990 8.0

1991 7.6

1994 5.4

1997 7.6

2001 7.5 42.7

2005 12.8 19.9

2009 14.5 3.0

2013 11.3

Romania PNL ALDE

1990 6.4

1992 2.6

1996 30.7 (CDR)

2000 6.9

2004 31.3 (+DP)

2008 18.6

2012 60.1 (USL)

2016 20.0 5.6

Slovakia ANO SDKU SaS

1990

1992

1994

1998

2006 1.4 18.4

2010

2012

2016

26.3 (SDK)

2002 8.0 15.1

15.4 12.1

6.1 5.9

0.3 12.1

Poland PD (UW) RP .N

1991

2001 3.1

2005 2.4

2007 13.2 (LiD)

2011

2015

Czech Republic ODA US-DEU ANO 2011

1990

1993

1992 5.9

1997 13.4

1998 6.4 8.6

2002 0.5 14.3

2006 0.3

2010

10.0 2013 18.7

Source: Parties and elections in Europe (http://www.parties-and-elections.eu/); official electoral results by country Note: The names in parentheses are the coalitions in which the party has participated, or the coalition partners (with a +).

7.6

2014 14.8 0.2

2017 9.2

Liberal parties in Central and Eastern Europe 233 election or were first-time voters (Cholova and De Waele 2013). Some exceptions are nevertheless observed, as in Bulgaria at the 2009 election, when the majority of the NDSV voters turned to the populist-­conservative party GERB (Grazhdani za Evropeisko Razvitie Balgariya – Citizens for European Development of Bulgaria).

Between success and failure: some explanations This section explores potential explanations for liberal parties’ relative weakness by focussing on the following elements: internal organisation, liberal ideology, and electoral base (voters’ profile and perceptions of parties, and electoral geography). Organisational instability and the issue of disloyalty The evolution of liberal parties over time reveals important dynamics of splits and mergers. The Polish case illustrates this trend, where one wing of the liberal movement changed its name and ideology several times but kept a clear filiation: UD (Unia Demokratyczna – Democratic Union) became UW, then PD; KLD (Kongres Liberalno-Demokratyczny – Liberal Democratic Congress) entered UW then founded PO, from which RP was created. Most liberal parties failed to create stable organisations that survived more than two legislatures. Even though a liberal-minded electorate existed in CEE countries, only in Romania did the PNL manage to overcome the splits, to survive the opposition, and to increase voters’ support over time. In all other cases, liberal parties attracted a relatively small share of the votes and were relatively short-lived due to internal disputes, further weakened by the competition with stronger conservative rivals: ODS (Občanská demokratická strana – Civic Democratic Party) in Czech Republic (Hanley 2008); AWS (Akcja Wyborcza Solidarność – Solidarity Electoral Action) then PiS (Prawo i Sprawiedliwość – Law and Justice) and PO in Poland; HZDS (Hnutie za demokratické Slovensko – Movement for a Democratic Slovakia) in Slovakia. This dynamic of party split often starts by a decline in the polls, which translates in both voters and party elites looking elsewhere and creating/ supporting a new actor with similar views but not (yet) tarnished by corruption scandals, poor government performance or unsuitable alliances (Sikk 2012). Party splits result from both voters and elites’ disloyalty. Without voters’ loyalty, liberal parties can hardly resist the centrifugal tendencies and internal in-fights, and often decline rapidly and disappear from the political scene. As explained below, voters’ disloyalty can partly be explained by the undefined nature of the liberal ideology, or by the fact that they often have a vague notion its principles and values. As far as party elites are concerned, they often prefer to abandon a ‘losing’ party and to create a new one rather than stay in the party and work for its

234  Blagovesta Cholova and Jean-Michel De Waele reconstruction and future success. This lack of loyalty results in the frequent emergence of new parties, which in turn incentivises voters’ volatility. In that regard, the Romanian case (with PNL) shows by contrast that, when a party is electorally strong over time, its splinter parties fail electorally and eventually return back to the main party. To endure, liberal parties should stabilise their organisation and build a network of grassroots associations that could act as socialising arenas to create more stable party preferences. A vague liberal ideology: the grassroots’ perceptions In CEE countries, like in Western Europe, different dimensions of liberalism can be identified: economic liberalism, related to free market as opposed to state regulation; civic liberalism, mostly associated with human rights, individual freedom, and libertarian values; and cultural liberalism, related to the protection of minorities, tolerance, and ethical issues such as abortion, same-sex marriage, and euthanasia. Liberal actors display ambiguous attitudes on these three dimensions (Cholova and De Waele 2013). Economic liberalism is often associated with neoliberal policies. Voters perceive the negative consequences of these policies, especially for the most fragile parts of the population. In the focus groups we have organised, most of the respondents preferred a limited form of economic liberalism and supported state regulation and welfare to reduce its negative impacts. These attitudes reflect the context of the transition period, when the social costs of the economic transformations were felt by many citizens. But survey data in fact reveals this paradox: Citizens support the welfare state, but at the same time defend the value of free market (DIMAP 2009). Civic liberalism is mostly associated with the rule of law and with the respect of individual rights. This is the most powerful aspect of liberalism in this area because it comes as the antithesis of the authoritarian rule. Civic liberal values are shared by most citizens, well beyond the liberal electorate. Their perceived negative aspects include a form of radicalism, the rejection of all types of authority and anarchism. Nevertheless, this form of liberalism is widely shared by the public (DIMAP 2009). Cultural liberalism is the most controversial aspect of liberalism in CEE and constitutes the only strong cleavage between the Liberal and the Conservative parties. The relevance of this cleavage varies across context and proves to be much more important in countries with strong attachment to traditional (or religious) values than in more secular countries (Poland and Slovakia vs. Romania and Bulgaria). However, even within the liberal electorate, voters who support cultural liberal values constitute a minority in most of the CEE countries, regardless of their religious or cultural specificity (DIMAP 2009). Overall, the liberal ideology in CEE is perceived as unclear and undefined. Interviews with voters show a lack of comprehensive understanding of what are liberal values and principles; and according to voters themselves, this

Liberal parties in Central and Eastern Europe 235 vagueness constitutes one explanation of the weakness of liberal parties. For many voters, being liberal means being centrist, not left or right. This can be an asset, but this is not sufficient to nurture electoral loyalty. Because they lack strong attachment to a concrete liberal ideology embodied by a party, liberal voters are highly volatile. Furthermore, when asked about which party embodies the main values of liberalism, many voters and party members identify as ‘liberal’ some parties that could be rather classified as conservative or populist. Liberal support: voters’ profile, electoral geography, and grassroots’ perceptions Liberal voters in most of the CEE countries analysed in this chapter have a similar profile. First of all, they are highly educated, much higher than rest of the electorate (Cholova and De Waele 2013). In Poland, education is the most stable factor predicting the liberal vote. Liberal ideas are indeed more popular among the intellectual and educated upper classes, where individual freedom and free enterprise, free circulation of goods and labour are seen as beneficial. This first common characteristic of the liberal electorate helps to understand its high volatility. According to the cognitive mobilisation theory (Dalton 1984), highly educated citizens tends to be more interested in politics than the average citizen, are more exposed to the media and the criticism of liberal parties therein and to the campaign messages of other political competitors. Besides, according to the cognitive mobilisation thesis, highly educated voters tend to be more critical towards the policies or the proposals of political parties, because they have the skills and cognitive resources to forge their own opinion on complex political issues. They are therefore more susceptible to rationally switch their votes. The second common characteristic of liberal voters in CEE countries is that they are younger than the average voter. This characteristic is partly correlated to their educational profile, as younger generations are generally better educated than the older ones (Cholova and De Waele 2013). The young tend to accept rather than reject economic liberal values; They tend to be less attached to social security and to perceive liberalisation as an advantage. Although civic liberalism is less controversial and its values are shared by most liberal voters and beyond, cultural liberalism is more popular among young well-educated urban dwellers. Young people tend also to be less nostalgic towards the old regime then their parents or grandparents, as they haven’t lived under or hardly recall anything from the communist regime. The relative youth of liberal voters can also partly explain why they vote more often for new parties than the rest of the population. Younger voters have a weaker attachment to the ‘traditional’ parties, and compared with the older generations, they are less attached to a system of values and common

236  Blagovesta Cholova and Jean-Michel De Waele references (Cholova and De Waele 2013). Hence, they can more easily cast their vote for newcomers. In addition, young voters were not socialised in a context of stable electoral support and did rarely inherit a traditional political legacy from their parents. The third common feature of liberal voters in CEE is that they live mostly in big cities. This is visible for most cases included in this chapter, except for RP and NDSV, which both attract voters in smaller cities and not in the capital (Cholova and De Waele 2013). Two factors explain this trend. First, people in smaller cities and villages are much more vulnerable to the effects of liberalised market, because their activities mainly rely on state subsidies or on small industries that can hardly compete in a globalised market. Voters living in these areas tend to opt for a socialist (Bulgaria, Romania) or a conservative-­protectionist party because their interests are threatened by the liberal agenda. Exceptions of liberal parties attracting votes in nonurban arena occur where a strong competitor on the right support a neoliberal agenda – e.g., PiS in Poland, ODS in Czech until 2010, GERB in Bulgaria, HZDS in Slovakia, and PDL (Partidul Democrat-Liberal – Democratic Liberal Party) in Romania. Second, urban voters are generally more secular and open to cultural change then voters in remote areas. This is important especially in countries such as Poland or Slovakia, where ethical issues are salient and where liberal parties promote progressive stances. Nevertheless, some specific cases do not fit these trends. One of them is DPS, an ethnic-based party in Bulgaria, whose electoral support is mostly concentrated in areas with predominantly Turkish or Muslim population, in the south and northeast of the country. Its voters have lower levels of education and to work in the agricultural sector. This party doesn’t fit the model of a liberal party but can be considered more as an ethnic-regionalist one. Examining the electoral geography of liberal parties in the CEE countries further helps to understand the context of their relative weakness. As mentioned above, the liberal electorate tends to be located in cities. In addition, and this pertains specifically to Poland and Slovakia, the liberal electorate is concentrated in the western part of the country. This geographical divide is first explained by the fact that western parts of these countries could benefit more from the open market than the more remote areas, where communist industries were not replaced by modern companies, and where a bigger percentage of the population is unemployed. Second, it relates to historical specificities. Poland is divided between the western part that matched the territories belonging previously to the Austrian (Austro-Hungarian) Empire, and the eastern part that used to belong to the Russian Empire. This created inequality already after World War I, when the country regained independence, between the less-developed and rural eastern territories, and richer and more urbanised western ones. After communism, this division reemerged in terms of electoral preferences. The western regions tend to vote for liberal and more progressive movements, while the eastern regions tend to vote for

Liberal parties in Central and Eastern Europe 237 conservative-nationalist or former communist – now socialist – movements. This matches the individual-­level trends mentioned above (education, urban versus rural, secularism versus religious values). Similarly, in Slovakia the western parts of the country (around Bratislava and Nitra) and the central areas (around Banska Bistrica) are more inclined to vote for the liberal parties. ANO however departs from that trend, since its electoral support is significant in the East around Preslov and Kosice, but this is due to its leader’s personal support (Kosice’s mayor, Rudolf Schuster). In the other countries included in this analysis, the geographical distribution corresponds to the urban/rural divide identified at the individual-level. In Romania, Bulgaria, and the Czech Republic, the Liberals are stronger in the capital and in the regional centres. The DPS attracts a regionalist vote from the southern and the northeastern parts of Bulgaria, populated mostly with Bulgarian Turks. This impacts the policy proposals of the DPS and the distribution of Ministerial seats. The party defends the interests of its constituency regarding notably the tobacco industry, which constitutes the main source of revenue in these regions (Todorov 2004). The lack of widespread support and of large territorial coverage weakens liberal parties as their network of support is limited to the big urban areas. Finally, the data collected through interviews and focus groups allows to dig deeper into how liberal parties are perceived by their supporters and to better understand the weakness of liberal parties from their perspective. When evaluating liberal parties, supporters point out three main reasons for their weakness. The first aspect is related to the program and the ideological stances of the party. According to the respondents’ view, parties’ positions should not only refer to liberalism in general, but should propose more specific and concrete policies connected to the needs of their constituents. This aspect is particularly relevant as liberal parties in CEE struggle to elaborate a distinctive message and to communicate their core principles. A clear ideological program is crucial for the reinforcement of party identity and loyalty and to compete with strong challengers in the right-wing spectrum that position themselves as the main representatives of the right-wing electorate. The second critical aspect relates to the lack of integrity and the corruption among the political elites. This tendency is not confined to liberal parties but can be very harmful in their case as well. This perceived lack of integrity explains why often a mandate in office can prove fatal for the image and the support of a party. The notion of integrity does not only imply being honest and not taking advantage of public goods for the sake of own interests, but also requires sticking to the liberal principles and keeping electoral promises. According to the interviewed supporters, responsiveness, accountability, and transparency, are the second most significant challenge for liberal parties. The third significant aspect underlined by the supporters relates to the coalition strategies adopted by liberal parties. Coalitions between liberal

238  Blagovesta Cholova and Jean-Michel De Waele and left-wing parties do not pay off at the subsequent election, as these alliances are perceived as ‘treason’ by the Liberals’ supporters. By the same token, coalitions with right-wing conservative rivals can lead to slow decay, as liberal parties become dominated by their bigger coalition partner.

Conclusion and perspectives: a strategic move towards the centre-left? In spite of the difficulties exposed in this chapter, there is a clear potential for strong liberal parties in the region. Liberal parties display several assets that can be crucial for their future development. First, there is a stable liberal electorate in CEE countries examined here. This group represents roughly 10 to 15% of the electorate in the region. This electorate is quite volatile, partly because of its disappointment with the existing liberal parties, but if socialised via a clear programme and with a sufficiently elaborated party strategy, these voters could develop a more long-term commitment. Second, liberal values, and especially civic liberalism, enjoy a large support. But the economic and cultural agenda of liberal parties need to circumvent two particularities of CEE societies: The latent support for a stronger state involvement in the economy and the conservative attitudes of these societies and of some of the liberal voters towards minorities and religious or ethical issues. Clarifying the articulation of these attitudes is crucial for building a long-term support among voters. This potential can be mobilised either by existing parties or by newly created ones and can constitute the milestone of the future revival of liberal parties in the region. Adopting a centre-left positioning in economic terms while keeping a progressive profile on the cultural axis in a region where right-wing conservatives and populists are dominant and where genuine left-wing parties are rather weak, could provide liberals with a niche on the political space or could enable them to play the role of pivot parties. In many of these countries, the centre-left political space is available and could be occupied by liberal parties.

Notes 1. In countries with strong religious heritage and culture like Poland, there is a structural cleavage between conservatives and liberals, while in other countries, such as Bulgaria or Romania, this is not the case. In Czech Republic and Slovakia, divisions occurred between nationalist and liberal parties, the nationalist parties embracing a more conservative strand. 2. We exclude Hungary, as the former members of the ALDE either became Conservative (Fidesz) or disappeared from the political scene before the past decade. The only present-day liberal party in Hungary, the MLP, is quite marginal. 3. Thirty in-depth interviews with voters and three focus groups with supporters and party members in each of the countries have been conducted between 2012 and 2013.

Liberal parties in Central and Eastern Europe 239

References Bozóki A., Ishiyama J.T. (2002), The communist successor parties of Central and Eastern Europe, Armonk, NY: M.E. Sharpe. Cholova B., De Waele J.-M. (2013), ‘The liberal parties in central and eastern Europe: weakness and potential’, Report to the Alliance of Liberals and Democrats for Europe Party. Dalton R.J. (1984), ‘Cognitive mobilization and partisan dealignment in advanced industrial democracies’, Journal of Politics 46(1), pp. 264–284. De Waele J.-M. (2004), Les clivages politiques en Europe centrale et orientale. Brussels: Editions de l’Université de Bruxelles. De Waele J.-M., Gueorguieva P. (2002), ‘La difficile émergence des partis libéraux en Europe centrale’ in Delwit P. (ed.), Libéralismes et partis libéraux en Europe. Brussels: Éditions de l’Université de Bruxelles, pp. 263–285. DIMAP Study on liberal values in Central and Eastern Europe (2009), https://www. infratest-dimap.de/en/ Grzymala-Busse A. (2002), Redeeming the Communist Past: The regeneration of communist parties in East Central Europe. Cambridge: Cambridge University Press. Hanley S. (2006), ‘Getting the right: Redefining the centre-right in post-communist Europe’ in Hanley S., Szczerbiak A. (2006), Centre-right parties in post-communist East-Central Europe. London: Routledge, pp. 9–28. Hanley S. (2008), The new right in the new Europe: Czech transformation and right-wing politics, 1989–2006. London: Routledge. Hanley S., Dawson J. (2016), ‘East Central Europe: The fading mirage of the ‘Liberal Consensus’, Journal of Democracy 27(1), pp. 20–34. Hanley S., Szczerbiak A. (2006), Centre-right parties in post-communist East-Central Europe. London: Routledge. Hanley S., Szczerbiak, A., Haughton, T., Fowler, B. (2007), ‘Explaining the success of centre-right parties in post-communist East Central Europe: a comparative analysis’, SEI working papers 94, pp. 1–53. Held J. (ed.). (1993), Right-wing politics and democracy in Eastern Europe in the 1990s. Boulder, CO: East European Monograms. Hloušek V., Kopecek L. (2016), Origin, ideology and transformation of political parties: East-Central and Western Europe compared. London: Routledge. Jansen T., Van Hecke S. (2011), At Europe’s Service. The Origins and Evolution of the European People’s Party. Heidelberg: Springer. Krastev I. (2007), ‘The strange death of the Liberal Consensus’, Journal of Democracy 18(4), pp. 56–63. Krastev I. (2016), ‘Liberalism’s failure to deliver’, Journal of Democracy 27(1), pp. 35–38. Lawson K., Römmele A., Karasimeonov G. (1999), Cleavages, parties, and voters: Studies from Bulgaria, the Czech Republic, Hungary, Poland, and Romania. London: Praeger. Mair P., Mudde C. (1998), ‘The party family and its study’, Annual Review of Political Science 1, pp. 211–229. Rogier A., De Waele J.-M. (2004), ‘La formation des clivages partisans en Europe centrale et orientale: À la recherche d’une méthode de comparaison’. In: De Waele J.-M. (ed.), Les clivages politiques en Europe centrale et orientale. Brussels: Editions de l’Université de Bruxelles, pp. 11–29.

240  Blagovesta Cholova and Jean-Michel De Waele Rybář M., Deegan-Krause K. (2008), ‘Slovakia’s communist successor parties in comparative perspective’, Communist and Post-Communist Studies 41(4), pp. 497–519. Sikk A. (2012), ‘Newness as a winning formula for new political parties’, Party Politics 18(4), pp. 465–486. Szacki J. (1995), Liberalism after communism. Budapest: Central European University Press. Szczerbiak A., Taggart P. (2008), Opposing Europe? The comparative party politics of Euroscepticism. Oxford: Oxford University Press. Sznajder Lee A. (2011), ‘After the party, the after-parties? The effects of communist successor parties on economic reform in Central and Eastern Europe’, Europe-Asia Studies 63(9), pp. 1697–1718. Tavits M., Letki N. (2009), ‘When left is right: Party ideology and policy in post-­ communist Europe’, American Political Science Review 103(4), pp. 555–569. Todorov A. (2004), ‘Un clivage centré sur le passé communiste’. In: De Waele J.-M. (ed.), Les clivages politiques en Europe centrale et orientale. Bruxelles: Editions de l’Université de Bruxelles, pp. 257–269. Vachudova M.A. (2005), Europe undivided: democracy, leverage, and integration after communism. Oxford: Oxford University Press. Vachudova M.A. (2008), ‘Centre-right parties and political outcomes in East Central Europe’, Party Politics 14(4), pp. 387–405. Whitefield S., Rohrschneider R. (2009), ‘Representational consistency: Stability and change in political cleavages in Central and Eastern Europe’, Politics & Policy 37(4), pp. 667–690. Zuzowski R. (1998), Political change in Eastern Europe since 1989: Prospects for liberal democracy and market economy. Westport, CT: Praeger.

13 Latvia, Lithuania, and Estonia Daunis Auers

Introduction The three Baltic states of Estonia, Latvia, and Lithuania renewed their independence, initially achieved after hard fought battles in the aftermath of World War I, following the collapse of the reactionary August 1991 Soviet putsch. All three states then embarked on the multiple challenges of building market economies, renewing bureaucracies and democratic institutions, and integrating with the West. A quarter century after independence, all three states are established members of the European Union (and are in the core group of states that are in both the Schengen area and have adopted the Euro as their national currency) and the North Atlantic Treaty Organisation (NATO) defence alliance. Estonia and Latvia have also joined the Organisation for Economic Cooperation and Development (OECD), essentially a think-tank for developed economies. However, while the end-destination of democracy, market economy, and integration with the West has been similar, the three states have adopted different roads to get there. Much of these differences can be explained by the influence of different political parties and, in particular, the sharply diverging impact of liberal parties in the three states. This chapter focusses on the different types of liberal parties and their different impact and influence on the politics of the three Baltic states. Liberal parties have dominated Estonian politics for most of this era. The Estonian Centre Party (Eesti Keskerakond – EK) has been a mainstay of Estonian politics since the early 1990s with the Reform Party (Eesti Reformierakond – ER) virtually monopolising executive office in the 21st century. In contrast, Latvia’s Way (Latvijas Ceļš – LC) dominated Latvian politics in the 1990s but liberalism then disappeared from mainstream Latvian politics in the 21st century. Lithuania in turn has seen a far greater turnover of liberal parties throughout the post-1991 era. However, these parties have been far less influential than LC, EK, or ER. It is also important to note that these parties’ understanding of liberalism has differed. In the case of LC and EK, liberalism was an instrument used to bind together the various disparate groups of what were essentially

242  Daunis Auers ‘catch-all’ parties. For others, liberalism was used for means of external identity and legitimation. For others still, it was an electoral tactic, to mark out a vacant territory in the political spectrum. Among the electorally successful liberal political parties discussed in this chapter, only ER can be clearly identified as a conventional European liberal party. This chapter initially considers the origins, development, electoral performance, and executive experience of liberal parties in the Baltic states. The development of the liberal parties in Estonia, Latvia and Lithuania is relatively straightforward, as they largely emerged from the centrist parts of the broad Popular Fronts that had helped sweep away the Soviet Union. However, the story gets more complicated when describing Lithuanian liberalism in the 21st century. The chapter then moves on to consider the ideological and policy positions of the parties before focussing on their structure and organisation. The conclusion focusses on the extent to which we can consider these parties as belonging to the broader European liberal party family.

Origins and development The basic contours of the Baltic states’ party systems were shaped by the battle for independence from the Soviet Union in the 1980s. There were three initial movements. First, the Popular Fronts (known as Są judis in Lithuania), pro-independence movements that served as broad tents uniting anti-Soviet political forces ranging from Baltic nationalists to moderate communists. Posed against them were Russophone Interfront movements (and hence much weaker in Lithuania with its far lower percentage of Russian-speakers than Estonia and Latvia and also with a communist party that was ‘lithuanized’ rather than ‘russified’ as was the case in Estonia and Latvia (Krupavicius 1998: 472)), that were rooted in the reactionary wing of the communist parties and aimed to keep together the existing Soviet state. These Interfronts were the starting point for the pro-Russian speaker Harmony Party in Latvia, pro-Polish minority parties in Lithuania, and the less influential pro-Russian speaker parties in Estonia. In Estonia and Latvia nationalists additionally formed radical right Congress movements that became the starting point for the radical right nationalist National Alliance (Nacionālā Apvienība – NA) and its predecessors in Latvia and the Conservative People’s Party of Estonia (Eesti Konservatiivne Rahvaerakond – EKRE) in Estonia. In Lithuania radical nationalists were absorbed into the right wing of the Są jūdis movement. Evolution of liberal parties in the Baltic states The origins of the liberal parties that emerged in the 1990s in all three Baltic states are to be found in the centrist Popular Fronts that led the push for independence in the late-1980s (Table 13.1). As broad coalitions, combining a great diversity of workers, professionals, intellectuals, and activists, the

Latvia, Lithuania, and Estonia 243 fronts essentially encouraged the type of pluralism that lies at the heart of liberal thinking. EK and ER are the two major liberal parties in Estonia. Both are members of the ALDE group in the European Parliament but their origins, ideological base, and executive experience differs quite significantly. EK is a direct successor of the Estonian Popular Front. Indeed, EK was founded in 1991 by Edgar Savisaar, who had been both the founder and leader of the Estonian Popular Front. In a sense EK is the most direct successor of the Estonian Popular Front, because by the time Savisaar founded the party almost all the other members of the Popular Front had deserted the Front and founded their own parties, often complaining that the alleged authoritarian style of Savisaar had driven them away from the party. Savisaar has been one of the dominant, albeit controversial, figures in Estonian politics for the past three decades and his impact on the party is discussed in more detail in the fourth section of this chapter. EK has been a stable part of the Estonian party system since the early 1990s. Although it has experienced a number of splits and breakaways, including when a number of liberals broke away to form the ultimately unsuccessful Progressive Party, following a scandal in the party in the mid-1990s (Pettai and Toomla 2003), it remains one of the two biggest parties in Estonia. ER was founded a few years later in 1994, but also has its roots in the Popular Front. The party was founded by a merger of the Reform Party, which was a splinter group that had left the Pro Patria National Coalition (Rahvuslik Koonderakond Isamaa – RKEI), and the Liberal Democratic Party (Eesti Liberaaldemokraatlik Partei – ELDP) that was founded in 1990 and was one of the first postcommunist liberal parties to be granted observer status at the Liberal International. ER was led by Siim Kallas, who had been the Governor of the Bank of Estonia and later went on to serve as Prime Minister, Foreign Minister, Finance Minister, and then as a Commissioner in the European Union before narrowly missing out on being elected president of Estonia in September 2016. See Table 13.1 at http://partyfamilies.eu/index.php/category/chapters/ chapter-13/ LC has been Latvia’s only electorally successful liberal party and it played a dominant role in Latvian politics between 1993 and 2002. It was formed at a meeting in the Latvian beach resort of Jurmala on 13 February 1993 in the run-up to Latvia’s first postcommunist elections in June 1993. The Jurmala meeting was attended by members of Club 21 (Klubs 21 – Club 21 was founded in 1991 by 21 leading members of the Latvian business and political communities. It brought together figures from the Latvian elite (to informally debate various economic, social, and political issues) and émigré Latvians unofficially representing the World Federation of Free Latvians (Pasaules Brivo Latviešu Apvieniba – PBLA), the main umbrella

244  Daunis Auers organisation for global émigré groups. Moderate ex-communists were later invited to join the party. The wide variety of these three factions prompted Anatol Lieven (1994: 301) to unforgettably describe it as a ‘political bouillabaisse’. The party used this diversity as a positive and central element of its 1993 election campaign slogan – ‘only those who can unify themselves can unify others’ (Dreifelds 1996: 88). The broad appeal translated into winning 36 of 100 seats in the 1993 parliamentary election, still a record margin of victory in a Latvian election. LC subsequently dominated the 1990s and was the only party represented in all eight government coalitions in this decade, half of which were led by LC prime ministers. However, the party’s influence was in decline by the beginning of the 21st century and it failed to win any seats in the 2002 parliamentary election. As a result, it formed an electoral coalition with Latvia’s First Party (Latvijas Pirmā Partija – LPP), a business-friendly conservative-clerical party in the 2006 election before formally merging the two parties in 2007. However, the party subsequently slumped in the polls leading it to form an electoral coalition with the conservative People’s Party – named “For a Good Latvia” (Par Labu Latviju – PLL) – and then finally being renamed the Šlesers LPP/LC Reform Party but failed to win any seats in the early election of 2011. The party was subsequently disbanded. More recently, Latvian Development (Latvijas Attīstībai – LA) was founded in December 2013. In contrast to the broad base of support for its liberal predecessor LC, LA attracted just 201 founders to its conference, thus only barely passing the 200 minimum member quorum needed to legally register a political party in Latvia at that time (the quorum has since been raised to 500 members for parties competing in national elections). However, despite filling a liberal niche in the Latvian party spectrum, the party polled less than 1% of the votes at the October 2014 parliamentary election. This is largely because the leadership of LA was a collection of tired-looking political hacks from a number of different political parties – Einars Repše, Edgars Jaunups, and Dans Titavs, the founders of New Era (Jaunais Laiks – JL), which had won the 2002 parliamentary election; Uldis Osis formerly of LC; and Emils Jakrins who had been a member of the radical right populist National Alliance (Nacionālā Apvienība – NA). Nevertheless, LA has not faded from the Latvian political spectrum since the election, it continues to hold regular conferences and seminars as well as articulate a consistent liberal position on political issues to the media. There are persistent rumours that the so-called ‘Liberal Six’ faction in the governing conservative Unity Party (Vienotība – V) may break away and form a liberal fraction in the parliament and eventually merge with LA. The liberal centre ground has been far more contested in Lithuania than in Estonia or Latvia and there are (and have been) several political parties in Lithuania that are identified as liberal through their affiliation with

Latvia, Lithuania, and Estonia 245 transnational political parties. Indeed, the first liberal party of the modern era in the Baltic states was founded in Lithuania in 1990, after Lithuania had become the first Baltic state to pass a law on political parties following its declaration of independence on 11 March 1990. The Lithuanian Liberal Union (Lietuvos Liberalų Są junga – LLS) was founded in 1990 and was broadly aligned with the Sajudis movement. Several years later, as was the case with EK in Estonia and LC in Latvia, a liberal party – the Centre Union (Lietuvos Centro Sajunga – LCS) – emerged from the political centre of the popular front Sajudis movement although it later merged with LLS and the Modern Christian Democratic Union (Moderniju Krikscioniu Demokratu Sajunga – MKDS) to establish the Liberal and Centre Union (Liberalų ir centro są junga – LiCS) in 2003. A conflict between LiCS leaders saw a group break away and form the Liberal Movement (Liberalų Są jūdis – LS) in 2006. However, LiCS folded in 2014 after failing to win any seats in the 2012 parliamentary election and was merged into the Lithuanian Freedom Union (Lietuvos Laisvės Są junga – LLS) along with YES-Homeland Revival and Perspective (TAIP – Tėvynės atgimimas ir perspektyva), which was founded in 2011 by the mayor of Vilnius, Arturas Zuokas. Liberal parties had mixed results in the 2016 parliamentary election, with DP failing to pass the 5% barrier for the proportional representation seats and winning just 2 (of 71) constituency seats, whereas LS raised its number of parliamentary seats from 10 to 14. LiCS and its predecessors were relatively small parties that ‘hibernated in the margins of the party system’ (Duvold and Jurkynas 2013: 127). The 2000 local government elections saw the emergence of two far more powerful and influential liberal parties. The Lithuanian Liberal Union (Lietuvos Liberalų Są junga – LLS) was founded and led by Rolandas Paksas, the then popular mayor of Vilnius. Paksas later abandoned the Liberal Union to establish the Liberal Democratic Party (Liberalų Demokratų Partija – LDP) that was then subsequently renamed Order and Justice (Partija tvarka ir teisingumas – PTT) and adopted a conservative rather than liberal identity. The same 2000 municipal election also saw the New Union (Social Liberals) (Naujoji Są junga (Socialliberalai) – NS) also emerge as a political force under the leadership of Arturas Paulauskas, a former Lithuanian Prosecutor-General who had come close to winning the popular vote in the first round of voting in the 1998 presidential election (although he subsequently lost in the run-off). The Labour Party (Darbo Partija – DP) was founded in 2003 by Viktor Uspaskich, a Russian-born businessperson and won the biggest share of the vote in both the 2004 European Parliament and national parliament election. The earliest post-1991 liberal parties in the Baltic states shared common origins in the Popular Front movements. However, the parties to emerge in Latvia and Lithuania in the 21st century had more diverse heritages. They have also experienced very different levels of electoral success.

246  Daunis Auers Electoral performance As Table 13.2 indicates, Estonia’s two liberal parties have been the two electorally best performing parties in Estonia in the 21st century. Indeed, in the last three parliamentary elections (2007, 2011, and 2015) they have jointly won more than 50% of the vote and more than half of the seats in the Estonian legislature. Indeed, the two parties have jointly won at least one-third of parliamentary seats since they both first participated in a parliamentary election in 1995. In contrast, the liberal LC experienced its greatest success in the early 1990s, gaining a record share of the vote in the first postcommunist elections in Latvia in 1993, before seeing its vote cut away to the extent that it failed to pass the 5% barrier in the 2002 election. Only an electoral coalition with LPP allowed it to reenter parliament in 2006. More recently, LA won less than 1% of the vote in the 2014 parliamentary election meaning that there are no liberal parties represented in the Latvian parliament between 2014–2018. See Table 13.2 at http://partyfamilies.eu/index.php/category/chapters/ chapter-13/ In the 1990s, Lithuanian politics was dominated by swings between big political parties on the left – the Lithuanian Democratic Labour Party (Lietuvos Demokratinė Darbo Partija – LDDP) – and right – the Homeland Union (Tėvynės Są junga – TS). However, the dominance of these two parties was broken in the 2000 election when LLS and NS came, respectively, second and third in the poll and formed a centrist government. Liberal parties again performed well in the 2004 election, and after a dip in 2008, won more than a quarter of parliamentary seats again in 2012. This electoral performance – of Latvian liberals performing well in the 1990s, but Estonian and Lithuanian liberals growing more dominant in the first two decades of the 21st century – is also mirrored in these parties’ executive experience. Executive experience Estonia’s ER has dominated the executive since January 2002 when an ER and EK minority coalition government took power under Prime Minister Siim Kallas (ER). This coalition came about after the two parties had agreed on creating a governing coalition at the Tallinn Municipal council, causing the then Prime Minister Mart Laar (IL) to resign and create the opportunity for ER and EK to form a new government. ER had also previously held government office (initially from October 1995 until November 1996), but 2002 marked the beginning of ER’s executive dominance. Indeed, it has held the Prime Minister’s post since 2005. In contrast, EK has only intermittently – and typically only for short periods of time – held government office.

