New Paths for Selecting Political Elites: Investigating the Impact of Inclusive Candidate and Party Leader Selection Methods 2020054377, 2020054378, 9780367901417, 9781003022893

Table of contents :
Cover
Half Title
Series Information
Title Page
Copyright Page
Table of Contents
Figures
Tables
Political parties
Contributors
1 Intra-Party Selection Methods and Political Elites: New Trends and Consequences
1.1 Introduction
1.2 The Selection of Political Elites: What we know so Far
1.3 Political Elites: Understanding the Interaction Between Party Organizations and Leadership
1.4 Rationales and Aims of this Volume
1.5 Plan of the Book
References
2 Choosing Party Leaders in Italy Between Personalization and Democratization
2.1 Introduction
2.2 Varieties of Party Models in Italy
2.3 Selection Methods Over Time
2.4 Demographics Characteristics
2.5 Conclusions
Note
References
3 Playing With Fire?: The Organizational Consequences of Party Primaries in Spain
3.1 Introduction
3.2 The Organizational Consequences of Introducing Party primaries On the Party Elites’ Strength
3.3 The Introduction of Party Primaries in Spanish Politics
3.4 The Organizational Consequences of First-Time Party Primaries in Spain
3.5 Discussion
3.6 Conclusions
Notes
References
4 The Outcomes of Party Primaries: (Dis)continuity in Top Candidates’ Political and Partisan Profiles
4.1 Introduction
4.2 The Theoretical Debate: Do Primaries Produce New Types of Top Candidates?
4.3 Using Spain to Disentangle the Role of Party Primaries
4.4 Results
4.5 Conclusion
Notes
References
5 Candidate Selection, Personalization of Politics, and Political Careers: Insights From Italy (2006–2016)
5.1 Introduction
5.2 The Italian Case in a Cross-Time Perspective
5.3 Candidate Selection, Personalization of Politics, and Patterns of Career: Is there a Connection?
5.4 Empirical Analysis
5.5 Conclusions
Notes
References
6 Impact of Intra-Party Democracy On Patterns of Political Career Within the Intra-Party-Pan-Hellenic Socialist ...
6.1 Introduction
6.2 Intra-Party Democratic Innovations Within Pasok: towards New Patterns of Candidates’ Selection
6.3 Intra-Party Democracy and Changing Patterns of Candidates’ Selection
Traditional Selection Models Within Greek Socialists
6.4 2004: The Primaries As a Shift? The Ambivalence of new Candidates’ Selection Patterns
6.5 To What Extent Has the Profile of Mps Changed Since 2004?
6.6 Conclusions
References
7 The Effect of Introducing Primaries On the Profiles of Candidates in the Turkish Context: The Local Experiences of ...
7.1 Introduction
7.2 Primaries and Political Experiences of Candidates
7.3 Primaries of RPP in Turkey
Abrief History of RPP
Primaries
7.4 Data, Analysis, and Results
7.5 Conclusion
Notes
References
8 Do Candidate Selection Modes Matter for the Gender Diversity Within Political Elites?: Evidence From the Belgian Case ...
8.1 Introduction
8.2 Gender Diversity On the Lists: Literature Review and Hypotheses
8.3 The Case of Belgium in 2014
8.4 Method and Empirical Results: Do Candidate Selection Processes Matter?
Operationalization of Variables
Method of Analysis
8.5 Empirical Analysis: How Candidate Selection Impacts Gender Diversity
8.6 Conclusions
Note
References
9 The Use of Primaries for the Selection of Party Leaders in the Uk Conservative and Labour Parties: Formal Rules and ...
9.1 Introduction and Research Questions
9.2 The Method: the Hypotheses of Organizational Convergence and Ideological Congruence
9.3 An Overview of Leadership Election Rules
9.4 Conclusions
Notes
References
10 Do Different Candidate Selection Methods Produce Different Types of Candidates?: An Analysis of the German Case
10.1 Introduction
10.2 Candidate Selection Methods, Party Ideology, and the Representativeness of Candidates
10.3 Candidate Selection in Germany
10.4 Data and Methods
Case Selection and Data
Operationalization: Outcome and Predictor Variables
10.5 Results
10.6 Conclusion
Notes
References
11 The ‘Belle Époque’ of French Primary: The Evolution of the National and Local Framework of Candidate Selection
11.1 Introduction
11.2 Research Question, Data, and Methods
11.3 The Socialist Experiment and the Development of the Primary in France
The First-Time Selection of the Presidential Candidate
The Stop and Return of the Primaries: the First Female Candidate
The Open Primaries for the Conquest of the Presidency
From Riches to Rags: the End of Socialist Primaries?
11.4 The Contagion Effects in the Adoption of Primaries By Communists, Ecologists, and Civil Society
The 2002 Presidential Elections
The 2007 Presidential Elections
The 2012 Presidential Elections
The 2017 Presidential Elections
11.5 The Primary of Others: From the Centre-Left to the Centre-Left-Centre-Right
The first attempts of the centre-right parties
11.6 The local primary from 2014 to 2020
11.7 The impact on social and political profiles of candidates
11.8 Conclusion
Notes
References
12 Political Elites and Party Primaries
12.1 Political parties between personalizationand intra-party democracy
12.2 Primaries and political elites: the impact of inclusive selectorates
12.3 Preliminary Results and Notes for a Research Agenda
Notes
References
Appendix
Index

Citation preview

New Paths for Selecting Political Elites

This book provides a cross-​country study of the consequences of the expansion of intra-​party democracy, the trend towards more inclusive methods of selection for party candidates and leaders, and the impact of these on political elites in terms of sociopolitical profile and patterns of careers. It explores the link between political organizations and political elites, by studying the role of parties in parliamentary and political selection and its impact on the political leadership appointed. Putting an emphasis on primary elections, it analyses the party elites that emerge from those selection processes and those democratized organizational settings. It focuses not only on the analysis of the processes through which party elites are selected and the consequences at the level of the party but also at the level of party elites themselves, i.e. what impact party primaries have on the characteristics of parties’ candidates and leaders. The book offers a theoretical, comparative, and empirical account of the internal electoral processes of parties and their impact on political recruitment. This book will be of key interest to scholars and students of political elites, political parties and party systems, electoral politics, democracy, populism, and leadership, and more broadly to comparative politics. Giulia Sandri is Associate Professor at the European School of Political and Social Sciences of the Catholic University of Lille, France. Antonella Seddone is Assistant Professor at the Department of Cultures, Politics and Society of the University of Turin, Italy.

Routledge Studies on Political Parties and Party Systems

This new series focuses on major issues affecting political parties in a broad sense. It welcomes a wide-​range of theoretical and methodological approaches on political parties and party systems in Europe and beyond, including comparative works examining regions outside of Europe. In particular, it aims to improve our present understanding of these topics through the examination of the crisis of political parties and challenges party organizations face in the contemporary world, the increasing internal complexity of party organizations in terms of regulation, funding, membership, the more frequent presence of party system change, and the development of political parties and party systems in under-​researched countries. Series Editors:  Ingrid van Biezen, Leiden University, the Netherlands, and Fernando Casal Bértoa, University of Nottingham, UK Informal Politics in Post-Communist Europe Political Parties, Clientelism and State Capture Michal Klíma Political Candidate Selection Who Wins, Who Loses, and Under-​Representation in the UK Jeanette Ashe Political Parties Abroad A New Arena for Party Politics Edited by Tudi Kernalegenn and Émilie van Haute Political Incivility in the Parliamentary, Electoral and Media Arena Crossing Boundaries Edited by Annemarie Walter New Paths for Selecting Political Elites Investigating the Impact of Inclusive Candidate and Party Leader Selection Methods Edited by Giulia Sandri and Antonella Seddone

New Paths for Selecting Political Elites Investigating the Impact of Inclusive Candidate and Party Leader Selection Methods Edited by Giulia Sandri and Antonella Seddone

First published 2021 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 605 Third Avenue, New York, NY 10158 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2021 selection and editorial matter, Giulia Sandri and Antonella Seddone; individual chapters, the contributors The right of Giulia Sandri and Antonella Seddone 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: Sandri, Giulia, editor. | Seddone, Antonella, editor. Title: New paths for selecting political elites : investigating the impact of inclusive candidate and party leader selection methods / Edited by Giulia Sandri and Antonella Seddone. Description: Abingdon, Oxon ; New York, NY : Routledge, 2021. | Series: Routledge studies on political parties and party systems | Includes bibliographical references and index. Identifiers: LCCN 2020054377 (print) | LCCN 2020054378 (ebook) | ISBN 9780367901417 (hardback) | ISBN 9781003022893 (ebook) Subjects: LCSH: Political parties–Cross-cultural studies. | Elite (Social sciences)– Cross-cultural studies. | Elections–Cross-cultural studies. | Government executives–Selection and appointment–Cross-cultural studies. Classification: LCC JF2051 .N49 2021 (print) | LCC JF2051 (ebook) | DDC 324.2/2–dc23 LC record available at https://lccn.loc.gov/2020054377 LC ebook record available at https://lccn.loc.gov/2020054378 ISBN: 978-​0-​367-​90141-​7  (hbk) ISBN: 978-​1-​032-​00579-​9  (pbk) ISBN: 978-​1-​003-​02289-​3  (ebk) Typeset in Times New Roman by Newgen Publishing UK

Contents

List of List of List of List of

figures  tables  political parties  contributors 

1 Intra-​party selection methods and political elites: new trends and consequences 

ix xi xiii xviii

1

G I U L I A S AN D R I A N D A N TO N ELLA SED D O N E

1.1  Introduction  1 1.2  The selection of political elites: what we know so far  2 1.3  Political elites: understanding the interaction between party organizations and leadership  5 1.4  Rationales and aims of this volume  7 1.5  Plan of the book  12 References  16

2 Choosing party leaders in Italy between personalization and democratization 

20

MARC O VAL B RU ZZI

2.1  Introduction  20 2.2  Varieties of party models in Italy  21 2.3  Selection methods over time  30 2.4  Demographics characteristics  37 2.5  Conclusions  41 References  43

3 Playing with fire? The organizational consequences of party primaries in Spain  OSC AR BARB E R À A N D JUA N RO D R Í G U EZ- T ​ ERUEL

3.1  Introduction  46

46

vi Contents 3.2  The organizational consequences of introducing party primaries on the party elites’ strength  47 3.3  The introduction of party primaries in Spanish politics  50 3.4  The organizational consequences of first-​time party primaries in Spain  54 3.5  Discussion  58 3.6  Conclusions  61 References  63

4 The outcomes of party primaries: (dis)continuity in top candidates’ political and partisan profiles 

66

JAV I E R MARTÍ N EZ- ​C A N TÓ A N D JAV I ER A STUDIL LO

4.1  Introduction  66 4.2  The theoretical debate: do primaries produce new types of top candidates?  67 4.3  Using Spain to disentangle the role of party primaries  70 4.4  Results  75 4.5  Conclusion  78 References  80

5 Candidate selection, personalization of politics, and political careers: insights from Italy (2006–​2016) 

82

B RU N O MAR I N O, N I C O LA MA RTO C C H I A DIODAT I, AN D LU C A V ER ZI C H ELLI

5.1  Introduction  82 5.2  The Italian case in a cross-​time perspective  84 5.3  Candidate selection, personalization of politics, and patterns of career: is there a connection?  88 5.4  Empirical analysis  90 5.5  Conclusions  97 References  101

6 Impact of intra-​party democracy on patterns of political career within the Pan-​Hellenic Socialist Movement(PASOK) in Greece, 2004–​2009  D I MI T RI OS KO SMO PO U LO S

6.1  Introduction  106 6.2  Intra-​party democratic innovations within PASOK: towards new patterns of candidates’ selection  108 6.3  Intra-​party democracy and changing patterns of candidates’ selection  111

106

Contents  vii 6.4  2004:The primaries as a shift? The ambivalence of new candidates’ selection patterns  114 6.5  To what extent has the profile of MPs changed since 2004?  115 6.6  Conclusions  120 References  121

7 The effect of introducing primaries on the profiles of candidates in the Turkish context: the local experiences of candidates 

124

FARU K   AK SOY

7.1  Introduction  124 7.2  Primaries and political experiences of candidates  125 7.3  Primaries of RPP in Turkey  128 7.4  Data, analysis, and results  132 7.5  Conclusion  138 References  139

8 Do candidate selection modes matter for the gender diversity within political elites? Evidence from the Belgian case in 2014 

142

AU D RE Y VAN D ELEEN E

8.1  Introduction  142 8.2  Gender diversity on the lists: literature review and hypotheses  143 8.3  The case of Belgium in 2014  148 8.4  Method and empirical results: do candidate selection processes matter?  148 8.5  Empirical analysis: how candidate selection impacts gender diversity  152 8.6  Conclusions  154 References  156