Latvia, Lithuania, and Estonia 247 See Table 13.3 at http://partyfamilies.eu/index.php/category/chapters/ chapter-13/ Table 13.3 indicates the extent to which LC was the dominant governing party in 1990s Latvia, in much the same way as ER has dominated Estonian politics since the mid-2000s. LC was a part of every government between 1993–2002, although it was admittedly at the peak of its powers in the early and mid-1990s. LC, now in coalition with LPP, fleetingly returned to power in the mid-2000s and it was an LC/LPP Prime Minister, Ivars Godmanis, who oversaw the first harsh economic austerity measures after Latvia’s deep economic recession from 2008–2009. Lithuania’s liberals first tasted executive office as minority partners in a centre-right government in the late 1990s and have been a constant presence in government since then, pivoting from the centre to form coalitions with parties both of the left and the right. However, in contrast to the other two Baltic states, Lithuania’s liberal parties have only once held the Prime Minister’s office – Rolandas Paksas (LLS) – despite liberal parties holding around 50% of ministerial posts between 2001 and 2006.

Ideology and policy positions The central tenet of liberalism is the promotion of individual freedom. A more nuanced definition from the Liberal International’s manifesto defines liberalism as: Freedom, responsibility, tolerance, social justice and equality of opportunity… These principles require a careful balance of strong civil societies, democratic government, free markets, and international cooperation. (Oxford Manifesto 1997) As a result, this section will focus on the ideological position of liberal parties in the Baltic states across three policy dimensions – minority policy, market economy, and European Union. Party positions will initially be plotted with data from the Comparative Manifesto Project (CMP) that estimates party positions on a range from –100 (left) to +100 (right) based on coded party manifestos (Budge et al. 2001; Jahn 2011; Laver and Budge 1992). In addition to the data this section will also explain how scholars and other observers have understood the Baltic liberal parties. After all, the policy approach of the parties can diverge from that laid out in the party manifesto because the manifesto may be used to attract voters while the nature of coalition governments in all three states means that the parties are unlikely to be in a position where they are expected to deliver on all the promises. Before moving on to the three policy dimensions, the general policy orientation of the Baltic liberal parties can be plotted on the CMPs RILE

248  Daunis Auers (right-left) indicator, a standard measure of party positions that summarises policy inclinations across whole party programmes. In the case of Estonia, the RILE indicates a rather large gap between the two Estonian liberal parties. In 1995, the gap was 28 points and, while it narrowed to just 3.1 points in 2003 (perhaps not coincidentally, the two parties had spent the previous year in a governing coalition), it returned to 36.3 in 2007. EK has clearly been consistently to the left of ER. This also reflects the way in which the parties are understood in Estonia, where ER is seen as a standard liberal party and EK is typically identified as either pro-Russian, left-wing, or both. This is also reflected in the difficulties that the two parties have had in cooperating with each other in government and helps to explain why they have spent only two short periods working together in government coalitions. For example, the 2003 Nations in Transit report on the formation of the new government including both EK and ER refers to EK as a ‘left wing party’ while in the same sentence ER is called a ‘centre-right party’ (Nations in Transit 2003). The same report also referenced the fact that the two parties were ‘ideologically different’. Similarly Pettai and Toomla (2003) also saw this as an unwieldy coalition combining liberals (the Reform Party) and a left-leaning Centre Party. This division between ER and EK still remains today. Before the 2015 election Prime Minister Taavi Roivas (ER) stated that he aimed to build Estonia into ‘a Nordic, Liberal Estonia. We want Estonia to be acknowledged as a Nordic country. A country with Nordic living standards and security, guiding the world, a leading country in terms of individual freedom and economic security’ (Deloy 2015: 2). At the same time, Roivas insisted of EK that the party’s ideology ‘in terms of security will never change as long as Edgar Savisaar chairs the Centre Party … and will never share our values’ (Deloy 2015: 2). In the case of Latvia only LCs manifestos have been coded in the CMP. Table 13.4 indicates that although LC has experienced some fluctuation of its RILE indicator, it has consistently stayed to the centre right in general. This fits in with the general conception of LC as a centre-right party. Indeed, an article by Andrejs Pantelejevs (1993: 2) in the second edition of the party’s own newspaper claimed that LC ‘simultaneously stands for both conservative and liberal values’. However, at its first party congress in 1994, the LC leadership announced that it had joined the Liberal International and the party subsequently was consistent in maintaining its external liberal identity. Finally, Lithuania’s liberal parties have differences in their RILE score, with, for example, the LS scoring +6.8 in the 2012 parliamentary election whereas DP scored -24.0 (which in itself was a 37.7 point swing from +13.7 in 2004). DP’s swings between elections reflect the observation that scholars see it as a leader-oriented anti-establishment party that has emphasises newness rather than liberalism as an ideology (Sikk 2004). Similarly to EK in Estonia and LC in Latvia, liberal ideology has largely been used for external legitimacy. This is true of a number of these Lithuanian liberal parties. Thus the policies of the NS in 2000 were described as ‘promising to fight

Latvia, Lithuania, and Estonia 249 Table 13.4  Liberal parties in the Baltic states’ CMP left-right scores Estonia Party EK ER

1992 0.3 -

1995 -15.3 12.7

1999 16.2 8.6

2007 -29.2 7.1

2011 13.0 3.2

2015 -21.8 0.5

Ave. -5.1 6.0

2010 -

2011 -

2012 -

Ave. 11.9

2004 -

2008 -

2012 -

2016 -

Ave. -3.4

5.1 13.7

5.3 -0.6 -3.3

6.8 -24.0

-

5.7 5.2 3.1 -4.5

2003 0.8 3.9 Latvia

Party LC

1993 15.7

1995 28.6

1998 4.4

2002 1.6

2006 9.0

Lithuania Party LCS NS LLS LiCS LS DP

1992 -

1996 1.7

-

-

2000 -8.6 -5.4 5.7 -

Source: Volkens et al. (2015)

corruption and promote ethical, open and accountable politics’, although it has also supported the reintroduction of capital punishment that could hardly be described as a liberal policy (Sikk 2004: 10). Thus Sikk sees NS as a party that is somewhere between social democracy and liberalism, with a populist anticorruption message at its core. However, LLS, LiCS, and LS fit more easily into the broader European liberal family. Freedom and human rights Individual freedom has been central to both traditional and contemporary liberal ideology. The CMP ‘freedom and human rights’ category tracks references to personal freedom and civil rights in party manifestos. Table 13.5 reveals that Baltic liberal parties show little significant variation on this policy dimension and tend to be clustered around the centre with average scores varying from 0.2 (EK) to 5.8 (LLS). One of the central ‘freedom’ issues in the post-1991 Baltic states has concerned the status of the Russian minorities in Estonia and Latvia and, to a far lesser extent, Lithuania. Estonia, and Latvia had experienced rapid russification in the Soviet period. For example, in the decade after World War II Latvia saw an influx of 650,000 people from other Soviet Republics, primarily Russia (Plakans 1995: 154). In the 1990s, Estonia and Latvia had to legislate Russian speakers’ citizenship status while in the 2000s the main focus has been on the role of the Russian language in public life. These issues were of little relevance in Lithuania, where liberal citizenship and language laws were swiftly passed largely because, in contrast to Estonia and Latvia, ethnic Lithuanians made up 80%-plus of the population.

250  Daunis Auers Table 13.5  Liberal parties in the Baltic states’ CMP freedom and human rights score Estonia Party EK ER

1992 0.0 -

1995 0.0 0.6

1999 0.4 6.2

2003 0.0 1.9

2007 0.0 0.9

2011 0.4 1.5

2015 0.5 1.5

Ave. 0.2 2.1

2006 0.0

2010 -

2011 -

2.8

2008 2.5 3.0 1.2

2012 1.1 0.0

2016 -

2.0 4.4 5.8 2.8 2.0 1.1

Latvia LC

1993 0.4

1995 11.9

1998 1.5

2002 0.0

Lithuania LCS NS LLS LiCS LS DP

1992 -

1996 2.1 -

2000 1.9 4.4 5.8 1.9 -

2004 4.1 2.1

Source: Volkens et al. (2015)

Estonia’s and Latvia’s liberal parties have generally been more open to more liberal citizenship and integration policies than the centre-right parties in parliament. However, the nature of Baltic politics in the 1990s was such that the citizenship debate was dominated by parties that had emerged from the radical Congress movements of the right-wing of the People’s Fronts. Nevertheless, in the late 1990s EK emerged as the major advocate for liberalising the Estonian citizenship law to grant automatic citizenship to the children of stateless citizens. The law was amended in December 1998 after ER gave its deputies a free vote on the motion and a majority voted together with ER (Nakai 2014). EK and ER parliamentary deputies similarly came together in 2000 to remove restrictions to the use of Russian in the workplace (ibid.). In its 1993 electoral programme, LC proposed granting citizenship to Russian speakers gradually, based on individual merit such as Latvian language fluency (Latvijas Vēstnesis 1993). Indeed, Mark Jubulis (2001: 8–9) has argued that LC was far more liberal than other Latvian parties on this issue: ‘LC has followed a cultural form of nationalism, which made citizenship primarily contingent upon integration into Latvian society through the acquisition of language skills… LC viewed non-citizens as potential citizens, capable of being integrated, while the radical right viewed the non-citizens as “occupiers” who should be barred from ever acquiring Latvian citizenship’. However, in contrast to EK, LC was never a policy-leader on citizenship and language policy, frequently siding with the nationalist consensus rather than risk losing voter support (Nakai 2014).

Latvia, Lithuania, and Estonia 251 Market economy Liberal economic theory has emphasised the importance of free markets – the famed ‘invisible hand’ of Adam Smith (1776/2003). Although this has been tempered by elements of state intervention in the 20th century, faith in the market economy remains central to modern liberal economic thinking. Support for the market economy is captured in the CMPs ‘markeco’ indicator that incorporates the CMPs ‘free market economy’ and ‘economic orthodoxy’ variables (Volkens et al. 2015). Table 13.6 shows that all the Baltic liberal parties are in a narrow point spread ranging, on average, from 0.6 (EK) to 5.8 (LC). In terms of Estonia it is not surprising to see ER to the right of EK. After all, EK has tended to be seen as rather more left-wing than ER. Latvia’s LC has the highest average score largely because of an evaluation of +21.0 in 1993. This reflects LC’s portrayal of itself as the party of economic reform, liberalisation, and modernisation (Dreifelds 1996: 87). LC’s economic policy was guided by the Latvia-2000 economic programme written in 1992 by the ‘Georgetown gang’ of Latvian economists who had briefly studied under the Latvian-American Georgetown economics professor, Juris Vīksniņš. Marju Nissinen (1999: 133) has compared it with the German soziale Marktwirtschaft economic model in that it sought to create a market economy that also emphasised the rights of workers, good working conditions and a generous state social welfare system. Lithuania’s liberal parties have also seen little variation in their economic policies, with all parties being grouped just to the right of the centre in terms of their evaluation of the Table 13.6  Liberal parties in the Baltic states’ CMP market economy score Estonia Party EK ER

1992 1.4 -

1995 1.4 5.5

1999 0.0 14.8

2003 1.1 5.4

2007 0.0 2.1

2011 0.4 2.1

2015 0.2 2.8

Ave. 0.6 5.4

2006 1.5

2010 -

2011 -

2012 -

5.8

2008 4.1 2.7 1.2

2012 3.3 1.2

2016 -

4.9 2.5 2.3 3.5 3.0 0.8

Ļatvia LC

1993 21.0

1995 2.4

1998 4.2

2002 0.0

Lithuania LCS NS LLS LiCS LS DP

1992 -

Source: Volkens et al. (2015)

1996 7.5 -

2000 2.4 2.5 2.3 2.4 -

2004 4.1 0.0

252  Daunis Auers market economy. This congruence in economic policy could be explained by support for integration and accession to the European Union and the common economic reform demands of the acquis communautaire. European Union integration Membership of the European Union promotes liberal values in terms of opening regional trade increasing personal freedom of travel and professional opportunities as well as the ‘conditionality’ of accession promoting human rights, rule of law, and liberal democracy (Gateva 2015). The CMPs European Community/Union positive indicator tracks favourable mentions of the organisation in party programmes. Table 13.7 indicates the broad liberal party degree of consensus on the support for and desirability of political integration into the European Union with average scores varying from 0.2 (NS) to 3.7 (LC). All the liberal parties have supported integration with the international community. For example, LC monopolised the foreign minister portfolio in Latvia from 1993 to 2002 and consistently worked towards integration with the West, rather than Russia and the East. Estonia’s EK may been as an exception in this respect. Although the party is formally pro-European, it has had a cooperation agreement with Vladimir Putin’s United Russia party since 2004. As a result, within Estonia EK has increasingly been seen as a pro-Russian party rather than a European liberal movement. This east-west issue has caused cracks within the party itself, with a more liberal wing of the party, led by a Table 13.7  Liberal parties in Baltic states’ CMP European Community/union positive score Estonia Party EK ER

1992 0.0

1995 0.0 1.8

1999 2.4 0.8

2003 5.8 5.1

2007 1.3 4.2

2011 1.3 0.7

2015 0.6 1.8

Ave. 1.6 2.4

2006 1.5

2010 -

2011 -

2012 -

3.7

2008 1.8 1.8 1.2

2012 1.3 1.0

2016 -

1.5 0.2 2.3 1.5 3.1 1.1

Latvia LC

1993 4.0

1995 4.8

1998 3.6

2002 4.7

Lithuania LCS NS LLS LiCS LS DP

1992 -

Source: Volkens et al. (2015)

1996 1.7 -

2000 1.3 0.2 2.3 1.3 -

2004 1.3 1.1

Latvia, Lithuania, and Estonia 253 new generation of politicians such as Kadri Simson and Mailis Repss, who have both challenged long-standing leader Edgar Savisaar’s leadership of the party and particularly the perceived reorientation to the east. How liberal are they? The Baltic liberal parties may have adopted liberal ideologies for different reasons. Some, such as LC, adopted it for pragmatic reasons – because contemporary liberalism is difficult to pin down and could thus be used as a label that incorporated a great deal of ideological and pragmatic diversity. It was also used as a vote-winning strategy at a time when liberal reforms had public support. Liberalism has been used as a source of identity and external legitimation through membership of international liberal party groups while it has also been adopted simply because there was perceived to be a space in the party system. Allan Sikk (2004) has argued that NS and LLS are best characterised as a ‘new’ party rather than an ideologically oriented party. In this sense, the party’s self-proclaimed liberalism is simply part of the party’s branding rather than an identifier of its actual ideological orientation. Nevertheless, the Baltic liberal parties have actually proved to be ideologically cohesive, tending to promote individual freedoms and human rights, market-economy oriented reforms and integration with the European Union.

Structure and organisation There is a clear differentiation in the rigour of party organisations across the Baltic states. Political parties in Estonia are the most organised – perhaps unsurprisingly, as both EK and ER have been in existence for more than two decades, in contrast to the far younger current crop of liberals in Latvia and Lithuania, all of whom were founded in the 21st century. Indeed, the Centre Party has been seen as one of the best-organised parties in Estonia. It is strongly rooted in metropolitan areas, but also has a regional membership. Reform Party is largely limited to the cities. However, liberal parties in Latvia and Lithuania have a less clearly defined supporter base, with voters in Lithuania in particular often defecting from one party to another in elections. This lack of roots has strengthened the importance of charismatic leaders in Lithuania (Jurkynas 2005). Nevertheless, leaders also play an important role in these parties. Estonia’s EK has been dominated by its founder, Edgar Savisaar, for more than two decades despite multiple corruption scandals and the leader’s failing health. Similarly, ER was founded by Central Bank Governor Siim Kallas in the mid-1990s and, two decades later, ER nominated Kallas for the 2016 presidency (although, in contrast to EK, ER has cultivated several other leaders over the past two decades). Similarly, Lithuania’s LLS and NS have been

254  Daunis Auers leader-oriented whereas LC’s initial success in Latvia was founded on the strength of the individual politicians that the party had collected under one roof rather than the ideology that bound them together (and the failings of LA may well be ascribed to its lack of charismatic leadership). Main party structures The Baltic liberal parties have broadly similar party organisations. They tend to have (1) regular party congresses that elect (2) the party chairperson, (3) a Council, and (4) a Board that meets on a weekly or monthly basis and runs the party together with (5) a General Secretary who also heads up a (6) party central office. Congresses are typically held every year, although Estonia’s EK has one but every two years (giving the leader a tighter grip on the party, with less opportunity to challenge Savisaar). EK also differs to other large parties in that it allows all its members to attend. Other parties have a delegate system, although LA in Latvia is so small that all members can attend its Congress. Party Congresses hear reports from party executives, debate policy, and elect the party chairman and the Board as well as voting on party manifestos. ERs Congress also votes on the party list for the parliamentary election. Congresses elect the chairperson of the party who then chairs the Council and the Board. Party Councils lie somewhere between the Congress and the Board and are formed to represent the broader interests of the party between Congresses (with Boards having more executive functions). The Councils are typically composed of representatives from the party’s central organs and regional groups and have programmatic and candidate selection functions. However, LA in Latvia is so small that it has no Party Council whereas the Boards in Lithuania are so large that they also fill the Council function. These boards vary in size across the Baltic states. In 2016 EK had 17 board members, ER had 15, LA had 9, DP had 48 (DP calls its board the Presidium), LLS had 29, and LS had 25. The Boards deal with ticklish political issues – such as political negotiations with other parties, naming candidates for political office as well as financial issues. In addition to this, parties often also have Ethics Commissions and can organise other ad hoc panels and commissions. The Boards also provide day-to-day oversight over the General Secretaries and central offices. Both EK and ER employed seven full-time staff in 2016 (Centre Party 2016, Reform Party 2016). Offices in Latvia and Lithuania are smaller, with the Latvian LC and Lithuanian parties typically using parliamentary resources for party management. Party leaders (chairpersons) are usually selected by the party Congresses and have mandates ranging from one to two years. The Estonian Centre Party stands out in terms of having a remarkably stable leadership, with Edgar Savisaar having been the party leader for almost all the two-decades plus of the party’s existence. He was forced to give up the leader’s post in the mid-1990s after it was leaked that he had recorded secret negotiations with

Latvia, Lithuania, and Estonia 255 other party leaders while negotiating a new governing coalition. However, just half a year after being forced out Savisaar had engineered a return to the post of party chairman. In the 2000s, Savisaar was again challenged for the cooperation deal he had signed with Vladimir Putin’s United Russia party. However, Savisaar again survived a leadership challenge at the Congress thanks to his popularity with the rank-and-file members who are free to attend (and vote) at EK party Congresses. ER has had three party chairmen since the party was founded in 1994. For the first decade the party was led by Siim Kallas, the party’s founder. However, after he left Estonia to take up a Commissioner’s post in Brussels in 2004 the baton was handed to Prime Minister Andrus Ansip. Like his predecessor, Kallas, in 2014, Ansip left Estonia to take up a post in the European Commission and the Prime Minister’s post as well as the ER chairmanship was taken by Taavi Roivas. Both leadership transitions were smooth and lacking controversy. LC had five different chairmen during its existence, reflecting the fact that it was more of a ‘team of stars’ than a one-person operation. Chairmen were elected at party congresses and the transfer of power was smooth and lacking the vitriolic language seen when there have been challenges against Savisaar. The first chairman was Valdis Birkavs (1993–1997), who was also LC’s first Prime Minister. He was followed in the post by a head of the party fraction (Andrejs Pantelejevs 1997–2000) and another Prime Minister (Andris Berzins 2000–2003). A number of Lithuania’s liberal parties are also rather leader-oriented. One observer noted of NS that it ‘came to power as a group of very different people who are more concerned about their personal interests and are not knowledgeable about party discipline… the foundation of its [NS] attractiveness lies not in its ideology but its leader Arturas Paulauskas’ (Novogrockiene 2000: 149). In the same way DP has been dominated by Viktor Uspaskich (a rare ethnic Russian in Lithuanian politics) who founded the party in 2003, spent part of 2006 and 2007 hiding from Lithuanian law enforcement in Russia after being investigated for tax violations for which he was eventually successfully prosecuted in 2013, although he has not served the sentence after claiming immunity as both a member of the Lithuanian and European parliaments. Disputes between Lithuanian party leaders have also contributed to the fragmentation of liberal parties – as was the case in the early 2000s when a conflict between Rolandas Paksas and Eugenijus Gentvilas led to the former leaving LiCS and setting up a new liberal party. Party membership As Table 13.8 indicates, there are significant differences in party membership between the Baltic states. Estonian and Lithuanian parties have extensive branch systems in the major municipalities and regions whereas Latvia’s liberal parties have been significantly smaller and thus their regional

256  Daunis Auers penetration has been less significant. Estonia boasts the highest levels of party membership, as a percentage of the population, in the region, with the two liberal parties, EK and ER, having long been the two biggest political parties in Estonia. In 2009, 4.84% of the Estonian electorate were party members in comparison to 2.66% in Lithuania and 0.74% in Latvia, which was the lowest indicator in the EU (Biezen et al. 2012). See Table 13.8 at http://partyfamilies.eu/index.php/category/chapters/ chapter-13/ Parties in Latvia are smaller largely because they choose to be. LC created the model that most Latvian parties have chosen to follow, of making parties relatively difficult to join, requiring three references from existing party members (no mean feat when there were a little more than 1,000 members) with applications later reviewed at party Conferences and, only if there were no objections from party members, was the application approved, a process that could take from two to six months. This was largely done to prevent the parties from ‘hostile take-overs’. However, other institutional factors are also at play. Until very recently, Estonian parties needed 1,000 members to be registered (now 500) whereas Latvian parties needed just 200 members (since raised to 500), and Lithuania was somewhere in-between with a requirement of 400 members that has since been raised to 1,000 and, from 2015, 2,000 members. Lithuania also has a mixed electoral system that encourages the formation of local membership groups. Finally, Lithuania’s Sajudis popular front movement, with its extensive membership, held together far longer in Lithuania than its equivalents in Estonia and Latvia thus forcing other parties to also attempt to build up significant party membership structures to compete with Sajudis.

Conclusion To what extent do the liberal parties in the Baltic states make up a party family? In terms of their origins, the early liberals (EK, ER, LC, LLS, and LCS) had common roots in the centrist elements of the Baltic Popular Fronts. However, while the Lithuanian liberal parties and ER were also committed to liberalism as an ideology, EK and LC adopted liberalism rather as a form of external (and internal) identity – a glue, to bind a rather disparate membership together. Some of the more recent liberal parties in Lithuania – NS and DP – also adopted a liberal identity for reasons other than a commitment to the ideology and are frequently identified as ‘left’, ‘anti-establishment’, ‘populist’, or ‘new’ parties rather than liberal. Nevertheless, all these parties have joined pan-European liberal organisations and, when elected to the European Parliament, sit in the ALDE group. Moreover, the parties also share a great deal of policy congruence across the dimensions discussed in the ideology and policy discussion

Latvia, Lithuania, and Estonia 257 of this Chapter. Indeed, the Baltic liberal parties occupy an ideological space in all three party systems that is in the political centre between the left and right parties (see, for example, Jurkynas 2005). Although there are differences in the size of party memberships as well as the degree of central office professionalisation, the basic organisation of the parties – with Congresses that appoint party leaders and other governing structures as well as powerful charismatic leaders front and centre – also point towards similarities. Certainly, this congruence can at least be partially explained by membership of ALDE and other international liberal organisations as well as some, albeit limited, cooperation at a Baltic level. Moreover, the political dominance of ER and EK in Estonia also provides a successful political model for other Baltic liberal parties to follow. Indeed, LA emphasises its links with liberals in Estonia and Lithuania and is actively following ERs organisational model and policy initiatives to attempt to replicate its success in Latvia.

References Auers D., Ikstens J. (2005), ‘Politisko partiju demokrātiskā loma’ [the democratic role of political parties]. In: Rozenvalds J. (ed.), Cik demokrātiska ir Latvija: Demokrātijas audits. [How Democratic is Latvia? Democracy Audit]. Rīga: Latvijas Universitāte. Berglund, S., Ekman, J., Deegan-Krause, K., Knutsen, T. (eds.), (2013), The handbook of political change in Eastern Europe. New York: Edward Elgar. Biezen I., Mair P., Poguntke T. (2012), ‘Going, Going,… Gone? The decline of party membership in contemporary Europe’. European Journal of Political Research 51(1), pp. 24–56. Budge I. et al. (2001), Mapping policy preferences: Estimates for parties, electors, and governments 1945–1998. Vol. 1. Oxford: Oxford University Press. Deloy C. (2015), ‘General elections in Estonia: A more uncertain election than forecast’, European Elections Monitor. Foundation Robert Schumann. Dreifelds J. (1996), Latvia in transition. Cambridge: Cambridge University Press. Duvold K., Jurkynas M. (2013), ‘Lithuania’. In Berglund S. et al. (eds.), The handbook of political change in Eastern Europe. New York: Edward Elgar, pp. 125–166. Estonian Centre Party (2016). Retrieved from http://www.keskerakond.ee/et/erakonnaliikmed/esimees.html Estonian Reform Party (2016). Retrieved from https://www.reform.ee/uldkogu Estonian National Election Committee (2016). Retrieved from https://www.vvk.ee Gateva E. (2015), European Union enlargement conditionality. Basingstoke, UK: Palgrave Macmillan. Jahn D. (2011), ‘Conceptualizing left and right in comparative politics. Towards a deductive approach’, Party Politics 17(6), pp. 745–765. Jurkynas M. (2005), ‘2004 general elections and changing left-right in Lithuania’, Lithuanian Political Science Yearbook 2004, pp. 11–30. Ikstens J. (2003), ‘Party organisations in Latvia’, National Council for Eurasian and East European Research, Washington, DC.

258  Daunis Auers Krupavicius A. (1998), ‘The post-communist transition and institutionalization of Lithuania’s parties’, Political Studies 46, pp. 465–491. Jubulis M. (2001), Nationalism and democratic transition: The politics of citizenship and language in post-soviet Latvia. Lanham, MD: University Press of America. Latvian Central Election Commission (2016). Retrieved from https://www.cvk.lv Laver M., Budge I. (1992), ‘Measuring policy distances and modelling coalition formation’. In: Laver M., Budge I. (eds.), Party policy and government coalitions. New York: St. Martin’s. pp.15–40. Liberal International (1997), ‘Oxford manifesto 1997: The liberal agenda for the 21st century’. Retrieved from http://www.liberal-international.org/resolutions/index. html Lieven A. (1994), The Baltic Revolution. London: Yale University Press. Lithuanian Central Election Commission (2016). Retrieved from https://www.vrk.lt Lithuanian Labour Party (2016). Retrieved from https://www.darbopartija.lt/apie/ prezidiumas/ Lithuanian Freedom Union (2016). Retrieved from http://lls.lt/ Lithuanian Liberal Movement (2016). Retrieved from http://www.liberalai.lt/ Lithuanian Ministry of Justice (2016), ‘Data on political parties’. Retrieved from http://www.tm.lt/tm/partijos/ Nakai R. (2014), ‘The influence of party competition on minority politics: A comparison of Latvia and Estonia’, Journal on Ethnopolitics and Minority Issues in Europe 13(1), pp. 57–85. Nations in Transit (2003), Estonia. Annual report. Washington: Freedom House. Nissinen M. (1999), Latvia’s transition to a market economy: Political determinants of economic reform. Basingstoke, UK: Macmillan Press. Novagrockiene J. (2001), ‘Elections to the Seimas 2000: Party system evolution or its transformation?’. In: Jankauskas A. (ed.), Lithuanian Political Science Yearbook 2000. Vilnius, Lithuania: IIRPS, pp. 138–150. Pantelejevs A. (1994), Mēs neatklājam jaunu Ameriku. [We’re not discovering a new America]. Latvijas Ceļš 2:2. Pettai V., Toomla R. (2003). ‘Political parties in Estonia’. Washington, DC: National Council for Eurasian and East European Research. Plakans A. (1995), The Latvians: A Short History. Stanford, CA: Hoover Press. Sikk A. (2004), ‘Successful new parties in the Baltic states: Similar or different?’ Paper prepared for the conference, ‘The Baltic States: New Europe or Old?’, University of Glasgow, Scotland, 22–23 January 2004. Smith A. (1776/2003), An inquiry into the nature and causes of the wealth of nations. New York: Dover. Smith A. (2012), ‘Institutional Design and Party Membership Recruitment in Estonia’, Demokratizatsiya 21(4), pp. 559–581. Volkens A., Lehmann P., Matthieß T., Merz N., Regel S., Werner A. (2015), The Manifesto Data Collection. Manifesto Project (MRG /CMP / MARPOR). Version 2015a. Berlin: Wissenschaftszentrum Berlin für Sozialforschung.

14 The Liberals in Europe The alliance of Liberals and Democrats for Europe Wouter Wolfs and Steven Van Hecke

Introduction The liberal party at European level – the Alliance of Liberals and Democrats for Europe Party (ALDE Party) – is not a party in the strict sense of the word, at least not in the same way as parties exist at the national (or regional) level. Most classic definitions of what constitutes a party include the element of competing for elections. Sartori (1976) argues that the criterion that distinguishes parties from other organisations is that parties present candidates at elections to the systems’ representative assembly. However, this does not apply to the European level: Elections to the European Parliament (EP) are held in national constituencies and national parties select candidates and make up the electoral lists. A European-wide constituency with electoral lists does not (yet) exist and therefore there is no role to be played by parties – often called Europarties – at European level. Nevertheless, one can reasonably argue that the liberal parties at national level, the ALDE Party, and the ALDE group in the EP are different components of one European–wide party organisation. Bardi et al. (2014) classify the national member parties as the ‘party on the ground’, the European political party as the ‘party in central office’, and the corresponding political group in the EP as the ‘party in public office’, together with the liberal ministers in the Council of Ministers and the liberal commissioners in the European Commission. Van Hecke (2010) argues that these components each operate at a different level. The main political arena of national political parties is the national level, were they influence EU politics through national governments and parliaments. The party groups in the EP operate at the supranational level: They act as an integrated organisation, independently from the national parties. Europarties are situated at the transnational level – the intersection of national (or intergovernmental) and EU (or supranational) spheres – and reflect this dual nature of the European Union. As such, Europarties provide partisan linkages between two levels and the three main EU institutions and try to influence European policymaking. In the same vein, Smith (2014) argues that the main role of ALDE is to serve as a platform for

260  Wouter Wolfs and Steven Van Hecke communication, to exchange ideas and best practices and to coordinate the positions of parties and politicians from the different political levels. In this chapter, we focus on the ALDE Party and the ALDE group in the EP, based on a literature review, party documents, and interviews with former party president Graham Watson, current party president Hans van Balen, and EP Group chairman Guy Verhofstadt. In the first part, the main developments of the ALDE Party from its establishment in 1976 to the 2014 European elections are discussed. The second part examines the ideological coherence of the liberals at the European level by analysing the content of their electoral manifestos and their voting behaviour in the EP. In the third part, the components of the ALDE Party organisation are presented. The chapter ends with an outlook of the main challenges the ALDE Party faces.

Origins and development of ALDE The development of the ALDE Party The predecessor of ALDE was founded in 1976 when the main ideological families established transnational party federations in the run-up to the first direct elections to the EP in 1979. These party federations could draw on two other traditions of transnational party cooperation. First, national parties of the same famille spirituelle were already organised in political Internationals. The first Liberal International – the Entente Internationale des Partis Radicaux et des Partis Démocratiques Similaires – was established in 1924 but did not survive World War II. In 1947, 19 liberal parties signed a manifesto in Oxford, which signified the start of a renewed Liberal International, the Liberal World Union. However, the level of activity in the organisation was very low until 1967, when a new Liberal Declaration was signed in Oxford, and the international liberal cooperation was rejuvenated (Hanley 2008: 118). Second, European liberal parties already cooperated in the representative body of the European Communities. The representatives in the Common Assembly of the European Coal and Steel Community – the predecessor of the EP – decided to form groups according to their political ideology and not their member state in the 1950s. The first liberal group – named Group of Liberals and Allies – was officially founded in June 1953. These two forms of cooperation ensured that a network of liberal politicians existed, which paved the way for a European liberal party federation. In 1976, the European liberal parties united in a transnational party organisation. However, there was significant disagreement among the leading liberal figures and parties on how this organisation should take shape. A first issue was related to the principle and strategy of this transnational organisation. Gaston Thorn, Prime Minister of Luxembourg and the provisional President, wanted a rather loose association encompassing parties from the whole political spectrum. In contrast, Jean Durieux, the leader of the liberal group in the EP, argued for a more unified and homogenous

The Liberals in Europe 261 organisation that could constitute a ‘third force’ as an alternative to the Socialists on the left and the Conservatives on the right. As a compromise, the parties opted for the term ‘Federation’, as it was sufficiently ambiguous to allow for both interpretations (Lodge and Herman 1982). A second dispute was related to the name. The word ‘Democratic’ was included in the name of the organisation as not all national parties – such as the Dutch VVD – could rally behind the term ‘liberal’ for domestic historical reasons (Smith 2014). Therefore, the new liberal party was officially founded as Federation of Liberal and Democrat Parties in Europe (ELD). In the run-up to the first direct elections of the EP in 1979, there was a struggle between the different party federations to recruit new members. The ELD was composed of 11 liberal parties from eight of the nine member states of the European Community (EC): two parties from Italy (Partito Repubblicano Italiano and Partito Liberale Italiano), two French parties (Parti Républicain and Parti Radicale Socialiste), two Belgian parties (the Dutch-speaking Partij voor Vrijheid en Vooruitgang and the Frenchspeaking Parti des Réformes et de la Liberté), the Danish Venstre, the Freie Demokratische Partei from (West-)Germany, the Parti Démocratique from Luxembourg, the Dutch Volkspartij voor Vrijheid en Democratie and the Liberal Party from the UK. Ireland was the only country of the EC that was not represented (Lodge and Herman 1982). Regarding the further development of the party, two factors were of particular importance. The first factor is the changing institutional context of the EC. The enlargements encouraged the Europarty to look for partners in the new member states. The second factor is the fragmented nature of liberalism at national level: Most member states have multiple parties that can be considered as liberal. Consequently, national dynamics and rivalry between parties have major consequences for the Liberal Europarty. The French Mouvement des Radicaux, for instance, refused to join the Europarty because they could not reconcile with the membership of the French Parti Republicain. Similarly, the Dutch Democraten 66 (D66) initially refused to enter ELD because of their rivalry with the VVD (Külahci and Van de Walle 2002). However, other parties did join. The enlargement of the EC in 1986 with Spain and Portugal enabled the ELD to appeal to the Portuguese Partido Social Democratico. To include them, the term ‘Reformists’ was added to the title of the party federation (Hanley 2008). ELD was now European Liberals, Democrats, and Reformists (ELDR). The Treaty of Maastricht that came into force in 1993 included for the first time legal recognition of these transnational party organisations. This ‘party article’ states that parties at European level ‘contribute to forming a European awareness and to expressing the political will of the citizens of the Union’. As a consequence, ELDR added the word ‘Party’ to its title, although the organisation and structure was not significantly changed (Sandström 2001).

BK-TandF-9780815372387_TEXT_CLOSE-181001-Chp14.indd 261

04/01/19 1:42 PM

262  Wouter Wolfs and Steven Van Hecke The subsequent EU enlargements in the 1990s and 2000s equally caused an expansion of the ELDR party. In the run-up to the EU enlargement of 1995 with three former EFTA countries – Austria, Finland, and Sweden – a number of Scandinavian parties joined ELDR: Liberaalinen kansanpuolue, Keskusta, and Svenska folkpartiet from Finland and Liberalerna from Sweden. At the end of the 1990s and the beginning of the 2000s, the ELDR party leadership – together with the Friedrich Naumann Foundation and the Liberal International – made great efforts to look for potential liberal parties in the candidate countries of Central and Eastern Europe (Hanley 2008). However, the changeability of ideology that characterised these new parties in the first decade after the fall of the communist regime made it a challenging task. Many parties were in frequent contact with various Europarties. A good example is the Hungarian Fidesz party that initially joined ELDR in the beginning of the 1990s but changed later to the European People’s Party (EPP) when it shifted to the right (Watson 2010). The most important event for the internal development of the ELDR party was the establishment of public funding for Europarties in 2004. This measure provides (most) parties with sufficient resources to build up their own organisation. Before the introduction of EU grants, Europarties were located in the EP with their respective political group and were dependent on them for offices, personnel, and funds. For example, the Secretary of ELDR in the 1990s, Christian Ehlers, was a staff member of the liberal group in the EP (Smith 2014). The idea of direct funding for Europarties dated back to the middle of the 1990s and was pressured further by a critical report of the Court of Auditors that questioned the lack of transparency and entanglement of the finances of the Europarties and political groups (Wolfs and Smulders 2018). As a consequence, ELDR was detached from its political group and purchased its own premises in Brussels. This had two ambivalent consequences. One the one hand, ELDR now had sufficient resources to develop its own party organisation and activities. On the other hand, it lead to an estrangement of the Europarty and the political group and has made the relations between the two more complex. The introduction of party funding also led to the establishment of the European Democratic Party (EDP), composed of a number of member parties whose members of European Parliament (MEPs) were part of the ALDE group but that were not a member of ELDR, ending the one-to-one relation between the ELDR and the liberal group in the EP. The subsequent years were characterised by a further build-up of the party organisation and an expansion of the party’s membership with parties of non-EU countries. In the past decade, the liberals welcomed 15 parties from Balkan and Caucasian countries, Moldova, Ukraine, and Russia. In 2012, the ELDR changed its name at the initiative of party leader Graham Watson. One of the priorities of his leadership was to unite the two factions in the liberal EP group – the members affiliated to ELDR and the members of the EDP – into one political party. To strengthen the relation

The Liberals in Europe 263 between the ELDR party and the ALDE group, the former strategically took on the name of the latter. This was only possible since a number of the initiators of the EDP had already left the group. The long-term goal of the name change was to eventually take in all remaining member parties of the EDP into the ALDE Party (interview with Graham Watson). In the run-up to the European elections of 2014, all major European political parties nominated their Spitzenkandidat, their candidate for the presidency of the European Commission. This significantly increased the role of European political parties, since it was the first recruitment process that fully took place at EU level. In the ALDE Party, two candidates were put forward, which reflected internal divisions: Olli Rehn, who was supported by 14 member parties, and Guy Verhofstadt, who was supported by the Benelux parties. To avoid an open conflict, the Dutch Prime Minister Mark Rutte and Christian Lindner of the German FDP were appointed as mediators and found a compromise: Verhofstadt would become the lead candidate for the Commission presidency and Rehn for another senior EU post (Put et al. 2016). The electoral performances of the liberal group in the EP The corresponding group of the Liberal Europarty in the EP is the Alliance of Liberals and Democrats for Europe. Table 14.1 gives an overview of the size of the liberal group in the different terms of the EP. Traditionally, liberalism has been strong at the party level in the United Kingdom, the Low Countries, Germany, and Scandinavia. However, even in these countries, the electoral support for liberal parties is characterised by significant volatility. In South, Central, and Eastern Europe, the number of established liberal parties is much lower. In countries in Southern Europe many parties were movements built around individuals instead of parties, and the anticlerical tradition was embodied by the Socialist parties, leaving little room for the liberals. The political dynamics in some countries – like France – lead to a polarised party system with a political left and right bloc, without much role to be played by liberal parties as a potential third force (Hanley 2008). Also in countries such as Austria, Hungary, the Czech Republic, and Slovakia, there is no strong liberal political tradition. However, in the EP the size of the political group matters more than the number of member states that are represented. The larger the group, the more resources and personnel it gets and the more influence it can exert on decision making. Consequently, the management of the EP Liberal group has always been a balancing act between keeping ideological coherence and pragmatically maximising the group size. After the first direct elections to the EP in 1979, the liberal group was formed with MEPs from all nine member states, except the United Kingdom. With 40 seats, the Liberal and Democratic Group (LD) became the fifth-largest group. The French delegation was by far the largest, with

Table 14.1  Seats of the liberal group in the European Parliament, 1979–2017 LD

LDR

1979-1984

1984-1989

ELDR 1989-1994

1994-1999

ALDE 1999-2004

2004-2009

2009-2014

2014-2019

Incoming Out­going Incoming Outgoing Incoming Outgoing Incoming Outgoing Incoming Outgoing Incoming Outgoing Incoming Outgoing Incoming June 2017 Austria Belgium Bulgaria Croatia Cyprus Czech Republic Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxemburg Malta

4

4

5

5

4

4

6

1 6

0 5

3

3

2

2

3

2

5

5

6

17 4

15 4

12 0 0

13 0 0

13 4 0

9 5 0

1 0 0

5 1 0 0

5 0 0 0

1 5

1 5

1 5

1 6

2 3

2 4

1 7

1 4

1 7

2

2

1

1

1

1

1

1

1

0 5

0 6

1 6 5

0 5 5 0 0

0 5 5 0 0 0

1 6 4 2 0 4

1 6 4 2 0 4

1 1

1 0

1 0

6 2 5 1 0 0 2 1 8 0 5 1 0

4 2 5 11 7 0 2 1 12 1 7 1 0

4 2 5 10 7 0 2 1 12 1 7 1 0

3 3 4 6 12 0 0 4 7 1 2 1 0

3 3 4 6 12 1 0 4 4 1 2 1 0

3 3 4 7 4 0 0 1 0 0 3 1 0

3 3 4 7 4 0 0 1 0 1 3 1 0

Netherlands Poland Portugal Romania Slovenia Slovakia Spain Sweden United Kingdom TOTAL ALDE TOTAL EP Share ALDE (%)

4

4

5

5

4

4

10

10

8

10

9

9

9

0

2

6

5

2

0

8 0 0

5 4 0 2 0 2 3 12

5 6 0 6 2 0 2 3 11

6 0 0 5 2 1 2 4 11

6 0 0 5 2 1 2 4 12

7 0 2 1 1 1 8 3 1

7 0 1 3 1 0 8 3 1

3 4 10

3 1 2 4 11

0

0

0

0

0

0

2

2 3 3

40

38

31

45

49

45

44

42

50

67

88

100

84

83

67

68

410 9.8

434 8.8

434 7.1

518 8.7

518 9.5

518 8.7

567 7.8

626 6.7

626 8.0

788 8.5

732 12.0

785 12.7

736 11.4

766 10.8

751 8.9

751 9.1

Source: European Parliamentary Research Service (2016); European Parliament (2017)

266  Wouter Wolfs and Steven Van Hecke 15 MEPs from the Union pour la Démocratie Française, a centre-right political alliance led by then President Valéry Giscard d’Estaing. In Greece, a country in which liberalism has been traditionally weak, the interim election in 1981 when the country joined the European Communities did not bring additional Liberal MEPs. This did not change during the subsequent terms: Greece became the only member state never to have MEPs that are part of the liberal group. Despite being only the fifth-largest group, it was LD member Simone Veil who was elected as the Parliament’s first president. The 1984 elections to the EP caused a decline of the liberal group, mainly because the German FDP came short of the electoral threshold and lost all its MEPs. Nevertheless, it remained the fifth-largest group in the European assembly. In 1985, the group changed its name into the Liberal and Democratic Reformist Group (LDR), to prepare for the new member parties from Portugal and Spain. The by-elections in Spain and Portugal in 1987 provided a boost for the liberal group. It welcomed 12 new members: two from the Spanish Convergència i Unió (CiU) and ten from the Partido Social Democrata (PSD), by far the largest Portuguese delegation in the EP. After the European elections in 1989, the LDR Group gained several seats, mainly because the FDP in Germany returned to the EP, and thanks to the Spanish party Centro Democratico y Social that joined the liberal group. Consequently, LDR became the third-largest group in the EP, a position it would keep until the European elections of 2014. The elections of 1994 resulted in a similar position of the LDR Group compared with the previous term. The LibDems returned into the EP and the Dutch (VVD and D66) and Danish (Venstre and Radikale Venstre) member parties had good electoral results. Also the six MEPs from the Italian Lega Nord joined LDR. Additionally, the liberal group got nine new MEPs following the 1995–1996 elections in the three new member states: two parties from Sweden (the Centerpartiet and the Liberalerna), two parties from Finland (Suomen Keskusta, and the Svenska folkpartiet) and one from Austria (Liberales Forum) joined the group. In the same period, however, the liberal group suffered from a number of defections. The Treaty of Maastricht had strengthened the position of the EP, among others through the introduction of the co-decision procedure. This new procedure required a two-thirds and absolute majority in the EP to amend or reject legislation, which created an incentive for expansion of (in particular the two biggest) political groups (Smith 2014). LDR suffered from the pragmatic expansion strategy of the EPP. In 1991, Valéry Giscard d’Estaing – although he was the group leader – left the Liberals with three of his party members to join the EPP group. Giscard d’Estaing’s assessment was that he could achieve more in one of the two biggest groups in the EP. In fact, he wanted to take the entire liberal group to the EPP, but all other member parties opposed this. In 1996, the LDR would suffer another loss, when the Portuguese PSD left the Liberals also to join the EPP.

The Liberals in Europe 267 The 1999 elections resulted in a bigger liberal group, mainly because of the good results of the British LibDems. The elections of 2004 were the most successful for the liberal group in the EP, for two reasons. First, it was the first election in which MEPs from Central and Eastern Europe were elected and the liberal parties did better than expected. The liberal group welcomed new MEPs from Cyprus, Estonia, Hungary, Latvia, Lithuania, Poland, and Slovenia. Particularly the results in Lithuania were positive: The liberal group obtained more than half of the total Lithuanian seats: five MEPs of the Labour Party and two from the Liberal and Centre Union. Second, the liberal group was able to attract some member parties from the EPP group. The French Union pour la Démocratie Française led by François Bayrou and the Italian umbrella party La Margherita of former Commission President Romano Prodi were disappointed with the political strategy of the EPP. The Christian democrats had accepted more centre-right and right-wing parties and had lost their ideological coherence and enthusiasm for European federalism (Hanley 2008). Particularly the EPP’s entry of the Rassemblement de la République and Forza Italia led to the defection of the French centrists and the Italian leftist Christian Democrats (Van Hecke and Alberti 2005). Liberal group leader Graham Watson played an important role in getting the two parties to come over from the EPP group. However, these new parties did not join ELDR because of the term ‘liberal’, but established a separate Europarty, the European Democratic Party (EDP), as a more socially progressive partnership (Hanley 2008). As a consequence, the liberal group changed its name to Alliance of Liberals and Democrats for Europe (ALDE) and was hereinafter composed of MEPS from members parties of two different Europarties. In other words, the direct, one-to-one relation between the liberal Europarty and its group in the EP was lost again. However, ALDE functioned as a unified group – no separate structures for the MEPs from the two parties – and the ideological differences between the two parties were limited. Overall, with more than 12% of all MEPs, the liberal group became the largest in its history. The ALDE group accomplished a similar strong result in the 2009 elections. The 2007 enlargement had already resulted in a significant liberal delegation from Romania and Bulgaria. Although in the European elections of 2009 the liberal group lost its representation in Austria (one MEP), Cyprus (one MEP), Hungary (two MEPs), and Poland (six MEPs) and suffered a loss of seats in the French, Italian, and Lithuanian delegations, this was partly compensated by a strong result of the German FDP and the entry of Fianna Fail from Ireland, a rather Eurosceptic party that was previously member of the conservative and Eurosceptic group Union for Europe of the Nations. Smith has labelled this expansion as a further ‘victory for geographical widening over ideological deepening’ (Smith 2014). ALDE remained the third-largest political group in the EP with 84 seats, behind the EPP and the socialist group.

268  Wouter Wolfs and Steven Van Hecke The European elections of 2014 were not good for the ALDE group. Particularly the liberal parties in the larger member states performed poorly: The LibDems were reduced to 1 MEP, losing 11 MEPs; the German FDP was reduced from 12 to 3 seats; and the Italian delegation disappeared from the EP. On the other hand, a number of new parties from member states where liberalism had been traditionally very weak were able to gain seats and join the ALDE group: in the Czech Republic ANO2011 gained four seats and also the Austrian party NEOS was able to win one seat. In Portugal, the Partido da Terra came in the EP for the first time and joined the ALDE group, making it the first Portuguese delegation since the Social Democratic Party left the liberal group in 1996 to join the EPP. Hence, ALDE became a group that is weak in the larger member states, but with a significant presence in the smaller and middle-sized member states. The liberal group also lost its third position in the EP to the European Conservatives and Reformists (ECR) Group.

Ideology and policy positions of ALDE ALDE is composed of various national political parties with different political and historical contexts. Consequently, it is not surprising that a number of cleavages run through the Liberal Europarty. The most described cleavage is the classic division between ‘market or economic liberalism’ – on the right of the political spectrum with a strong focus on individual and market freedoms – and ‘social liberalism’ – on the political left that dedicates more attention to social issues (Smith 2014: 109–110). Examples of the first type of liberalism include parties such as the Dutch Volkspartij voor Vrijheid en Democratie (VVD) and the German FDP, whereas Democraten 66 (D66) and the Danish Radikale Venstre are considered to be more social liberal parties. It is a cleavage that can be traced back to the origins of the Liberal Europarty. When the party federation was established in 1976, there was a stronger emphasis on classic continental liberalism. This is the reason why five of the 14 liberal parties that were invited did not join ELD. The French Movement des Radicaux de Gauche, for example, was part of a left-wing political coalition at the time and declined membership. Also the Dutch Democraten 66, the progressive wing that split from the VVD, opted out from membership until 1994 (Hanley 2008). In the subsequent decades, it remained an ideological cleavage within the party. However, on a number of issues, the geographical cleavage with differences across national lines is much more profound. In the different ELDR/ ALDE bodies, the national member parties try to build a coalition with like-minded parties. For example, in 2009 the LibDems cooperated with the Centrepartiet and the Folkpartiet to achieve an ambitious ALDE position on climate change (Smith 2014). Environmental issues can indeed be a source of friction between parties that support far-reaching environmental protection

The Liberals in Europe 269 and parties that are more industry-oriented (Hanley 2008). Other issues on which disagreement exists are EU enlargement and security policy (Külahci and Van de Walle 2002). In 1977, for example, the Danish Radikale Venstre left ELD because it objected to its position for a stronger EU security policy (Külahci and Van de Walle 2002). In general, however, disagreements among the member parties in ALDE should not be exaggerated and consensus among the member parties seems rather high (Hanley 2008). Sandström has described how the party has developed a sophisticated mechanism for consensus-building (2001). On certain political issues, there is a strong consensus among the member parties. Although ALDE contains liberal parties from the right and the left of the political spectrum, both groups attach much importance to personal and collective liberties (Smith 1988; Hanley 2008). Another core issue is the push for more European integration. Throughout its history, the party has always advocated strong European level policies, such as a common environmental, migration, and foreign policy, as well as stronger community decision-making mechanisms, such as Qualified Majority Voting in the Council of Ministers and more powers for the EP (Külahci and Van de Walle 2002; Hanley 2008). Lord (1998) argued that all liberal member parties ‘shared the ideal of a united Europe’. A rare dissident voice in this respect is the Scandinavian agrarian parties, who are more eurosceptic. Moreover, in recent years, the strong pro-European consensus within the party has been challenged by a number of member parties, including the Dutch VVD. One of the most prominent federalist liberals, Andrew Duff, experienced increasing difficulties to find support within the party for his federalist positions, such as the establishment of a transnational electoral list (interview with Graham Watson). The fundamental values of ALDE are described in the Stuttgart Declaration, the founding document of the Liberal Europarty that was adopted on 26 March 1976. The declaration is characterised by a distinct integrationist character, both regarding the institutional set-up and the competences that should be conducted at the European level. The signatories argue for increased powers for the EP and majority voting in the Council of Ministers; the European Commission should be directly accountable to the EP and every citizen should have the right to appeal to the European Court of Justice; the European project should be based on the promotion of the fundamental rights and freedoms of the individual and the rights of minorities and regions should be protected. Concerning the division of competences, the declaration calls for an economic and monetary union through greater harmonisation of the member states’ economic and financial policies and their currencies, including the creation of a joint central bank. Furthermore, the signatories encourage a European redistributive policy through the Social Fund and the Regional Development Fund and push for a common environmental, energy, and foreign policy.

BK-TandF-9780815372387_TEXT_CLOSE-181001-Chp14.indd 269

14/01/19 2:50 PM

270  Wouter Wolfs and Steven Van Hecke The Stuttgart Declaration provided an impetus for a common electoral manifesto of the Liberal Europarty in the run-up to the European elections. Since the first elections in 1979, the member parties of ALDE have agreed upon a common manifesto negotiated by the national delegations of the party. Whereas the first manifesto for the 1979 elections was a short document in line with the Stuttgart Declaration (ELD 1977), the 1984 electoral manifesto was much more detailed (ELD 1983). In a document of more than 50 pages, the Europarty meticulously set out its political positions in various policy fields, ranging from a common foreign and security policy to a European cultural and education policy. Over time, however, the electoral manifestos of the Liberal Europarty have become shorter and less detailed. The manifesto for the elections of 2009 for example was only three pages long (ELDR 2008). This illustrates the growing difficulties to adopt a common electoral platform for the elections for the EP, particularly as the number of member parties increases. The debates on drafting the electoral manifestos were often fierce and the votes on different policy issues to be included in the manifesto were sometimes close, particularly for issues on the balance between economic and social liberalism (interview with Graham Watson). Furthermore, although the Liberal Europarty always developed a common manifesto, the national member parties were free to use it and for the most part just ignored it (Hanley 2008). Even during the elections for the EP, the Europarties still have a secondary role compared with their national counterparts, which is also evident in the electoral campaigns. In 2004, there barely was a coordinated European campaign of ELDR and in 2009 this was even less the case. Party President Annemie Neyts preferred a short document with the liberal key points to a comprehensive electoral manifesto, because the national member parties only scarcely made use of it. She wanted to avoid that in-depth discussion on a manifesto would reveal and intensify internal party divisions. However, even in the debates on the key point document, disagreements between the different member parties became clear. The British LibDems and the German FDP, for example, were conflicted over the reference to a European army (Smith 2014). The ALDE manifesto for the European elections of 2014 has a strong market liberalist character. The main focus is on job creation and the fight against unemployment by removing obstacles for trade and excessive national rules and regulation. It argues for the expansion of the single market in energy, digital single market, financial services, and the transport and health-care sectors. The statements on the institutional set-up and competences of the European level are also described from a market-liberalist perspective: A strong unified foreign policy should help to create jobs and improve people’s lives, and the EU institutions should focus on a smaller EU administration with less agencies and on less bureaucracy and simpler rules.

The Liberals in Europe 271 When looking at the voting behaviour of the liberal group in the EP, there are no significant ideological cleavages. In the last four terms of the EP, the cohesion rate of ELDR-ALDE has varied between 85 and 90%. In other words, in most (registered) votes, the liberal group has voted as a bloc. This cohesion is very comparable with most of the other political groups in the EP. Only the Eurosceptic group in the EP shows a significantly lower level of cohesion (about 50%). Similar to the position of many liberal parties at the national level, the liberal group has regularly presented itself as the ‘kingmaker’ in the EP, which means that despite their relative smaller size, they can decide who wins the vote. Since the EP does not have a fixed majority and opposition like most national parliaments, a majority needs to be found for each vote on a case-by-case basis. As the liberal group is in the ideological centre of the EP, they can either vote together with the EPP to form a centre-right majority or swing to the socialists to supply a centre-left majority. This becomes clear when looking at the winning coalitions in the EP. During the 7th term of the EP (2009–2014), ALDE was on the winning side of the vote in 86.4% of the times, which was slightly more than the S&D Group (83.2%) and significantly better than the Greens-EFA Group (67.9%) and the ECR Group (56.7%). Only the EPP group (89.4%) had a better track-record. When considering the different policy domains, ALDE was part of the winning coalition in more than 80% of the times for all policy fields with the exception of Budget Control (76.1%) and Agriculture (67%) (Hix 2013: 7). The picture for the first half of the 8th term of the EP (2014–2019) has been comparable: ALDE has been part of the winning coalition in 88.1% of the votes, which was close to the S&D Group (88.8%) and more than the EPP Group (85.9%), the ECR (58.6%), and the Green–EFA Group (67%). The policy fields with the lowest winning scores for ALDE have been Internal Market and Consumer Protection (76.5%) and Legal Affairs (77.7%). In all other domains the winning rate has been more than 80% (VoteWatch 2016). This illustrates that, although ALDE has lost its position of third-largest group to the ECR in 2014, it still has pushed far above its electoral weight in the EP. The main reason is that EP has been characterised by a will to form the largest-possible coalition (to have a stronger position vis-à-vis the Council of Ministers). In the 7th term of the EP, almost twothirds of the winning coalitions were a ‘Super Grand Coalition’, consisting of the EPP Group, the S&D Group, and the ALDE Group; in 12% a centre-right coalition (EPP and ALDE) won; and in 12% a centre-left coalition (S&D and ALDE) was the victor. Only in less than 10% of the votes was ALDE outvoted by the EPP Group and S&D Group, mostly in the field of Agriculture and Fisheries (Hix 2013: 4). A similar situation can be recognised in the first half of the 8th term of the EP. ALDE has voted in coalition with the S&D Group and the EPP Group in more than 80% of the times (VoteWatch 2016).

272  Wouter Wolfs and Steven Van Hecke

Party structure and organisation Main party structures The ALDE Party is predominantly an umbrella organisation of national parties. The ALDE Party differentiates between full and affiliate member parties. Affiliate member parties can apply for full membership after two years (Hanley 2008: 119). The main difference between the two is that affiliate members do not have voting rights in the party bodies. ALDE currently has 36 full members and six affiliate members from 25 EU countries. It has no member parties from France, Slovakia, and Malta. In most EU member states, however, there is more than one national party affiliated with ALDE. This can lead to the ambiguous situation in which one member party is part of the government and the other(s) are in the opposition. This is, for example, the case in Denmark, where Venstre is in government and Radikale Venstre is in opposition, or in Finland were ALDE member Svenska Folkpartiet is in opposition for the first time in 35 years and the Suomen Keskusta is part of the government coalition. This can also cause tensions within the liberal family. Even if they are not from the same country, member parties that are part of the government have different expectations about ALDE than opposition parties. Liberal parties that are used to be part of the government – such as the Dutch VVD and the Belgian Liberals – take part in the meetings of the European Council and the Council of Ministers and have a minimalist view about the Liberal Europarty: They consider it primarily to be a platform for information sharing. Parties in opposition without representation in parliament tend to attach much more importance to ALDE because it constitutes their only connection to European decision making. When the German FDP was not represented in the EP between 1994 and 2004, and disappeared from the German government in 1998, for instance, it invested heavily in ELDR (Hanley 2008). ALDE’s members are not limited to liberal parties from EU member states as the party has full and affiliate members from other European countries. For these parties, ALDE membership is a good opportunity to come into contact with their EU counterparts and to exchange ideas and best practices. For the EU parties it is a good way to defend their interests beyond the EU’s borders (Smith 2014: 116). The ALDE Party includes parties from the EFTA countries: Venstre from Norway, the Freisinnig-Demokratische Partei from Switzerland, and the Björt Framtið, a new Liberal, pro-European party from Iceland. ALDE also has members from the Caucasian countries: two parties from Georgia and one party from Armenia and Azerbaijan. Furthermore, a number of national parties from the Balkan countries are affiliated with ALDE: the New Kosovo Alliance and the Liberal Party of Kosovo, Our Party from Bosnia Herzegovina, the Liberal Party of Montenegro, and the Liberal Democratic Party from Serbia. Finally, ALDE

The Liberals in Europe 273 has two member parties from Ukraine and Russia and members from Moldova and Belarus. In total, the ALDE Party has 61 national member parties from 41 European countries. Aside from these national member parties, ALDE – as the last of the Europarties – has a developed form of individual party membership (Hertner 2017). Whereas in other Europarties individual EU citizens can only become a member through membership of one its national member parties, ALDE gives the possibility to become a direct individual member. This individual membership was long resisted by the national member parties, because it could cause rival loyalties among their supporters (Hanley 2008) and challenge the dominance of the national parties over ALDE. Mair (2000) emphasised that national parties, who form the constituent components of the Europarties, act as gatekeepers to preserve their own influence and interests. Eventually, the possibility of individual membership was introduced in the Party’s statutes in 2004 but remained unimplemented for a long time. Party leader Graham Watson took up the issue in 2011. According to Hertner (2017), by April 2016, ‘ALDE had approximately 2,000 individual members in 45 countries, including the United States, where Liberal European Expats set up a group’. Moreover, political rights such as voting rights at the congress have been granted to representatives of individual members (Hertner 2017). The initiative has an ideological as well as a strategic motive. As one of the most pro-European Europarties, ALDE wants to contribute to the development of a European public sphere and a European democracy and considers individual membership as an important measure to achieve this goal. From a strategic perspective, individual membership allows citizens that are prone to liberal ideas to affiliate with ALDE, even if there is no (significant) national liberal party in their country or they do not want to be a member it. For example, ALDE does not have a member party in Greece but was able to recruit many Greeks that identify with liberal ideas as individual members. Citizens can become an individual member if they recognise the fundamental values of the EU and of the founding Stuttgart Declaration, and pay an annual membership fee. Currently, there are 1,700 individual members of the ALDE Party. At the level of the ALDE Party, individual members are represented by a Steering Committee that is directly elected every two years ahead of the ALDE Party Congress. These delegates are involved in the internal party decision making and also have voting rights in the main party bodies. The leadership of the party is in the hands of the Bureau, which is composed of the party President, seven Vice-Presidents, and the Treasurer. The Bureau is responsible for the day-to-day management of the party. The President of ALDE also shapes intra-organisational policy and priorities of the party. The subsequent leaders all had different priorities for the party (Table 14.2). Annemie Neyts – Party President from 2005 to 2011 – focussed primarily on stimulating the debate among the member parties. She wanted

274  Wouter Wolfs and Steven Van Hecke Table 14.2  ALDE Party leaders Dates

Name of party leader

1978–1981 1981–1985 1985–1990 1990–1995 1995–2000 2000–2005 2005–2011 2011–2015 2015–present

Gaston Thorn Willy De Clercq Colette Flesch Willy De Clercq Uffe Ellemann-Jensen Werner Hoyer Annemie Neyts-Uyttebroeck Graham Watson Hans van Baalen

to determine a clear liberal political position on Europe and to use that in the policy of the Party (Hanley 2008). Graham Watson invested in the party organisation and communication, which was made possible because of the growing EU party subsidies. He wanted to turn ALDE into a proper transnational party. An important element in this respect was the operationalisation of the individual party membership. Finally, he wanted to unite the two liberal factions – the ELDR and the EDP – into one European party (interview with Graham Watson). It is no coincidence that he was the driving force of the name change to ALDE to make the relation with the EP group stronger and to try to make the EDP join the Liberal Europarty. Up till now, however, this has not happened. The current party leader, Hans van Baalen, seems less occupied with the European level party and more focussed on the electoral success of the national parties to strengthen the liberal position in the EU institutions (interview with Hans van Baalen). Officially, the most important internal decision-making body is the annual Party Congress. It brings together delegates from all the member parties of ALDE, representatives from the EP political group, the liberal European Commissioners, Ministers, members of parliament (MPs) and delegates from the youth organisation. It has an important organisational and ideological function. Every two years, the Congress elects the party leadership. Furthermore, it sets out the long-term ideological course of the party. It is the main platform for political debates and the members vote on resolutions and adopt a common party programme for the European elections. Interestingly, the decision making within the Congress has evolved from a unanimity rule to increasing majority voting. The ALDE Party Council is composed of delegates from all full member parties and has generally two meetings a year. Its responsibilities are of a more administrative and financial nature. The Council appoints the Secretary General and approves the party’s annual budget and accounts. Furthermore, the Council discusses the applications for membership of ALDE and voted the membership fees. The presence of these two bodies – a party congress and a party council – might seem rather plethoric.

The Liberals in Europe 275 Hanley (2008) attributes the existence of a separate council as a symbolic gesture towards the rank-and-file. However, there is also a more bureaucratic reason: Belgian law requires their ‘associations without lucrative purpose’ – the legal status of ALDE – to have a separate Council that must oversee party finances. The most important political events of the party are the leadership summits, bringing together key liberal figures in the EU institutions and heads of governments. These meetings do not serve as decision-making bodies, but as a platform for coordination and networking. They point to an important function of Europarties as linking bodies for the different actors and levels in the EU’s decision-making system (Van Hecke and Johansson 2013). These liberal leadership summits were established in 1999 after the practice of the two bigger Europarties EPP and PES and have been a regular feature of ALDE. Under the leadership of Annemie Neyts, these summits were doubled: One type of summit was organised for liberal Prime Ministers, Commissioners, and the Presidents of the ELDR Party and ALDE group in the EP; a second meeting to bring together all other liberal leaders. However, because of the separate meeting, enthusiasm among liberal leaders to attend the second type of meeting was rather limited (Smith 2014). Party leader Graham Watson took a new initiative ‘Liberals in Government’ in 2012 together with LibDem leader and then UK Deputy Prime Minister Nick Clegg, bringing together all liberal Ministers from EU governments for in-depth discussions on different policy issues (Smith 2014). These summits help to coordinate the positions among the liberal government leaders and high-raking EU officials. In 2007, most of the European political parties established political foundations, after the image of the German political Stiftungen, when EU funding for this type of organisation was established. The main role of these foundations is to act as a network of think-tanks, foundations and experts of the same ideological background and contribute to the political debates at EU level. The foundation affiliated to the ALDE Party is the European Liberal Forum (ELF). Although each foundation is formally linked to a Europarty – a condition to receive EU funding – the ELF enjoys a high degree of independence from ALDE. The ELF provides scientific input and liberal policy proposals and supports the programmatic work of ALDE. However, due to the limited resources, it does not conduct a lot of own research, but depends on the contributions of its different members. The ELF thus functions as a liberal network to exchange and pool expertise from the member organisations (Gagatek and Van Hecke 2014). It is open to all liberal actors that endorse the values of the European Liberal as drawn up in the Stuttgart Declaration. ELF currently has 37 member organisations from all EU member states, except Cyprus, Ireland, Latvia, Luxembourg, Malta, and Slovakia. These include thinktanks from members parties of ALDE, as well as more independent liberal research organisations.