9 The use of primaries for the selection of party leaders in the UK Conservative and Labour parties: formal rules and ideological congruence  AG N È S AL E XAN D R E- C ​ O LLI ER A N D EMMA N UE L L E  AVRIL

9.1  Introduction and research questions  160 9.2  The method: the hypotheses of organizational convergence and ideological congruence  161 9.3  An overview of leadership election rules  165 9.4  Conclusions  172 References  177

160

viii Contents

10 Do different candidate selection methods produce different types of candidates? An analysis of the German case 

180

E L I S A D E I SS-​H ELBI G

10.1  Introduction  180 10.2  Candidate selection methods, party ideology, and the representativeness of candidates  182 10.3  Candidate selection in Germany  184 10.4  Data and methods  186 10.5  Results  192 References  201

11 The ‘Belle Époque’ of French primary: the evolution of the national and local framework of candidate selection  204 MARI N O D E  LU C A

11.1  Introduction  204 11.2  Research question, data, and methods  204 11.3  The socialist experiment and the development of the primary in France  206 11.4  The contagion effects in the adoption of primaries by communists, ecologists, and civil society  208 11.5  The primary of others: from the centre-​left to the centre-​right  210 11.6  The local primary from 2014 to 2020  213 11.7  The impact on social and political profiles of candidates  215 11.8  Conclusion  224 References  225

12 Political elites and party primaries 

228

AN T ON E L LA SED D O N E A N D G I U LI A  SA N DRI

12.1  Political parties between personalization and intra-​party democracy  228 12.2  Primaries and political elites: the impact of inclusive selectorates  233 12.3  Preliminary results and notes for a research agenda  235 References  239

Appendix  Index 

244 249

Figures

2.1

Personalization and level of intra-​party democracy in Italian political parties (2019)  22 2.2 Control over policy choices between members and leadership in the main Italian parties (2017–​19)  24 2.3 Control over party policy choices and left-​right position of Italian parties (2018)  24 2.4 Party leader selection methods in Italy, by decade (1960–​2020)  31 2.5 Party leader selection method in Italy, by ideological orientation (1960–2020)  32 2.6 Trend in intra-​party competitiveness and inclusiveness in Italian political parties, 1945–​2020  34 2.7 Intra-​party competitiveness and inclusiveness of the Italian political parties, 1945–​2020  35 2.8 Intra-​party competitiveness and inclusiveness of the Italian political parties, by decade  36 2.9 Age of the Italian party leaders at the time of the first selection (by class age), 1945–​2020  38 2.10 Italian party leaders by level of education, 1945–​2020  38 3 .1 Ideal types of sequences linking party primaries with party elites’ strength and other indirect outcomes  49 4.1 Distribution of party primaries in Spain according to political party and five-​year periods  72 4.2 Spanish top candidates’ social, partisan, and political features across time and selection method  73 4.3 The effects of party primaries on top candidates’ social, partisan, and political features  77 5.1 Intra-​party personalization of politics (the control of parties by their leaders) in Italy, 2006–​16  86 5.2 MPs’ career classes in Italy, 2006–​16  87 5.3 Average marginal effect of PATTERN OF CAREER (Peripherals and Insider Party Politicians only) on the different levels of OFFICE conditioned to SELECTORATE  95

x Figures 5.4 Average marginal effect of SELECTORATE on the different levels of OFFICE conditioned to PARTY PERSONALIZATION  5.5 Average marginal effect of PATTERN OF CAREER on the different levels of OFFICE conditioned to PARTY PERSONALIZATION  7.1 Average marginal effects: local experience  7.2 Average marginal effects: experience in central organization  10.1 Share of immigrant-​origin (IO) and female candidates on party lists/​promising positions  10.2 Type of selectorate –​district nominations (in percent of all district nominations)  10.3 District nomination chances of female and immigrant-​origin (IO) aspirants (binomial regressions)  10.4 Nomination chances in winnable districts of female and immigrant-​origin (IO) aspirants (binomial regressions)  10.5 Inclusiveness of the selectorate and women’s share on party lists (OLS regressions)  10.6 Inclusiveness of the selectorate and IOs’ share on party lists (OLS regressions) 

96 98 136 137 189 192 195 196 197 198

Tables

2 .1 Italian party leaders, 1990–​2020  2.2 Characteristics of the Italian party leaders and selection method, 1990–​2020  2.3 Operationalization of the variables ‘inclusiveness’ and ‘intra-​party competitiveness’  2.4 Occupation of the Italian party leaders, by decade (%), 1945–​2020  2.5 Features of Italian party leaders by selection method, 1945–​2020  3.1 First time party primaries in mainstream Spanish parties  3.2 Spanish mainstream parties: Party primaries main features by 2015  3.3 The internal consequences of party primaries in mainstream Spanish parties: a first qualitative assessment  3.4 Winner’s strength and party cohesion in Spain: a qualitative assessment  3.5 Winner’s strength, party cohesion and future use of party primaries in Spain: a qualitative assessment  5.1 Descriptive statistics  5.2 Ordinal logistic regression on Lower House MPs’ career, Italy (2006–​16)  6.1 Share of MPs according to their birth decade 1996–​2009 (% values)  6.2 Evolution of women share within PASOK and Parliament  6.3 Evolution of selected professions within PASOK’s MPs  6.4 Political seniority and experience of MPs within socialist parliamentary group  7.1 Multinomial regression results without the year of birth  7.2 Multinomial logistic regression results including the year of birth  8.1 Summary of the operationalization of the variables included in the model and of the hypotheses 

26 28 33 39 40 51 53 55 60 60 91 92 117 117 118 119 134 135 152

xii Tables 8.2 Explaining the gender diversity among candidates on the realistic list positions (unstandardized coefficients)  9.1 Conservative leadership election, 2001  9.2 Conservative leadership election, 2005  9.3 Conservative leadership election, 2016  9.4 Conservative leadership election, 2019  9.5 Labour leadership election, 2010 (fourth and final round)  9.6 Labour leadership elections, 2015  9.7 Labour leadership election, 2016  9.8 Labour leadership election, 2020  10.1 Operationalization of variables (party lists)  10.2 Operationalization of variables (district nominations)  10.3 Immigrant-​origin (IO) and female aspirants/​candidates at the district level (only non-​incumbents)  10.4 Type of selectorate district nominations  11.1 The presidential primary elections, 1995–​2017  11.2 The 2014 mayoral primary elections  11.3 The 2020 mayoral primary elections  11.4 Political career and office of presidential nominees before primary elections (1995–​2017)  11.5 Political career and office of mayoral nominees before primary elections (2014)  11.6 Political career and office of mayoral nominees before primary elections (2020)  A.1 List of Italian party leaders and their length in office, 1945–​2020  A.2 List of acronyms, presence and party ideology of the selected parties  A.3 The effects of party primaries on top candidates’ social, partisan and political features according to the strong definition of primaries  A.4 The effects of party primaries on top candidates’ social, partisan and political features according to the soft definition of primaries 

153 166 166 167 168 169 169 170 171 190 190 191 193 212 214 216 218 222 223 244 246 247 248

Political parties

Acronym

Name

AKP (AK Parti)

Adalet ve Kalkınma Partisi

AN AP AP BNG C’s CC CCD CD&V cdH CDU

CHA CHP CiU/​ PDECat

CNIP

Translation

Justice and Development Party Alleanza Nazionale National Alliance Alianza Popular People’s Alliance Adalet Partisi Justice Party Bloque Nacionalista Galician Galego Nationalist Block Ciudadanos Citizens Coalición Canaria Canarian Coalition Centro Cristiano Democratic Democratico Christian Center Christen-​ Christian Democratisch en Democratic & Vlaams Flemish Centre démocrate Humanist humaniste Democratic Centre Christlich Christian Demokratische Democratic Union Union Deutschlands Chunta Aragonese Aragonesista Assembly Cumhuriyet Halk Republican Partisi People’s Party Convergència i Unió Convergence and /​Partit Demòcrata Union/​Catalan Europeu Català European Democratic Party Centre national des National Centre indépendants et of Independents paysans and Peasants

Country

Notes

Turkey Italy Spain Turkey Spain Spain Spain Italy

State-​wide Galicia State-​wide Canary Islands

Belgium

Flanders

Belgium

French-​ speaking Belgium All German states but Bavaria

Germany

Spain

Aragon

Turkey Spain

France

Catalonia

xiv  Political parties Acronym

Name

Conservative Conservative Party CSU DC DL DP DS DSP DYP Ecolo EELV ERC EUiA FAC FD FDF FDG FDI FI FN Greens Groen ICV IDV IU IV Labour Lega LEU LFI LibDem LN LR

Translation

Country

Notes

Conservative Party United Kingdom Christlich-​Soziale Christian Social Germany Bavaria Union in Bayern Union Democrazia Christian Italy Cristiana Democracy La Margherita –​ Daisy –​ Italy Democrazia è Democracy is Liberta Freedom Demokrat Parti Democrat Party Turkey Democratici di Democrats of the Italy Sinistra Left Demokratik Sol Democratic Left Turkey Parti Party Doğru Yol Partisi True Path Party Turkey Ecolo Ecology Party Belgium French-​ speaking Belgium Europe Écologie Les Europe Ecology –​ France Verts The Greens Esquerra Catalan Spain Catalonia Republicana de Republican Left Catalunya Esquerra Unida i Alternative and Spain Catalonia Alternativa United Left Foro Asturias Asturias Forum Spain Asturias Front démocrate Democratic Front France Fédéralistes Francophone Belgium French-​ Démocrates Democratic speaking Francophones Federalists Belgium Front de gauche Left Front France Fratelli d’Italia Brothers of Italy Italy Forza Italia Go Italy Italy Front national National Front France Greens Greens United Kingdom Groen Green Belgium Flanders Iniciativa per Catalan Spain Catalonia Catalunya Verds Initiative-​Greens Italia dei Valori Italy of Values Italy Izquierda Unida United Left Spain Statewide Italia Viva Italy Alive Italy Labour Party Labour Party United Kingdom Lega League Italy Liberi e Uguali Free and Equal Italy La France insoumise Unbowed France France Liberal Democrats Liberal Democrats United Kingdom Lega Nord Northern League Italy Les Républicains The Republicans France

Political parties  xv Acronym

Name

Translation

Country

LREM

La République en marche Movimento 5 Stelle

The Republic On the Move Five Star Movement Democratic Movement Reformist Movement

France

Italian Social Movement New Democracy New Flemish Alliance Open Flemish Liberals and Democrats Aragonese Party Panhellenic Socialist Movement Christian Democratic Party Communist Party of Spain French Communist Party Italian Communist Party Democratic Party People of Freedoms Democratic Party of the Left Ecologist Party Left Party Italian Liberal Party Monarchist National Party Basque Nationalist Party We can Popular Party Italian Popular Party Riojan Party Regionalist Party of Cantabria

Italy

M5S MoDEM MR MSI ND N-​VA Open Vld PAR PASOK PCD PCE PCF PCI PD PDL PDS PE PG PLI PNM PNV Podemos PP PPI PR PRC

Mouvement démocrate Mouvement Réformateur Movimento Sociale Italiano Nea Dimokratia Nieuw-​Vlaamse Alliantie Open Vlaamse Liberalen en Democraten Partido Aragonés Panellinio Sosialistiko Kinima Parti chrétien-d ​ émocrate Partido Comunista de España Parti communiste français Partito Comunista Italiano Partito Democratico Partito delle libertà Partito Democratico di Sinistra Parti écologiste Parti de gauche Partito Liberale Italiano Partito Nazionale Monarchico Partido Nacionalista Vasco Podemos Partido Popular Partito Popolare Italiano Partido Riojano Partido Regionalista de Cantabria

Notes

Italy France Belgium

French-​ speaking Belgium

Greece Belgium

Flanders

Belgium

Flanders

Spain Greece

Aragon

France Spain

State-​wide

France Italy Italy Italy Italy France France Italy Italy Spain Spain Spain Italy Spain Spain