276  Wouter Wolfs and Steven Van Hecke Relation to social movements As opposed to the two biggest Europarties, the network of civil society organisations that are affiliated with ALDE is much less developed. The party has established a European Liberal Women Network in 1999 to promote equal opportunities for men and women. Over the years, the network has undertaken several actions to increase women’s participation in political life, equal gender treatment, and the promotion of women’s rights. When the Liberal Europarty changed its name to ALDE in 2012, the Network also adopted the new name Gender Equality Network. The youth wing of ALDE, the European Liberal Youth (LYMEC), was established in 1976, the same year as the ELD. It has a similar organisation to its mother party: It consists of full and associate member youth organisations and individual members. In fact, LYMEC introduced individual membership before ALDE did, in 1997. Currently, the liberal youth movement has 47 members organisations from 33 European countries.

Discussion As one of the oldest Europarties, ALDE has been able to play a role within the EU decision-making process, despite its modest size. The main challenge for the Liberal Europarty will be the ability to support and strengthen national liberal parties. To remain relevant, a significant liberal presence in the different EU institutions is essential. This can only happen if ALDE has strong member parties. Although the increased electoral volatility has been particularly challenging for liberal parties in the member states, there are positive signs for the liberal family. In several member states without a strong liberal tradition, newly established liberal parties retain growing support. However, the more members the ALDE Party and group in the EP contains, the more difficult it becomes to guard party discipline and ideological coherence. The ambiguous situation in some member states in which one liberal member party is part of the government and the other(s) are in opposition negatively affects the relations in the ALDE Party and group. Furthermore, the liberal group has always been characterised by a certain degree of disagreement between social and market liberals and along geographical lines. One of the issues of the ALDE Party that all member parties could rally behind in the past – a plea for a stronger and federal Europe – has become more controversial in the last years. Some of the national parties – such as the Dutch VVD – have become increasingly lukewarm towards more European integration and it remains to be seen to what extent ALDE can present itself as the federalist Europarty. Nevertheless, it is clear that organisationally ALDE is evolving into a transnational party, among others through the development of its individual party membership. This feature allows connecting with a young, urbanised,

The Liberals in Europe 277 and politically active class that is pro-European. However, it could be challenging to conciliate a larger role for individual members with the traditionally strong position of the national member parties that will certainly try to defend their interests within ALDE.

References ALDE (2013), A Europe that works. Adopted at the Congress in London. Bardi L., Bressanelli E., Calossi E., Cicchi L., Gagatek W., Pizzimenti E. (2014), Political parties and political foundations at European level. Challenges and opportunities, Brussels: European Parliament. ELD (1983), Euro election manifesto 1984: For a liberal and democratic Europe. Adopted at the Congress in Munich. ELD (1977), Electoral manifesto 1979. Adopted at the Congress in Brussels. ELDR (2008), ELDR manifesto for the European elections 2009. Adopted at the Congress in Stockholm. European Parliament (2017), MEPs by member state and political group–8th parliamentary term. Retrieved from http://www.europarl.europa.eu/meps/en/crosstable. html European Parliamentary Research Service (2016), Review of European and national election results. Brussels: European Parliament. Gagatek W., Van Hecke S. (2014), ‘The development of European political foundations and their role in strengthening Europarties’, Acta Politica 49(1), pp. 86–104. Hanley D. (2008), Beyond the nation state: Parties in the era of European integration. London: Palgrave Macmillan. Hertner I. (2017), ‘Members are everything. Or not? Europarties and their individual membership schemes’, unpublished paper. Hix S. (2013), ‘Why the 2014 European elections matter: Ten key votes in the 2009–2013 EP’, Swedish Institute of European Policy Studies. Külahci E., Van de Walle C. (2002), ‘La quête d’identité du parti européen des libéraux, démocrates et réformateurs’. In Delwit P. (ed.), Libéralismes et partis libéraux en Europe. Brussels: Ed. de l’Université de Bruxelles, pp. 277–285. Lodge J., Herman V. (1982), Direct elections to the EP: A community perspective. London: Macmillan. Lord C. (1998), ‘The untidy right in the EP’. In Bell D., Lord C. (eds.), Transnational parties in the European Union. Aldershot, UK: Ashgate, pp. 117–133. Mair P. (2000), ‘The limited impact of Europe on national party systems’, West European Politics 23(4), pp. 27–51. Put G., Van Hecke S., Cunningham C., Wolfs W. (2016), ‘The choice of spitzenkandidaten: A comparative analysis of the Europarties’ selection procedures’, Politics and Governance 4(1), pp. 9–22. Sandström C. (2001), ‘Le parti européen des libéraux, démocrates et réformateurs: De la co-opération à l’intégration’. In Delwit P. et al. (eds.), Les Fédérations Européennes de partis. Brussels: Ed. de l’Université de Bruxelles, pp. 123–140. Sartori G. (1976), Parties and party systems: A framework for analysis. Cambridge: Cambridge University Press. Smith J. (2014), ‘Between ideology and pragmatism: Liberal party politics at European Level’, Acta Politica 49(1), pp. 105–121.

278  Wouter Wolfs and Steven Van Hecke Smith G. (1988), ‘Between left and right: The ambivalence of European liberalism’. In: Kirchner E. (ed.) Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 16–28. Van Hecke S. (2010), ‘Do transnational party federations matter? (… and why should we care?)’, Journal of Contemporary European Research 6(3), pp. 395–411. Van Hecke S., Alberti P. (2005), ‘Les groupes parlementaires au parlement européen: changements et continuités’. In: Delwit P. (ed.), Parlement puissant, électeurs absents? Les élections européennes de juin 2004. Brussels: Ed. de l’Université de Bruxelles, pp. 273–294. Van Hecke S., Johansson K.M. (2013), ‘Gipfelpolitik politischer parteien auf Europäischer Ebene: Eine vergleichende Analyse’. In: Poguntke T., Morlok M., Merten H. (eds.), Auf dem wege zu einer Europäischen Parteiendemokratie? Baden-Baden, Germany: Nomos, pp. 175–190. VoteWatch Europe (2016), Voting records of European political groups in the Parliament, Retrieved from https://www.votewatch.eu/en/term8-political-groups-votes.html. Watson G. (2010), Building a liberal Europe: The ALDE Project. London: John Harper. Wolfs W., Smulders J. (2018), ‘Party finance at the level of the European Union’. In: Mendilow J., Phélippeau E. (eds.), Handbook of political party funding. Northhampton, UK: Edward Elgar.

Part II

Comparative perspective on liberal parties in Europe

15 Liberal parties and elections Electoral performances and voters’ profile Caroline Close and Pascal Delwit

Introduction In terms of electoral performances and electoral sociology, liberal parties display diverging configurations across Europe. To get a better overview of their fate, this chapter first examines the electoral dynamics of liberal parties across specific spatial ‘blocs’. The chapter then examines the individual-level determinants of the liberal party vote, and questions the distinctiveness of the liberal electorate across Europe. The analysis is based on data from the latest rounds of the European Social Survey (ESS) and explores the impact of social characteristics and ideological preferences across 19 countries. The findings suggest a great heterogeneity among liberal party voters, which reflects their diversity in terms of origins, ideology, and electoral support.

The electoral performances of liberal parties The electoral assertion of liberal parties in the Benelux countries Historically, the liberal parties of the Benelux countries competed primarily on the church-state cleavage against the Catholic parties (allied with the Protestants in the Netherlands). Above all, liberal parties promoted modern parliamentarianism and constitutionalism (Delwit 2017). The promotion of economic liberalism did not constitute a dominant element of their identity. In terms of competition, the Liberals had to face challenging opponents on the right, the confessional parties, and by the end of the 19th century, on the left, the (often anticlerical) Social Democratic parties. The political landscapes that emerged after World War II confirmed the difficulty of liberal parties to attract substantial segments of the electorate. This challenge was particularly salient in Belgium and in the Netherlands, where liberal parties performed only modestly, attracting around 10% of the vote. However, since the 1960s and 1970s, liberal parties managed to increase their electoral scores, reaching around 20%. This change is mainly due to two parallel evolutions: the gradual weakening of Catholic or Christian parties due to

282  Caroline Close and Pascal Delwit the ongoing secularisation, the diversification of electoral preferences among the believers, and the depolarisation on the church-state cleavage (Delwit 2003); and the political realignment of the Liberals as the main right-wing contenders on the increasingly salient economic cleavage (Roussellier 1991). In the absence of conservative parties, liberal parties developed as the main right-wing organisations of the political spectrum. As such, they attracted right-wing voters, believers and nonbelievers alike. By the end of the 20th century and beginning of the 21st century, the Belgian Liberals – Mouvement Réformateur (Reformist Movement – MR) and Vlaamse Liberalen en Democraten (Flemish Liberals and Democrats – VLD) – and the Dutch Liberals – Volkspartij partij voor vrijheid en democratie (People’s Party for Freedom and Democracy – VVD) – reached the levels of the two historical party families, the Christian Democrats and the Social Democrats. They even became the first political force in their country. This process is less clear-cut in Luxemburg. The Christian Democrats (Chrëschtlech Sozial Vollekspartei – CSV) faced less electoral disaffection and the country remained divided along a significant philosophical divide. The contemporary period is characterised by two evolutions. The first one relates to the emergence of a new liberal party in the Netherlands, Democraten 1966 (Democrats 1966 – D66), structured around societal issues and libertarian positions, as well as around more universalist stances than most other parties that have moved towards more ethnocentrist positions. D66 joined the liberal family while sustaining tense relationships with its brother party, the VVD (Daalder and Koole 1988). From the mid-1960s until today, D66 faced unstable electoral results. Its average score revolves around 5 to 6%. Yet, D66 managed to exceed 10% three times: in 1981, in 1994, and in 2017 (see Table 15.1). The second noticeable evolution relates to the emergence of new challengers: radical right parties, or a form of unconstrained right on socioeconomic or immigration issues, such as the Partij voor vrijheid (Party for freedom – PVV) in the Netherlands and the Vlaams Belang (Flemish Interest – VB) and the Parti Populaire (People’s Party – PP) in Belgium; and Conservative parties, such as the Nieuw-Vlaamse Alliantie (New Flemish Alliance, N-VA) in Belgium, which has become in the past 10 years the leading party in Flanders and in Belgium, or the Alternativ Demokratesch Reformpartei (Alternative Democratic Reform Party – ADR) in Luxemburg. Hard-pushed liberals in the centre of Europe By contrast to Belgium and the Netherlands, and to lesser extent to Luxemburg, the Swiss Liberals have for long appeared as a very stable force in the political landscape. Two parties have embodied liberalism: the Parti Libéral Suisse (Swiss Liberal Party – PLS) and the Parti RadicalDémocratique (Radical Democratic Party, PRD). The first has been quite

BK-TandF-9780815372387_TEXT_CLOSE-181001-Chp15.indd 282

14/01/19 2:20 PM

Table 15.1  Electoral performances of liberal parties in the Benelux countries, Austria, Germany, and Switzerland

1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1961 1963 1965 1967

Austria

Germany

LIF NEOS Total

FDP

Belgium PLP-PVV MR 8.9

11.9

9.5

7.7

13.6 11.3

12.2

OpenVLD Total 8.9 13.6 11.3

Netherlands

Luxemburg

Switzerland

VVD D66 Total

DP

PLS PRD Total

6.4

6.4 8.0

8.0

8.8

8.8

12.2

11.1

11.1

12.8

12.3

12.3

9.5

21.6

21.6

18.0 11.6 20.9 12.3

3.2

23.0

26.2

2.6

24.0

26.6

2.2

23.3

25.5

2.3

23.7

26.0

23.9

26.1

23.2

25.5

8.8

8.8

12.2

12.2

10.3

10.3

2.2

15.2

2.3

10.7

4.5

20.3

(Continued )

Table 15.1  Electoral performances of liberal parties in the Benelux countries, Austria, Germany, and Switzerland (Continued )

1968 1969 1971 1972 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1989

Austria

Germany

LIF NEOS Total

FDP 5.8 8.4 7.9

10.6

Belgium PLP-PVV MR

OpenVLD Total

Luxemburg

Switzerland

VVD D66 Total

DP

PLS PRD Total

20.9

20.9 5.7

9.5

15.2

5.6

9.6

15.2

7.0 6.0

8.5 10.3

15.5 16.3

18.0

8.6

12.9

21.5

17.3 23.1

7.0

9.1

Netherlands

10.2

10.8

21.0

9.4

11.6

21.0

10.3 14.5

6.8 4.2

5.4

17.1 18.7

18.1 23.3

21.8

24.0

2.4

22.2

24.6

2.8

24.1

26.9

2.8

23.4

26.2

2.7

23.0

25.7

23.4 21.9

11.1 4.3

2.2

28.4 27.4 20.4

17.4

6.1

23.5

14.5

7.9

22.4

17.2

1990 1991 1994 1995 1998 1999 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2017

11.0 6.0 5.5

6.0 5.5

3.7 1.0

3.7 1.0

6.9 6.3 7.4

8.1

12.0

20.1

10.3

13.2

23.5

10.1

14.3

24.5

11.4

15.4

26.8

12.5

11.8

24.4

9.3

8.6

17.9

9.8 2.1

2.1

5.0

5.0

14.6

4.8

9.7

9.8

20.0

15.5

35.5

24.7

9.0

33.7

15.4 17.9

5.1 4.1

20.5 23.0

14.7

2.0

16.7

20.5

7.0

22.5

26.6

8.0

34.6

21.3

12.2

33.5

19.4

19.3 21.6 14.9

3.0

21.0

24.0

2.7

20.2

22.9

2.3

19.7

22.0

2.2

17.3

19.5

1.9

15.8

17.7

15.1

15.1

16.4

16.4

12.9

18.3

286  Caroline Close and Pascal Delwit weak electorally throughout its history (around 2.5% of the vote). The second has steadily attracted about 22 to 24% of the vote during half a century. In 2009, these two parties merged to form the Parti Libéral-Radical (Radical Liberal Party – FDP), at a time when the PRD had been losing ground for a decade. The merge did not halt this decline (Table 15.1). Like other Swiss parties, the Liberals have had to cope with the electorally successful transformation of the Union Démocratique du Centre (Democratic Union of the Centre – UDC) from a Conservative-agrarian party to a radical right party (see Chapter 6 in this volume). In this context, the Swiss Liberals struggle to find their political space between an ethnocentric right and the universalism promoted by the Social Democrats and the Greens. In Germany, the Freie Demokratische Partei (Free Democratic Party – FDP) displayed for long two characteristics: modest electoral results (between 5 and 11% of the vote until the mid-2000s), and a role of pivotal party or kingmaker in the German political system (Abedi and Siaroff 2011). However, it lost its status of kingmaker in 1998. The entry of Die Grunen (The Greens), and then of the Partei des Demokratischen Sozialismus (Party of Democratic Socialism – PDS) modified the electoral and political offer. After a second experience of grand coalition between 2005 and 2009, the FDP attracted 14.6% of the vote in October 2009, its best historical result, and occupied power again (Table 15.1). But in 2013, the party failed to attain the 5% threshold and consequently lost its parliamentary representation. The 2014 European election was disastrous. In the follow-up, the FDP lost representation in the Länders of Bavaria, Brandenburg, Saxony, and Thuringia. It failed to reconquer representation in Mecklenburg-Pomerania, Saarland, and Saxony-Anhalt. However, more recently, the FDP won in Berlin, Bremen, and Rhineland-Palatinate. For the party, the September 2017 elections are crucial. A new failure to get representation in the Bundestag could well seal its fate. In Austria, the history of political liberalism is complex. To a certain extent, the Freiheitliche Partei Österreichs (Freedom Party of Austria – FPÖ), which emerged from the Verband der Unabhängigen (Federation of Independents – VdU), incarnated liberalism for a while.1 The FPÖ had for long a double identity, expressed through two wings: a radical right wing nostalgic of the 1930s, and a liberal wing. In 1983, under the leadership of Jorg Haider, the former wing took over. The liberal wing left and formed the Liberales Forum (Liberal Forum – LiF). Despite attracting nearly 6% of the vote in 1994 and 1995, the LiF failed to establish as a relevant force, caught between the Conservative Österreichische Volkspartei (Austrian People’s Party – ÖVP) and the FPÖ. After losing parliamentary representation, the party almost disappeared, before joining a new political organisation, Das Neue Österreich (New Austria – NEOS). NEOS was formed in 2012, mainly by a dissidence of the liberal wing of the ÖVP (see Chapter 7 this volume). At its first election, NEOS attracted around 5% of the vote and subsequently

Liberal parties and elections 287 captured 8.1% of the vote and one seat at the 2014 European elections. For this new party, the most difficult task lies ahead: to confirm these results and to impose itself in the long run against the FPÖ. The electoral adversities of the Scandinavian liberal parties In the Scandinavian or Nordic countries, the Liberals present various profiles. A significant part of these parties were historically related to agrarian movements (Ivaldi 2002). Among them are the Danish and Norwegian Venstre (Left – V), the Icelandic Framsóknarflokkurinn (Progressive Party – FSF), and the Centre parties of Finland (Suomen Keskusta – KESK) and Sweden (Centerpartiet – C). Originally, these parties promoted universal suffrage and parliamentarianism, as well as the introduction of mandatory education, and the public regulation of social issues. These parties have occupied a centrist position, often between major Social Democratic and Conservative parties. In Denmark and Sweden, the Liberals position themselves to the centre right as an attempt to secure the position of main right-wing party, by opposition to the Social Democrats. In Denmark, this strategy has worked: Today Venstre now attracts between 25 and 30% of the vote and supplants the Conservative People’s Party, while the latter dominated in the 1980s and 1990s. By contrast, in Sweden, this strategy did not work: The Moderate Party is the main political force on the right. The Centre Party for long secured around 15% of the vote, but this score has severely decreased since the end of the 1990s, and today barely reaches 6 to 7%. The Norwegian Venstre lost its status of relevant party at the end of the 1960s. Until then, the Norwegian Liberals attracted around 9% of the vote at each election. Since then, their average score has considerably dropped. The party even failed to secure any seat at the 1985 and 1989 elections. Nowadays, the party remains stuck at about 5 to 6% (Table 1.2a). In Finland, KESK competes with the Conservatives from the KOK, but this competition is not entrenched within a right-wing bloc. To the contrary, regularly, the presence of KESK in the government is mutually exclusive of that of KOK. However, 2015 appears as an exception: An unusual rightwing government was formed, which aggregated KESK, KOK, and the radical right True Finns (Perussuomalaiset). The electoral performances of KESK have been stable over time (about 20 to 22%), with a lower interlude in the 1970s and 1980s (Table 15.1). In Iceland, the FSF remained electorally stable (between 24 and 28%) until the 1970s. During the 1980s, the party endured a net decline, well below 20%, and the rebound in 1995 was only ephemeral. The decline carried on, and the party’s average score is nowadays closer to 10%. In a context of virtual bankruptcy of the state and of rejection of the Independence Party, the FSF experienced a boom at the 2013 election, but again that success

288  Caroline Close and Pascal Delwit was only temporary. Three years later, the party obtained the second worse electoral score of its history. Other profiles exist among Nordic liberal movements. In Sweden, the Folkpartiet Liberalerna (Liberal People’s Party – L) belongs to the rightwing bloc and supports economic liberalism tempered by more social stances. The party competed to become the first party of the right-wing bloc. From 1945 to the mid-1950s, the party relied on the support of 20 to 25% of the electorate. But the 1958 election launched a downwards dynamic. Until 1970, the party stabilised around 17 to 18% of the vote. Then, from 1974, a new backlash decreased the party’s support down to 10%. In 1998, the party reached a record low of 5%. Since then, Liberalna has partially recovered, but remains stuck in a position of junior governmental partner and of modest electoral player in the right-wing bloc (Table 15.1). The Danish Radikale Venstre (Radical Left – RV) stands out within the Nordic liberalism. Founded in 1905 by a split from the Danish Venstre, RV belongs to the centre-left bloc with the Socialdemokraterne (Social Democratic Party – SD). The party’s electoral support has been stable over time, between 5 and 10% (Table 15.1). The Suomen ruotsalainen kansanpuolue (Swedish People’s Party of Finland – SFP) also stands out. The Swedish minority in Finland massively supports this party. This support is all-the-more relevant given that the party has regularly been part of the government. Its political support has slightly and slowly declined, following the sociodemographic evolution of the Swedishspeaking population of the country. From 8.5% in 1945, the party’s electoral support has evolved towards a stable score of 5% nowadays. Finally, we should emphasise the emergence of a new political party in Denmark, the Liberal Alliance (Liberal Alliance – LA). The party was originally founded in 2007 under the label New Alliance by some dissidents from the RV and the Conservative People’s Party (see Chapter 2). The ambition was to establish a centrist party in the context of the right-wing turn of the Venstre. Attracting 2.8% of the vote in 2007, the LA confirmed its parliamentary representation in 2011 and improved its performances in 2015 (7.5%) (Table 15.2). Despite internal tensions, the LA accepted to join the minority government, supported by the radical right Dansk Folkeparti (Danish People’s party – DF). Dissimilar British and Irish Liberals The Liberals in the United Kingdom and in Ireland appear extremely dissimilar. In the UK, following a period when they alternated in power with the Conservatives, the Liberals almost disappeared after World War II. Progressively though, bipartisanship waned and the Liberal Party could revive. This upturn was obvious in the 1970s, even though the party’s strength in terms of seats still appeared marginal. At best, the party secured 13 seats. The cartel, followed by the merge with the Social Democratic Party (SDP)2

Table 15.2  Electoral performances of liberal parties in Nordic countries Denmark

1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1956 1957 1958 1959 1960 1961 1962 1963 1964

V

RV

23.4

Sweden

Total

KESK

SFP

Liberals

Total

8.2

31.5

21.3

8.4

5.2

34.9

27.6

6.9

34.5

24.2

7.7

3.9

35.9

21.3

8.2

29.5

23.3

7.6

6.0

36.8

22.1 23.1

8.6 7.8

30.7 30.9

24.1

7.0

7.9

39.0

25.1

7.8

32.8

23.1

6.7

6.2

36.0

21.1

5.8

LA

Finland

26.9 23.0

20.8

5.3

26.1

6.4

6.5

FP

C

Iceland Total

22.8

12.3

35.1

24.4

14.4

38.8

23.8

17.1

40.9

18.2

12.7

30.9

17.5

13.6

31.1

17.0

13.2

30.2

35.8

Progressive Party

Norway BF

Total

V 13.8

23.1

23.1

24.5

24.5

13.1

21.9

21.9

10.0

15.6

15.6

27.2 25.7

27.2 25.7

9.7

8.8 28.2

28.2 (Continued )

Table 15.2  Electoral performances of liberal parties in Nordic countries (Continued ) Denmark

1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1981 1982 1983 1984 1985 1987 1988

V

RV

19.3 18.6

LA

Finland

Sweden

Total

KESK

SFP

Liberals

Total

7.3

26.6

21.2

6.0

6.5

33.7

15.0

33.5

15.6

14.3

30.0

12.3

11.2

23.5

23.3

7.1

30.4

12.0

3.7

15.6

12.5 11.3

5.4 5.1

18.0 16.4

12.1

5.5

17.6

10.5 11.8

6.2 5.6

16.8 17.4

17.1

5.7

6.0

28.8

16.4

5.4

5.2

26.9

17.6

17.3

5.0

4.6

17.6

4.6

17.6

5.3

4.4

3.7

27.0

25.5 22.2

1.0

23.9

FP

C

Iceland Total

Progressive Party

Norway BF

Total

V 10.4

14.3

15.7

29.9

16.2

19.9

36.1

9.4

25.1

34.5

11.1

24.1

35.2

10.6

18.1

28.7

5.9

15.5

21.4

14.2

12.4

26.7

12.2

11.3

23.5

28.1

28.1

25.3

25.3

24.9

24.9

16.9 24.9

16.9 24.9

18.5

18.5

18.9

18.9

9.4

3.5

3.2 3.9

3.1

1989 1990 1991 1993 1994 1995 1997 1998 1999 2001 2002 2003 2005 2006 2007 2009 2010 2011 2013 2014 2015 2016

15.8

3.6

19.3

23.3

4.6

27.9

24.0

3.9

27.9

31.3

5.2

36.4

29.0

9.2

38.2

26.3

5.1

26.7 19.5

9.5 4.6

2.8 5.0 7.5

31.4 36.2 31.6

3.2 24.8 19.6 22.4 24.7 23.1 15.8 21.1

5.8 5.5 5.1 4.6 4.6 4.3 4.9

0.8 0.6 0.2 0.3 0.1

31.4 25.7 27.7 29.6 27.8 20.0 26.0

9.1

8.5

17.6

7.2

7.7

14.8

4.7

5.1

9.8

13.2

6.1

19.3

7.4

7.7

15.2

7.1

6.6

13.6

5.4

6.1

11.5

18.9

18.9

23.3

23.3

18.4

18.4

17.7

17.7

11.4 14.8

11.4 14.8

3.9 5.2

24.4

8.3

32.7

11.5

7.2

18.7

3.6 4.5 3.9 5.9

292  Caroline Close and Pascal Delwit modified the liberal support. At the 1983 election, the upcoming Liberal Democrats (LibDem) attracted one quarter of the vote. Nevertheless, their number of seats remained tenuous: 23. The Liberals failed to maintain this level. For 15 years, their average electoral score swayed around 17 to 18%. Yet at the 1997 legislative election the party significantly increased its number of seats: from 20 (in 1992) to 46. The LibDems even reached 62 seats in 2005, when they crossed again the 20% threshold. Despite increasing their vote share in 2010, they lost five seats. For the first time since the end of World War II, they entered a governmental coalition under the lead of the Conservative David Cameron (Bale 2011). The price to pay was cruel. At the 2015 election, the party’s support dropped below 8%. Following the early election called by Theresa May in June 2017, the LibDems’ electoral performances slightly declined again. In Ireland, the party belonging to the ALDE is the major historical party of the country, the Fianna Fail (FF). The identity of FF is entrenched in the centre-periphery cleavage in the Irish context: The party opposed the 1921 Treaty but accepted to sit in the Irish parliament (Dunphy 1995). From the end of World War II until the early 2000s, FF established itself as the first political force in the Republic, attracting around 45 to 50% of the vote and constituting the main governing party. In the last ten years however, FF has been in trouble, largely because of the severe financial, economic, and social crisis that hit the country after 2008 and for which it was held responsible. As a result, FF was defeated in the 2011 election, reaching a record low score of 17.5%. After a legislature in which the Labour (L) and Fine Gael (FG) governed together, FF only partially recovered at the 2015 election: 23.2%. Another liberal organisation has participated in the Irish political life between 1985 and 2007, the Progressive Democrats (PD). Founded by personalities coming from FF and FG, the PD have displayed a double identity: very conservative on socioeconomic issues – by far the most right-wing party in the country, and liberal on societal issues, in a country where the Catholic imprint remains deeply embedded. Its first electoral participation was quite a success: the PD attracted 11.9% of the vote. However, the party did not manage to consolidate that performance. In 1989, the PD lost half of its support, and declined even further afterwards. Securing only two seats in 2007, the party was dissolved in 2009 (Table 15.3). The electoral evanescence of liberal organisations in Southern Europe In Portugal, Greece, Cyprus, Malta, and France, no liberal party can be identified as being electorally relevant (Massart 2006). Admittedly, liberal wings have always existed within conservative parties, but the ALDE

Liberal parties and elections 293 Table 15.3  Electoral performances of liberal parties in Ireland and the United Kingdom Ireland FF 1945 1948 1950 1951 1954 1955 1957 1959 1961 1964 1965 1966 1969 1970 1973 1974 1977 1979 1981 1982 1982 1983 1987 1989 1992 1997 2002 2005 2007 2010 2011 2013 2015 2017

PD

United Kingdom Total

41.9

41.9

46.3 43.4

46.3 43.4

48.3

48.3

43.8

43.8

47.7

47.7

45.7

45.7

48.8

48.8

50.6

50.6

45.3 47.3 45.2

45.3 47.3 45.2

44.2 44.2 39.1 39.3 41.5

11.9 5.5 4.7 4.7 4.0

56.0 49.6 43.8 44.0 45.5

41.6

2.7

44.3

17.5

17.5

24.4

24.4

Libdem 9.1 9.1 2.8 2.7 5.9 11.2 8.5 7.5 19.3 18.3 13.8

25.4 22.6 17.8 16.8 18.3 22.0 23.0 7.9 7.4

doesn’t include any relevant actor in these States. In this context, the decision of the République en Marche (The Republic Forward – LREM) to join the European liberal party and parliamentary group appears crucial for the ALDE. In Italy, political liberalism was central at the turn of the 19th and 20th centuries. But the establishment of universal suffrage, the end of the

294  Caroline Close and Pascal Delwit non-expedit, and the advent of the Mussolini regime, put an end to its relevance. After World War II, the Liberals got back on their feet, but they lacked a solid base. The Partito Liberale Italiano (Italian Liberal Party – PLI) never managed to attract beyond 6 to 7% of the vote and suffered from the spectacular organisational strength of the Democrazia Cristiana (Christian Democracy – DC) and of the Partito Comunista Italiano (Italian Communist Party – PCI) (Delwit 2016). Nonetheless, their political role was significant: The Liberals often exercised power jointly with the DC. Similar observations can be done regarding the Partito Repubblicano Italiano (Italian Republican Party – PRI). In the early 1990s, both the PLI and PRI were damaged by the Manu Pulite investigation. As a result, the Liberals lost their representation. The former judge Antonio Di Pietro attempted to bring back liberalism in the form of a personalised party, Italia dei Valori (Italy of Values – IdV) (Musella 2014) but this episode didn’t last. No liberal organisation has sent deputies to the Chamber in the last ten years. In Spain, political liberalism was partially incarnated by the Centro Democrático y Social (Democratic and Social Centre – CDS) during the democratic transition, but the presence of the CDS was ephemeral. At the 1986 and 1989 national elections, the CDS managed to obtain honourable results: respectively, 9.3% and 8%. Afterwards, the party merged into the Partido Popular (People’s Party – PP). These performances were followed by a long absence of the Liberals. In 2008, a new political organisation ran for the election, Unión, Progreso y Democracia (Union, Progress and Democracy – UPyD). UPyD performed modestly, but in the follow-up of the financial and economic crisis and of the early election called in 2011, the party attracted 4.8% of the vote and managed to get parliamentary representation. In the context of the erosion of the two dominant parties, the Partido Socialista Obrero Español (Spanish Socialist Workers’ Party – PSOE) and the PP, UpyD appeared as a potential alternative. But UPyD failed to capitalise on this potential. With the emergence of Ciudadanos (The Citizens – C’s), UPyD collapsed and disappeared from the political spectrum. Originally founded in Catalonia, C’s realised a remarkable breakthrough in the 2015 and 2016 national elections (Bosch and Duran 2017). With 14% and 13.2%, respectively, the party contributed to break the bipartisan dynamics that existed since the end of the 1980s (Table 15.4). In Malta, Cyprus, Greece, and Portugal, there are no relevant liberal parties. Yet in Portugal, the Partido Social Democrata (Social Democratic Party – PSD), the main right-wing party of the country, joined the Federation of Liberal and Democratic Parties (ELD) in 1986. The party participated in initiating the renaming of the Federation into the ‘Federation of Liberal, Democratic and Reformist Parties’ (ELDR) (Sandström 2001). However, the PSD left the Federation to join the Conservative and Christian Democratic parties gathered in the European People’s Party (EPP).

Liberal parties and elections 295 Table 15.4  Electoral performances of liberal parties in Italy and Spain Italy

1946 1948 1953 1958 1963 1968 1972 1976 1979 1982 1983 1986 1987 1989 1992 1993 2000 2001 2004 2006 2008 2011 2015 2016

Spain

PRI

PLI

IdV

Total

4.8 2.5 1.6 1.4 1.4 2.0 2.9 3.1 3.0

6.8 3.9 3.0 3.5 7.0 5.8 3.9 6.8 2.0

11.5 6.4 4.6 4.9 8.3 7.8 6.7 9.8 5.0

5.1

2.9

8.0

3.8

2.1

6.0

4.4

2.9

7.3 0.5

0.5

2.3 4.4

2.3 4.4

C’s

14.0 13.2

UPyD

1.2 4.8 0.6 0.2

CDS

Total

2.9

2.9

9.3

9.3

8.0

8.0

1.8 0.1

1.8 0.1

0.1

0.1 1.2 4.8 14.7 13.4

Liberal parties in the Baltic and CEE countries In the Baltic and Central and Eastern European countries, the Liberals look like a patchwork (Table 15.5). Estonia displays a number of strong liberal parties. The Eesti Reformierakond (Estonian Reform Party – ER) is a classical liberal party adopting right-wing stances on socioeconomic issues. Founded in 1994, the party quickly asserted itself as a relevant actor. During its first decade of existence, ER attracted between 16 and 17% of the vote. At the 2007 election, ER’s score jumped to 26 to 27% and the party became one of the most successful liberal parties in Europe. Eesti Keskerakond (Estonian Centre Party – EK) displays a more leftwing profile on socioeconomic issues. But what makes the party distinctive is its relationship to the Russian-speaking minority of the country. Accordingly, the party steadily attracts about a quarter of the vote at each election. The contrast with Latvia is striking. Latvijas Ceļš (Latvian Way – LC), a conservative-liberal party, joined the international liberal organisations and dominated Latvian politics in the 1990s. After an electoral breakdown in 2002, LC formed an electoral coalition in 2006 with Latvijas Pirmā Partija (Latvia’s First Party – LPP), a party entrenched in conservatism and

296  Caroline Close and Pascal Delwit Table 15.5  Electoral performances of liberal parties in the Baltic countries Estonia

Latvia

Lithuania

ER EK Total LC LPP LA Total DP LCS LICS LLS LRLS Total 1992 1993 1995 1996 1998 1999 2000 2002 2003 2004 2006 2007 2008 2011 2012 2014 2015 2016

12.2

32.4 16.2 14.2 30.4 14.6 15.9 23.4 39.3 17.7 25.4 43.1 27.8 26.1 53.9

18.0 4.9 9.6

14.5

8.6*

8.6

28.6 23.3 51.9 27.7 24.8 52.5

0.9

0.9

2.5

2.5

8.7

8.7

2.9

2.9 37.6

28.4

9.2

9.0

5.3

5.7

20.0

20.7

2.1

9.0

31.8

9.5

16.6

4.9

2.2

Note: *LPP/LC electoral coalition.

Christian democracy. In 2007, the two parties merged, but this new party was short-lived. Recently, a new liberal party has emerged, Latvijas attīstībai (Latvian Development – LA), but remains quite marginal. In Lithuania, liberalism is quite fragmented. In the 1990s and 2000s, it was partly embodied by the Lietuvos Centro Sajunga (Centre Union – LCS), which formed the Lietuvos liberalų ir centro są junga (Liberal and Centre Union – LICS) in 2003 with other parties. In 2004, LICS attracted 9.2% of the vote. Yet, LICS failed to impose itself in the long run. In 2006, a dissident group created a new party, Lietuvos Respublikos Liberalų są jūdis (Liberals’ Movement of the Republic of Lithuania – LRLS). This organisation managed to reach an average score of 9% in the 2012 and 2016 elections. Conversely, the Lietuvos Laisvės Są junga (Lithuanian Freedom Union – LLS) founded by Vilnius’ former mayor Artūras Zuokas, and stemming from LICS, failed to attract a substantial share of the vote at the 2016 election. Among the parties that joined the ALDE, the most successful one in electoral terms is Darbo partija (Labour Party – DP). Founded in 2003, DP has a more social profile than the other liberal organisations. Its electoral performances are volatile. In 2004, DP attracted almost 30% of the vote, but four years later the party suffered a severe setback. A similar dynamic occurred at the two following elections of 2012 and 2016. In Central Europe, liberal parties have struggled to establish themselves against the Conservative and Radical right parties (Table 15.6).

Table 15.6  Electoral performances of liberal parties in CEE countries Czech Republic

Hungary

Poland

Slovenia

Croatia

Bulgaria

Romania

ANO2011 DEU/US Total SZDSZ DP UW UD KLD Total SMC LDS ZaAB DL Total HNS HSLS IDS Total DPS NSDV Total 1990 1991 1992 1993 1994 1996 1997 1998 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2013 2012 2014 2016 2017

21.4

19.7 8.6 14.3

1.5 14.3

7.6 5.6

4.8

18.7

2.4

12.3 7.5

19.8

10.6 4.0

14.6

13.4

13.4

3.1

3.1

2.4

23.5

23.5

27.0

27.0

36.3

36.3

22.8

31.6

5.6

5.6

1.5

18.7 34.6

8.4 4.3

9.9

0.6 39.5

8.0

6.7

4.1

6.4

0.6

8.0 7.6

8.0 7.6

5.4

5.4

7.6

7.6

7.5

42.7

50.2

12.8

19.9

32.7

14.5

3.0

17.5

12.1

1.5 14.6

0.6

11.3

11.3 14.8 9.2

0.2

15.1 9.2

PNL 6.4 2.6 30.7 6.9

31.3

18.6

60.1 20.0

298  Caroline Close and Pascal Delwit In the Czech Republic, liberalism was initially incarnated by Demokratická unie (Democratic Union – DEU) and its enlargement to Unie Svobody (Freedom Union – US). However, in 1998, DEU did not break through. In 2002, the electoral cartel between DEU/US and Křesťanská a demokratická unie-Československá strana lidová (Christian and Democratic Union – Czechoslovak People’s Party – KDU-CSL) did not perform as expected and DEU/US was dissolved. During the following ten years, no liberal party participated in the elections. One of the richest men in the country, Andrej Babiš, has recently founded a new party, ANO2011 (Yes2011). Its main goal was to challenge Občanská demokratická strana (Civic Democratic Party – ODS) and to unlock the political system marked by the confrontation between ODS and the Social-Democrats (CSSD), and by the stigmatisation of the Communist Party of Bohemia and Moravia (KSCM). ANO2011 won its bet by gleaning 18.6% of the vote and by forming a government with the Christian Democrats and the Socialists. In Slovakia, no liberal party has managed to emerge since the independence of the country3. The current situation in Hungary is not better. Initially, Szabad Demokraták Szövetsége (Alliance of Free Democrats – SZDSZ) established itself as a relevant actor, but after two promising results at the 1990 and 1994 elections, it collapsed and could not recover from its electoral crash of 1998. The party, confronted with the radical-right turn of the FIDESZ and the rise of JOBBIK, dissolved in 2013. Liberal parties faced more or less the same fate in Poland. In the early years after the democratic transition, Unia Demokratyczna (Democratic Union – UD) and Kongres LiberalnoDemokratyczny (Liberal Democratic Congress – KLD) – at that time led by Donald Tusk – obtained promising results. These two parties merged into the Unia Wolności (Freedom Union – UW), personified by Borislaw Geremek, at the price of enduring internal tensions (Millard 2001). UW did not consolidate its 1997 performances (13.4%), and soon disappeared from the political scene, marked by the rise of Prawo i Sprawiedliwość (Law and Justice – PiS) in competition with the conservative Platforma Obywatelska (Civic Platform – PO). In the past 15 years, there has not been any deputy sitting under a liberal label. In Slovenia, Liberalna demokracija Slovenije (Liberal Democracy of Slovenia – LDS) constituted a stable liberal party during 15 years. Stemming from the Communist youth organisation, the party presented itself as the main party of the country, gathering 25 to 35% of the vote (Jou 2011). In 2004, the party was sent back in the opposition. In 2008, it did not attract more than 5%, and, in 2012, it vanished from the parliamentary scene. At that time, the disappearance of liberal forces appeared as a credible scenario given the strength of the Conservatives on the right and of the Social Democrats on the left. However, both underwent a marked decline in 2014 to the benefit of the new liberal party Stranka modernega centra (Modern Centre Party – SMC). Founded only a couple of weeks before the elections, SMC won with 34.6% of the vote and became the main governmental party. Its challenge will be to sustain its strength in a context of intense volatility.

Liberal parties and elections 299 At the same time, another formation has joined the ALDE, Zavezništvo socialno-liberalnih demokratov (Alliance of Social Liberal Democrats – ZaAB), founded the same year by outgoing Prime Minister Alenka Bratušek, whose resignation led to early elections. This party was much less successful and attracted only a bit more than 4%. In Croatia, the fate of the Liberals is much trickier. Since the democratisation, two formations have borne their colour: Hrvatska narodna stranka – liberalni demokrati (Croatian People’s Party-Liberal Democrats – HNS) and Hrvatska seljačka stranka (Croatian Peasant Party – HSS). The former is entrenched in a Western-based liberal vision on economic issues; the latter is rooted in agrarianism. In the 2003 and 2007 elections, both parties ran independently, attracting a total of 12% of the vote. Afterwards, HSS considerably lost its influence and almost disappeared in 2011. The two parties decided to join electoral cartels. HNS opted for an alliance with the main left-wing party, Socijaldemokratska partija Hrvatske (Social Democratic Party of Croatia – SDP). HSS chose to coalesce with the main Conservative party, Hrvatska demokratska zajednica (Democratic Union – HDZ). In Romania, the Partidul Național Liberal (National Liberal Party – PNL) claims to be one of the oldest liberal parties in Europe and the successor of a party founded in 1875. Initially, the PNL performed relatively modestly (Preda 2004), but it has progressively increased its electoral performances, either in coalition or independently. Characterised by positions oscillating between conservatism and liberalism, the PNL left the ALDE in 2014 and joined the EPP soon after, a choice that was ideologically coherent but which also resulted from strategic considerations. In Bulgaria, no liberal party managed to structurally emerge in the post1990 political landscape. Only one formation can pretend to have done so, Dvizhenie za prava i svobodi (Movement for Rights and Freedoms – DPS). DPS is a community-based party, supported by the Turk-speaking minority of Bulgaria. The party actually stems from the Turkish National Freedom Movement founded in 1986 (Spirova 2005). Again, this largely explains its stable electoral performances. In the first half of the 2000s, the National Movement Siméon II was founded by the last King of Bulgaria. In 2002, the party was remarkably successful and its founder became Prime Minister for one legislature. After the withdrawal of its founder from the political life, the organisation was renamed National Movement for Stability and Progress (NACV), but it never recovered the same aura. In 2014, (marking its last electoral participation), the party attracted only 0.2% of the vote. In sum, electoral dynamics vary greatly for liberal parties in Europe. This can be explained by several elements. First, party systems are quite fluid in some countries, especially in Central and Eastern Europe. Second, the right side of the political spectrum has been on the move in the past 25 years. Christian Democratic parties are in electoral decline, most often to the benefit of Conservative and Radical Right parties. Finally, the liberal family is singularly ideologically heterogeneous compared to other political families

300  Caroline Close and Pascal Delwit (see Chapter 17). Liberal parties have emerged and evolved around different cleavages. This diversity is also reflected in the heterogeneous profile of the liberal voters, as shown in the next section.

The profile of liberal party voters The ‘invisible’ liberal party voter? Comparative analyses of voters’ profile in Europe usually point out the lack of distinctiveness of the liberal electorate (Vandermotten and Médina Lockhart 2002). Using the 2004 European Election Study data, van der Brug et al. (2009) investigated the effect of sociodemographic and ideological variables on the vote for the different ideological families across 20 European countries. In their models, the percentage of explained variance for the liberal family is the lowest of all families: only 4% of explained variance, compared to 22% for the conservative family, and 15% for the Social Democrats. Only two sociological traits are consistent across national contexts (Kirchner 1988): education and socioeconomic status. Liberal party voters come significantly more from the middle and upper middle classes than from the working class (van der Brug et al. 2009), and are more likely to be self-employed, employers, nonmanual employees, and highly skilled workers than manual workers (Knutsen 2008). Other sociodemographic variables have also been considered, but no clear trends emerge. Regarding the impact of religion, the evidence is mixed and relates to the national cleavage structure. Where liberal parties emerged on the secular side of the state-Ccurch cleavage, they attracted nonreligious voters. Yet, as in the Belgian case, some of these liberal parties realigned and their electorate is today hardly distinctive in terms of religiosity (Close 2017). In other contexts, as in Germany, the liberal party attracted both secular and Protestant voters (Kirchner and Broughton 1988); but the party also managed to gain support among Catholics (Müller-Rommel 2000). Analysing the effect of religion comparatively, van der Brug et al. (2009) found that attendance to religious services is inversely correlated to liberal party voting, but that belonging to specific denomination (Protestant and Orthodox) is positively correlated to liberal party voting. Testing the effect of age, van der Brug et al. (2009) show that liberal party voters are slightly older than other voters, whereas liberal parties in CEE countries attract younger voters (Chapter 12 in this volume). Regarding residence, liberal party voters are often depicted as urban, but the history of liberal parties has also been linked with agrarian movements (Ivaldi 2002). Finally, in certain contexts, liberal parties have been associated with the protection of regional or ethnic identity; in these cases, liberal party voters are likely to come from specific ethnic groups. Regarding the ideological profile of liberal party voters, studies show that their self-placement leans to the centre-right (van der Brug 2009;

Liberal parties and elections 301 Vandermotten and Médina Lockhart 2002). De Winter and Marcet (2000) show that this self-placement combines a clear rightist positioning on socioeconomic issues with more centre-left views on the cultural or societal dimension. However, this can vary greatly across countries and parties, especially when several liberal parties compete in the same party system. Regarding the ‘new’ cleavages, support for European integration seems to be widespread among liberal party voters, although weaker than among Social-Democratic and Green party voters. Liberal party voters’ position towards immigration is less clear: on the one hand, they value individuals’ civil rights and should therefore adopt positive stances on immigration; on the other hand, some liberal voters and parties have become more rightist on this dimension (De Winter 2000). Finally, their position on environmental issues has not been extensively examined. We expect that they favour economic growth at the expense of environmental protection. However, in Scandinavia, the environmental dimension appears quite important for the liberal electorate (Ivaldi 2002; Chapters 1 and 2 in this volume). Data and Method The analysis relies on data from the 6th round of the ESS (2012) and, when available, from the 7th round (2014). Our dataset includes 19 countries and 29 liberal parties that are all members of the ALDE, with some notable exceptions: the Social Democratic Party in Portugal is now a member of the EPP, but is a former member of the ELD and of the Liberal International; the Liberal Alliance in Denmark has no European affiliation; and the Progressive Party in Iceland is part of the Liberal International (Table 15.7). Liberal party voters are identified with the question asking the respondents for which party they voted in the last national election. Abstainers and respondents who did not answer the question are excluded from the analysis; the statistical models thus compare liberal party voters with voters for all other parties. With regard to sociodemographic characteristics, we include age, gender, and education (primary, secondary, and higher education), religion, occupation, type of residence and belonging to a minority ethnic group. Religion is measured with two indicators: religious denomination and religious practice (measured as the frequency of religious service attendance apart from special occasions). The type of residence distinguishes between rural (countryside or small village), semi-urban (small town), and urban (big city). To measure respondents’ political preferences, we use their answers to Likert-type questions related to issues. For the socioeconomic dimension, we use the proposition ‘the government should reduce differences in income level’. For the cultural dimension, we use the proposition ‘gays and lesbians should be free to live life as they wish’. High scores (on a 5-point scale) on these questions indicate a right-wing or conservative position. For the environmental issue, we use a 10-point scale where 10 indicates a great importance given ‘to caring about the nature and the environment’. The position on

302  Caroline Close and Pascal Delwit Table 15.7  Selected cases and frequency distribution Country

ESS round

Party

Austria Belgium

2014 2014

Czech Republic Denmark

2014 2014

Estonia

2014

Finland

2014

Germany Iceland Ireland Lithuania

2014 2012 2014 2012

Netherlands

2014

Norway Poland Portugal Slovenia

2014 2014 2012 2014

Spain Sweden

2012 2014

Switzerland United Kingdom Total

2014 2012

NEOS OpenVLD MR ANO 2011 RV V LA EK ER KESK SFP FDP FSF FF DP LiCS LRLS VVD D66 V RP PSD ZaAB SMC UPyD FP C FDP LibDems

N liberal vote (%) N total (valid) 59 (5.4%) 104 (7.9%) 144 (11.0%) 315 (28.1%) 134 (11.4%) 311 (26.4%) 48 (4.1%) 284 (27.6%) 321 (31.2%) 242 (18.0%) 62 (4.6%) 55 (2.7%) 74 (14.3%) 326 (22.1%) 273 (27.3%) 16 (1.6%) 62 (6.2%) 109 (8.3%) 177 (13.5%) 65 (6.0%) 16 (2.1%) 328 (35.4%) 19 (3.3%) 218 (37.9%) 47 (4.1%) 112 (7.8%) 95 (6.6%) 123 (18.8%) 256 (18.5%) 4,395 (20.5%)

1,089 1,311 1,119 1,179 1,029 1,345 2,077 518 1,476 1,001 1,310 1,083 769 926 575 1,130 1,437 653 1,382 21,410

Source: ESS 2012 and 2014 Note: For Croatia, Greece, Latvia, Romania, and Luxembourg, data was not available in the 2012 and 2014 rounds. In the case of Bulgaria (NDSV) and Hungary (SZDSZ), too few liberal party voters could be identified. These countries were thus dropped from the sample; * When using the ‘second vote’: vote for FDP = 4.7%.

the authoritarian/libertarian dimension is measured with an additive index of two propositions: the ‘importance that government is strong and ensures safety’ and the ‘importance to follow traditions and customs’ (a high score indicates more libertarian attitudes). For positions on immigration, we use a 10-point scale question about whether immigrants make their country’s economy worse or better (a high score indicates ethnocentrism). Respondents’ positions on these issues barely correlate, thus we kept them as separate items. For positions on EU integration, we use the respondents’ average trust in the EU parliament (10-point scale, a high score indicating a high level of trust), which is highly correlated with respondents’ opinion on European integration.

Liberal parties and elections 303 Analyses The multivariate analyses presented in Table 15.8 examine the profile of liberal party voters across our sample. Model 1 tests the effect of sociodemographic variables alone. Model 2 focusses on ideological predictors. Model 3 is the full model, while Model 4 controls for country effects (the country coefficients are not included in the table for reason of space). The models take into account the hierarchical structure of the database, by clustering the data by country. The displayed coefficients are the odds-ratio: a value above 1 indicates that the variable increases the probability of voting for a liberal party, a value below 1 indicates that it decreases it. The value of the adjusted pseudo R2 is quite weak (especially in model 1), but corresponds to that found in previous studies (Van der Brug et al. 2009). The effect of education and occupation are in line with previous findings: Self-employed and higher educated voters are significantly more likely to vote for a liberal party. Liberal party voting is significantly more likely for nonreligious voters. The positive effect of Eastern Orthodox is mainly due to Estonia, which concentrates 64% of the Eastern Orthodox respondents included in the database, among which 84% voted for the Centre Party (49% of the party’s voters declare themselves orthodox). Attendance to religious service does not appear as a discriminant factor. Interestingly, belonging to an ethnic minority group increases the probability of liberal party voting, although the significance level varies across models. Age, gender, and residence have no clear and significant effect, except in model 3, where women appear significantly more likely to vote for a liberal party. These general trends need to be nuanced for country specificities. The relationship between belonging to an ethnic group and liberal party voting is supported only in a few cases: on the one hand, in ethno-regionalist parties, the Swedish People’s Party and the Estonian Centre Party (appealing to the Russian-speaking minority); on the other hand, in two parties that are the most ‘universalist’, the Swedish Liberals and the Norwegian Venstre (respectively at 3.3 and 3.6 on the immigration dimension). As regard residence, former agrarian parties such as the Danish Venstre, the Icelandic Progressive Party, and the Finnish and the Swedish Centre parties, perform better among rural residents. This is also the case for other parties: the VVD, VLD, FF, PSD, and SFP. Liberal parties that attract more respondents from urban areas include the RV, Swedish Liberals, Norwegian V, D66, EK, LRLS, and ZaAB. Regarding the effect of age, education, and religion, a group of parties present different dynamics: the EK, KESK, FF, and PSD. Voters of these parties appear older, attend religious services more often, and display lower education levels than other voters in their country (except the PSD, for which the effect of education is not significant). Contrary to the general trend that religious denomination has either no effect or that nonreligious voters are more likely to vote for a liberal party (i.e., this is the case for ANO2011, LibDem, D66, SMC, ZaAB, and overwhelming for the Palikot’s

304  Caroline Close and Pascal Delwit Table 15.8  The determinants of the liberal party vote: Logistic models (odds ratio) (1) Sociodemographics Age Women Education Religious denomination (ref. not religious)

Attend religious services (ref. never) Occupation (ref. self-employed)

Residence (ref. urban)

(3)

(4)

Primary Higher

1.0 1.1 1.3 1.3*

1.0 1.2* 1.4 1.3**

1.0 1.1 1.2 1.2+

Roman Catholic

0.8

0.8

1.0

Protestant Eastern Orthodox Other Christian Other Sometimes

0.9 6.0*** 0.9 0.4*** 1.2

0.8 5.2*** 0.9 0.5** 1.1

1.0 3.3*** 1.1 0.7+ 1.0

Often

1.0

0.8

0.8

Employee in public sector Employee in private sector Student Unemployed Retired Housework Other Semi-urban Rural

0.8**

0.8*

0.8*

0.7**

0.8*

0.8*

0.9 0.7* 0.8+ 0.6** 0.7* 1.0 1.1

0.8 0.7 0.8+ 0.6** 0.8+ 1.1 1.1

1.6+

1.6*

0.9 0.7* 0.8+ 0.7* 0.8 1.1 1.2 0.2 1.5+

1.1* 1.1 1.2 1.1 1.0 1.0

1.2** 1.1 1.2+ 1.0 1.0 1.0

1.2** 1.1+ 1.0 1.0 1.0* 1.0

1.1*** 18,565 0.024

1.1** 18,261 0.042

1.0+ 18,261 0.157

Belong to a minority ethnic group Ideological preferences Left-right self-placement Socioeconomic right Cultural conservatism Libertarian Ethnocentrism Environment important issue Trust in EP Observations McFadden’s Adjusted R2

21,121 0.019

(2)

Source: ESS round 2012 and 2014 Note: Design weights applied. Standard errors adjusted for 19 country clusters. + p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001.

Liberal parties and elections 305 Movement), these parties attract religious respondents: Eastern Orthodox (as seen above, EK), Catholics (FF and PSD), or Protestants (KESK). Protestant respondents are also overrepresented among voters of the Danish Venstre and of the Swedish Liberals. Moreover, rural residents are overrepresented among their electorate (except for EK). Note that binary analyses conducted country by country show that gender has no effect, except in two cases: the Danish Liberal Alliance and the Palikot’s Movement, which are both overwhelmingly supported by male respondents. The effect of ideological attitudes indicates that, across countries, a more rightist self-placement increases the probability of liberal vote. In the same way, the higher respondents’ trust in the European parliament, the higher their probability to vote for a liberal party. These effects are consistent across the models. However, few other effects are observed across the models, even on the traditional socioeconomic and cultural cleavages. This lack of relationship is due to important cross-country and cross-party variations. To illustrate these variations, we aggregate voters’ attitudes at the party level. Figure 15.1 draws the distribution of these aggregated positions along the socioeconomic left-right and the cultural dimensions. The results show that right-wing preferences on socioeconomic issues are inversely related to

R2 Quadratic = 0,398

5,0

Progressive-conservative dimension

LiCS 4,0 DP_lith LRLS EK 3,0 PSD KESK 2,0

ER ZaAB

ANO11

SFP RP LibDem FDP UPyD FF NEOS FDP_swi MR VLD D66 FSFC_swe RV FP V_nor VVD

SMC

1,0 1,5

2,0

3,0 3,5 2,5 Left-right socio-economic dimension

V_den LA

4,0

Figure 15.1  Distribution of voters’ aggregated ideological preferences. Source: ESS round 2012 and 2014

306  Caroline Close and Pascal Delwit right-wing preferences on cultural issues. This is probably one of the main characteristics of liberal parties (see Chapter 17). The figure also illustrates the diversity of liberalisms in Europe, along the two traditional cleavages. A first group of voters tend to hold progressive views and clear right-wing economic preferences (e.g., V and LA in Denmark, FDP in Germany and Switzerland, etc.). A second group of voters holds similar preferences along the cultural axis, but are positioned close to the centre on the socioeconomic axis (e.g., NEOS, Libdem, UPyD, FSF, etc.). A third group is situated around the centre of the socioeconomic axis but hold more conservative views (e.g., LiCS, DP, LRLS, EK, PSD, etc.). To examine the existence of these different groups, we compute a k-means cluster analysis, using six ideological dimensions. Table 15.9 displays the final cluster centres and lists the parties included in each group. The parties included in cluster 1 distinguish themselves by their more conservative stances combined with centrist socio-economic positioning (as suggested in Figure 15.1.). Liberal parties included in cluster 2 are the most right-wing on socioeconomic issues, hold more progressive and libertarian views, and more universalist and pro-European stances. Liberal parties included in cluster 3 are more centrist on socioeconomic issues (closer to cluster 1), although they still hold relatively progressive stances. This group of parties also attracts voters with more ethnocentric and less pro-European positions; this is particularly driven by the voters of the VVD and ANO2011, which both score between 6 and 7 on the ethnocentric dimension. Besides, the voters of the VVD are particularly less pro-European, scoring around 2.5. Other parties included in cluster 3 are positioned lower on the pro-European scale compared to other liberal parties: NEOS, FSF, UPyD, ZaAB, and SMC score on average around 3. Table 15.9  The diversity of liberal parties: Average positioning of their voters (final cluster centres) Cluster 1 Socioeconomic dimension Cultural dimension Libertarian Ethnocentrism Environment important issue Trust in EP Countries/ parties included

Cluster 2

Cluster 3

1.9

2.8

2.1

3.3

1.5

1.9

2.5 4.8 4.9

2.9 4.3 4.9

2.6 5.3 4.9

4.8 EK, ER, KESK, DP, LiCS, LRLS

5.0 MR, VLD, FDP (SWI), FDP (GER), LA, RV, V (DK), SFP, D66, V (NOR), C (SE), FP

3.7 NEOS, ANO, UPyD, LibDem, FF, FSF, VVD, Palikot, PSD, SMC, ZaAB

Liberal parties and elections 307 These divisions follow an East/West divide: Cluster 2 includes only Western European cases; cluster 1 has mostly Eastern European cases; and cluster 3 gathers a mix of Western and Eastern European cases. This suggests that voters’ preferences are context-driven. Note that, where several liberal parties compete within the same national party system, they still belong to the same cluster – except in two instances: Finland, and the Netherlands.

Conclusion In terms of electoral success and electoral sociology, liberal parties display diverging configurations across Europe. This chapter has examined this heterogeneity at two levels. First, the chapter has provided an overview of the electoral fortunes of liberal parties across specific spatial ‘blocs’ and revealed diverging fates. Second, the chapter has examined the profile of liberal party voters. The minimum common denominator among liberal party voters is their position as self-employed and their level of education. The analysis has also uncovered the existence of at least three ideological profiles of liberal party voters. A first group is characterised by their positions that combine rightwing socioeconomic preferences with progressive, libertarian, universalist, and pro-European values. This cluster mainly includes the electorate of ‘old’ liberal parties from Western Europe. A second group of voters combine progressive stances with more centrist socioeconomic preferences and less marked pro-European attitudes. This group is more heterogeneous in terms of geographical scope, and gathers both older (FSF, FF, LibDem, VVD) and newer parties (ANO11, NEOS, UPyD, ZaAB). The third group includes more conservative liberal voters, mostly found in Estonia, Lithuania, and Finland (KESK). These findings suggest a great heterogeneity among voters, and question the existence of a coherent and unified liberal party family at the level of the ‘party in the electorate’. The heterogeneity of voters’ profile reflects the diversity of the liberal parties in terms of origins, ideology, and electoral support. These findings further help to understand the disparate and volatile fates of liberal parties across European contexts, in which they often struggle to position themselves in the political landscape between strong challengers on the centre-left and on the conservative or radical-right side.

Notes

1. The FPÖ was a member of the Federation of Liberal and Democratic parties for a while, and it was included as a liberal party in the volume edited by Kirchner (1988). 2. The SDP is a pro-European dissidence from the Labour Party, and was founded in 1981. In 1983, they formed an electoral cartel with the Liberals, then they merged with them. 3. This is however debated. See Chapter 12 in this volume.

308  Caroline Close and Pascal Delwit

References Abedi A., Siaroff A. (2011), ‘The kingmaker is dead, long live the kingmaker: Examining the degree of influence of small parties in the coalition-formation process in Germany’, German Politics 20(2), pp. 243–259. Bale T. (2011), ‘I don’t agree with Nick: Retrodicting the Conservative–Liberal Democrat coalition’, Political Quarterly 82(2), pp. 244–250. Bosch A., Duran I.M. (2017), ‘How does economic crisis impel emerging parties on the road to elections? The case of the Spanish Podemos and Ciudadanos’, Party Politics (Online First). Close C. (2017), ‘La sociologie électorale des libéraux francophones’. In: P. Delwit (ed.), Le mouvement réformateur. Brussels: l’Université de Bruxelles, pp. 219–236. Daalder H., Koole R. (1988), ‘Liberal parties in the Netherlands’. In: Kirchner E.J. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 151–177. Delwit P. (ed.). (2003), Démocraties chrétiennes et conservatismes en Europe. Une nouvelle convergence? Brussels: l’Université de Bruxelles. Delwit P. (2016), Les gauches radicales en Europe. XIXe et XXIe siècles. Brussels: l’Université de Bruxelles. Delwit P. (ed.) (2017), Du parti libéral au MR. 170 ans de libéralisme en Belgique. Brussels: l’Université de Bruxelles, pp. 17–65. De Winter L., Marcet J. (2000), ‘Liberalism and liberal parties in the European Union’. In: De Winter L. (ed.). (2000), Liberalism and liberal parties in the European Union. Barcelona, Spain: ICPS, pp. 11–24. Dunphy R. (1995), The making of Fianna Fáil power in Ireland 1923–1948. Oxford: Clarendon Press. Ivaldi G. (2002), ‘Les partis libéraux Scandinaves. Unité et diversité du courant libéral au Danemark, en Finlande, en Norvège et en Suède’. In: Delwit P. (ed.), Libéralismes et partis libéraux en Europe. Brussels: l’Université de Bruxelles, pp. 141–178. Jou W. (2011), ‘Left–right orientations and ideological voting in new democracies: A case study of Slovenia’, Europe-Asia Studies 63(1), pp. 27–47. Kirchner E.J. (1988), Liberal parties in Western Europe. Cambridge: Cambridge University Press. Kirchner E.J., Broughton D. (1988), ‘The FDP in the Federal Republic of Germany: The requirements of survival and success’. In: Kirchner E.J. (ed.), Liberal parties in Western Europe. Cambridge, Cambridge University Press, pp. 62–92. Knutsen O. (2008), Class voting in Western Europe. A comparative longitudinal study. Plymouth, MA: Lexington Books. Massart A. (2006), ‘Y-a-t’il un parti libéral en France? ’. In: Delwit P. (ed.), Libéralismes et partis libéraux en Europe. Brussels: l’Université de Bruxelles, pp. 75–92. Millard F. (2001), ‘Party development and the Polish presidential election of October 2000’, Perspectives on European Politics and Society 2(3), pp. 379–400. Müller-Rommel F. (2000), ‘The F.D.P.: Small but beautiful’. In: De Winter L. (ed.), Liberalism and liberal parties in the European Union. Barcelona, Spain: Institut de Ciències Politiques I Socials, pp. 83–118. Musella F. (2014), ‘How personal parties change: party organisation and (in)discipline in Italy (1994–2013)’, Contemporary Italian Politics 6(3), pp. 222–237. Preda C. (2004), ‘La société roumaine entre rupture et continuité. Le système politique roumain après quatre élections’, Transitions 43(1), pp. 127–152.

Liberal parties and elections 309 Roussellier N. (1991), L ‘Europe des libéraux. Brussels: Complexe. Sandström C. (2001), ‘Le parti Européen des libéraux, démocrates et réformateurs. De la coopération à l’intégration’. In: Delwit P., Kulahci E., Van de Walle C. (eds.), Les fédérations Européennes de partis. Organisation et influence. Brussels: l’Université de Bruxelles, pp. 123–140. Spirova M. (2005), ‘Political parties in Bulgaria: Organizational trends in comparative perspective’ Party Politics 11(5), pp. 601–622. Van der Brug W., Hobolt S., de Vreese C.H. (2009), ‘Religion and party choice in Europe’, West European Politics 32(6), pp. 1266–1283. Vandermotten C., Médina Lockhart P. (2002), ‘La géographie du libéralisme Belge et Européen’. In: Delwit P. (ed.), Libéralismes et partis libéraux en Europe, Brussels: l’Université de Bruxelles, pp. 57–74.