Basque Country State-​wide Statewide La Rioja Cantabria

xvi  Political parties Acronym

Name

Translation

PRG

Radical Party of France the Left Republican Italian Italy Party

PS

Parti radical de gauche Partito Repubblicano Italiano Parti Socialiste

PS

Parti socialiste

PRI

PSC PSDI PSI PSLI PSM/​Mes PSOE RAD RC RP SC SHP SNP SODEP sp.a SPD UDB UDC

Socialist Party

Country

Belgium

French Socialist France Party Partit dels Socialistes Catalan Socialists Spain de Catalunya Party Partito Italian Democratic Italy Socialdemocratico Socialist Party Italiano Partito Socialista Italian Socialist Italy Italiano Party Partito Socialista dei Socialist Party of Italy Lavoratori Italiani Italian Workers Partit Socialista de Socialist Party of Spain Mallorca /​Més Majorca/​More per Mallorca for Majorca Partido Socialista Spanish Workers’ Spain Obrero Español Socialist Party Radicali Radical Party Italy Rifondazione Communist Italy Comunista Refoundation Refah Partisi Welfare Party Turkey Scelta Civica Civic Choice Italy Sosyaldemokrat Social Democrat Turkey Halkçı Parti People’s (Populist) Party Scottish National Scottish National United Party Party Kingdom Sosyal Demokrasi Social Democracy Turkey Partisi Party Socialistische Partij Socialist Party Belgium Anders Different Sozialdemokratische Social Democratic Germany Partei Party of Deutschlands Germany Union démocratique Breton Democratic France bretonne Union Unione dei Union of the Italy Democratici Center Cristiani e Democratici di Centro

Notes

French-​ speaking Belgium Catalonia

Balearic Islands Statewide

Flanders All 16 German states since 2008: Unione di Centro

Political parties  xvii Acronym

Name

Translation

Country

UDE

Union des démocrates et des écologistes Union des démocrates et indépendants Union pour un mouvement populaire Unión del Pueblo Navarro Vlaams Belang)

Union of Democrats and Ecologists Union of Democrats and Independents Union for a Popular Movement Navarrese People’s Union (Flemish Interest

France

UDI UMP UPN VB

Notes

France France Spain

Navarre

Belgium

Flanders

Contributors

Faruk Aksoy is a PhD candidate in the Political Science department at Sabancı University. He received his BA in Political Science and his minor degree in Psychology from Bilkent University in 2013. He completed his MA in Political Science at Sabancı University in 2015. His primary research interests are political psychology, institutions of democracy, and voting behaviour. Agnès Alexandre-​Collier, a former visiting scholar at St Antony’s and Pembroke Colleges (Oxford), is Professor of British Civilisation and Politics at the University of Burgundy (Dijon, France). Her main research interest is in the British Conservative Party with a special focus on the organizational impact of European integration and more extensively party organizational changes in relation to populism. She is the author of several articles and books including recently:  Leadership and Uncertainty Management in Politics, Leaders, Followers and Constraints in Western Democracies, co-​edited with François Vergniolle de Chantal (Palgrave, 2015)  and Politics Reinvented. When Innovations Reshape Representative Democracy, co-​edited with Alexandra Goujon and Guillaume Gourgues (Routledge, 2020). Javier Astudillo is Associate Professor of Political Science at Universitat Pompeu Fabra (Barcelona, Spain). He is the author of various works on interest groups and political parties published peer-​review international journals and edited books, and with Prof. Tània Verge, the Spanish coordinator of the Political Party Database Project (www.politicalpartydb.org/​). He is currently working on the relationship between party organization and chief executives at national and regional levels in Western parliamentarian democracies. Emmanuelle Avril is Professor of Contemporary British Politics and Society at the Sorbonne Nouvelle University in Paris. Her main line of research is intra-​party democracy, organizational change, political leadership, and mobilization within the Labour Party, which has come to include the

Contributors  xix study of the impact of new technologies on party membership, activism, and political participation in general. She is the current director of the Centre for Research in the English-​ speaking world (CREW). Recent publications include Labour United and Divided from the 1830s to the Present (Manchester University Press, 2018, co-​edited with Y.  Béliard) and Democracy, Participation and Contestation. Civil Society, Governance and the Future of Liberal Democracy (Routledge, 2016, co-​edited with J. Neem). Her latest book, co-​authored with Pauline Schnapper, was Où va le Royaume-​Uni? Le Brexit et après? (Odile Jacob, 2019). Oscar Barberà is Associate Professor at the Universitat de València (UV). He has also been teaching and visiting at several Spanish and European Universities. His main areas of interest are party politics, decentralization, and political elites. His PhD won the award of the Spanish Political Association and was published in 2011 by the Spanish Centro de the Investigaciones Sociológicas. He has published extensively on Spanish party politics. His latest contributions on Routledge are chapters in the following books:  Party Primaries in Comparative Perspective and Party Members and Activists. Elisa Deiss-​Helbig is a postdoctoral researcher at the University of Stuttgart. Her main research interests are party research (electoral pledges, candidate selection, and far right parties), and political representation with a focus on descriptive representation. Her current research focuses on how parties’ electoral pledges and their fulfillment reflect patterns of political inequality. Marino De Luca is Marie Sklodowska-​ Curie Research Fellow at the Department of Politics –​School of Law, Politics, and Sociology –​University of Sussex. His main research interests concern political parties, elections, intra-​party democracy, political communication, and populism. His works have been published in French Politics, Italian Political Science Review, Communist and Post-​Communist Studies. He is currently the Secretary of the Italian Electoral Studies Association (SISE). Dimitrios Kosmopoulos holds a PhD in Political Science from the Paris-​ Dauphine University-​ PSL. He is attached to the Institute for Interdisciplinary Research in Social Sciences (IRISSO). He was a visiting research fellow at the Centre for Southeast European Studies of the University of Graz during the winter semester 2019/​2020. His doctoral research focused on the political upheaval in Greece between 2010 and 2014, a period marked by economic crisis and the implementation of structural and economic adjustment programmes. His research interests include the local dimension of party politics, the study of party systems and political crisis, the sociology of political elites, as well as the study of European politics in a comparative perspective.

xx Contributors Bruno Marino has been a post-​doctoral researcher at Unitelma Sapienza University in Rome. He has published articles in journals such as Electoral Studies, European Political Science Review, Regional and Federal Studies, Italian Political Science Review/​Rivista Italiana di Scienza Politica. His main research interests are related to parties and party systems in comparative perspective, political elites, and the personalization of politics. Javier Martínez-​Cantó is a post-​doctoral researcher in the research group on Comparative Politics at the University of Konstanz. His research interest includes party organization, candidate selection, legislative behaviour, and federalism. Before he had been a doctoral fellow at the Bamberg Graduate School of Social Sciences (University of Bamberg) and he has been a visiting researcher at Princeton University. He obtained a BA and a MA in political science at the University of Valencia and Pompeu Fabra University. His work has been published in Journal of Elections, Public Opinion and Parties and Regional and Federal Studies. Nicola Martocchia Diodati is a data scientist and has carried out teaching activies in Quantitative Methods and Machine Learning courses in several universities. He holds a PhD in Political Science and Sociology from the Scuola Normale Superiore and has published articles in journals such as West European Politics, European Union Politics, Electoral Studies, European Political Science Review. His main research interests are related to quantitative methods, machine learning, electoral behaviour, and political elites. Juan Rodríguez-​Teruel is Associate Professor at the University of Valencia. He has also been lecturer at the Autonomous University of Barcelona and visiting fellow at the LSE, Edinburgh, Leiden, the ULB, and Nottingham. His main areas of research are political elites and party politics. He is author of Ministers in Democratic Spain (Linz Award 2007, in Spanish), and co-​author of Party Finance and Democratic Transparency. He has published on political elites and Spanish party politics. His latest contributions for Routledge are chapters in the following books:  Party Primaries in Comparative Perspective and Party Members and Activists. Giulia Sandri is Associate Professor at the European School of Political and Social Sciences of the Catholic University of Lille. She was previously Research Fellow at Christ Church and at the DPIR of the University of Oxford. Her main research interests are party politics, intra-​party democracy, and political behaviour. She recently published in Politics and Policy, Acta Politica, Comparative European Politics, Religion, State and Society, Ethnopolitics, Italian Political Science Review, and Regional and Federal Studies. She also recently co-​edited with Luca Tomini the book Challenges of Democracy in the 21st Century:  Concepts, Methods, Causality and the Quality of Democracy (Routledge, 2018).

newgenprepdf

Contributors  xxi Antonella Seddone is Assistant Professor at the Department of Cultures, Politics and Society of the University of Turin. Her research interests focus on political parties and organizational changes. She also researches on political communication, with specific reference to election campaign, public debate, and populist communication. She is co-​convenor of the Italian Politics Specialist Group within the Political Studies Association and she is co-​chair of the Conference Group of the Italian Politics and Society of the American Political Science Association. Recently she published in: International Journal of Press/​ Politics, Acta Politica, Parliamentary Affairs, European Politics and Society. She co-​edited (with G. Sandri and F. Venturino) the volume Primaries in Comparative Perspective (Ashgate, 2015). Marco Valbruzzi is an Assistant Professor of Political Science at the University of Naples –​Federico II. His current research interests focus on intra-​party politics and the transformation of West European party systems. He has recently co-​edited (with R. Vignati) Il vicolo cieco. Le elezioni del 4 marzo 2018 (Il Mulino, 2018). His latest book is A Changing Republic:Politics and Democracy in Italy (Epoké, 2015, with G Pasquino). Audrey Vandeleene is a senior researcher at the Department of Political Science of Ghent University (Belgium). She graduated from UCLouvain (Belgium) and was a postdoctoral research fellow at Lund University in Sweden. Her research interests cover party politics, intra-​party democracy, electoral systems, women-​in-​politics, federalism, and political discourse. Some of her works were published in Representation, American Behavioral Scientist, Politique et Sociétés, Basic Income Studies. She was co-​editor of the Palgrave Macmillan book Candidates, Parties and Voters in the Belgian Partitocracy. Luca Verzichelli is Professor of Political Science at the University of Siena. He holds a PhD in Political Science from the University of Florence and he has extensively published in the fields of Political Elites, Comparative Political Institutions and Budgetary Politics. He is the author of a few monographs, and about 40 articles on international journals, such as European Journal of Political Science, West European Politics, South European Societyand Politics, Journal of Legislative Studies, Swiss Political Science Review.

1  Intra-​party selection methods and political elites New trends and consequences Giulia Sandri and Antonella Seddone

1.1  Introduction A crucial function of political parties is the recruitment of political elites and political leadership (Bartolini and Mair 1990; Norris 2004). If voters are the central players in elections, political parties play a major role before the elections by determining who will run in their name –​i.e. the candidates –​and who will lead the party campaign and electoral platform –​i.e. the party leader. However, so far, this key function of parties has been studied mostly with regard to parliamentary recruitment, usually by examining the background of elected members of the legislatures or the ministers (Matthews 1984; Best and Cotta 2000; Dowding and Dumont 2014). Yet, the relation between the recruitment of political elites and the types of selection procedures set up by parties in the final stages of the recruitment processes requires a special attention. This is especially true in times of populist claims where elites in general, and political elites in particular, are constantly blamed for the malfunctioning of democratic processes. Who rules matter, and how they are selected, thus, matters too. Contemporary political parties in advanced democracies have undergone dramatic changes (Dalton 2014). Recent research suggests that party organizations have changed and, more specifically, they have become more internally democratic (Hazan and Rahat 2010). Participatory and deliberative democratic innovations have been a central research object in political science for decades now (Fishkin 2009), but the study of their use within party politics is relatively recent. The expansion of intra-​party democracy, the adoption of new forms of e-​participation and the introduction of new types of party membership are nowadays characterizing party politics in many advanced democracies (Scarrow 2014). These internal democratization processes affect all the functions performed by parties and particularly those of selection of political leadership and of recruitment of political elites. This edited volume focuses precisely on candidate and leader selection, rather than more broadly addressing the topic of candidate or leader recruitment. Indeed, political recruitment and elite selection constitute two different intra-​party processes, which are however interconnected (Norris 1997). For

2  Giulia Sandri & Antonella Seddone example, candidate selection in a primary election can be open and democratic, but where those candidates come from –​their recruitment –​and how they get to be on the primary ballot will not necessarily be open or democratic (see for instance, Vandeleene 2016). Rather, recruitment may involve stringent party acceptance, precluding non-​enrolled supporters, for instance, from walking into a party office and becoming a viable party candidate. In this book, the theoretical, methodological, and empirical scope of research is focused on candidate and leadership selection processes, with the aim to assess whether and to what extent different procedures may have a diverse impact on the very features of the political elites selected. The most innovative and inclusive method for selecting party candidates and leadership is represented by primary elections. However, the dynamics and consequences of such selection methods vary substantially from case to case, according to country, party, election, and territorial level parameters. More specifically, previous literature has shown that there is significant variation in the types of primary institutions that can be used to select party leaders and elites as well as variation with respect to the types of leaders or elites who emerge on the winning side of a primary election. For example, recent empirical research on primary elections in comparative perspective has shown that such inclusive types of candidate and leadership selection vary considerably with regard to several dimensions: from their regulation (party or state), to the dimension of candidacy and deselection rules, to the degree of inclusiveness of the selectorate, to the decentralization, as well as to the level of competitiveness of the process (Kenig 2009; Hazan and Rahat 2010; Sandri et al. 2015; Cross et al. 2016; Boatright 2018). In addition, other studies have empirically explored the significant variations in terms of representativeness (with respect to sociopolitical profiles) and of responsiveness (with respect to legislative behaviour in particular) of the leaders or candidates selected through primaries or other inclusive selection methods. However, the existing literature has done so mostly by exploring case studies or by the way of small N comparisons. Most research so far focused in particular on the primary processes and their peculiarities. However, understanding the possible impact in terms of the sociopolitical features and careers patterns of the political elites selected through different methods remains underinvestigated (Cordero and Coller 2014; Smith and Tsutsumi 2014; Spies and Kaiser 2014; Rombi and Seddone 2017; Alexandre-​Collier 2016; Ahler et  al. 2016; Akirav 2018 Cordero and Coller 2018; Coller et al. 2018).