16 Governmental participation and alliances of liberal parties in Europe Johan Hellström and Daniel Walther

Introduction As this chapter shows in detail, in most European democracies liberal parties are not only represented in parliament but have also been in office at some point in time. In fact, liberal parties have been unusually successful at getting into government and have been in power to an extent that far exceeds their electoral or parliamentary strength (Kirchner 1988). More specifically, in most countries, they have been in the majority of governments despite typically having a parliamentary seat share of between 5 and 15%. This ability to successfully transform a relatively modest electoral standing into government participation, also means that liberal parties have been unusually successful in obtaining ministerial portfolios. In this respect, liberal parties have often held major portfolios and had opportunities to influence salient policy areas, in particular Finance and Foreign Affairs. Accordingly, in this chapter we investigate the government participation and alliances of liberal parties and specifically try to provide answers for the following questions: How often have liberal parties participated in national government and with which preferred partners? Under which circumstances do liberal parties get into government? And when getting into office, which ministerial portfolios do they usually obtain? A central question is whether this relative success of liberal parties in ensuring political influence stems from particular skills at playing the political game or whether there are more general reasons behind it. Liberal parties are, for example, often centrally located on the ideological spectrum, which opens up for collaborations both to the right and to the left (Laver and Schofield 1990). Moreover, these parties have been active for a long time in most party systems and have therefore had ample time to develop sound relations with other parties in the system. Their long history and experience of government also make them safe(r) choices because other parties (and voters) know what to expect from them (Grotz and Weber 2012). Thus, in this chapter we use theoretical and empirical insights from studies in coalition politics to offer some potential explanations as to why liberal parties

Alliances of liberal parties in Europe 311 have been disproportionately successful at ensuring formal representation in the European democratic machinery. The chapter is organised as follows. We start by providing a comparative examination of the patterns of government participation, or lack thereof, of liberal parties in 28 European democracies from 1945–2015. We then proceed to look at the control of ministries, and thus potential policy impact. For this purpose, we use a new and updated dataset on ministerial portfolio allocation to be able to investigate which kind of portfolios Liberal parties tend to get and how frequently they have obtained portfolios in the European countries under study. The chapter closes by summarizing the findings and discussing what might lie ahead for Europe’s liberal parties.

Liberal parties in government Although liberal parties have on average been successful in getting into power, a closer look at the empirical record shows that they have been better able to get into government in some countries than in others. Figure 16.1 identifies the cases of liberal parties in government, with several parties spending more than one subsequent spell in government. In some countries, liberal parties have never been in government (i.e., Greece and Spain) or Austria Belgium Bulgaria Croatia Czech Republic Denmark Estonia Finland France Germany Greece Hungary Iceland Ireland Italy Latvia Lithuania Luxembourg Netherlands Norway Poland Portugal Romania Slovenia Spain Sweden Switzerland United Kingdom 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

Figure 16.1  Participation of liberal parties in Europe, 1945–2015

312  Johan Hellström and Daniel Walther have only rarely been part of the government (i.e., Austria, Czech Republic, Latvia, and UK). In other countries, they have been more successful, for instance in Denmark, Finland, Germany, and Switzerland where they have been included in a vast majority of the cabinets that existed. Why are liberal parties often so successful in getting into government, even though many of them are not particularly electorally successful? To answer this question, previous theoretical and empirical work that focusses on government participation or “who gets in” is of particular importance (Franklin and Mackie 1984; Warwick 1996; Mattila and Raunio 2004; Döring and Hellström 2013). In this literature a set of factors at both the party and the party system/country levels has been singled out as important determinants of a party’s chance of entering government. One of the most important factors in this respect is electoral performance or, more precisely, the number of seats held by a party in parliament (Döring and Hellström 2013; Mattila and Raunio 2004). Particularly, obtaining majority control in parliament guarantees government participation and at the same time also strongly reduces the incentives to invite coalition partners. For liberal parties this is an exception, as only a few of them have ever become the majority winner (i.e., Bulgarian NDSV in the 2001 elections, the Irish Fianna Fáil in late 1950s to 1970 and the Portuguese PSD in the 1987 and 1991 elections). Generally, liberal parties have instead predominantly been included in coalition governments. However, even in minority situations, parties’ chances of getting into a coalition government or gaining the opportunity to form a minority single-party government depend on their success in the elections. In most countries, the party that gained the highest number of seats (or the largest party) in parliament is given the first chance to form a government. It becomes the formateur in the process of coalition formation (Warwick 1996; Bäck and Dumont 2008). Thus, the largest party is likely to become a government member, but is also likely to get the Prime Ministership. Table 16.1 shows the share of number of cabinets (both single-party and coalition cabinets) that liberal parties participated in, as well as the share of Prime Ministers, their average seat share (in general and when in office), and how frequently they are the median legislator, or in other words, the median legislative party (i.e., the party with the median position that divides the legislature into two halves). The table also shows how often liberal parties are ideologically adjacent to the median legislator. When looking at the liberal parties from the 28 European countries examined in this study, we find that being the largest party does increase the likelihood of becoming a government member. Of all parliamentary elections between 1945–2015, liberal parties were the largest party in about 10% of the elections and became cabinet members in about 85% of those cases. Table 16.1 also indicates that liberal parties do a lot better in getting into government than they do in the parliamentary elections. Comparing seat share and share of cabinets, the table shows that liberal parties have been

Alliances of liberal parties in Europe 313 Table 16.1  Liberal parties in government in 27 European countries, 1945–2015 Share of cabinets Average Share of Median where liberal Average seat share median legislator parties have been Share of seat when in legislator or adjacent represented % PMs % share % office % % % Austria Belgium Bulgaria Croatia Czech Republic Denmark Estonia Finland France Germany Greece Hungary Iceland Ireland Italy Latvia Lithuania Luxembourg Netherlands Norway Poland Portugal Romania Slovenia Spain Sweden Switzerland United Kingdom

6.1 56.2 23.1 66.7 12.5

0.0 10.4 7.7 0.0 0.0

5.7 9.5 15.4 3.9 10.5

6.6 11.6 22.0 4.3 13.7

42.4 0.0 53.8 0.0 0.0

57.6 18.7 69.2 55.6 12.5

69.2 73.3 79.6 38.1 68.0 0.0 36.4 56.2 72.0 52.3 8.7 33.3 42.9 72.7 21.9 19.0 52.0 40.9 81.2 0.0 22.6 100.0 4.0

25.6 46.7 14.8 12.7 0.0 0.0 0.0 28.1 72.0 3.1 4.3 0.0 9.5 9.1 0.0 14.3 44.0 13.6 56.2 0.0 3.2 0.0 0.0

13.1 23.6 5.0 12.8 9.6 0.0 10.9 24.4 35.5 2.6 10.0 13.8 18.1 11.6 6.5 11.1 36.7 15.5 21.8 3.6 19.3 100.0 4.0

13.9 24.6 7.2 12.3 10.1 -8.4 24.5 37.7 2.1 10.0 15.7 21.1 14.2 8.4 9.3 40.3 14.7 22.9 -9.3 22.5 8.8

59.0 13.3 16.7 15.9 0.0 0.0 0.0 96.9 96.0 1.5 17.4 22.2 0.0 3.0 3.1 9.5 32.0 0 43.7 0.0 12.5 10.8 3.0

74.4 66.7 51.8 49.2 60.0 0.0 45.4 100.0 100.0 75.4 17.4 22.2 100.0 39.4 6.2 9.5 96.0 0.0 81.2 30.8 38.7 26.3 48.0

Source: Döring and Manow (2016) Note: The total number of cabinets is 771, and 666 of these were minority situations (i.e., no single party has the majority of seats in parliament). A full list of the liberal parties included in the analysis is in the Appendix.

particularly successful in Switzerland (FDP), where they have been included in every single government. In Slovenia (mostly LDS), Finland, Estonia, the Netherlands, Denmark, Croatia, Austria, Iceland, Belgium, and Italy, liberal parties have been in office in more than half of the cabinets that have ever existed. The reference to liberal parties as ‘small’ or ‘minor’ parties is therefore, for many liberal parties, unfitting considering their frequent government participation and the political influence that this brings. This relative success in securing formal representation in government can also be seen in the share of ministerial portfolios held by liberal parties in

314  Johan Hellström and Daniel Walther 71.5

Share (%) of portfolios

60

40

39.839.1 34.6 26.3

20

23.9 18.6 17.9

16.1

14.6 13.3

13

10.6

9.3 9.2

7 6.3

4.5 3.2 3

1.7 1.1

1

0

0

Ire la Sl nd ov e C ze Po nia ch rtu Re ga pu l bl i I c N e c et la he nd r la G nd Lu erm s xe an m y bo Ro urg m D ania en m Be ark lg iu Fi m nl an Es d to Sw nia ed H en un g Bu ary lg ar ia Ita l Po y la n La d tv N ia or wa G Fra y re at nce Br ita Au in st G r ia re ec e Sp ai n

0

Country

Figure 16.2  Share of portfolio allocation to liberal parties in Europe Source: Woldendorp et al. (2000), Seki and Williams (2014), with own updates

the different countries. Data on this is presented in Figure 16.2. In some countries, liberal parties have been among the most successful parties at securing ministerial positions. Ireland is a prominent example, but liberal parties have also held more than one third of all ministerial positions in Portugal, Slovenia and the Czech Republic. Even in a country such as Germany, where the CDU and SPD have dominated the party system since the World War II, almost one fifth of all ministers have belonged to the liberal FDP. At the other end of the spectrum, we have countries such as Greece and Spain where liberal parties have never held any ministerial positions, or a very low share such as in Austria, France, or the UK. What can then explain liberal parties’ successes in this regard in many European countries? In other words, why do they play such an important role in the formation of governments in some countries? As mentioned above their role as formateurs is certainly an important factor. But probably even more importantly, liberal parties often have a favourable ideological position in the party system (see Chapter 17 in this volume). Thus, party ideology is another factor that needs to be taken into consideration. According to the median legislator or median party theorem, coalition formation is assumed to take place in a one-dimensional political space, and the median legislator should have a high likelihood of becoming a member of a cabinet (De Swaan and Rapoport 1973; Laver and Schofield 1990: 110–113). This is also the case for parties that are ideologically adjacent or proximate to the median legislator, which is often the case with liberal

Alliances of liberal parties in Europe 315 parties. Given that parties have policies that they want to implement, they will try to find similar ideological partners. Parties that are positioned close to the median party would be more likely to become government members. As governments are typically formed by and of political parties (but less so in semi-presidential regimes), it involves bargaining between political parties with varying policy priorities and goals. Nonetheless, depending on the particular party in question, they are more or less inclined to accept ideological compromises. Whereas fringe left- and right-wing parties are less willing to make policy concessions, centrist and moderate parties can make less costly ideological compromises to be included in a coalition. This happens not least because ideological compromises are easier to tolerate the closer a party is to the likely outcome of a jointly agreed policy position. In a coalition government, a policy compromise is likely to be close to the preferred position of the median legislative party. As seen in Table 16.1, the fact that liberal parties often hold the median legislator themselves or are ideologically close to the median legislator probably helps explain why liberal parties have a disproportionate success in entering governments despite their relatively small share of the parliamentary seats. When cabinets are formed, liberal parties are also rather successful in becoming the head of governments. About 13% of all cabinets considered in this study were controlled by liberal Prime Ministers (PMs). But being the largest party only accounts for 57 of the 95 cases. Other factors are also likely to play a role, such as being the median party, or having a liberal head of state that can nominate the PM from within his or her own party, or is free to appoint the formateur (Glasgow et al. 2011). This goes beyond what we can investigate in this chapter. The centrality of liberal parties is certainly important in this respect, but getting the opportunity to form a single-party government is a guarantee. Thus, a related question concerns what type of governments liberal parties tend to participate in. And in particular, when they are able to form governments without including any other parliamentary parties. When in cabinet, as indicated above, liberal parties generally participate in coalition cabinets. However, Ireland and Denmark stand out in this respect, as Fianna Fáil ruled alone in half of the cabinets they participated in and Venstre in about one of five cabinets. To illustrate this, Table 16.2 shows all single-party liberal governments and their seat share in parliament. In Table 16.2 the three Danish, Portuguese and Swedish parties stand out as having quite low parliamentary support, but still being able to form a government without including any other parties. These governments have two features in common. Firstly, in none of these countries must an incoming cabinet (or Prime Minister) win a parliamentary majority vote to assume power (i.e., ‘positive parliamentarism’), making the formation of minority cabinets more likely (Bergman 1993; Bergman 1995). In Sweden and Portugal, a new cabinet does not have to win such a parliamentary majority vote, but simply has to pass a negative vote in the parliament

316  Johan Hellström and Daniel Walther Table 16.2  Single-party governments, 1945–2015 Time in office Denmark 1945 Nov. – 1947 Oct. 1973 Dec. – 1975 Jan. 2015 Jun. – Estonia 1996 Dec. – 1997 Feb.

Ireland Portugal

Spain Sweden

Cabinet name

Seat share %

Kristensen Hartling Rasmussen L II Vahi III

25.5 12.3 19.0 40.6

Party

Venstre Venstre Venstre Eesti Koonderakond 40.6 Eesti Koonderakond 46.9–56.8 Fianna Fáil

1997 Mar. – 1999 Siimann Mar. 1951 – 1989 (11 of 17 Valera VIII-IX, cabinets) Lemass I-III, Lynch I-III, Haughey I-III 1978 Nov. – 1979 Jun. Mota Pinto 27.8 Partido Social (caretaker cabinet) Democrata 1985 Nov. – 1995 Oct. Silva I, II and III 35.2–59.2 Partido Social Democrata 1977 Jul. – 1982 Oct. Suarez I, II and 47.4–48.0 Unión de Centro Calvo-Sotelo Democrático 1978 Oct. – 1979 Sep. Ullsten 11.2 Folkpartiet

Source: Andersson et al. (2014) with own updates

(i.e., not having an absolute majority voting against it). Danish governments, on the other hand, can be formed without any vote at all as the country lacks formal inauguration rules. Secondly, all these cases have in common the absence of a viable and strong potential alternative government to the ones that were formed, mainly due to divisions between otherwise ideologically close parliamentary parties (and potential coalition partners). A good example of this is the Swedish single-party minority cabinet led by the Liberal Party leader Ullsten. The Ullsten cabinet was one of Europe’s “weakest”, with a parliamentary support of only about 11%. The government came to power in October 1978, a little more than two years after the parliamentary elections in 1976. It succeeded the previous right-wing coalition government between the Centre Party, the Liberal Party, and the Conservatives that resigned because of disagreement on further expansions of nuclear plants in Sweden (Petersson 1979). When the Prime Minister Fälldin from the Centre Party handed over his resignation, this decision was also supported by the Conservative Party leader, as the latter expected a two-party government including the Conservatives and the Liberal Party to take over. However, Ullsten chose to present a single-party Liberal government, not least because the party leadership felt that they had more in common with the Social Democrats than the Conservatives. The Liberal Party had held discussions with the Conservatives to form government, but has also held parallel discussions with the Social Democrats to form

Alliances of liberal parties in Europe 317 Christian Democrats/ Conservatives Social Democrats

Coalition partners

Agrarians Communists/ Left socialists Others Greens Liberals (smaller parties) Radical right/ Populist radical right 0

10

20

30

40

50

Percent

Figure 16.3  Coalition partners of liberal parties in Europe

a single-party government themselves. Also, the parliamentary seat share balance gave the Liberal government the choice to achieve a parliamentary majority with both the Social Democrats and, alternatively, with the Conservatives and the Centre Party. At the vote of investiture, the Social Democrats abstained making a single-party Liberal government possible (Bergstrand 2010; Petersson 1979). Turning to the question of the partners with whom liberal parties generally form government, Figure 16.3 shows the distribution of coalition partners based on party family. The figure shows that liberal parties are usually included in centre-left or centre-right coalitions. The most frequent coalition partners are Christian Democratic/Conservative parties, followed by Social Democratic and Agrarian parties. There is a slight tendency for liberal parties to form or be invited in right-wing governments over more left-wing governments, but the difference is not very pronounced. As party ideology matters for whom parties choose to invite for negotiations or to form governments with, it should come as no surprise that liberal parties, which are ideologically positioned near the centre or the centre-right, cooperate in governments with parties with similar positions. Liberal parties that focus more on social liberalism should see cooperation with Social Democratic parties as more viable, whereas liberal parties that put a greater emphasis on economic liberalism should prefer right-wing coalition partners.

318  Johan Hellström and Daniel Walther

Portfolio allocation and liberal parties As we saw in Figure 16.2, liberal parties have held a sizeable chunk of ministerial positions in most countries, even though the precise number differs quite significantly across countries in our data. Portfolio allocation is one of those rare cases in political science where a single mechanism, namely the principle of proportionality, can explain a lot of what is going on. Already in the 1970s, Browne and Franklin (1973) demonstrated that government parties tend to receive a share of portfolios that is roughly proportional to the share of seats they control in parliament. That is, if a particular coalition member has control over 15% of the government’s seats in parliament, it also tends to get about 15% of the available portfolios (Bäck et al. 2009). However, not all ministerial portfolios are of equal importance. A few ministries tend to be highly desired by most parties (such as Finance or Foreign Affairs), whereas others tend to be more important to particular parties depending on issues that they emphasise and prioritise. For example, the portfolio of Agriculture has traditionally been more important to agrarian parties, as has the portfolio of Labour been to Social Democratic parties. When it comes to liberal parties and portfolio allocation, the question is then first if their central ideological position in the party system (and the increased bargaining power that comes with that position) gives them access to more highly desired portfolios. And second, when given the choice, which portfolios do liberal parties tend to prioritise? Expectations for portfolio allocation Previous research has suggested that liberal parties have actually found it difficult to use their central ideological position or increased bargaining power to get qualitatively more important ministries. Warwick and Druckman (2006) looked at whether the resources that a party brings to the coalition are not only its legislative size, but also its bargaining power as measured by the Shapley-Shubik (1954) index or the Banzhaf index (Banzhaf 1964). In this context, Warwick and Druckman (2006) relied on country experts to assign qualitative values to the different portfolios to have a measure of which ministries were most highly valued within each country. They then tested whether parties with higher bargaining power were more likely to obtain a share of the most valuable portfolios. As it turns out, refining the analysis in this way to account for other aspects of party bargaining power than their share of parliamentary seats does not improve our ability to predict portfolio allocation. The general rule of a proportional relationship between legislative seats and portfolios therefore seems to apply to liberal parties as well. The central ideological position of liberal parties thus makes them more likely to end up in government, but it does not help them obtain quantitatively more or qualitatively better posts than the other parties in the government. The main

Alliances of liberal parties in Europe 319 reason why liberal parties get more ministerial posts than other parties of equal size is that they are more likely to end up in government in the first place. Clearly, parties that get a seat at the table also obtain more portfolios than those that do not. However, once in government, liberal parties do not get a greater number of portfolios than what the law of proportionality would suggest. Turning our attention to the preferences of liberal parties in the bargaining over ministerial portfolios – which portfolios are they likely to be the most interested in obtaining? Kirchner (1988) argued that they are mostly interested in portfolios related to the defence of individual rights (such as Justice or Home Affairs) or related to economic management, such as Finance or Trade (also see Budge and Keman 1990), as these portfolios are in line with the classic liberal ideological profile. Issues related to law and order, as well as market-liberal and centre-right economic policies, are important aspects of the policy platforms of most European liberal parties. It thus makes sense that they would want to work actively on these questions while in power. Additionally, many liberal parties focus on international cooperation (including EU integration and migration) and sociocultural issues connected to individual and social rights, such as, for example, marriage equality, and abortion. As liberal parties have increasingly emphasised these issues to distinguish themselves from the crowded centre-right spectrum of the left-right continuum, it seems likely that they have also been interested in securing these portfolios. Dividing the pie: which ministerial portfolios do liberal parties get? To examine whether liberal parties actually obtain the portfolios that match their ideological profile, we use a dataset on ministerial positions in Europe that covers 26 countries between 1945 and 2015. For the governments that were in office up until 1998 we use the dataset created by Woldendorp et al. (2000) and for the years 1999 to 2012 we use the data collected by Seki and Williams (2014). For the years 2012–2015 we have updated the data. In total, the updated dataset contains information on more than 16,000 ministerial positions, out of which liberal parties held about 1,900. Figure 16.4 shows the distribution of portfolios among the total of 1,944 ministerial positions held by liberal parties. The figure shows the 17 most popular portfolios and these account for approximately 65% of all positions held by liberal parties. The other 35% (not reported here) are made up of less frequently held positions that mainly tend to be smaller or country specific portfolios that lack clear cross-country equivalents. In Figure 16.4, we can see that the most popular positions tend to be in line with what we could expect based on liberal parties’ ideological priorities. The most common portfolio has been Justice, which was held by liberal politicians on 119 occasions, or 6.1% of all the portfolios held by liberal parties. Given the strong liberal commitment to the rule of law and due process, and the fact that the

320  Johan Hellström and Daniel Walther 6.1% 5.5% 5.4% 100

5.1% 4.7% 4.3% 4.2%

Count

75

3.9%

3.7% 3.5% 3.5% 3.1% 3.1% 3%

50

2.1% 1.6%

1.4%

25

Ju st ic e Fi Fo na re nc ig e n Af fa irs D Pu efe n bl ce i Pr c W or im ks e M in is te In r te Ed rior u So cat io ci n al D Af ep f ai ut rs y Pr H ea im l e M th in Ag iste r In ricu du ltu s r e tr Ec on y/Tr om ad e ic En Affa i vi ro rs nm en H ou t si ng La bo ur

0

Portfolio

Figure 16.4  Portfolios of liberal parties in Europe Note: Overview of the portfolios that liberal parties most frequently hold. The percentage is the share of times that portfolio has been held out of all portfolios held by liberal parties.

Justice portfolio is usually not the most sought after, this is not a surprising finding. The second and third most common ministerial positions are Finance and Foreign Affairs. As liberal parties are generally interested in economic management and are committed to international issues and global cooperation, it is not surprising that these positions are also actively bargained for in the government formation process. What is more surprising is that liberal parties have so frequently been able to obtain two positions with such high saliency despite the fact that they tend to be one of the smaller coalition partners. This is particularly interesting as it goes against Warwick and Druckman (2006)’s finding that even qualitatively more important portfolios tend to be distributed according to the law of proportionality. This suggests that liberal parties place particular weight on these two posts and might therefore be willing to accept other, far less salient portfolios as a trade-off to ensure that they do not get a qualitatively disproportionate share overall. Moreover, this finding cannot be explained by country differences. The position of Minister of Foreign Affairs, for example, was monopolised by a liberal party (FDP) in Germany, where it held the position in every government for 29 consecutive years between 1969 and 1998. However, liberal parties have been successful in gaining the portfolio in almost every country in

Alliances of liberal parties in Europe 321 our dataset including Eastern Europe. In Slovenia, for example, the liberal LDS has held the position in three governments since 1996. The fact that two high-saliency positions have been repeatedly secured by liberal parties in many different countries suggests that they tend to put significant weight on them in coalition negotiations. Further down the list in Figure 16.4 we also find portfolios that are less associated with the classic liberal profile. Defence, for example, the fourthmost common portfolio is usually of greater importance to conservative or nationalist parties. Liberal parties have nonetheless managed to secure that position 100 times (or 5.1% of all portfolios held liberal parties). However, in general, if we look at the 17 most common portfolios depicted in Figure 16.4, most of them are in line with what we would expect on ideological grounds: Home Affairs, Industry and Trade, or Economic Affairs. What stands out is that a number of portfolios less associated with the liberal ideology, such as Public works, Social Affairs, or Health, also score relatively high. These findings taken together seem to suggest that liberal parties often get some of their preferred portfolios (even high-saliency ones such as Finance and Foreign Affairs), but generally being smaller coalition partners, they frequently also accept positions that are less closely aligned with their core ideological profile.

Conclusion One of the characteristics of many liberal parties in Europe is the significant difference between their extensive government participation and their average electoral and parliamentary strength. Nonetheless, as discussed in this chapter, liberal parties occupy a special position in many party systems in Europe. In particular, their central ideological position and their pragmatic relationship with parties on both sides of the political spectrum have ensured that they can play a significant role in the formal democratic processes despite their often modest electoral size. However, although liberal parties have been unusually successful at getting into government in many countries, this does not imply that once in government, they necessarily get more or more important ministerial portfolios than their electoral size would allow (Warwick and Druckman 2006). They do, however, have particular preferences about which portfolios they want to take responsible for. In particular, Justice, Finance, and Foreign Affairs seem to be highly valued. There are, however, considerable between-country differences in liberal parties’ abilities to uphold the pivotal role in government formation. Even in some countries where government incumbency has been the rule rather than the exception, liberal parties are starting to struggle. Most notably, FPD in Germany, which has been in the vast majority of German governments since World War II, failed to pass the parliamentary threshold in 2013 and faces an uncertain position as to its future. Similarly, the Swedish

322  Johan Hellström and Daniel Walther Folkpartiet (called Liberalerna since 2015) received only 5.42% in the 2014 election and 5.49% in the latest election in 2018, its second- and third-worst results since the party was founded in 1934. On the other hand, liberal parties seem to be making strides in Eastern Europe and have been successful in gaining both parliamentary seats and office in all but a few of the volatile democratic party systems in the region. However, it is still reasonable to suggest that liberal parties are likely to continue to have an above average success rate at getting into government. As long as the parties do not deviate too much from the centre of the ideological landscape, their pivotal role in forming coalitions is likely to continue to earn them a seat in government, provided of course that they manage to secure parliamentary representation. In that regard, a complicating factor is the increasing importance of identity politics and the second dimension in politics (Hooghe et al. 2002), the so-called Green/alternative/libertarian (GAL) – traditional/authoritarian/ nationalist (TAN) dimension, which could present an obstacle for liberal parties in the future. Although they are at present and in most systems, located at the centre on left-right issues, they do not necessarily enjoy the same centrality on issues related to ‘new politics’ or cultural issues. The growing importance of issues associated with this dimension could therefore mean that liberal parties need to adapt and change to retain their pivotal role in government formation processes and their place in future governments.

Appendix In the chapter, we examine a rather long time-period, from 1945 until 2015. Thus, we need to consider that some parties might have developed into liberal parties over time, whereas others experienced such fundamental ideological changes that they can no longer be considered as such. To account for temporal changes in parties’ ideological profiles in this chapter, we have limited the time period for which we include some of the parties and have also included some additional parties not covered in the other chapters. More specifically, we included the Austrian FPÖ, but only for a shorter time-period. FPÖ with its roots in national liberalism, long considered as the party of old Nazis, was founded in 1956. Over time this party gradually developed to become what, at least on the surface, could be classified as a liberal party. In 1979 it joined the Liberal International and in 1983 the party participated in its first government. Thus, we include FPÖ between 1983 until 1986 before the party started the reverse ideological turn towards right-wing populism under the party leader Jörg Haider, even though the party was a member in Liberal International until 1993 (Luther, 1998). In addition, the former Agrarian Finish Centre party (KESK) and Swedish Centre Party (C) was coded liberal from the year 1992 to 1995, respectively. That is, when they

Alliances of liberal parties in Europe 323 Table 16.3  Appendix. Parties included in the comparative analysis Political parties included Austria Belgium Bulgaria Croatia Czech Republic Denmark Estonia Finland France Germany Greece Hungary Iceland Ireland Italy Latvia Lithuania Luxembourg Netherlands Norway Poland Portugal Romania Slovenia Spain Sweden Switzerland United Kingdom

FPÖ 1979–1986; NEOS; LIF PVV/PLP with its predecessors and successors (i.e., PL, LP-PL, PRL, PVV, VLD) NDSV, DPS HNS, HSLS, IDS-DD, LS A02011, LIDEM/VIZE 2014, US-DEU RV, V, I EK, ER KESK from 1992, Ålandic Centre, SFP, L-LKP Most of UDF until 1994. FDP Drassi SZDSZ, MLP FSF, BF FF, PD PRI,PLI, IdV, ApI, PR-R LPP/LC, LA DP, LiCS, LLS, LRLS DP VVD, D66 V DP, RP, KLD, UD, UW PLD, PDR, PSD, MPT PNL LDS, ZaAB, SMC, DL CDC,C’s, UPyD, CDS C after 1994, FP/L FDP, LPS, LdU-ADI LibDem

Source: Andersson et al. (2014); Döring and Manow (2016) Note: The parties in italics have “disappeared” or lost their relevance. They are included in longitudinal analysis, but are not relevant anymore today.

become affiliated with or members of the ELDR. Finally, for France we include the parties, in the federation of parties of UDF (Frears, 1988), that was members of ELDR (and its predecessors) until they left the organisation in 1994. Table 16.3 lists all parties that we examine in the chapter.

References Andersson S., Bergman T., Ersson S. (2014), The European Representative Democracy Data Archive, Release 3. Main sponsor: Riksbankens Jubileumsfond (In 20070149:1-E). Retrieved from https://www.erdda.se Bäck H., Dumont P. (2008), ‘Making the first move’, Public Choice, 135(3–4), pp. 353–373.

324  Johan Hellström and Daniel Walther Bäck H., Meier H. E., Persson T. (2009), ‘Party size and portfolio payoffs: The proportional allocation of ministerial posts in coalition governments’, Journal of Legislative Studies, 15(1), pp. 10–34. Banzhaf J. F. (1965), ‘Weighted voting doesn’t work: A mathematical analysis’, Rutgers Law Review, 19(2), pp. 317–343. Bergman T. (1993), ‘Formation rules and minority governments’, European Journal of Political Research, 23(1), pp. 55–66. Bergman T. (1995), ‘Constitutional rules and party goals in coalition formation: An analysis of winning minority governments in Sweden’, Umea, Umea University, Department of Political Science. Bergstrand M. (2010), ‘Ola Ullsten’. In Ohlsson M. B. P. T. (ed.), Sveriges statsministrar under 100 år. Stockholm: Bonniers. Browne E. C., Franklin M. N. (1973), ‘Aspects of coalition payoffs in European parliamentary democracies’, American Political Science Review, 67(2), pp. 453–469. Budge I., Keman H. (1990), Parties and democracies. Coalition formation and government functioning in 20 states. Oxford: Oxford University Press. De Swaan A. (1973), Coalition theories and cabinet formations: A study of formal theories of coalition formation applied to nine European parliaments after 1918, Amsterdam: Elsevier Scientific. Döring H., Hellström J. (2013), ‘Who gets into government? Coalition formation in European democracies’, West European Politics, 36(4), pp. 1–21. Döring H., Manow P. (2016), Parliaments and governments database (ParlGov): Information on parties, elections and cabinets in modern democracies. Development version. Retrieved from http://www.parlgov.org/ Franklin M. N., Mackie T. T. (1984), ‘Reassessing the importance of size and ideology for the formation of governing coalitions in parliamentary democracies’, American Journal of Political Science, 28(11), pp. 671–692. Frears J. (1988), ‘Liberalism in France’. In Kirchner E. J. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 124–150. Glasgow G., Golder M., Golder S. N. (2011), ‘Who “wins”? Determining the party of the Prime Minister’, American Journal of Political Science, 55(4), pp. 937–954. Grotz F., Weber T. (2012), ‘Party systems and government stability in Central and Eastern Europe’, World Politics, 64(4), pp. 699–740. Hooghe L., Marks G., Wilson C. J. (2002), ‘Does left/right structure party positions on European integration?’ Comparative Political Studies, 35(8), pp. 965–989. Kirchner E. J. (1988), ‘Western European liberal parties: Development since 1945 and prospects for the future’. In Kirchner E. J. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 470–496. Laver M., Schofield N. (1990), Multiparty governments: The politics of coalition in Europe. Oxford: Oxford University Press. Luther K. R. (1998), ‘The Freiheitliche Partei Österreichs: Protest party or governing party?’ In Kirchner E. J. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 151–177. Mattila M., Raunio T. (2004), ‘Does winning pay? Electoral success and government formation in 15 West European countries’, European Journal of Political Research, 43(2), pp. 263–285. Petersson O. (1979), ‘The government crisis in Sweden1’, Scandinavian Political Studies, 2(2), pp. 171–178.

Alliances of liberal parties in Europe 325 Seki K., Williams L. K. (2014), ‘Updating the party government data set’, Electoral Studies, 34, pp. 270–279. Shapley L. S., Shubik M. (1954), ‘A method for evaluating the distribution of power in a committee system’, American Political Science Review, 48(3), pp. 787–792. Warwick P. V. (1996), ‘Coalition government membership in West European Parliamentary Democracies’, British Journal of Political Science, 26(4), pp. 471–499. Warwick P. V., Druckman J. N. (2006), ‘The portfolio allocation paradox: An investigation into the nature of a very strong but puzzling relationship’, European Journal of Political Research, 45(4), pp. 635–665. Woldendorp J. J., Keman H., Budge I. (2000), Party government in forty-eight democracies (1945–1998), Dordrecht, The Netherlands: Springer.