1.2  The selection of political elites: what we know so far The topic covered by this edited book represents an important aspect of the political transformations of contemporary democracies. In addition, it refers to one of the increasingly observed phenomenon in party and electoral politics: the personalization of politics and its effects on political elites. Recent contributions and ongoing research projects prove this is a growing

Intra-party selection methods  3 research avenue; see for instance Rahat and Kenig’s recent book on personalization (2018), or Bennister et al.’s (2017) book on leadership power, both published by OUP. The diffusion of intra-​party democracy and in particular the increasing use of inclusive methods for selecting candidates and party leaders stimulated a fertile literature that, while aiming to understand the kind of organizational changes occurring within political parties, also attempted to provide an understanding about the consequences of such changes. We build our volume and its theoretical framework on a series of research pieces that in recent years have defined the research agenda on candidate selection methods, but we also aim to develop the knowledge on his field. The academic investigation on these issues developed quite recently outside US. On this regard, it is necessary to recall the Special Issue of Party Politics that in 2001 provided new insights on candidate selection methods –​and more in general about primaries –​without a specific reference to US cases. Indeed, the main concepts and theoretical assumptions needed to be re-​adapted to a different sociopolitical contexts, where these inclusive tools have been adopted and declined in various ways and have determined diversified consequences at the political system and party organization level. The edited volume by Siavelis and Morgenstern (2008), which covers candidate recruitment and selection in Latin America, provides another crucial contribution to the literature. The strength of this book lies in the clarity of its aims. It is clear from the beginning that this work focuses on a particular region, with theoretical and empirical considerations tailored thereupon. The main weakness of this volume is its narrow focus on one region, in which the generalizable lessons to other parts of the democratic world only apply given a similarity in conditions relative to cases like Brazil or Argentina (or any of the cases in this book). The seminal volume by Hazan and Rahat (2010), in this sense, provide the most comprehensive framework of analysis on this issue. The two authors suggest interpreting the consequences of candidate selection methods in the light of four different dimensions: participation, representation, competition, and representation. Their theoretical argument contributed to developing a set of empirical studies aimed to deepen the investigation on candidate selection procedures and party organizational change in particular. We should thus consider Hazan and Rahat’s work as a seminal inspiration for this volume. Hazan and Rahat effectively discuss the consequences of varying degrees of inclusiveness and selection processes, and their effects on elites and political behaviour. More recently, William Cross and Jean-​Benoit Pilet (Pilet and Cross 2014, Cross and Pilet 2015)  coordinated a research network (COSPAL) aimed at collecting data on party leader selection methods. Their extensive empirical research allowed to clarify the mechanisms of party leadership selection and their impact at various levels, such as the effects in terms of sociopolitical profile of party leaders, the effects in terms of electoral performance of

4  Giulia Sandri & Antonella Seddone leadership renewal, or the consequences in terms of gender representation, thus widening the theoretical and empirical perspectives of research on this field. Cross et al. (2016) in particular have addressed the analysis of selection methods and primary elections in a recent book that adopts a comparative approach, focusing in particular on Canada and Israel as case studies. They explore dimensions such as elites’ selection rules, financing, and effects at organizational party level. More specifically, they provide a theoretical background on candidate selection methods, clarifying the most central concepts and analytical categories used in this subfield of the literature. Sandri et al. (2015) on the contrary took into consideration the effects of candidate and leader selection methods on electoral performances and membership rates, with the aim to assess whether and to what extent primaries could keep their promises in helping parties to respond to citizens’ dissatisfaction and improving their public image. Recently, the volume edited by Scarrow, Webb, and Poguntke (2017) presents the data collection carried out by the research network Political Party Database, which also devoted attention on candidate selection methods. The edited book identified the latter as crucial aspects of party functioning and as one of the most relevant organizational changes that scholars have to take into account when approaching the study of political parties and democracies in general. The crucial reading on the topics covered by the present book is the Routledge Handbook on Primary Elections (Boatright 2018). This book represents the first academic work offering to scholars a complete overview on primary elections and candidate selection methods around the world. The presentation of the most relevant theoretical and analytical approaches for understanding primary elections is combined with an extensive analysis of interesting case studies. The strength of this book is the clarity of its aim and its coverage of theoretical and methodological issues surrounding the field of candidate selection and primary elections, along with its illustration of these issues through country case studies from around the democratic world (ranging from Canada to Iceland, Italy, and Ghana). Finally, the volume by Coller, Cordero, and Jaime-​Castillo (2018) focuses on the issue of candidate selection methods by taking into consideration new parties that emerged after (and as a consequence) of the Great Recession. They provide an assessment of organizational features of new parties and they investigate the potential ‘contagion’ effect. They also provide an exploratory analysis on the impact of these inclusive methods on the social structure of the parliamentary elites, deriving some interesting conclusions about the potential renewal driven by such new (and inclusive) procedures for selecting candidates. Cordero et al.’s book takes a similar approach than the Boatright handbook by first considering the effect of economic and political triggers on the dynamics of candidate selection from a theoretical point of view using illustrations from mostly European cases, but also featuring analyses from Latin America and the United States. The weakness of this volume lies in only addressing candidate selection and not leadership selection. Additionally, it

Intra-party selection methods  5 emphasizes the post-​Great Recession time frame, and thus does not address the nature of relative ‘starting points’ for each of the selected case studies. Clearly, the topic of candidates and party leaders’ selection methods has raised new and significant scholarly attention in recent years. This is due to the increasing diffusion of such methods beyond the traditional US context. Party politics and elections scholars required new paradigms and new analytical frameworks to clearly assess the causes and consequences related to the adoption and use of these innovative inclusive procedures for selecting the political elites outside the US. The issue could be explored by several and different viewpoints. However, the theoretical reflection so far has only seldom be combined with proper and defined empirical research, which up to now has been mainly focused on a limited set of dimensions (i.e. the electoral bonus, impact on membership rates). The impact of these new procedures on the political elites has been investigated mainly in terms of representation, and only rare articles or research papers attempted to explore dimensions such as the internal competition, the incumbency effects or the impact on career patterns. Our volume attempts to enlarge the focus so far developed in literature in order to cover different dimensions besides the classical approach on representation. This would allow party politics and elites’ scholars to benefit from a more in depth understanding of the functioning of these methods.

1.3  Political elites: understanding the interaction between party organizations and leadership As mentioned above, this volume aims to deepen the investigation of the link between political organizations and political elites, by studying the role of parties in parliamentary and political selection at all government levels (national regional, local) and its impact on the political leadership appointed. The empirical chapters aim to offer a new and original understanding on how the kind of tools, rules, methods, and processes implemented for selecting candidates or party leaders may affect the features of political elites in terms of sociopolitical profile and patterns of careers as well. The trend towards more inclusive methods of selection of party candidates and leaders in several countries across the world, and the increasing data availability on intra-​ party democratic innovations (on primary elections for instance) allow us to explore the relation between party organizations and the recruitment of political elites. Developing new research on the consequences of these parallel trends seems now both empirically relevant and feasible. This volume bridges the two sub-​disciplines, party politics and political elite studies, by focusing on the political selection to party mandates and to elective offices of the state at all levels from the point of view of intra-​party organizational settings. This volume builds upon a broad stream of literature in the two sub-​ disciplines of party organizational studies and of political leadership. In particular, the research questions addressed by the book stem from literature on political personalization and presidentialization of political parties, which

6  Giulia Sandri & Antonella Seddone links organizational challenges currently faced by parties in Western democracies with the increasing centrality of party leaders and candidates in electoral and political processes (Poguntke and Webb 2005; Blondel and Thiébault 2009; Rahat and Kenig 2018). These different strands of research contribute to the growing and somewhat blurred body of literature on the ‘personalization’ of politics (Karvonen 2010; Rahat and Kenig 2018). Within this peculiar approach to the study of party politics and political leadership, the issues of leadership selection, change, turnover, and renewal are seldom explored. The second stream of literature upon which this volume is built upon is the burgeoning one on party organizational changes and intra-​party democratization (for an overview, see Cross and Katz 2013; Gauja 2016; Alexandre-​ Collier et al. 2020). Candidate and leadership selection processes and primary elections in particular are a recurrent theme in the debate about parties and their organizational changes (Katz and Mair 1995; Scarrow et  al. 2000; Bolleyer 2012). The debate on the consequences of intra-​party democracy, more in general, arose within the early, classical works on party politics (Ostrogorski 1902; Michels 1962 [1911]; APSA 1950) and has resurfaced in the last two decades as a spinoff on the literature on party politics (Cross and Katz 2013; Hazan and Rahat 2010; Kenig et al. 2016). The arguments and theories developed so far for explaining the consequences of intra-​party democratization are different, but the empirical evidence is overall mixed (Scarrow 2005; Scarrow et al. 2017). More specifically, the volume builds upon comparative literature on candidate and leadership selection procedures and primary elections in particular. Part of this literature has analysed primary elections as an instrument of organizational adaptation of parties that combines processes of political personalization with new forms of intra-​party democracy and participation (Katz and Mair 1995; Kenig 2009; Sandri et  al. 2015; Cross et  al. 2016). The variety of primary elections is argued to create a direct link between candidates and leaders selected through such inclusive methods and their selectors, namely voters, which undermines the role of party organizational structures in electoral processes (Katz and Mair 1995; Ware 2002). This suggests the existence of a close relation between the democratization of leader and candidate selection procedures  –​and specifically the use of primaries –​and processes of political personalization. Rahat and Hazan (2010) have developed a theoretical framework for assessing the consequences of the variety of candidate selection processes that needs further empirical testing but which constitutes an excellent starting point for cross-​country research. In particular, this study elaborates a useful analytical framework for exploring the consequences of the variation in the types of selection processes (and specifically primary institutions) that can be used to choose candidates on the types of elites who emerge on the winning side of internal elections, mostly in terms of representativeness of political elites. This stream of the literature proves very useful for this volume since the latter aims at analysing to what extent more inclusive procedures of candidate

Intra-party selection methods  7 (and leaders’) selection ensure a greater representativeness of political elites. The main focus of research has however remained on gender representation (Rahat and Hazan 2010; Wauters and Pilet 2015; Gauja and Cross 2015), while this volume aims at developing further the analysis of the consequences of elites’ selection processes and primary elections in particular on other dimensions of political representation (political profiles, career patterns, seniority, etc.). With regard to this point, the volume also builds upon the rich American-​ centred literature (both US and Latin America related) on the ideological positioning and political profiles of candidates selected through various types of primary elections (Ranney 1972; Marshall 1978; Norrander 1989, 1993; Care and Polga-​Hecimovich 2006; Steger 2007; Kemahlioglu et al. 2009; Serra 2011). In particular, the volume provides an opportunity to further examine –​ with regard to different case studies and cross-​national analysis –​the research questions developed by Gerber and Morton (1998) and Kaufmann et al. (2003) on the ideological extremism of candidates selected through more open and inclusive types of primaries. More recently, studies by Cross and Pilet (Pilet and Cross 2014; Cross and Pilet 2015), specifically dedicated to the selection processes of party leaders, has collected a relevant amount of comparative data on procedures and rules managing this dimension of intra-​party politics and also on the individual sociopolitical features of the selected leaders.

1.4  Rationales and aims of this volume The volume aims at bridging the above-​mentioned strands of political science literature by focusing on the relation between primary elections and patterns of personalization of politics, and in particular by developing theoretical and empirical analyses of the following research question: What effects do primaries have on the selection of political elites? This main research question can in turn be broken down to a set of subquestions: Do parties with primary elections choose as candidates and leaders people with different characteristics than ones without primaries? How do primaries affect the way parties are run by their elites? How do primaries affect the way leaders control the strategy and policy of parties? Do primaries enhance processes of renewal of the political elites? How do primaries differ from more exclusive and centralized candidate selection methods? The added value of this edited volume to the literature on party politics and political elites is twofold. First, the book constitutes an original attempt not only at bridging two important strands of the scholarly literature, but also at developing a theoretically grounded and analytically structured comparative analysis of the consequences of the use of much debated democratic innovations such as primary elections. The scholarly literature on party democratic innovations and primary elections has been rapidly evolving recently (see for instance: Sandri et al. 2015; Gauja 2016; Cross et al. 2016; Scarrow et al. 2017). How­ ever, the study of the consequences of the adoption of intra-​party democracy

8  Giulia Sandri & Antonella Seddone instruments, especially from a comparative perspective, is still understudied and/​or mostly limited to case studies (Ramiro 2016; Indriðason and Kristinsson 2015; Gomez and Ramiro 2019; Deseriis and Vittori 2019). Second, this book collects a range of different case studies and is based on recent and original empirical observations providing in-​depth national case studies, which analyse the origins and effects of these organizational innovations on the reshaping of political elites and thus on representative institutions. Belgium, France, Germany, Greece, Italy, Spain, Turkey, and UK represent some pertinent cases for investigating the change that intra-​party democracy brought by within parties and its impact on political elites. These eight country cases include both more advanced and transitional democracies, both parliamentary and presidential systems, both majoritarian and RP electoral systems and include some degree of geographical variation within the European continent (while focusing on the Southern part of Europe). What makes the case selection of this volume particularly fitting is the different degree of institutionalization of inclusive candidates’ selection methods used in the countries considered. Indeed, we investigate case studies as the Belgian one where party (closed) primaries are recognized within the political routine, but we also focus on cases such as France, Germany, Italy, Spain, and the United Kingdom where the introduction of primaries is more recent, and became common at different levels with interesting signs of contagion among parties form different party families. We also examine cases such as Turkey and Greece where primaries are at an embryonal and experimental stage (see below for a more detailed description of case selection justification). This case selection of course does not allow carrying out comparative analyses, but the chosen cases permit to assess the potential effects of changes in candidate and party leader’s selection methods in different settings, taking into account the different institutionalization of the procedures within a given political system. Moreover, the analysis covers the use of open and closed primaries at different territorial levels of government, from local to national level. Both the party elites and parliamentary elites are empirically investigated allowing for better understanding of the kind of effects that inclusive procedures for selecting candidates and party leaders may have on political elites, in terms of their sociopolitical profile, career patterns or competition strategies within and between parties. In particular, for those countries, such as Italy and Spain, that present a quite extensive use of inclusive procedures for selecting both candidates and party leaders we offered two different chapters exploring the two dimensions. Each country offers a different angle on the general research question detailed above, so that the volume as a whole offers a complete overview on the issue. With regard to the empirical grounds and scope of this edited book, there are three points that need to be clarified further: the rationales for case selection, the differences in the empirical study of leadership and candidate selection processes and the generalization potential of this volume.

Intra-party selection methods  9 First, candidate and leadership selection processes are similar but essentially separate processes. Fundamentally, the difference between candidate and leadership selection is that the first set of processes is meant to select a multitude of individuals who will be running for several elected mandates at once, while the second is aimed at choosing one or several nominees but for a monocratic party mandate (or elected mandate, in the case of selections of PM candidates). Moreover, candidate selection draws from existing or prospective rank-​and-​file members or activists, which can more easily be mobilized from among non-​members or outsiders. Leadership selection is only rarely based upon input from party outsiders–​and where it is, such as in the United States, those candidates are usually vetted by party insiders before being allowed to compete in open primaries. At any rate, while acknowledging these differences, the present book aims at developing an analytical framework for empirically assessing the two sets of processes at the same time. This has been done before in the literature on primaries (see for instance Sandri et al. 2015; Cross et al. 2016) and several primary election experts have shown that this is feasible from a theoretical and methodological point of view, albeit it requires a few additional justifications in the analytical model (Kenig et  al. 2015). Moreover, in this book closed and open primaries are considered and analysed altogether, while taking into account the potential differential outcomes in terms of their consequences on parties and their elites (Sandri et al. 2015; Bernardi et al. 2017). Crucially, this volume assembles work on candidate and leadership selection concurrently–​ in some chapters combining the two –​thus providing potential readers with a very general overview of the introduction of primaries and their effects on the makeup of party leadership and rank-​and-​file. Second, with regard to the empirical chapters’ case selection, any comparative, cross-​country research design for studying party primaries is strictly limited by the fact that, unlike general elections, primaries are present only in some political systems across the world and only within some parties within those countries. This entails a significant lack of data and the general need for intensive case studies. The unequal, asymmetric use of primaries –​ which varies across time, is limited in geographical terms and can be totally absent in some political systems –​significantly constrains the possibilities of selecting cases in the framework of comparative research designs. It compels to limit the scope of analysis to the empirically available cases of political systems where at least one party has adopted such peculiar instruments of democracy. That said, among the empirically available cases of political systems where at least one party uses primaries, we decided to limit the geographical scope of our comparative research design to European countries. That is why existing and relevant cases elsewhere in the world, i.e. in the United States, but also in Iceland, Korea, Japan, and Canada and in many countries in Latin America, are not included in the analysis. This choice is aimed at allowing the inclusion of case studies covering both advanced and transitional democracies, both

10  Giulia Sandri & Antonella Seddone parliamentary and presidential systems, both majoritarian and RP electoral systems, but without integrating political systems with excessive variation in terms of state institutions, cultural traditions, and party politics. Within the European continent, the case selection is implemented based on one main contextual variable:  primary elections’ institutionalization in a given party system, both in terms of the number and proportion of parties using them and the length of time during which primaries have been organized. The differences in the degree of diffusion and institutionalization of primary elections in various political systems constitute a pertinent explanatory factor that is taken into account in the analytical framework of this study. Therefore, the case selection in the empirical chapters in terms of countries, parties and periods covered is operated based on the variation in the degree of diffusion, institutionalization, and development of primaries in European countries (within which we integrate Turkey, which is a rather understudied case on this topic). This actually contributes to the explanatory potential of our main independent variable, namely the degree of inclusiveness of the selectorates and primary elections processes. This also explain the focus of the empirical parts of the book on Southern Europe, where –​with the exception of Iceland and Finland, which are excluded for other reasons –​ can be found most of the party systems where primary elections are more institutionalized. Consequently, the book’s research design includes case studies that provide empirical examples of party systems in which primary elections, both for candidate and leadership selection, are fairly institutionalized: Italy and Spain. Then, the book also covers case studies in which primaries are less institutionalized in the respective party systems: Greece (in which primaries are used since the early 2000s, but essentially by one party, the PASOK) and Turkey. The case selection of this book covers also case studies located in Northern and Continental Europe and which are endowed with different degrees of institutionalization of party primary systems. This is done by including one case of primaries for candidate selection in a party system where they are fairly institutionalized (Belgium, where closed primary elections are used by most parties for selecting party leaders candidates since the 1990s) and one case where primaries for selecting leaders are relatively new but have attracted significant political and media attention (the United Kingdom). Moreover, the Belgian case is contextually comparable to Spain in terms of being a consociational democracy with distinct regions that engender a certain degree of political tension. Also, the British case is included because of the institutional configuration of the political and party system in the United Kingdom, which is completely different from the rest of continental Europe, and allows thus to test for contextual differences and their impact on the interaction between primaries and political elites. The book also includes two other selected cases of Continental Europe party systems using primaries in a rather institutionalized way: France and Germany, where primaries have been used quite extensively and at different

Intra-party selection methods  11 territorial levels of government (local, state/​regional, and national), but in a more irregular and less steady manner than in Italy and Spain. These two cases also allow controlling for other features of the political system and government type (e.g. federal and regionalized unitary systems, semi-​presidential systems, extensive state regulation of parties and party finance). Finally, the case of Turkey is included because, albeit clearly an outlier given the state of democratic backsliding in that country, it provides some relevant similarities in the way primary elections function even though the political development path is remarkably different from any of the other cases featured in this study. Overall, in Germany, France, Italy, and Spain the use of party primaries is more diffused and at least partially institutionalized, while in Turkey or Greece they represent more experimental organizational features of some specific parties (often of only one party in the selected country). In the study of primaries, few countries, such as Italy or Spain, are characterized by a more systemic and institutionalized use of such instrument, which is both intensive (in terms of the number of parties using them) and extensive (in terms of cross-​time analysis and of the territorial levels involved). This makes them the most relevant cases in the framework of comparative analyses focused on Europe. We are not including cases in which the use of primaries, and/​or the internal functioning of parties, is regulated by state law such as Finland and outlier cases such as the Icelandic one. There are four empirical chapters that cover the cases of Spain and Italy, following a two-​by-​two matrix. Two chapters focus on leadership selection (with the distinction between party and electoral leaders in the Spanish case), one per each case, and two chapters focus on candidate selection, one per each case. Each empirical chapter covers similar independent variables (degree of inclusiveness of the selectorates and primary elections processes) and the same dependent variables (elites’ profiles and/​or political behaviours). All chapters are framed based on the same research question: to what extent the selection methods for leadership or candidates affects the selected political elites. We cover all the (relevant) cases within the selected countries, even though no homogenous party*country case selection can be performed since primary elections are not developed equally in each country and the organizational culture differs significantly between the cases. All in all, while the spread in the empirical chapters might seem uneven, in terms of the parties covered (e.g. one chapter on one party for Turkey, vs. two chapters on Italy and Spain, for the latter even including regional parties) or period (Belgium in 2014 vs. all Italian elections since 1990 and all French presidential primary elections since 1995, or all regional Spanish elections since 1980), the case studies have been selected following a specific rationale. Some chapters only focus on one issue (candidate selection or leadership selection), others on both. However, we would like to stress here two points: first, the logic of the case selection is based on specific rationales linked to party primaries literature –​as explained above; and second, the theoretical and analytical framework is the same for all the empirical chapters.

12  Giulia Sandri & Antonella Seddone Third, the generalizability of the empirical findings is of course limited due to the case selection’s specificities. In this regard, at the end of each empirical chapter and in the conclusions of the book, the authors discuss to what extent the results of the different chapters may produce generalizations beyond the specific countries analysed. Overall, though, the authors opt to focus on specific cases in a specific part of the world, thereby precluding assumptions about generalizability. Even so, many takeaways from these case studies are relevant for the understanding of the effects of primaries on party membership and leadership from a more global point of view thanks to the common analytical framework applied to each empirical chapter. Moreover, the organization given to the chapters in a specific logical format also contributes in showing how the findings can be generalized beyond the particular case studies. All in all, most recent studies on parties’ organizational democratization serve as a pretext to analyse whether some formal scope for intra-​party democracy, especially in candidate and leader selection, affects actual intra-​party democracy, party’s organizational strength, cohesion, and electoral success, namely focus on the consequences of primaries on the party as a whole. On the contrary, the research questions at the basis of this volume explore the party elites that emerge from those selection processes and those democratized organizational settings. Putting an emphasis on primary elections in particular, the chapters collected in this volume focus not only on the analysis of the processes through which party elites are selected and the consequences at the level of the party (organization and electoral success) but also on the consequences that can be observed at the level of party elites themselves, i.e. what impact party primaries have on the characteristics of parties’ candidates and leaders. This is quite innovative with regard to existing studies in the sub-​field of the discipline.

1.5  Plan of the book Ultimately the core of the volume comes down to the ten empirical ­chapters –​ three on party leadership selection processes, seven on candidate selections. Each empirical chapter presents a country case study selected by the editors based on the degree of comparability of the features at political system level and of the features of the primary election processes. The two case studies where party primaries have been used more extensively and at multiple levels of government in the last decade, namely Spain and Italy, are covered by two chapters each, one focusing on primaries for selecting leader and one of candidate selection. As mentioned above, the independent variables (degree of inclusiveness of the selectorates and primary elections processes), the context variables (degree of institutionalization of primary processes) and the dependent variables (elites’ profiles and political behaviours) are the same in each chapter. All chapters are framed based on the same research question:  to what extent

Intra-party selection methods  13 the selection methods for leadership or candidates affect the profiles and behaviours of political elites? This research problem will be addressed by different angles. However, the empirical chapters will all focus on one of the three following dimensions of analysis for understanding changes in political elites: • • •

sociodemographic profiles (age, gender, seniority, ethnic origin); political profiles (ideological positions, careers); political strategies (candidates and/​or incumbent leader’s decision/​ability to run).