17 The liberal party family ideology Distinct, but diverse Caroline Close

Introduction The objective of this comparative chapter is to assess the existence of a liberal party family on the basis of the parties’ ideological or policy orientation, which constitutes one of Mair and Mudde’s (1998) main approaches to delineate party families. The chapter departs from the criteria of transnational affiliation (belonging to the ALDE or former ELDR), and assesses the ideological distinctiveness and homogeneity of this group of parties (Elff 2013; Freire and Tsatsanis 2015)1. In that way, this chapter aims to answer the question: Can we talk about a Liberal party family when looking at the ideological criteria? The chapter examines the distinctiveness and homogeneity of the liberal party family cross-sectionally and longitudinally, by using data from the Chapel Hill Expert Survey (CHES) and from the Comparative Manifesto Project (CMP). Using data on parties’ position on a variety of policy issues, the chapter estimates their placement on the two main ideological dimensions that have structured most party systems in Europe: the economic dimension, and the cultural dimension. Both the cross-sectional and longitudinal analyses proceed in two steps: First, the analysis evaluates the extent to which liberal parties have a distinct ideological placement compared with the other main European families; second, the analysis examines the degree of homogeneity of the family. This strategy reveals that liberal parties have had and still display a distinct ideology that combines a marked right-wing position on economic issues with centre-left preferences on cultural matters. It also points out the great diversity among liberal parties and examines the existence of three distinct enduring liberal traditions within the liberal ideology: the classical, the social, and the conservative traditions.

The liberal party family ideology in comparative perspective In terms of ideological placement, the position of the liberal party family in the left-right space has been described as ‘ambivalent’ (De Winter and Marcet 2000; Smith 1988). Liberal parties tend to have a rather central

The liberal party family ideology 327 position on the general left-right scale, being more rightist than the SocialDemocrats, Greens, and radical left; but more leftist than the Christian Democrats, Conservatives, and radical right (Camia and Caramani 2012). This ‘between left and right’ (Smith 1988) position is translated into a differentiated positioning of the liberal parties on the economic and cultural dimensions: They tend to promote right-wing economic policies (lower taxes, free-market economy, etc.) while being on the centre-left or progressive side of sociocultural policies (e.g., permissive or libertarian positions on moral issues such as euthanasia, same sex marriage; as well quite positive stances on multiculturalism) (Camia and Caramani 2012; Elff 2013; Ennser 2012). This ambivalent position illustrates well Smith’s statement that ‘above all else . . . liberal parties maintain their connections with the two original tenets of liberal ideology, the two freedoms: economic freedom and the liberties of the individual’ (Smith 1988: 19). However, the picture is slightly more complex. Comparative researches on party families’ ideology tend to emphasise the heterogeneity of the liberal party family in terms of ideology, which echoes the lack of distinctiveness of the liberal electorate (see Chapter 15, this volume). Based on expert survey data collected across Western Europe, Ennser (2012) finds that the liberal party family is the least homogeneous of all families (see also Smith 1988: 16); a finding that leads him to question ‘to what extent there is a common ideological core in the liberal party family’ (Ennser 2012: 167). Using comparative attitudinal data of party candidates across Western European countries, Freire and Tsatsanis (2015) arrive at a similar conclusion that ‘the Liberal party family is by far the most heterogeneous of all the European party families’ (Freire and Tsatsanis 2015: 17–18) on both the economic and cultural dimensions. Camia and Caramani’s (2012) research brings another perspective. The authors examine the homogeneity of party families in Europe through a longitudinal perspective (1945–2009) using longitudinal data collected in the frame of the Comparative Manifesto Project, across Western and Central and Eastern Europe (CEE). Over the whole time period, liberal parties do appear less homogeneous than most families, but not more than the Conservatives and the radical right (Camia and Caramani 2012: 61). The authors also show that the degree of homogeneity of the liberal party family on the two economic and cultural dimensions has barely varied over time; and that liberal parties are not more heterogeneous on either dimension. Their results therefore slightly counterbalance those of Freire and Tsatsanis (2015) and Ennser (2012); as the liberal party family does not appear particularly more heterogeneous than other party families. Beyond examining the general degree of homogeneity of the liberal party family, comparative research also highlights the existence of distinct ‘groups’ or ‘traditions’ within the liberal ideology. In his foundational volume, von Beyme (1985) arguably distinguishes between three subcategories: the Radical Liberals, the Conservative Liberals, and the agrarian parties.

328  Caroline Close More recently, Ennser (2012) identified two groups: on the one hand, the Conservative Liberals, which are ideologically close to the Christian Democrats and the Conservatives and include for instance the German FDP and Irish Progressive Democrats, as well as the Agrarian Liberals; on the other hand, the Social Liberals, which are close to the Social Democratic parties and include, for instance, the Danish Radikale Venstre and the Dutch D66 (Ennser 2012: 163). Camia and Caramani (2012) also identify two distinct groups, although slightly different from Ennser’s categories: the Social Liberals, which have on average a centrist position on the economic axis but a left-wing placement on the cultural dimension; and what they call Liberals, which are centrist on the cultural dimension but right-wing on the economic axis. Although Freire and Tsatsanis (2015) do not identify specific groups within the liberal family, they develop the argument that the regional or national context in which parties emerge in fact undermines the ideological distinctiveness and cohesion of European party families. In their own words, ‘region appears to often trump party family’ (Freire and Tsatsanis 2015: 15). In the case of liberal parties, they observe that depending on the countries, liberal parties appear more to the left or to the right on the economic dimension than the Conservatives. In a similar extent, they show that liberal parties’ dispersion on the cultural dimension (defined in terms of a libertarian/authoritarian scale) depends on the religious cultural context, with, for instance, Catholic regions producing more conservative liberal parties, and Scandinavian countries producing more progressive liberal parties (Freire and Tsatsanis 2015: 17–18). Hence, they suggest that context might produce different types of liberal parties. On the whole, these existing studies seem to disagree on the existence or not of a distinct liberal party family ideology; and scholars’ answers to this question seems to have rested on their perception of the liberal ideology’s relative degree of heterogeneity. The present chapter aims to contribute to these debates. First, by using more recent data on liberal parties’ policy orientations, it provides a modest update. Second, it extends the geographical scope by including data from CEE countries, in that way allowing examining the impact of CEE context on the liberal party family’s ideological orientations and homogeneity. Third, it combines two types of data on parties’ policy orientation: expert survey and manifesto data. This is crucial, as researches using different sources and approaches to assess party families’ ideological distinctiveness and homogeneity seem to have reached different conclusions (see above). Fourth, the chapter provides both a cross-sectional and a longitudinal perspectives. Fifth, the analysis examines more systematically the existence of different groups or traditions within the liberal party family ideology. More importantly, it doesn’t proceed a priori but inductively: Parties are not classified into preexisting categories (Radicals, Conservatives, etc.) on the basis of their origins, but latent categories are identified on the basis of the parties’

The liberal party family ideology 329 placement in the two-dimensional ideological space, through clustering analyses (see Ennser 2012, and Chapter 15 in this volume).

Data and method The Chapel Hill Expert Survey (CHES) and the Manifesto Project As stated by Mair and Mudde (1998: 220) in their founding article, party ideology ‘is difficult to define and to specify with any precision’. As argued by the authors, the concept taps much into what parties are rather than to what parties do, and in that sense, party ideology greatly appeals to the party’s origins. Examining party ideology across a large number of cases would ‘require a comprehensive approach’ and would therefore ‘prove to be highly time-consuming and conceptually demanding’ (Mair and Mudde 1998: 219–220). Given these limitations, this comparative chapter rather relies on a policy-based approach and uses parties’ policy orientations as a proxy for parties’ ideology. A variety of sources can be used to assess parties’ policy orientations: party programmes or manifestos (Budge et al. 2001; Elff 2013; Klingemann et al. 2006, expert surveys (Benoit and Laver 2006; Ennser 2012), policy statements or campaign issues reported in the media (Kriesi et al. 2006), mass opinion data or elites surveys (Camia and Caramani 2012; Freire and Tsatsanis 2015; Leimbruger et al. 2010). To assess the distinctiveness and coherence of the liberal party family ideology, this chapter mainly uses two of these sources: expert surveys and manifesto data. First, I use the last round of data collected in the frame of the Chapel Hill Expert Survey (2014) (Bakker et al. 2015). As explained on their website2, ‘the Chapel Hill expert surveys [CHES] estimate party positioning on European integration, ideology and policy issues for national parties in a variety of European countries’. These data are used to assess the ideological distinctiveness and homogeneity of the liberal party family through a cross-sectional perspective. Although the Chapel Hill project has collected data on parties’ position over time (every four years since 1999), I do not use it in a longitudinal perspective, for two reasons. First, the number of countries (and parties) included in the survey greatly varies from one round to the other (from 16 countries and 143 parties in 1999, to 31 countries and 268 parties in 2014). Second, the content of the questionnaire has also evolved quite substantially: Although originally aimed at measuring parties’ position on European integration and EU policies, the survey subsequently developed a broader set of questions on non-EU policy issues. Second, I mobilise data collected in the frame of the Manifesto Project (Volkens et al. 2015a). The project is based on a content analysis of electoral party manifestos, and ‘covers over 1000 parties from 1945 until today in over 50 countries’3. It therefore allows conducting a longitudinal study of the liberal party family’s ideology. As put by Mair and Mudde (1998: 218),

330  Caroline Close one of the advantages of this type of data compared with expert survey is that parties’ ideology or policy position is ‘derived from what the parties themselves [chose] to emphasise in their election programs rather than from some position that [has] to be imputed to them by an external researcher’. In that way, manifesto data also overcome the problematic methodological assumption underlying cross-national expert surveys ‘that the same policy means the same thing’ (Mair and Mudde 1998: 218) in different political systems. Yet, manifesto data can also be problematic (see, for instance, Dinas and Gemenis 2010). First, one can reasonably question whether the positions published in party manifestos truly reflect the parties’ ideological identity (Mair and Mudde 1998: 219), or if they are the product of the specific context of electoral competition. Second, the leading scholars of the project themselves acknowledge that their data mostly captures the saliency of policy issues in manifestos rather than parties’ position on these issues (Benoit and Laver 2006: 99–101; Budge 2001). Finally, as highlighted by Freire and Tsatsanis (2015: 3), both types of data ‘overstate a unitary nature of political parties’, and in that way ‘tend to overstate the distinctiveness and the cohesion of the party families’ (Freire and Tsatsanis 2015: 3). Nevertheless, the databases used in this chapter provide extensive and resourceful information on liberal parties’ policy orientations, across countries and time. Case selection Before running any analysis, the content of the two databases have been examined closely. The ‘party family’ variable has been harmonised across the databases and adapted to fit the classification adopted in this edited volume based on the transnational affiliation of parties. Table 17.1 below lists the liberal parties identified in each database. For the Manifesto Project, the election years for which data is available are also given. Note that while the units of analysis in the CHES are the parties at the year for which the data was collected (in this case, 2014), the units of analysis for the Manifesto Project are the parties at each election. Table 17.2 displays the distribution of cases across party families in the two datasets that were recoded for the purpose of this chapter. The 2014 CHES dataset includes 221 parties, among which 34 liberal parties. The Manifesto Project includes 2,325 cases (parties*election); 20% of which are coded as liberal. Parties are classified in eight families plus an ‘other’ category that includes single-issue parties, ephemeral parties built around one candidate or leader, Agrarian parties that did not join the ALDE, etc. Note that the comparisons between the liberal and other families focus on the seven most representative families (radical left, Green, Social Democrat, Liberal, Christian Democrat, Conservative, and radical right), thus excluding the regionalist family – mostly because it displays extremely diverse policy positions, except on the decentralisation issue.

The liberal party family ideology 331 Table 17.1  Case selection Country

Party abbreviation

CHES 2014

Austria

NEOS LIF OpenVLD MR NDSV DPS HNS HSLS IDS-DD LS AN02011 LIDEM/VIZE 2014 US-DEU RV V LA EK ER ESE - EDP-LEE KESK Ålandic Centre SFP L-LKP FDP Drassi SZDSZ MLP FSF BF FF PD IdV ApI PR-R LPP/LC LA DP (LiCS)* LLS LRLS DP VVD D66 V DP RP KLD UD UW

Yes / Yes Yes / Yes Yes Yes Yes / Yes / / Yes Yes Yes Yes Yes / Yes / Yes / Yes / / / / / Yes / / / / / / Yes / / Yes Yes Yes Yes Yes / Yes / / /

Belgium Bulgaria Croatia

Czech Rep. Denmark Estonia Finland

Germany Greece Hungary Iceland Ireland Italy Latvia Lithuania

Luxembourg Netherlands Norway Poland

Manifesto Project (years covered) / 1994-1995 1971-2010 1971-2010 2001-2005 1991-2013 1992-2007 1992-2003 1992-2007 2000 2013 / / 1945-2011 1945-2011 2007-2011 1995-2015 1995-2015 / 1966-2011 / 1945-2011 1966-1991 1949-2013 / 1990-2006 / 1946-2013 2013 1948-2011 1987-2007 2001-2008 / 1946-1992 1993-2006 (2010)* / 2004-2012 (1996) 2004-2008 / 2008-2012 1959-2013 1946-2012 1967-2012 1945-2009 / 2011 1991 1993 1997 (Continued)

332  Caroline Close Table 17.1  Case selection (Continued) Country

Party abbreviation

CHES 2014

Manifesto Project (years covered)

Portugal

PDR PSD MPT PNL LDS ZaAB SMC DL CDC C's UPyD CDS FP C FDP LPS LdU-ADI LibDem

/ Yes Yes Yes / Yes Yes / / Yes Yes / Yes Yes Yes / / Yes

/ 1979-2011 / 1990-2008 1990-2008 / / / / 2015 2011 1982-1989 1944-2010 1958-2010 1947-2011 1991-2003 1947-1999 1945-2015

Romania Slovenia

Spain

Sweden Switzerland UK

Sources: CHES 2014 (Bakker et al. 2015); Manifesto Project (Volkens et al. 2015a) * For LPP/LC, Data is available for the period 1993–2006, and for 2010 For LiCS, data is available in 1996, and for the period 2004–2008

Measuring the positions of liberal parties in a two-dimensional space For the sake of simplicity and comparability, this chapter focuses on liberal parties’ positions on the two central ideological dimensions that have structured most European party systems in the past decades (Kriesi et al. 2006: 923): the traditional left-right socioeconomic dimension and Table 17.2  Party families: Frequency distribution CHES 2014

Radical left Green Social-Democrat Regionalist Other Liberal Christian-Democrat Conservative Radical right Total

Manifesto Project 1945–2015

N parties

%

N parties*election

19 22 35 15 23 34 17 39 17 221

8.6 10.0 15.8 6.8 10.4 15.4 7.7 17.6 7.7 100.0

259 107 433 143 199 462 322 248 148 2,321

Sources: CHES 2014 (Bakker et al. 2015); Manifesto Project (Volkens et al. 2015a)

% 11.2 4.6 18.7 6.2 8.6 19.9 13.9 10.7 6.4 100.0

The liberal party family ideology 333 the sociocultural dimension, which originally emerged as the churchstate or religious cleavage but from the 1960s onwards has gradually transformed into the ‘new value’ (Inglehart 1977, 1990), the ‘new politics’ (Franklin 1992), or the ‘GAL-TAN’ dimension (green/alternative/libertarian versus tradition/authority/nation) (Hooghe et al. 2002). Hence, it does not cover the entire spectrum of ideological dimensions. Party systems across Europe and individual parties may have emerged and are still structured around other relevant lines of conflict such as the centre-periphery cleavage or the urban-rural cleavage (see, for instance, liberal parties in Scandinavian countries in Chapters 1, 2, and 3 in this volume). Yet, it is a necessary trade-off for simplicity and comparisons. The Chapel Hill and Manifesto projects considerably differ as regards the type and number of policy issues they include, as well as the scales used to position parties. The methods explained below therefore aim at (1) determining the broad policy orientations of parties on the two main ideological dimensions on the basis of their positions on various policy issues; (2) harmonising the position of parties on the two dimensions on a 0–10 scale, where 0 stands for the most leftist position (or, on the cultural dimension, the most progressive stance), and 10 represents the more rightist (or conservative) position. In the case of the CHES, I selected several socioeconomic and cultural policy issues4 on which experts had to position parties between 0 and 10. These issues include: redistribution of wealth from the rich to the poor, deregulation of markets, reducing tax versus improving public services, promoting civil liberties versus supporting tough security measures, support for liberal lifestyle policies (e.g., homosexuality), position on the role of religious principles in politics, support for tough policies on immigration, support for multiculturalism versus assimilation, and support for environmental protection at the expense of economic growth (or vice versa). I used a factorial analysis to reduce the number of issue dimensions. As expected, the principal component analysis (PCA, Promax rotation) detected two latent factors, which account for 85.3% of the variation. As shown in Table 17.3 and Figure 17.1, the positions of parties on sociocultural issues highly load on the first factor (eigen value 5.7), whereas positions on socio­ economic issues highly load on the second factor (eigen value 1.9). Hence, the two components largely correspond to the economic and cultural conflicts that have structured party competition in many European contexts. These two factors, which range between -3 and 3, were transformed into two 0–10 point scales, with higher values indicating a right-wing/conservative position. The two final variables have a mean at 5 and a standard deviation of 1.7. In the case of the Manifesto Project, I replicated the method developed by Camia and Caramani (2012)5. To create the left-right economic and cultural dimensions, manifesto items were first classified into ‘left-wing’ and ‘right-wing’ categories for each dimension (see Table 17.4, reproduced from Camia and Caramani 2012). Second, I ‘calculated the number of quasi-sentences in favour of right-wing categories as a percentage of the

334  Caroline Close Table 17.3  Structure matrix Components Position Spend vs. tax Deregulation Redistribution Civil liberties vs. law and order Social lifestyle Role of religious principles in politics Immigration policy Multiculturalism Environment

1

2

0.454 0.326 0.453 0.946 0.914 0.852 0.927 0.939 0.738

0.972 0.970 0.962 0.430 0.242 0.360 0.443 0.416 0.574

Source: CHES 2014 (Bakker et al. 2015) Note: Extraction method: PCA analysis. Rotation method: promax rotation with Kaiser normalisation.

total number of quasi-sentences in favour of both left- and right-wing categories’ (Camia and Caramani 2012: 84). The result was then divided by 10 to obtain a 0–10 point scale. The economic dimension variable has its means at 3.8; the cultural dimension variable has its means at 5.0. In addition to the left-right position of parties on these two dimensions, the analysis below also refers to their ‘general’ left-right position (0–10 scale), as Factor plot after rotation 1,2 deregulation spendvtax redistribution

1,0

Component 2

0,8 0,6 0,4

environment immigrate_policy civlib_laworder

0,2 0,0

religious_principlemulticulturalism sociallifestyle

–0,2 –0,2

0,0

0,2

0,4 Component 1

0,6

0,8

1,0

Figure 17.1  Principal component analysis of issue positioning: factor plot after rotation (Promax) Source: CHES 2014 (Bakker et al. 2015)

The liberal party family ideology 335 Table 17.4  Manifesto Project left and right items Dimensions

Left-wing items

Right-wing items

Economic

per404 per405 per409

Cultural

per401 per402 per414

Free enterprise Incentives Economic orthodoxy

per412 per413 per415

Economic planning Corporatism Keynesian management Controlled economy Nationalisation Marxist analysis

per505 per702 per704

per504 per701 per103 per107 per503 per602 per604 per607

Welfare expansion Labour groups Anti-imperialism Internationalism Social justice Anti-nationalism Against traditions Multi-culturalism

_ _ per109 per305 per601 per603 per605 per608

Welfare limitation Against labour groups Middle class and professional groupings _ _ Anti-internationalism Political authority Nationalism Traditions Law and order Against multi-culturalism

Source: Table reproduced from Camia and Caramani (2012: 85). For details about the single items, see the Manifesto Project Codebook (Volkens et al. 2015b)

given by experts in the Chapel Hill survey, or derived from the ‘rile’ index6 included in the Manifesto Project that I transformed into a 0–10 scale (see the method applied by Camia and Caramani 2012: 84).

Analysis Assessing the ideological distinctiveness and homogeneity of the liberal parties in 2014 Based on the 2014 CHES data, Table 17.5 compares the policy orientations of the main party families on the general left-right scale and on the two main ideological dimensions. Statistically significant differences (according to Scheffé test) with the liberal party family are specified by the p values (*). With an average left-right position at 5.9, the liberal party families appear significantly more to the right than the three left-wing families (radical left, Green, and Social-Democrat), while the liberal party families appear significantly more to the left than the Conservative and radical right families. Thus, according to experts’ coding, the liberal family in 2014 stands at the centre/centre-right of the ideological space. As shown by the liberal party family’s position on the economic and cultural dimensions, this central position combines a marked right-wing position on economic issues (along with the Conservatives, the Liberals display the highest score on this dimension) with a centre-left position on sociocultural issues. Figures 17.2 and 17.3 graphically represent these tendencies (the dotted lines represent the mean value of both dimensions, i.e., 5.0).

336  Caroline Close Table 17.5  Party families’ positions and dispersion on the general, economic, and cultural left-right dimensions (CHES 2014) Party family Radical left Green Social-Democrat Liberal Christian-Democrat Conservative Radical right Total

Mean Std. Mean Std. Mean Std. Mean Std. Mean Std. Mean Std. Mean Std. Mean Std.

General left-right

Economic left-right

Cultural left-right

1.3*** 0.7 3.4*** 1.2 3.6*** 0.9 5.9 1.3 6.2 1.0 7.4** 1.0 8.2*** 1.4 5.3 2.3

2.2*** 0.5 3.8*** 1.0 4.2*** 0.9 6.3 1.1 5.8 0.8 6.3 1.2 5.0 1.3 5.0 1.7

3.3* 0.9 2.9** 0.7 4.2 0.9 4.4 0.9 5.9** 0.8 6.4*** 0.9 7.6*** 0.5 5.0 1.7

Source: CHES 2014 (Bakker et al. 2015) Note: p value: *0.05, **0.01, ***0.001.

10

UKIP

Economic left-right

8

6

4 XA ZL SP/PS_swi

2

ATAKA

0 Radical left

Social-democrat Christian-democrat Radical right Green Liberal Conservative

Figure 17.2  Party families’ position on the economic dimension (CHES 2014) – Boxplot Source: CHES 2014 (Bakker et al. 2015)

The liberal party family ideology 337 9 8

Cultural left-right

7 6

UNPR KSCM

PSD_rom

5 4 3 ZL 2 Radical left

Social-democrat Christian-democrat Radical right Green Liberal Conservative

Figure 17.3  Party families’ position on the cultural dimension (CHES 2014) – Boxplot Source: CHES 2014 (Bakker et al. 2015)

These two figures clearly illustrate the ambivalent position of the liberal family. Liberals can be classified as right-wing on the economic dimension, but they are closer to the left-wing group on the cultural dimension. Their distinctiveness resides specifically in this combination (Elff 2013: 20). Another specificity of the liberal family is its high degree of heterogeneity. The standard deviations reported in Table 17.5 and the dispersion shown in Figures 17.2 and 17.3 indicate that the liberal party family is amongst the least homogeneous families, along with the Conservative (on both dimensions) and the radical right (on the economic dimension). This finding is thus more in line with Camia and Caramani’s (2012) findings than with those of Ennser (2012) and Freire and Tsatsanis (2015). This difference can probably be explained by the fact that these latter studies do not include countries in CEE. To dig deeper into the ideological diversity of the liberal party family, Figure 17.4 examines the position of individual liberal parties in the two-dimensional space, in a similar way as presented in Chapter 15. Figure 17.4 distinguishes liberal parties in Western Europe from those in CEE. The figure illustrates the ideological diversity within the liberal family: on the economic dimension the position of parties vary from 3.9 (MPT – the Portuguese Party of the Earth, which is arguably more a Conservative Green

338  Caroline Close

7

CEE Western Europe R2 Quadratic = 0,056

KESK FF

6

V_den PSD

MPT

Cultural left-right

DP_lith EK

5

PNL

DPS SMC

FDP/PLR

ANO2011 C’s HSLS

VLD ER MR

FDP DP_lux LRLS V_nor RKP/SFP NEOS ZaAB FP

UPyD

4

VVD

LibDem

LA

IDS C_swe HNS RV

3

D66 RP

2 3

4

5

6 7 Economic left-right

8

9

Figure 17.4  Liberal parties in the two-dimensional ideological space Source: CHES 2014 (Bakker et al. 2015)

party) to 8.5 (LA – the Danish Liberal Alliance); on the cultural dimension the positions vary from 2.5 (the Polish Palikot’s Movement) to 6.0 (KESK, the Finnish former agrarian party). As in Chapter 15, I compute a k-means cluster analysis, here around the two ideological dimensions, to identify the existence of groups or tendencies within the liberal party family.7 A k-means cluster analysis suggests the existence of three groups of liberal parties, which could be described as: the economic or ‘Classical’ Liberals, the Social Liberals, and the Conservative Liberals. The first group8 is composed of liberal parties whose primary identity is defined by their right-wing position on the economic axis (Elff 2013: 6), with a centrist or indecisive positioning on the cultural axis9: among them are found the Swiss and German FDP, the Belgian MR and OpenVLD, the Dutch VVD, the Luxemburgish DP, the Danish Venstre and Liberal Alliance (LA), and in CEE the Croatian Social-Liberal Croatian Party (HSLS), the Lithuanian Liberal Movement (LRLS), and the Romanian National Liberal Party (PNL). The second group10 is composed of liberal parties that combine a leftwing position on social-cultural issues with a centre-right position on the economic axis: D66, the LibDems, NEOS, the two Swedish liberal parties (FP and C), the Social-Liberals of Denmark (RV), the Norwegian Venstre,

The liberal party family ideology 339 the Slovenian ZaAB, the Croatian Liberal Democrats (HNS), and the Polish Palikot’s Movement (RP); as well as a few parties that could be classified as ‘regionalist’ (IDS in Croatia, FSP in Finland). The third group11 is characterised by a centrist position on the economic axis combined with a more centre-right position on the cultural axis. This group includes the Finnish KESK, the Estonian EK, the Irish Fianna Fail, the Lithuanian DP, the Slovenian Modern Centre Party (SMC) as well as the Bulgarian Movement for Rights and Freedoms (DPS), the Czech ANO2011, the Portuguese MPT; and, more surprisingly, the two Spanish liberal parties (Ciudadanos and UPyD). The liberal party family ideology through time: increasingly less liberal? Using the Manifesto Project data, this section investigates the ideological distinctiveness and diversity of the liberal party family over time, in two steps. As in the previous section, I first consider the policy orientations of the liberal party family in comparison with other party families on the two main dimensions (economic and cultural). Then, I investigate the internal ideological dynamics of the liberal family through an inductive clustering analysis. As in the previous section, three subgroups are identified within the family, and I examine their respective evolution in terms of policy orientations and relative weight in the liberal party family. Figure 17.5 displays the placement of party families on the general, economic, and cultural left-right dimensions over time. Left-wing families (radical left, Green, and Social Democrat) and right-wing families (Christian Democrat, Conservative, and radical right) are aggregated into two broad categories12. This allows highlighting the peculiarities of the liberal party family. Overall, although the general left-right placement of the families has not much changed throughout the period, their placement on the two dimensions has changed quite substantially. Three periods can be identified. For each period, Table 17.6 reports the average position of the liberal party family compared with the other two groups. Statistically significant differences (according to Scheffé test) with the liberal party family are specified by the p values (*). During the period (1945 to mid-1970s), the placement of party families on the two dimensions is quite distinct: The right-wing parties are positioned on the centre-right of both dimensions; the left-wing parties are positioned on the left of the economic dimension, and centre-left of the cultural dimension; although the liberal parties combine right-wing/centre-right economic preferences with centre-left cultural preferences. During this first period, the average placement of liberal parties thus fits with the classic understanding of liberalism. On average, they are significantly more to the right on the economic (and cultural) dimension than the left-wing parties, but they are significantly more leftist than the right-wing parties on the cultural

340  Caroline Close Rile index Economic left-right Cultural left-right

Left-wing families

8 6 4 2 8

4 2

Right-wing families

8

Party families

Liberal family

6

6 4 2 1945– 1955– 1965– 1975– 1985– 1995– 2005– 1954 1964 1974 1984 1994 2004 2015 Time

Figure 17.5   Party families’ positioning on the general, economic, and cultural dimensions, 1945–2015 (Manifesto Project) Source: Manifesto Project (Volkens et al. 2015a)

dimension. In terms of homogeneity, liberal parties appear more heterogeneous than the left-wing bloc on all dimensions, and more heterogeneous than the right-wing bloc on the cultural dimension. From the mid-1970s to the mid-1990s, all political families moved to the right on the cultural dimension, especially the right-wing parties (from 5.0 to 6.4) (see also Camia and Caramani 2002: 69–71). In the case of the liberal parties, the tendency is less marked, but we observe that the discrepancy between their economic and cultural preferences tends to decrease, both moving towards the centre. Yet, during this period, liberal parties remain distinct from the other families: On the economic axis, they appear significantly more to the right than the left-wing parties; and significantly more leftist than the right-wing parties on the cultural dimension. Interestingly, during this period characterised by the emergence and structuring of the materialist/postmaterialist cleavage, the average placement of left-wing parties on the cultural dimension (4.0) comes closer to that of the liberal parties (4.4), and liberal parties appear more homogeneous on this dimension than the left-wing bloc. From the mid-1990s until present times, we can observe parties moving towards the left on the economic dimension in both the right-wing group

The liberal party family ideology 341 Table 17.6  Party families’ positioning and dispersion on the general, economic, and cultural left-right dimensions, 1945–2015 (Manifesto Project) 1945–1974 Liberal parties Left-wing parties Right-wing parties Total

Rile index 5.2 1.1 168 3.8*** 1.0 202 5.4 1.1 187 4.8 1.3 557

Economic left-right 5.5 2.5 167 1.7*** 1.5 200 5.2 2.7 185 4.0 2.8 552

Cultural left-right 4.0 3.1 166 3.0** 2.9 197 5.0** 2.6 180 4.0 3.0 543

Rile index

Economic left-right

Cultural left-right

Mean Std. N Mean Std. N Mean Std. N Mean Std. N

5.2 1.0 145 4.4*** 1.1 298 5.5** 0.9 226 4.9 1.1 669

5.3 2.8 145 2.1*** 2.0 293 5.2 2.2 222 3.9 2.7 660

Mean Std. N Mean Std. N Mean Std. N Mean Std. N

Rile index 5.0 0.8 154 4.2*** 0.7 299 5.5*** 0.9 304 4.9 1.0 757

Economic left-right 4.7 2.0 154 2.0*** 1.5 299 4.4 2.1 302 3.5 2.2 755

Mean Std. N Mean Std. N Mean Std. N Mean Std. N

1975–1994 Liberal Left-wing parties Right-wing parties Total

1995–2015 Liberal Left-wing parties Right-wing parties Total

4.4 2.7 144 4.0 3.0 296 6.4*** 2.2 223 4.9 2.9 663 Cultural left-right 5.7 2.3 154 4.0*** 2.2 299 7.3*** 1.8 304 5.7 2.6 757

Source: Manifesto Project (Volkens et al. 2015a) Note: p value: *0.05, **0.01, ***0.001.

and the liberal group. Liberal parties seem to have changed their policy orientations: They have combined a more centrist position on the economic dimension with a right-of-centre placement on the cultural dimension. Yet, they haven’t done so to the extent observed in the case of right-wing parties.