The theoretical framework is based on the main existing theories for explaining the variation in terms of representativeness (with respect to sociopolitical profiles) and of responsiveness (with respect to legislative behaviour in particular) of political elites by looking at the degree of inclusiveness and institutionalization of their respective selection methods, and party primaries in particular (Hazan and Rahat 2010; Smith and Tsutsumi 2014; Spies and Kaiser 2014; Akirav 2018; Rahat and Shapira 2017). Given that the case selection is elaborated on the basis of the main context variable of this comparative study, namely the degree of institutionalization of primaries in a given party system, both in terms of number of parties using the instruments and of time period in which it is used, the chapters are structured as follows. The chapters are organized regionally, first focusing on Spain and Italy (which feature across multiple chapters), then moving onto Greece, Turkey, Belgium, the UK, Germany, and France. Also, chapters are organized based on the distinction between Northern and Southern Europe, in order to take into account the differences in the institutional contexts of the case studies. In its first part, the edited volume focuses on Southern Europe and presents chapters based on case studies that provide empirical examples of party systems in which primary elections, both for candidate and leadership selection, are fairly institutionalized: Italy and Spain. Two chapters focus on candidate selection and two on leadership selection processes in each country case. Then, in its second part, the book also focuses on Southern Europe but rather on case studies in which primaries are less institutionalized in the respective party systems: Greece (in which primaries are used since the early 2000s, but essentially by one party) and Turkey. Both chapters cover candidate selection processes. In its third part, the book explores case studies located in Northern and Continental Europe and endowed with different degrees of institutionalization of party primary systems. This part studies three case of primaries for candidate selection in a party system where they are fairly institutionalized (Belgium, France, and Germany) and one case where primaries for selecting leaders are relatively new (the United Kingdom). Therefore, the first part of the book focuses on Southern Europe and this on more institutionalized primary systems. First Marco Valbruzzi explores

14  Giulia Sandri & Antonella Seddone the consequences of primary elections on Italian party leaders. His chapter provides an examination of political party leadership selection in Italy since the collapse of the ‘first party system’ in the early 1990s. In particular, it analyses the consequences of different types of leadership selection for the characteristics, the quality, and the degree of turnover of the main Italian political elite. Valbruzzi examines selection methods, the factors that result in parties changing their selection rules, the types of leaders chosen and, particularly, the impact that different types of leadership selection have had on the characteristics of Italian political elite in terms of sociodemographic features. More specifically, the chapter analyses the procedures and the results of the leadership ballots within the most important political parties during the last twenty years: Left, ecology and freedom (Lef), Democratic Party (PD), Five Star Movement (F5M), People of Freedom/​Go Italy (Pof), and the right-​wing Northern League (NL). In the third chapter of this collection, Oscar Barberà and Juan Rodríguez-​ Teruel explore the organizational consequences of party primaries for selecting leaders in Spain. The analysis focuses on both party leaders and top candidates at the regional and national level considering the Spanish mainstream parties (i.e. PSOE, IU, PSC, EUiA, ERC, ICV). After exploring the adoption of these processes and their main formal features, the chapter examines their internal consequences on party cohesion, party leadership strength and on their upcoming use once they have been implemented for the first time. The chapter points out that, despite common knowledge, in many cases their introduction has strengthened the party leadership and reinforced party cohesion. On the other hand, the results also show some evidences of a relationship between shrinking party cohesion and the restricted use of party primaries in the future. In the fourth chapter of this book, Javier Martínez-​Cantó and Javier Astudillo study the effects of primaries on the nomination of electoral leaders in Spain. They elaborate fine-​grained results on the effect of primaries by clearly focusing on electoral leaders’ and not party leaders’ selections. They show that primaries are more likely to elect (a) outsiders and (b) fast-​trackers as electoral leaders. They focus on the Spanish case at the sub-​national level. They use a novel dataset containing information on personal, partisan, and political experience of 625 candidacies since 1980, regarding all main Spanish political parties (PP, PSOE, IU, Podemos, C’s, and the main regionalist parties) and their methods of selection. The fifth chapter of this book, by Bruno Marino, Nicola Martocchia Diodati, and Luca Verzichelli, analyses the relationship between more open candidate selection rules and the personalization of politics. By focusing on the Italian case between 2006 and 2016, this chapter aims at explaining whether relevant changes in the candidate selection rules and the intra-​ party-​personalization levels might foster the selection of politicians for more relevant parliamentary or governmental offices. They show that more open

Intra-party selection methods  15 candidate selection rules do not have a boosting effect either as an independent or as a conditioning variable. Conversely, their empirical analysis shows a significant effect of a face of intra-​party personalization (the control of party organizations by their leaders), that, if interacted with the candidate selection rules variable, makes wider candidate selection rules a boost for more relevant offices, and also favours the obtaining of higher offices by MPs without a classical party or institutional career. In Chapter 6, Dimitrios Kosmopoulos examines to what extent the introduction of intra-​party democracy norms impacts on party elites’ profile. It is based on PASOK as a case study, since 2004. The chapter explores to what extent the implementation of participatory models into party structure influences the sociodemographic and political profile of socialist MPs during the 2004–​09 period by combining quantitative and qualitative data. The seventh chapter of this book by Faruk Aksoy explores the effect of primaries on the profiles of candidates in Turkey. This chapter explains the effect of decision of Republican People’s Party (RPP) in Turkey to hold primaries as a candidate selection method for national parliamentary elections on candidates’ profiles. In this chapter, Aksoy analyses the differences in candidates’ profiles in İstanbul constituency across two consecutive national elections –​in 2011 and 2015. The analyses cover standard demographic characteristics of the candidates such as age, sex, the education level, occupation, but also the candidates’ political seniority. Chapter 8 by Audrey Vandeleene aims at investigating how the modes of selection may affect the characteristics of the candidates selected, focusing in particular on gender diversity of the lists. The chapter focuses on the Belgian case that provides a privileged angle for addressing this issue. Gender quotas have existed for about 20 years and this has led Belgian politics to be rather gender-​balanced although men still hold a majority of political mandates, and often the most visible ones. This chapter aims to assess whether and to what extent intra-​party democracy might enhance the quality of representation, more precisely the purpose is to clarify whether who obtains which spots on the list is a consequence of how they are selected. Eleven political parties that competed for the regional, federal, and European elections of May 2014 are included in the analyses. The findings demonstrate that selection matters but not always in the expected direction. Chapter 9 by Agnès Alexandre-​Collier and Emmanuelle Avril explores the use of primaries for the selection of party leaders in the UK Conservative and Labour parties. In recent years both the Conservative and Labour parties have experimented with primaries (open, closed, or partial) for the selection of their party leaders. Thus, they investigate the impact of selection rules changes on the ideological profile of successful candidates. An intriguing effect of the experimental selection methods, designed to reinforce party cohesion by attempting to establish a more direct link between party leaderships and the wider electorate and thus to bypass the activist stratum,

16  Giulia Sandri & Antonella Seddone has been to paradoxically favour the emergence of more candidates whose ideological positioning marks a departure from that of the mainstream of the parliamentary party. In Chapter 10, Elisa Deiss-​Helbig studies how candidate selection methods affect the parliamentary representation of ethnic minorities in the German Bundestag. This chapter explores thus the consequences of the use of more inclusive selection methods on the profile of the political personnel, while focusing on a specific sociodemographic trait of political elites: their ethnic origins. The operationalization of the selection methods is done by drawing on the parties’ statutes as well as the parties’ minutes of the nomination meetings. Deiss-​Helbig’s results show, first, a strong homogeneity in the way the main five German parties select their candidates. However, she demonstrated inter-​ (between, on the one hand, the two major German parties and, on the other hand, the three smaller parties) as well as intra-​party (between the different German Länder) differences. As regards the selection of ethnic minority candidates, she shows that party ideology seems to play an important role because ethnic minority candidates are mainly nominated by left-​wing parties. However, more open candidate selection methods seem to be more favourable to ethnic minority candidates than more exclusive selection methods. In the last empirical chapter, Chapter 11, Marino De Luca looks at the use of primaries for selecting candidates in France, both for presidential and local elections. This chapter analyses how intra-​party democracy became rooted in the French political life and how it affected the profiles of the selected political personnel, both in the case of primaries used for candidate selection and chief executive candidate selection. It focuses on the political, social, and cultural milieu that led to the development of primary elections. In addition, the chapter analyses the consequences of different types of candidate selection methods on the sociodemographic and political profiles of political elites in France. Finally, in the last chapter the two editors summarize the main results of the empirical chapters, providing a general interpretation of the political consequences of primary elections on political elites and representative institutions and drawing together the main themes that have emerged from the empirical work so far. They provide also a brief explanation on how the study contributes to the scientific and political debate on primary elections and how the present work fits into the existing scholarly literature. Moreover, the concluding chapter is designed to frame further analyses on the topic on the basis of the findings of the book. This chapter provides therefore a framework for a major long-​term research agenda and it identifies new research questions (with associated possible indicators) of the impact of primary elections on parties and political personnel.

References Ahler, D. J., Citrin, J., and Lenz, G. S. (2016). Do open primaries improve representation? An experimental test of California’s 2012 top-​two primary. Legislative Studies Quarterly, 41(2): 237–​268.

Intra-party selection methods  17 Akirav, O. (2018). Candidate selection methods, cooperation and legislative effectiveness. Review of European Studies, 10(1): 46. Alexandre-​Collier, A. (2016). The‘open garden of politics’: The impact of open primaries for candidate selection in the British Conservative Party.The British Journal of Politics and International Relations, 18(3): 706–​723. Alexandre-​ Collier, A., Goujon, A., and Gourgues, G. (eds) (2020). Innovations, Reinvented Politics and Representative Democracy. London: Routledge. APSA, Committee on Political Parties (1950). Towards a more responsible two-​party system. American Political Science Review. September 1950, supplement. Bartolini, S., and Mair, P.  (1990). Policy competition, spatial distance and electoral instability. West European Politics, 13(4): 1–​16. Bennister, M., Worthy, B., and Hart, P. T. (eds) (2017). The Leadership Capital Index: A New Perspective on Political Leadership. Oxford: Oxford University Press. Bernardi, L., Sandri, G., and Seddone, A. (2017). Challenges of political participation and intra-​party democracy: Bittersweet symphony from party membership and primary elections in Italy. Acta politica, 52(2): 218–​240. Best, H., and Cotta, M. (eds) (2000). Parliamentary Representatives in Europe, 1848-​ 2000:  Legislative Recruitment and Careers in Eleven European Countries. Oxford: Oxford University Press Blondel, J., and Thiébault, J. L. (eds) (2009). Political Leadership, Parties and Citizens: The Personalization of Leadership. London: Routledge. Boatright, Robert G. (ed.) (2018). Routledge Handbook of Primary Elections. New York: Routledge. Bolleyer, N. (2012). New party organization in Western Europe: Of party hierarchies, stratarchies and federations. Party Politics, 18(3): 315–​336. Carey, J. M., and Polga-​ Hecimovich, J. (2006). Primary elections and candidate strength in Latin America. The Journal of Politics, 68(3): 530–​543. Coller, X., Cordero, G., and Jaime-​Castillo, A. M. (eds) (2018). The Selection of Politicians in Times of Crisis. London: Routledge. Cordero, G., and Coller, X. (2014). Cohesion and candidate selection in parliamentary groups. Parliamentary Affairs, 68(3): 592–​615. Cordero, G., and Coller, X. (eds) (2018). Democratizing Candidate Selection:  New Methods, Old Receipts? Basingstoke:  Palgrave.Cotta, M., and Best, H. (2000). Parliamentary Representatives in Europe 1848–​2000. Oxford: Oxford University Press. Cross, W. P., and Katz, R. S. (2013). The Challenges of Intra-​ Party Democracy. Oxford: Oxford University Press. Cross, W. P., Kenig, O., Rahaṭ, G., and Pruysers, S. (2016). The Promise and Challenge of Party Primary Elections: A Comparative Perspective. Montreal: McGill-​Queen’s Press-​MQUP. Cross, W., and Pilet, J. B. (2015). The Politics of Party Leadership. A Cross-​National Perspective. Oxford: Oxford University Press. Dalton, R. J. (2014). Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies. Washington: Cq Press. Deseriis, M., and Vittori, D. (2019). The impact of online participation platforms on the internal democracy of two Southern European parties: Podemos and the Five-​ Star Movement. International Journal of Communication, 13: 5696–​5714. Dowding, K., and Dumont, P.  (eds) (2013). The Selection of Ministers around the World. New York: Routledge.. Fishkin, J. (2009). When the People Speak:  Deliberative Democracy and Public Consultation. Oxford: Oxford University Press.

18  Giulia Sandri & Antonella Seddone Gauja, A. (2016). Party Reform:  The Causes, Challenges, and Consequences of Organizational Change. Oxford: Oxford University Press Gauja, A., and Cross, W. (2015). The influence of party candidate selection methods on candidate diversity. Representation, 51(3): 1–​12. Gerber, E. R., and Morton, R. B. (1998). Primary election systems and representation. Journal of Law, Economics, & Organization, 14(2): 304–​324. Gomez, R., and Ramiro, L. (2019). The limits of organizational innovation and multi-​speed membership: Podemos and its new forms of party membership. Party Politics, 25(4): 534–​546. Hazan, R. Y., and Rahat, G. (2010). Democracy within Parties:  Candidate Selection Methods and Their Political Consequences. Oxford: Oxford University Press. Indriðason, I. H., and Kristinsson, G. H. (2015). Primary consequences: The effects of candidate selection through party primaries in Iceland. Party Politics, 21(4): 565–​576. Karvonen, L.  (2010). The Personalisation of Politics:  A Study of Parliamentary Democracies. Colchester: ECPR Press. Katz, R. S., and Mair, P. (1995). Changing models of party organization and party democracy the emergence of the cartel party. Party Politics, 1(1): 5–​28. Kaufmann, K. M., Gimpel, J. G., and Hoffman, A. H. (2003). A promise fulfilled? Open primaries and representation. Journal of Politics, 65(2): 457–​476. Kemahlioglu, O., Weitz-​Shapiro, R., and Hirano, S. (2009). Why primaries in Latin American presidential elections? The Journal of Politics, 71(1): 339–​352. Kenig, O. (2009). Democratization of party leadership selection: Do wider selectorates produce more competitive contests? Electoral Studies, 28(2): 240–​247. Kenig, O., Rahat, G., and Hazan, R. (2015). Leadership selection versus candidate selection: similarities and differences. In G. Sandri, A. Seddone, and F. Venturino (eds), Party Primaries in Comparative Perspective. Farnham: Ashgate, pp. 21–​39. Kenig, O., Cross, W., Pruysers, S., and Rahat, G. (2016). Party primaries: Towards a definition and typology.Representation, 51(2): 147–​160. Marshall, T. R. (1978). Turnout and representation:  Caucuses versus primaries. American Journal of Political Science, 22(01): 169–​182. Matthews, D.R. (1984). Legislative recruitment and legislative careers. Legislative Studies Quarterly, 9(4): 547–​585. Michels, R. ([1911] 1962). Political Parties:  A Sociological Study of the Oligarchical Tendencies of Modern Democracy. New York: Collier Books. Norrander, B. (1989). Ideological representativeness of presidential primary voters. American Journal of Political Science, 33(3): 570–​587. Norrander, B. (1993). Nomination choices: Caucus and primary outcomes, 1976–​88. American Journal of Political Science, 37(2): 343–​364. Norris, P. (ed.) (1997). Passages to Power:  Legislative Recruitment in Advanced Democracies. Cambridge: Cambridge University Press. Norris, P. (2004). Electoral Engineering:  Voting Rules and Political Behavior. Cambridge: Cambridge University Press. Ostrogorski, M. (1902). Democracy and the Organization of Political Parties (Vol. 1). London: Macmillan and Company Limited. Pilet, J. B., and Cross, W. (2014). The Selection of Political Party Leaders in Contemporary Parliamentary Democracies: A Comparative Study. London: Routledge. Poguntke, T., and Webb, P.  (eds) (2005). The Presidentialization of Politics:  A Comparative Study of Modern Democracies. Oxford: Oxford University Press.

Intra-party selection methods  19 Rahat, G., and Kenig, O. (2018). From Party Politics to Personalized Politics?: Party Change and Political Personalization in Democracies. Oxford:  Oxford University Press. Rahat, G., and Shapira, A. (2017). An intra-​party democracy index: Theory, design and a demonstration. Parliamentary Affairs, 70(1): 84–​110. Ramiro, L. (2016). Effects of party primaries on electoral performance: The Spanish Socialist primaries in local elections. Party Politics, 22(1): 125–​136. Ranney, A. (1972). Turnout and representation in presidential primary elections. American Political Science Review, 66(1): 21–​37. Rombi, S., and Seddone, A. (2017). Rebel rebel. Do primary elections affect legislators’ behaviour? Insights from Italy. Parliamentary Affairs, 70(3): 569–​588. Sandri, G., Seddone, A., and Venturino, F. (eds) (2015). Party Primaries in Comparative Perspective. Farnham: Ashgate. Scarrow, S. E. (2014). Beyond Party Members:  Changing Approaches to Partisan Mobilization. Oxford: Oxford University Press. Scarrow, S. E., Webb, P. D., and Farrell, D. M. (2000). From social integration to electoral contestation: The changing distribution of power within political parties. In R. J. Dalton, and M. P. Wattenberg (eds), Parties without Partisans: Political Change in Advanced Industrial Democracies. Oxford: Oxford University Press, pp. 129–​151. Scarrow, S. E., Webb, P. D., and Poguntke, T. (eds) (2017). Organizing Political Parties: Representation, Participation, and Power. Oxford: Oxford University Press. Siavelis, Peter M., and S. Morgenstern (eds) (2008). Pathways to Power:  Political Recruitment and Candidate Selection in Latin America. University Park: Pennsylvania State University Press. Smith, D. M., and Tsutsumi, H. (2014). Candidate selection methods and policy cohesion in parties:  The impact of open recruitment in Japan. Party Politics, DOI: 1354068814549347. Spies, D. C. and Kaiser, A. (2014). Does the mode of candidate selection affect the representativeness of parties? Party Politics, 20(4): 576–​590. Steger, W. P. (2007). Who wins nominations and why? An updated forecast of the presidential primary vote.Political Research Quarterly, 60(1): 91–​99. Vandeleene, A. (2016). Does candidate selection matter? A  comparative analysis of Belgian political parties’ selection procedures and their relation to the candidates’ profile (Doctoral dissertation, UCL-​Université Catholique de Louvain). Ware, A. (2002). The American Direct Primary:  Party Institutionalization and Transformation in the North. Cambridge: Cambridge University Press. Wauters, B., and Pilet, J.-​B. (2015). Electing women as party leaders: Does the selectorate matter? In W. P. Cross and J. B. Pilet (eds), The Politics of Party Leadership:  A Cross-​National Perspective. Oxford: Oxford University Press, pp. 73–​89. Webb, P., and Poguntke, T. (2005). The Presidentialization of Politics: A Comparative Study of Modern Democracies. Oxford: Oxford University Press.

2  Choosing party leaders in Italy between personalization and democratization Marco Valbruzzi

2.1  Introduction For about twenty years, the literature on party politics has been questioning the transformations of political parties and their consequences for the processes of candidate recruitment and for the composition of the political class, including party leaders (Cross and Pilet 2016; Gallagher and Marsh 1988). Yet, there is a country where these transformations have been more sharp, sudden and, in some ways, more dramatic than others. The reference goes, obviously, to Italy which, at the beginning of the 1990s, experienced the collapse of its party system and, at the same time, saw the rebirth of another, much more unstable and fluid party system (Pasquino 2019). In the light of these profound changes, which have affected both the party system and the individual party organizations, Italy has become, almost unwillingly, a privileged observatory for the analysis of political parties. Indeed, it was in Italy that the first ‘relevant’ personal party in the European context (Berlusconi’s Forza Italia) took shape (Calise 2010; McDonnell 2013). Again, it was Italy, especially thanks to some left-​wing parties, that opened the way in Europe to the introduction of more inclusive mechanisms (such as ‘open’ primary elections) for the selection of leaders and candidates (Valbruzzi 2005; Sandri et  al. 2015). Finally, Italy gave birth to the first (successful) digital party experiment, with the formation and then the institutionalization of the Five Star Movement (Gerbaudo 2019; Calise and Musella 2019). In other words, if Italy has been a latecomer in the process of nation building, as far as the processes of transformation/​degeneration/​innovation of the party system and its components are concerned, it is instead ahead of processes that only today we are observing in other European countries. Besides a peculiarity linked to the timing of these transformations, another factor that makes the Italian experience absolutely exceptional concerns the multiplicity (or diversity) of these changes. If in the first Italian party system (1946–​92) there was a single-​party model (the mass party) towards which all political actors strived to converge, since the 1990s there has not been a single reference model, but different party organizations have arisen in competition with each other. The ‘personal party’, the ‘digital party’, the ‘open (or light)

Choosing party leaders in Italy  21 party’ (through the implementation of primary elections), and others, are all examples of party models that emerged in the fire of the Italian political laboratory. However, the interesting aspect that we will investigate throughout this chapter is that, below this relevant plurality of party models, some common trends emerge for all the party under investigation. These are, as we shall see, only apparently conflicting tendencies. On the one hand, there is a rising process of personalization of politics and, in particular, of party politics (Rahat and Kenig 2018; Musella 2018; Bittner 2011; Karvonen 2010; Poguntke and Webb 2005). Indeed, parties are increasingly identified with their leaders and, oftentimes, the trajectory of consensus for a leader ends up coinciding with the trajectory of his/​her own party. However, alongside this process of personalization, parties have also started to make their organizations more ‘inclusive’ and open to the voice of their members, sympathizers, or even ordinary voters (Cross and Pilet 2016; Pilet and Cross 2014; Cross and Katz 2013; Pennings and Hazan 2001). These two processes –​intra-​party personalization and ‘democratization’ by definition are not unrelated. On the contrary, as some scholars have recently suggested, it is likely that the push towards a further personalization of the parties has led to a greater openness of the internal decision-​making (Katz 2018). And vice versa, in the sense that the trend towards greater inclusiveness of political parties may have eventually laid the foundations for an increasingly central –​and crucial –​role for party leaders. In any case, the analysis of the Italian case, especially with regards to the process of leadership selection, allows us to better focus on these political phenomena and to control more closely some hypotheses circulating in the literature.

2.2  Varieties of party models in Italy In order to understand the transformations that have taken place inside and outside of party organizations in Italy, it is useful to start with as faithful an overview as possible of the current situation, particularly with regard to the level (or process) of personalization and the inclusive decision-​making mechanisms within parties (Musella 2015; Pasquino 2014; Sandri, Seddone, and Venturino 2015). To this end, the data reported in Figure 2.1 allow for the analysis of the main Italian parties along two dimensions. On the one hand, it shows the level of personalization of Italian parties (and, therefore, the degree to which the parties are characterized by a more or less personal leadership) and, on the other hand, their level of internal democracy, understood as a measure of how involved members and supporters are in the internal decision-​making processes (and their effectiveness). As can be seen, apart from the exception of a recently formed party and others that have recently been dissolved (LEU: Free and Equal), the majority of the main ‘relevant’ Italian parties –​in the Sartorian sense (Sartori 1976) –​ are characterized by a marked dose of personalism or personalization. This

10

22  Marco Valbruzzi

FI

8

LN

M5S

6

PD

4

Personalization

FdI

0

2

LEU

0

2

4

6

8

10

Intra-party democracy

Figure 2.1 Personalization and level of intra-​ party democracy in Italian political parties (2019). Source: author’s own compilation based on data by Meijers and Zaslove (2020).

is clearly the case for Forza Italia (Go Italy), founded and owned by Silvio Berlusconi, but it is not its exclusive characteristic. In fact, the ‘second Italian party system’, emerged since the mid-​nineties, is filled with personal parties, that is to say, organizational vehicles for use or consumption by a sole lider maximo, or personalized parties, which have strongly centralized or vertical power structures around the figure of a single leader, with ever-​increasing personal resources and ever-​decreasing challengers or internal opponents. This applies, for example, to the ‘new’ Lega (Nord) that became strongly personalized in 2012 with the leadership of Matteo Salvini who also applied an ideological shift to the party (Albertazzi et al. 2016; Passarelli and Tuorto 2018). A similar process can also be observed in the other radical right-​wing party (Fratelli d’Italia  –​Brothers of Italy) where the internal leadership of Giorgia Meloni is becoming stronger and more centralized. Nevertheless, this tendency towards increasing personalization is not exclusive to centre-​right parties (nor, it is useful to reiterate here, to Italian parties). Even parties with a seemingly ‘horizontal’ and democratic rhetoric, created on the wave of the ‘uno vale uno –​one equals one’ mantra, have partly adapted to this trend. And this is true first and foremost for the Movimento 5 Stelle (Five Star Movement). This is a party which, despite being defined as leaderless (Tronconi 2015), has, over the last few years (and especially after entering into government), increased its level of personalization around some