342  Caroline Close As shown in Table 17.6, liberal parties still occupy a specific position in the ideological space relative to other families: on the right of left-wing parties on the economic dimension, and between the left and right-wing parties on the cultural dimension. Thus, despite their shifts in the two-dimensional space we see that the Liberals have maintained their distinctiveness over time compared with other party families. However, they still appear slightly more heterogeneous than the two other blocs, even if their overall heterogeneity has decreased. To investigate the internal evolution of the liberal party family, I have computed for each decade a k-means cluster analysis along the two ideological dimensions13. For every decade, the cluster analysis reveals the existence of three groups within the liberal party family, which could easily be identifiable as the Classical, Social, and Conservative Liberals. Figure 17.6 displays the average position of each of these groups on the main ideological dimensions (corresponding to the centres of the clusters) over time. Classical Liberals have combined through the whole period a marked rightwing placement on the economic dimension with centre-left preferences on the cultural dimension – except during the 1995–2004, when they moved Rile index Economic left-right Cultural left-right

10 Classical Lib

8 6 4 2

Social Lib

8 6 4 2

Conservative Lib

0 10

3 clusters 1945–2015

0 10

8 6 4 2 0 1945– 1955– 1965– 1975– 1985– 1995– 2005– 1954 1964 1974 1984 1994 2004 2015 Time

Figure 17.6  Classical, social, and conservative liberals’ positioning on the general, economic, and cultural dimensions, 1945–2015 (Manifesto Project) Source: Manifesto Project (Volkens et al. 2015a)

BK-TandF-9780815372387_TEXT_CLOSE-181001-Chp17.indd 342

14/01/19 2:23 PM

The liberal party family ideology 343 towards the centre-right. Social Liberals are identified by their centre-left placement on both dimensions, which has been more pronounced on one or the other dimensions depending on the decades. Conservative Liberals distinguish themselves by their rightist placement on the cultural dimension, while their economic preferences have varied through time. Figure 17.7 shows for each period the proportion of cases that were identified in the three categories, and thus allows grasping the relative weight of each subgroup within the liberal party family across time. Social liberalism has decreased over time, except in the past decade while Conservative liberalism has increased, especially since the mid-1990s. By contrast, Classical liberalism has sharply declined in the recent decades. It is important to keep in mind here that the unit of analysis is the ‘electoral’ party and that the clustering analysis was run for each decade separately. As a consequence, one particular party can belong to different clusters over the whole timespan. This actually illustrates the fluctuation, hesitation, and division of individual parties over which ‘liberal’ orientation to emphasise. Some parties are characterised by higher levels of instability, for instance in the UK, in Luxembourg, Italy, Belgium, and Sweden (especially regarding the Centre Party). In these countries and throughout the period, liberal parties have oscillated between Classical, Social, and Conservative liberalism. By contrast, some parties have maintained a clear and stable identity through time. This is the case of the German FDP, where Classical liberalism has remained the dominant ideological trend; 3 clusters 1945–2015

50

Classical Lib Social Lib Conservative Lib

%

40 30 20 10 0 1945– 1955– 1965– 1975– 1985– 1995– 2005– 1954 1964 1974 1984 1994 2004 2015 Time

Figure 17.7  Proportion of classical, social, and conservative liberals within

the liberal party family, 1945–2015 (Manifesto Project)

Source: Manifesto Project (Volkens et al. 2015a)

BK-TandF-9780815372387_TEXT_CLOSE-181001-Chp17.indd 343

14/01/19 2:23 PM

344  Caroline Close of D66, which is the archetypical Social Liberal party; and of the Finnish Liberals (LKP), which was dominated by Conservative liberalism throughout the period. The recent evolution towards less Classical liberalism and more Conservative liberalism and (to a lesser extent) Social liberalism can be attributed for a part to the development of liberal parties in CEE countries. Comparatively to their Western counterparts, these liberal parties have adopted more conservative stances on the cultural axis and more centrist economic positions. Examples of these conservative-liberal parties include the Bulgarian National Movement for Stability and Progress (NDSV), the Czech ANO11, Latvia’s First Party (LPP/LC), the Lithuanian Liberal and Centre Union (LiCS), and the Romanian National Liberal Party (PNL). Other liberal parties from CEE have oscillated between Conservative and Social liberalism: see, for instance, the two Estonian liberal parties, the Lithuanian Labour Party and the Slovenian Liberal Democracy (LDS). The changes observed in recent decades can also be attributed to the emergence of new liberal parties in the Western countries, such as the Social liberal Union, Progress and Democracy (UPyD) and Ciudadanos in Spain. In addition, some liberal parties that were classified as Classical Liberals throughout the 1945–1995 period have shifted to Conservative Liberals in the period 2005–2015: This is the case of the Dutch VVD, the Swedish Liberals and the Swiss FDP, which have adopted more conservative stances (see case studies in this volume). Interestingly, other parties classified as Conservative Liberals in previous decades have oscillated between Conservative and Social liberalism since the 1990s: This is the case of the Portuguese Social Democratic Party (PSD), the Irish Fianna Fail and the Icelandic Progressive Party (FSF). Overall, the past two decades are characterised by a general decline of the relative weight of Classical liberalism within the liberal party family ideology.

Conclusion and perspectives The objective of this comparative chapter was to assess the existence of a liberal party family on the basis of the parties’ ideological or policy orientations. The chapter has examined the distinctiveness and homogeneity of the liberal party family in a cross-sectional and longitudinal perspective, by using data from the Chapel Hill Expert Survey and from the Manifesto Project. The chapter has shown that the liberal party family has displayed and still displays an ideology that is distinct from that of the other main families. This ideology appears as centrist on the general left-right scale, but has better been described as ambivalent given that it combines a right-wing position on economic issues with centre-left preferences on cultural matters. As argued by Elff (2013), it is precisely this ambivalent position of the liberal family ‘that brings the existence of more than one “ideological dimension” to the surface’ (Elff 2013: 20).

The liberal party family ideology 345 But if the liberal party family ideology meets the criteria of distinctiveness, the analysis has also pointed out the great diversity of the family and has identified the existence and persistence of three distinct liberal traditions, which could be labelled as Classical, Social, and Conservative liberalisms. The cross-sectional analysis has classified current liberal parties into these categories and has identified their respective placement in the two-dimensional ideological space. The longitudinal analysis has gone a little further. Given that it relied on a different unit of analysis (party*election), the analysis shows how the belonging of parties to one or the other tradition has varied from one election to the other. This has allowed examining the relative weight of each tradition within the liberal party family throughout time. On the whole, since the mid-1990s, increasingly fewer parties could be identified as Classical Liberals on the basis of their manisfestos. Such a trend should inevitably bring us to question the very existence of a liberal ideology nowadays, as liberalism was originally understood as Classical liberalism, thus founded on the defense of ‘the two freedoms’ (Smith 1988). Are liberal parties doomed to disappear? If the decline of Classical liberalism is to be confirmed in the near future, a split of the family could well materialise between the Social and Conservative Liberals, with Social Liberals joining the (increasingly Liberal) Social Democrats on the centre-left, and the Conservative Liberals merging with the Christian Democrats and moderate Conservatives on the centre-right of the ideological spectrum. To the contrary, a more optimistic perspective could conclude that several liberal traditions have always existed and have nourished the ideological and political debates in Europe, with varying success through time and across countries. The inherent flexibility of the liberal ideology could thus be seen as an asset that has made liberal parties more adaptable to societal evolutions. It remains to be seen how the liberal party family will adapt to and shape the big questions of the 21st century.

Notes

1. Whereas Freire and Tsatsanis talk about the ideological cohesion or cohesiveness of the party families, I use the term homogeneity. See Close and Nunez (2017), for further discussion on the two concepts. 2. Retrieved from http://chesdata.eu/. 3. Retrieved from https://manifestoproject.wzb.eu/information/documents/ information. 4. I excluded issues related to European integration and European policies, as well as issues pertaining to decentralisation/centralisation policies as they usually bring additional lines of conflicts. Preliminary analyses have shown that liberal parties are amongst the most supportive of the EU (6.1) along with the Christian Democrats; and the difference is significant when compared with the radical left (3.1) and radical right parties (2.0). On the decentralisation issue, the only significant result concerns the Regionalist family, which is significantly more in favour of decentralisation than other families, including the Liberals.

346  Caroline Close



5. The only difference being that I used a 0–10 scale, whereas Camia and Caramani use a 1–10 scale. 6. The Manifesto Project’s rile index is based on Laver and Budge (1992) (see Volkens et al. 2015b: 28). 7. Interestingly, the placement of parties as measured by their voters’ preferences (ESS data, see Chapter 15) and as measured by the CHES do significantly correlate, revealing congruence: economic positions correlate with a coefficient of 0.472 (significant level 0.05); cultural positions correlate with a coefficient of 0.579 (significant level 0.05). 8. Final cluster centres at 7.4 (economic dimension) and 4.9 (cultural dimension). 9. See, for instance, the Belgian Liberals (Chapter 5); or the Danish Venstre, which is not libertarian but adopts progressive stances on family life issues (Chapter 2). 10. Final cluster centres at 6.0 (economic dimension) and 3.2 (cultural dimension). 11. Final cluster centres at 5.2 (economic dimension) and 5.1 (cultural dimension). 12. Quite similar tendencies were observed within these two groups, especially within the right-wing group. For a more detailed evolution of party families, see Camia and Caramani (2012: 70). 13. I also used a two-step cluster analysis to double-check the findings. Although for two decades (1945–1954 and 1965–1974) the two-step cluster analysis only identified two groups instead of three, similar conclusions as those presented below could be drawn.

References Bakker R., Edwards E., Hooghe L., Jolly S., Koedam J., Kostelka F., Marks G., Polk J., Rovny J., Schumacher G., Steenbergen M., Vachudova M., Zilovic M. (2015), ‘1999–2014 Chapel Hill Expert Survey Trend File. Version 1.13’. Chapel Hill, NC: University of North Carolina. Retrieved from https://chesdata.eu Benoit K., Laver M. (2006) Party Policy in Modern Democracies. London: Routledge. Budge I., Klingemann H.-D., Volkens A., Bara J., Tanenbaum E. (eds) (2001) Mapping policy preferences: estimates for parties, electors, and governments 1945–1998. Oxford: Oxford University Press. Camia V., Caramani D. (2012), ‘Family meetings: Ideological convergence within party families across Europe, 1945–2009’, Comparative European Politics 10(1), pp. 48–85. Close C., Núñez L. (2017), ‘Preferences and agreement in legislative parties: Testing the causal chain’, Journal of Legislative Studies 23(1), pp. 31–43. Dinas E., Gemenis K. (2010), ‘Measuring parties’ ideological positions with manifesto data: A critical evaluation of the competing methods’, Party Politics 16(4), pp. 427–50. De Winter L., Marcet J. (2000), ‘Liberalism and Liberal Parties in the European Union’. In: De Winter L. (ed.), Liberalism and liberal parties in the European Union. Barcelona, Spain: Institut de Ciències Politiques I Socials, pp. 11–24. Elff M. (2013), ‘On the distinctiveness of party families’. Paper presented at the 71th Annual Conference of the Midwest Political Science Association, Chicago, April 11–14. Ennser L. (2012), ‘The homogeneity of West European party families: The radical right in comparative perspective’, Party Politics 18(2), pp. 151–171. Franklin M. (1992), ‘The decline of cleavage politics’. In: Franklin M.N., Mackie T.T., Valen H. (eds.), Electoral change: Responses to evolving social and attitudinal structures in Western countries. New York: Cambridge University Press, pp. 383–431.

The liberal party family ideology 347 Freire A., Tsatsanis E. (2012), ‘Party families, ideological distinctiveness and cohesion: A strong test of the heuristics of the concept of “familles spirituelles”’. Paper presented at the ECPR General Conference, Montreal, August 26–29. Hooghe L., Marks G., Wilson C.J. (2002), ‘Does left/right structure party positions on European integration?’ Comparative Political Studies 35, pp. 965–989. Inglehart R. (1977), The silent revolution: Changing values and political styles among Western publics. Princeton, NJ: Princeton University Press. Inglehart R. (1990), Culture shift in advanced industrial society. Princeton, NJ: Princeton University Press. Klingemann H.-D., Volkens A., Bara J., Budge I., McDonald M.D. (2006) Mapping policy preferences II: estimates for parties, electors and governments in Central and Eastern Europe, European Union and OECD, 1990–2003. New York: Oxford University Press. Kriesi H., Grande E., Lachat R., Dolezal M., Bornschier S., Frey T. (2006), ‘Globalization and the transformation of the national political space: Six European countries compared’, European Journal of Political Research 45, pp. 921–56. Laver M., Budge I. (eds.). (1992), Party policy and government coalitions. Basingstoke, UK: MacMillan. Leimbruger P., Hangartner D., Leeman L. (2010), ‘Comparing candidates and citizens in the ideological space’, Swiss Political Science Review 16(3), pp. 499–531. Mair P., Mudde C. (1998), ‘The party family and its study’, Annual Review of Political Science 1(1), pp. 211–229. Smith G. (1988), ‘Between left and right: The ambivalence of European liberalism’. In: Kirchner E.J. (ed.), Liberal parties in Western Europe. Cambridge: Cambridge University Press, pp. 16–28. Volkens A., Lehmann P., Matthieß T., Merz N., Regel S., Werner A. (2015a), The Manifesto Data Collection. Manifesto Project (MRG /CMP / MARPOR). Version 2015a. Berlin: Wissenschaftszentrum Berlin für Sozialforschung (WZB). Volkens A., Lehmann P., Matthieß T., Merz N., Regel S., Werner A. (2015b), The Manifesto Project Dataset – Codebook. Manifesto Project (MRG / CMP / MARPOR). Version 2015a. Berlin: Wissenschaftszentrum Berlin für Sozialforschung (WZB). Von Beyme K. (1985), Political parties in Western Democracies. New York: St. Martin’s Press.

18 How liberal parties organise Stefanie Beyens, Emilie van Haute, and Tom Verthé

Introduction What differentiates the organisations of liberal parties from those of other party families? Answering that question usually entails a study of party labels, of transnational federations parties may be members of or, more commonly, investigating parties’ origin and sociology together with their ideology and policy positions (Mair and Mudde 1998). This is certainly true for liberal parties, as liberalism refers more directly to an ideology than to a type of organisation. The literature on party politics does develop some expectations about liberal party organisations, albeit sometimes indirectly. Two main arguments link party organisations to party families: A direct argument that relates a party’s core ideology and values to its organisational implementation, and an indirect argument that connects a party’s origins to its current organisational structure. First, as Enyedi and Linek point out (2008), some aspects of a party’s ideology may directly refer to conceptions of how to manage communities, groups, and individuals within these. These conceptions may be reflected in how parties decide to set up their organisation. In the same vein, Close argues that ‘a party family’s core values and principles … are reflected in the party’s organisational structure… . Party families thus carry specific models of democracy’ (2016: 2). As an ideology, liberalism puts an emphasis on individual freedom and rejects state regulation and excess of political authority (see Chapter 17 in this volume). Consequently, authors have stressed organisational features of liberal parties that would embody these characteristics: general reservations with regard to organisation and a focus on individuals and on individual participation and direct democracy rather than collective bodies, delegation, and representation (Gauja 2013; Close 2016). Second, classic theoretical studies on models of party organisations seek to explain ‘how contemporary parties bear the marks of their origins, and how organisational differences reflect institutional contexts and ideological (party family) similarities’ (Poguntke et al. 2016: 662). Duverger (1954) famously distinguishes between internally created cadre parties and

How liberal parties organise 349 externally created mass parties. Panebianco (1988) links these features to party development, stressing two paths for party development: penetration or diffusion, where diffusion entails that a number of local associations autonomously spring up and later unite in a (lose) national organisation. According to him, ‘both conservative and liberal European parties are for the most part internally created (parliamentarily), and … many liberal parties develop through diffusion” (Panebianco 1988: 51). These origins would lead liberal party organisations to be characterised by more decentralisation, more autonomous and turbulent leadership, and constant struggle for party control from the dominant coalition leading to complex internal power dynamics and factionalism. However, these general characteristics have rarely been empirically tested in a systematic way. In his comparative study of liberal parties in Western Europe, Kirchner (1988: 8) argues that ‘the organisational potential of liberal parties … can be regarded as low in comparison with other parties… . liberal parties never built up the solid party organisations or membership support that is characteristic of their main competitors’. The author describes liberal parties as characterised by a lack of internal organisation and tightly controlled structure and underdeveloped support among leading interest groups or particular groups. In line with Duverger and Panebianco, he associates liberal parties to cadre parties with a direct appeal to voters rather than members. In their country-case studies on How parties organize, Katz and Mair (1994: 121) highlight three organisational features of liberal parties (mainly in reference to the liberal Democrats in the UK): federal structures, the direct selection of the party leader by the rank-and-file and a more democratic process in defining party policy, and the absence of mass membership but a valuation of individual membership. These features are in line with the theoretical expectations above, although the authors point to organisational convergence leading liberal parties to become more membership-based as their organisational development progressed (p. 338). More recently, Poguntke et al. (2016: 662) have systematically analysed the link between institutional context (country) and party family on various organisational characteristics of parties, based on the new Political Party Database (PPDB) dataset. They find that party finance and the number of paid staff in the central party headquarters (HQ) are heavily determined by country (institutions) and not by party family. But other party characteristics such as membership figures, leadership strength, or inclusiveness of decision-making processes are affected both by the institutional context and by party family, giving grounds to the idea that, despite organisational developments and strong institutional constraints, parties from the same family do share some common features. This chapter intends to systematically look at organisational specificities of liberal parties. In line with the above-mentioned literature, we expect liberal parties to display the following organisational characteristics,

350  Stefanie Beyens, Emilie van Haute, Tom Verthé structured around Janda’s (1970) two dimensions of degree of organisation (or intensiveness) and level of centralisation (or extensiveness): 1 Degree of organisation: Generally, we expect liberal parties to have developed weaker organisations than Conservative or Social Democratic parties. This should especially stand out when it comes to resources and relations to special interest groups. We expect liberal parties to be characterised by • •

Fewer staff, fewer financial resources, fewer members; and Fewer rights or less guaranteed representation to groups.

2 Level of centralisation: We expect liberal parties to be characterised by: • • • •

Openness and inclusiveness of membership; More rights to individual members in the main decision-making processes (leadership and candidate selection, policy formulation); More power for parliamentary party group; and More contested leaderships.

Country chapters in this volume have emphasised some organisational features, most often pointing in these directions (with the exception of liberal parties with agrarian roots that resemble mass parties). Yet these are not systematic accounts of organisational features of liberal parties compared with other party families. Using newly available datasets on political parties (PPDB, MAPP, COSPAL)1 this chapter offers the largest comparative analysis of liberal party organisations so far.

Degree of organisation As parties that originated inside parliament, liberal parties would be more voter- and election-oriented. They are expected to have a weaker organisation beyond electoral cycles. Therefore, we expect liberal party organisations to rely on fewer resources, be it in terms of staff, finances, or members. Professionalisation In scrutinising liberal parties’ organisation, we must not overlook the personnel providing consistency between elections. The mandate of elected politicians is usually secure for a couple of years at a time; this is not necessarily the case for staff working at parties’ headquarters. Although the core business of political parties is to get their representatives elected and, in case of success, to govern the polity, parties do not just need staff in the run-up to elections. The organisation itself needs continuity between elections, which can be provided by the personnel working at HQ. Likewise, once the party has managed to elect representatives to parliament, staff are needed to support members of parliament (MPs) in their daily tasks.

How liberal parties organise 351 These two indicators, staff working at HQ and staff working for the parliamentary party group (PPG), are used to capture the levels of professionalisation of liberal parties and to compare them with other party families. We use data from the PPDB project to assess the level of professionalisation of parties. This allows us to compare 15 liberal parties to 68 other parties in ten European countries2. Contrary to our expectations, Table 18.1 shows that party family does not matter when it comes to the professionalisation of parties; only size does. No matter what we take as indicator – party HQ staff or PPG staff (i.e., support staff in MPs’ offices) – all result in the finding that party family does not affect the number of staff employed by a party. When we standardise our data on seat share (in the lower house), there is no significant difference in numbers of PPG staff members between party families. Similarly, we do not find that liberal parties display fewer staff in their HQ compared with other party families. Even if they were born in parliaments, they display similar levels of staffing in their extra-­ parliamentary organisation as other party families. Overall thus, in terms of staff resources, liberal parties do not display weaker organisations than other party families. Let us look at a couple of possible explanations for these findings. A first possible reason for this lack of effect of party family on professionalisation focusses on what parties do instead of what they are: The functions parties need to perform (e.g., in parliament) are more important than the ideologies to which they adhere. Another way of interpreting these null-findings is to look at it from a resources angle: Parties of a certain size receive state subventions in accordance to, for example, the number of votes or seats they have won in parliament. It could be this money that drives how many employees are hired. A third alternative explanation considers the country level: Although party families may not drive professionalisation, regulation at the country level and diffusion of practises across parties from the Table 18.1  Mean staff members HQ and PPG (in full-time equivalent) Party family Christian Democrats/Conservatives Social Democrats Liberals Greens Left Socialists Right-wing (populists) Far right (extreme right) Total N ANOVA F P

Staff HQ mean FTE

Staff PPG mean FTE

2.5 4.7 6.8 3.5 7.4 1.3 2.5 4.4 55 0.61 0.72

13.0 8.2 8.7 12.2 14.7 1.5 2.1 10.4 64 0.45 0.84

Source: PPDB round 1, standardised on seat share in lower house

352  Stefanie Beyens, Emilie van Haute, Tom Verthé same polity may. This is what Kirchheimer (1966) had anticipated cadre parties to do: to adopt similar organisations as mass parties. This is also what Poguntke et al. (2016) find: In a comparison of party organisations in 19 countries, they uncover multiple indications that intercountry variation is more important than interfamily variation. Next, we look at the different ways parties can acquire income, and we finish with an exploration of campaign expenditures. Finances State subsidies are the main source of revenue for parties of all families and make up roughly half of their reported income3 (see Table 18.2). Even though the mean shares of revenue streams differ – with liberal parties being on average when it comes to membership contributions (19.2%) and higher than average in terms of private donations (19%) – there is overall more variation within the party families than between them. The only significant differences can be found in the share of state subsidies. A pairwise comparison of means (Tukey test) shows that extreme-right parties and right-wing populists draw considerably higher revenue shares from the state than other party families. Between the other party families there are no significant differences. One of the ways in which party family could drive campaign expenditures is in allocation to either party or individual candidates. We could expect that liberal parties focus more on individual candidates because of an ideological preference for individuals over community. However, the data shows this expectation is not met (see Table 18.2). All party families prioritise money for the party itself rather than for individual candidates during election campaigns. Table 18.2  Mean share party revenue and campaign expenditure (in %) Party family

Members

Christian Democrats/ 20.1 Conservatives Social Democrats 18.0 Liberals 19.2 Greens 22.4 Left Socialists 28.0 Right-wing (populists) 3.4 Far right (extreme right) 8.4 Total 19.0 N 123 ANOVA F 1.36 P 0.235 Source: PPDB round 1

Private donations

State subsidies

13.5

54.4

12.7 19.0 4.2 12.5 0.3 3.4 12.3 113 1.57 0.164

49.2 0.9 0.8 56.2 92.0 80.7 54.8 129 3.55 0.003

National party 0.9 0.8 0.1 0.2 0.8 1.0 0.9 37 0.38 0.860

Candidates 0.1 0.2 0.2 0.0 0.1 37 0.38 0.860

How liberal parties organise 353 The results here are remarkably similar to those related to professionalisation: Party families do not differ much on their revenue distribution or on how they distribute campaign money. So, can we then conclude that liberal parties are professional organisations so in tune with what is needed to perform that they do not display any specificities related to their origins or ideology? No. First, although the results do not reach conventional significance levels, the trend shows that liberal parties distinguish themselves by their higher reliance on private donations, which tends to match their ideological orientations. Secondly, we should point out that we are taking a birds-eye view of the situation: We are comparing averages. So again, one of the reasons we cannot find (many) significant between-group differences is that there is too much within-group variation. That means that party regulation at the country level would trump ideological differences within-­ country, just as was the case when we explored why there was little significant difference between party families’ degrees of professionalisation. It seems that when it comes to the distribution of resources (in terms of staff and in terms of income), parties converge within the specific institutional context of their country, not with ideological counterparts in other political systems (Poguntke et al. 2016). Membership size Membership figures expressed as average ratio of the entire electorate per party family per decade since World War II, show that liberal parties tend to have fewer members than the Social Democrats or the Conservatives and Christian Democrats. This is in line with Kirchner’s general observation that ‘Although they were among the first to emerge, liberal parties never built up the solid party organisations or membership support which is characteristic of their main competitors – the conservatives/Christian democrats and the socialist/labour parties’ (1988: 8). However, there has been convergence over time. Figure 18.1 focusses on membership figures in the past two decades, for which membership figures are deemed more reliable (van Haute et al. 2017). It distinguishes between parties in Western Europe and in Central and Eastern Europe, given the different age structure of the party systems, a variable affecting membership levels. For Western Europe, the figure confirms that the Liberals are located between the traditional mass membership parties (Social Democrats and Conservatives/ Christian Democrats) and the smaller new parties (Greens, Radical Right, Regionalists). It also confirms the rather stable membership base of liberal parties compared with other party families in the past two decades. Figure 18.1 also illustrates the organisational weakness of liberal parties in Central and Eastern Europe, especially in the 1990s.

Evolution of ME ratios by party family in Western Europe 3.00

(Former)Communist Social Democratic

Conservative

Christian Democratic/ Religious

Ethnic and Regional

Ecology

Liberal

Other

National Special Issue

M/E %

2.00

1.00

0.00 1990

1995

2000 2005 2010 Year Western Europe: Austria, Belgium, Denmark, Finland, Germany, Ireland, Italy, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom Evolution of ME ratios by party family in Easterm Europe

(Former)Communist

M/E %

7.50

Social Democratic

Conservative

Christian Democratic/ Religious

Ecology

Ethnic and regional Liberal

Other

National Special Issue

5.00

2.50

0.00 1990

1995

2000 Year

2005

2010

Eastern Europe: Croatia, Czech Republic, Estonia, Hungary, Lithuania, Poland, Romania, Slovenia.

Figure 18.1  Evolution of Members/Electorate ratio by party family, 1990–2010

How liberal parties organise 355 Overall thus, in terms of party size measured as membership, liberal parties are not mass parties. They are characterised by a midrange size and a capacity to maintain their membership at a more stable rate than the Social Democrats and Conservatives/Christian Democrats. Rights and guaranteed representation to groups We expect liberal parties to pay less attention to group representation in the structure of their party organisation, building on two assumptions. First, when we look at the origin of party organisations, we see that liberal parties originated inside parliaments. This differs from other parties that originated outside these institutions to defend group interests within civil society. Therefore, we do not expect that liberal parties will put as much emphasis as other party families (especially Christian and Social Democrats) on group representation, even if they might have preferential links to some groups, such as employers’ organisations or self-employed unions. The second assumption is based on the ideology shared by liberal parties, which tends to prefer rights linked to individuals instead of rights based on identification of group identities. Connecting the assumptions, if we take as a given that parties are founded to represent a certain constituency (religious or shared norms for Conservatives and Christian Democrats and shared socioeconomic background for Social Democrats), then it would be hard to put a label on the constituency that the Liberals were formed to represent. We therefore expect less attention to group representation in liberal parties. To test this expectation, we examine the presence of identity-based subgroups within the party and the institutionalisation of rules requiring a more balanced distribution in terms of gender. Eyeballing Table 18.3 gives us a rather more complicated picture than the one we had envisioned. When it comes to the mere existence of a women’s and a youth organisation, we see that liberal parties barely differ from Social Democrats (Christian Democrats lead the pack when it comes to identity-­ based groups within the larger party). However, when it comes to having rules in place that would ensure a more equal gender balance on the party executive, the Liberals lag behind Christian Democrats/Conservatives, Figure 18.1  (Continued) Note: Proxies for missing observations calculated based on a linear interpolation between existing observations. Source: M: subset of MAPP dataset (www.projectmapp.eu) – 55 liberal parties in 22 countries: NEOS and LIF (Austria), OpenVLD and MR (Belgium), HNS, HSLS, IDS-DD, and LS (Croatia), US-DEU and ANO2011 (Czech Republic), RV, I, and V (Denmark), ER, EK, and ESE-EDP-LEE (Estonia), KESK, L-LKP, and SFP (Finland), FDP (Germany), SZDSZ (Hungary), FF since 2009 and PD (Ireland), PLI, PRI, PR-R, and IDV (Italy), LICS, LLS, LRLS, DP, and NS (Lithuania), VVD and D66 (Netherlands), V (Norway), RP, KLD, UD, and UW (Poland), PLD, PSD until 1996 (Portugal), PNL (Romania), LDS, Zares, and DL (Slovenia), UCD, CDS, UPyD, and Cs (Spain), C and FP (Sweden), FDP, LPS, and LdU-ADI (Switzerland), LibDem (United Kingdom). Van Haute and Paulis (2016)

356  Stefanie Beyens, Emilie van Haute, Tom Verthé Table 18.3  % of parties with specific rights or guaranteed representation for groups (in %) Rules about representation of women on Women’s Youth Senior the party organization organization organization executive Conservatives / Christian Democrats Social Democrats Liberals Greens

N

70.6

100.0

41.2

37.5

16-17

50.0 53.3 11.1

90.9 86.7 66.7

27.3 13.3 0.0

63.6 13.3 33.3

11 15 9

Source: PPDB round 1

Social Democrats and Greens: Only 13.3% of liberal parties have these guarantees in place. If groups do exist within liberal parties, their role is not as institutionalised as in other party families, reflecting their general ideological preferences for equal opportunities rather than equal representation (Klausen and Maier 2001).

Level of centralisation Openness and inclusiveness of membership One correlate expectation regarding liberal parties is that, although they may have fewer members, they would still value individual membership. This would reflect in a more open conception of membership where individual citizens can easily engage in the party either as a member (low barriers to entry) or as any alternative form of ‘membership lite’ (Scarrow 2015). This would also reflect in a more inclusive conception of membership, in which members are granted more say in the party’s main internal decision-making processes. We use PPDB data (15 liberal parties) and scales computed by Kosiara Pedersen et al. (2017) to measure relative financial costs of membership and procedural barriers to entry, to assess whether parties offer an alternative affiliation option to standard party membership, and to measure the extent of voting rights granted to members on the above-mentioned three key decision-making areas (leadership selection, candidate selection, and policy formulation). Table 18.4 presents the distributions across party families. It partly confirms our expectations: Liberal parties tend to have lower procedural costs than some party families, although they are very close to the levels of the

How liberal parties organise 357 Table 18.4  Distribution of costs, benefits, and alternative affiliation by party family Party family Christian Democrats/ Conservatives Social Democrats Liberals Greens Total ANOVA F P ETA SQ

Financial costs

Procedural costs

Plebiscitary benefits

Alternative affiliation

N*

0.09

0.39

0.32

0.47

16-17

0.08 0.10 0.06 0.08 0.620 0.776 0.86

0.36 0.37 0.42 0.39 0.334 0.960 0.46

0.47 0.35 0.60 0.39 2.045 0.049 0.232

0.45 0.40 0.44 0.38 1.075 0.394 0.133

10-11 15 8-9 49-52

Source: PPDB round 1 Note: The scale for all measures is 0–1, where 0 means lowest costs/benefits or no alternative affiliation, and 1 means higher costs/benefits or existence of alternative affiliation option. * Shown as a range if the number of cases varies for each indicator.

Christian Democrats/Conservatives and Social Democrats. However, they display on average higher financial costs. This may be more in line with the sociological roots of these parties than with their ideological affinity with freedom and laissez-faire. Surprisingly, they tend on average to make less use of alternative affiliation options to traditional membership than the Greens, who have always advocated for participatory models of party organisations (van Haute 2016), but also than the Conservatives/Christian Democrats and the Social Democrats. This may be due to the fact that these parties have faced larger membership decline than the Liberals and have used alternative affiliation options as attempts to renew themselves (Scarrow 2015). Moreover, going back to the sociological roots of the Liberals, their raison d’être was not as tied to group representation and the related mass membership as was the case for the Christian and Social Democrats. This implies that regaining members is less urgent for them, hence potentially their relative lack of attention for alternative affiliation options. Similarly, the expectation that liberal parties would offer more benefits to their members than other party families is only partly confirmed. Social Democratic, Green, and even Regionalist parties display higher averages on the plebiscitary benefits scale. Some liberal parties may have been at the forefront of the introduction of plebiscitary benefits for their members in their national contexts (e.g., the Flemish Liberals in Belgium when it comes to leadership selection). But as Scarrow et al. (2016) pointed out, how parties organise faces strong within-country diffusion effects. When it comes to membership benefits, diffusion may be enhanced when new competitors such as Green or Regionalist parties appear, advocating new politics.

358  Stefanie Beyens, Emilie van Haute, Tom Verthé Openness and inclusion in leadership selection How a party selects its leader illustrates who holds the power to make important decisions about the party’s identity and its future goals (Figure 18.2). Especially in first-past-the-post electoral systems, a party leader is automatically a candidate for prime ministership. The party leader is also the public face of the party, making a change at the helm of the party an important event. The selection of a new party leader is then a good indicator of how power is distributed within a party. If only the party council or the parliamentary party group decide on leadership selection, we call this a top-down approach. If all party members – or even nonmembers – are allowed to vote in primaries, we call this a bottom-up approach. Letting only party delegates vote is the intermediate option. Although the bottom-up approach has been hailed for being more democratic, studies show that widening the selectorate does not always equal a retreating party elite (Hazan and Rahat 2010). It can signify that the important decision is moved from the actual selection of a leader to selecting the candidates for the leadership election (Aylott and Bolin 2017), both in terms of the number of candidates allowed to compete (which affects competitiveness) and in terms of selecting those candidates who share the elite’s preferences. Regardless of what broadening the selectorate means, we do see a move from top-down to bottom-up approaches in choosing a new party leader. This move occurs both in political parties in general (Cross and Pilet 2016) and in liberal parties in particular. If we look at data from the Cospal project (Cross and Pilet 2016), we can see that the preferred leadership selection method in liberal parties was by party delegates (at least two thirds of cases) from 1965 until 1990. Although it remains the preferred procedure in half of all cases, we do see some experimentation with allowing the full party membership to vote as early as 1988. In a third of all parties in the dataset, all members could vote in leadership elections from 1992 up until and including 2012 (end of data). Comparing liberal parties on the one hand with all other party families on the other, shows us that convergence between party organisational forms at the country level appears to be less of an issue when we look at leadership selection methods. Although it is the case that the first drop in party delegates forming the selectorate happens simultaneously in all party families (mid-1960s), there is a difference in what replaces it. Liberal parties give a bigger role to the PPG, whereas the other parties give more power to the Party Council. More importantly, we see that this trend continues as leadership selection methods evolve: Liberal parties are the only family to witness a drop in using party delegates as the main selectorate. Again, party delegates are replaced by party members in liberal parties, whereas other families use a plethora of alternative selection methods. This shows that liberal parties were not only early but also the most enthusiastic adopters of full membership voting rights in party leadership elections. The replacement of party delegates with party members in the selectorate of liberal

Liberal Parties 1

Percentage

.8 .6 .4 .2 0 1960

1970

1990

1980

2000

2010

Year Open Primaries Party Delegates

Full Membership Votes Party Council

Parliamentary Party Group Other

Mixed

Other Parties 1

Percentage