Choosing party leaders in Italy  23 key figures of the Movement, from the founder Beppe Grillo, to the former ‘political head’ (capo politico) Luigi Di Maio, up to the shadow-​leader Davide Casaleggio who controls many of the internal organizational and information technology ‘grey areas’ of the party (Tronconi 2019; Bordignon and Ceccarini 2019). Even the main centre-​left party (PD  –​Democratic Party) could not resist the push towards greater personalization, particularly under the leadership of former Prime Minister Matteo Renzi. So much so that during that period (2013–​18), many analysts and academics began defining the PD as the PDR (Partito Democratico di Renzi –​or Renzi’s Democratic Party), modelled on his image and his likeness (Diamanti 2016; Pasquino and Valbruzzi 2017; Bordignon 2013). Subsequently, particularly after Renzi left the party and created his own party organization centred around him (Italia Viva), the degree of personalization within the PD decreased, even if –​as we shall see –​ clear traces of personalization remained within the organizational structure (Rombi and Serricchio 2020). In addition to personalization, a second element deserves to be emphasized in Figure 2.1 regarding the specific level of democracy within Italian parties. From this perspective, there are significant differences between the different organizational actors. Put simply, the room for internal decision-​making and involvement of members or supporters is reduced to a minimum with Forza Italia, where the voice of the leader prevails over everyone and everything, while more inclusive decision-​making mechanisms exist in the PD, thanks to the primary elections tool for voting for the leader or candidates, or in the drafting of program proposals. Between these two extremes are the other parties, such as both the 5 Star Movement (where internal participation occurs through the channels of the online platform called Rousseau), as well as the Lega, particularly with regards to the role of members in core local/​grassroots activities promoted by the party. Moreover, again with regard to the rights or ‘powers’ of decision-​making and internal involvement of Italian parties, the data reported in Figure 2.2 show substantial stability over the course of the last few years. In this case, the data refer specifically to the balance of power between the members and the leader in the internal decision-​making processes of the party. As is clear, on a scale of 0 to 10 (where 0 indicates a maximum role played by members and 10 indicates maximum power for the leader), virtually all Italian parties have maintained a stable power structure with few notable variations. The only exception that occurred between 2016–​17 and 2019 concerns the 5 Star Movement, where Di Maio’s role as recognized political leader was restricted after his resignation, thereby enlarging the space for greater internal dialogue between the different factions within the party. In any case, beyond the (few) variations that have occurred in the meantime in the internal balance of power between members and leaders, the association between inclusivity/​incisiveness of the decision-​making processes within the parties and their ideological orientation remains stable. As can be seen in Figure 2.3, in which the balance of power between members and leaders is linked to the traditional left-​right ideological scale, it is mainly the progressive

24  Marco Valbruzzi 10

Members vs Leader

9 8 7 6 5 4 3 2 1 0

LN

FI

FDI

LEU

2017

PD

M5S

2019

10

Figure 2.2 Control over policy choices between members and leadership in the main Italian parties (2017–​19). Source: author’s own compilation based on data by Bakker et al. (2020) and Polk et al. (2017).

FI FdI

M5S

6

PD

4

LEU

0

2

Members vs Leader

8

LN

0

Left

2

4

6

8 Right

10

Figure 2.3 Control over party policy choices and left-​right position of Italian parties (2018). Source: author’s own compilation based on Polk et al. (2017) data.

Choosing party leaders in Italy  25 parties that are more likely to include or involve members in the decision-​ making process, while, on the other end of the scale, there is a more dominant and overriding role played by the leadership. The data presented thus far have served to make sense of the recent transformations in the methods of selection (and deselection) of Italian party leaders in relation to a broader more general background, that is able to hold together phenomena of different types and range, such as the processes of personalization and democratization within parties. However, in order to conduct a more detailed analysis of party leaders and their ways of gaining access to office, it is useful to first show the changes in party leadership over the course of the last 30  years, i.e. since the collapse of the so-​called First Republic onwards (see Table 2.1). Also in this case, partly as a consequence of the processes outlined above, a difference in structure can be noted or, better said, in the survival of the leadership between parties according to their ideological position. In other words, the centre-​right parties, also due to the party organization that they have built over time, are more likely to have stability in terms of leaders, regardless of their electoral performance. From this perspective, the case of Forza Italia is a prime example; founded by Berlusconi in 1994, there has never been a change of guard of the presidency of the party (Ignazi 2014). Like in a ‘proprietary monarchy’, Berlusconi remained the undisputed and indisputable leader of Forza Italia and thereby became the longest-​standing party leader (in terms of survival in office) in Italian history. Even with a totally different organizational model (Passarelli and Tuorto 2012; Biorcio 1998), even the Lega (Nord) has demonstrated notable stability in its leadership over the last three decades, first with Bossi (leader for more than twenty years) and, currently, with Matteo Salvini. A similar description applies to Giorgia Meloni’s Brothers of Italy, even if the party has only recently been formed (2013). That being said, the discourse of the Democratic Party is very different (including in its different form as the PDS –​the Democratic Party of the Left), where leaders remained in office for an average of three years and were constantly subjected to the scrutiny of electoral consensus (both at a national as well as a local level) and voters’ intentions were measured with daily polls. In this case, the effects of internal party democratization and, therefore, the involvement of the members of the internal decision-​making processes are evident. Greater inclusivity and openness of the party produced as a consequence –​not inevitable –​often constant change in national leadership. Finally, with regard to the M5S, it is still too early to define a stable and long-​lasting leadership (and replacement) model. For as long as the founder of the party was also its (main) leader (Beppe Grillo), his leadership was not questioned, and his survival had no statutory expiry date (Bordignon and Ceccarini 2013). On the contrary, at the beginning of 2017 with the elections of the new political leader by the ‘online registered members’, even the internal dynamics of the Five Star Movement leadership became more agitated and unstable.

26  Marco Valbruzzi Table 2.1 Italian party leaders, 1990–​2020 PDS/​DS

PD

DL-​M

M5S

FI/​PDL

1990

LEGA

AN

FDI

Bossi Occhetto

D’Alema

Berlusconi

Fini

1995

Veltroni

2000 Fassino Rutelli

2005

Veltroni Franceschini Bersani

Grillo

2010

Maroni Epifani

Salvini

Renzi

La Russa Meloni

2015

Di Maio Martina Zingaretti 2020

Crimi

Source: author’s own compilation based on data by Sandri, Seddone, and Venturino (2015). Note: Boxes indicate dates of formation and termination; grey areas indicate government participation or support.

Choosing party leaders in Italy  27 As will be discussed in the following sections, the different lengths of time in office of party leaders depends, to a large extent, on their selection and de-​ selection methods, as well as the level of competitiveness that characterizes the recruitment process. In addition to the main demographic characteristics of Italian party leaders from 1990 to 2020, Table 2.2 also shows the respective selection methods and the level of competitiveness (zero, low, medium, and high) of the selection process. The general picture shows a fair amount of variability between the Italian parties both with regards to the selection methods as well as to the degree of competitiveness. By way of illustration, on the one hand, there is the ‘extreme’ case of selection through the decision of a single individual (Forza Italia) and, on the other hand, the use of ‘open primaries’ for supporters and voters to elect the Secretary of the Democratic Party. Between these two cases that are poles apart in terms of inclusiveness in internal decision-​making processes, are more nuanced cases, in which the different actors are involved in selecting a leader. In many cases it is the limited elite or the central committee of the party that chooses the party candidates and a leader. This is especially true for parties deriving from traditional mass organizations (the PPI and the PDS), that chose their first secretaries through procedures involving only a very limited inner party circle. At the same time, there are ­examples –​such as the initial case of the Margherita (Daisy) party –​where the party leader was selected with the involvement of all components of the parliamentary group. In other cases, particularly at the beginning of the period being examined, it was, above all, the delegates to the party congresses, selected with the participation of local members, who would choose their party leader. Finally, more recently, a different way of selecting party leaders with the direct involvement of party members has emerged. This procedure was used, both by the Five Star Movement in 2017 with the introduction of ‘closed primaries’ reserved only for registered and certified members of the party to elect a political leader, as well as by the two radical extreme-​right parties (the Lega and Brothers of Italy), that chose their leaders (in 2012 and 2013 respectively) through direct membership ballots (in the case of Brothers of Italy this is also usually extended to sympathizers). As can be observed, therefore, the transformations that have occurred in the Italian party system have not only affected some formal aspects of party organization or marginal mechanisms of party life. These changes have had a direct impact on what still remains one of the main party functions, that is to say, the recruitment of candidates and, more specifically, the selection of internal leadership. Once again, as previously stated, a single reference model for leadership selection was not established or imposed. And, in fact, the different party models (at different points in time) preferred to adapt their selection processes to their needs and to their specific organizational requirements. Nonetheless, one common trend does prevail over the others, that is to say a demand from the base or a push from above (from the competing groups within the party) to open up and make decision-​making

newgenrtpdf

Name of leader

Party

Year Gender National political seniority

Age at time of selection

Length in office (in months)

Selection method

Competitiveness

Umberto Bossi Umberto Bossi Roberto Maroni Matteo Salvini Matteo Salvini Matteo Salvini Walter Veltroni Dario Franceschini Pier Luigi Bersani Guglielmo Epifani Matteo Renzi Matteo Renzi Maurizio Martina Nicola Zingaretti Silvio Berlusconi Silvio Berlusconi Silvio Berlusconi Silvio Berlusconi Ignazio La Russa Giorgia Meloni Giorgia Meloni Beppe Grillo Luigi Di Maio Vito Crimi Achille Occhetto Massimo D’Alema Massimo D’Alema

LN LN LN LN LN League PD PD PD PD PD PD PD PD FI PDL PDL FI FdI FdI FdI M5S M5S M5S PDS PDS PDS

1989 1995 2012 2013 2017 2020 2007 2009 2009 2013 2013 2017 2018 2019 1994 2008 2009 2013 2013 2014 2017 2009 2017 2020 1991 1994 1997

50 56 57 40 44 47 52 51 58 63 38 42 40 54 58 72 73 77 66 37 40 61 26 48 55 45 48

272

Party delegates Party delegates Party delegates Closed primaries Closed primaries Party delegates Open primaries Party delegates Open primaries Party delegates Open primaries Open primaries Party delegates Open primaries Single leader Party elite Party elite Single leader Parliamentary group Open primaries Party delegates Parliamentary group Closed primaries Party delegates Party elite Party elite Party elite

Absent Absent Absent Low Low Absent Low Low High Absent Medium Medium Absent Medium Absent Absent Absent Absent Absent Absent Absent Absent Low Absent Absent High Absent

Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Female Female Male Male Male Male Male Male

Yes Yes Yes Yes Yes Yes Yes Yes Yes No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes

17 141 17 8 41 6 48 8 12 314

11 73 96 28 2 41 53

28  Marco Valbruzzi

Table 2.2 Characteristics of the Italian party leaders and selection method, 1990–​2020

newgenrtpdf

DS DS DS DS DL DL DL DL UDC UDC PPI PPI PPI PPI CCD CCD CCD AN AN IDV

1998 2000 2001 2005 2001 2002 2004 2007 2002 2005 1994 1995 1997 1999 1994 1995 2001 1994 1995 1998

Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes

45 47 52 56 47 48 50 53 48 54 46 65 64 54 39 40 47 42 43 48

37 72 65

35 12 18 33 27 14 76 18 12 162 184

Source: author’s own compilation based on data provided by Sandri, Seddone, and Venturino (2015).

Party delegates Party delegates Closed primaries Closed primaries Parliamentary group Party delegates Party delegates Party delegates Party delegates Party elite Party delegates Party elite Party delegates Party delegates Parliamentary group Party delegates Party delegates Party elite Party delegates Party elite

Absent Medium High Medium Absent Absent Absent Absent Absent Absent High Absent High Medium Absent Absent Absent Absent Absent Absent

Choosing party leaders in Italy  29

Walter Veltroni Walter Veltroni Piero Fassino Piero Fassino Francesco Rutelli Francesco Rutelli Francesco Rutelli Francesco Rutelli Marco Follini Lorenzo Cesa Rocco Buttiglione Gerardo Bianco Franco Marini Pierluigi Castagnetti Pierferdinando Casini Pierferdinando Casini Marco Follini Gianfranco Fini Gianfranco Fini Antonio Di Pietro

30  Marco Valbruzzi processes more inclusive when choosing candidates for the leadership. But what were the consequences of this progressive ‘opening-​up’ of selection methods for the parties and their leaders? This question will be addressed in the next section.

2.3  Selection methods over time For every political party, the choice of its leader represents a fundamental moment that will have consequences on both the internal organization (including the balance of power) as well as the message being conveyed to voters or public opinion in general. Moreover, it is now generally known that, over the course of the last few decades, the figure of party leader has become increasingly important both inside and outside of the party (Cross and Pilet 2016). The media spotlight is increasingly focused on the leader who, not only represents the party’s position, but also ‘embodies’ its message and indicates its political-​ideological orientation. Very often, the same internal party organization resources, starting with communication resources, are controlled from above and thus, more or less directly, by the leader (Musella 2018). When the party then gains access to government roles or positions of power, institutional resources (economic, staff, communications, etc.) can be added to the internal organizational resources, which allow the leader to further increase their power over the organization of the party. In other words, when the personalization process of politics overlaps with and is strengthened by the various phenomena of becoming president, the role of party leader is destined to become increasingly central and significant (Poguntke and Webb 2005). From one point of view, this also explains why over the last few years academics have paid greater attention to selection and de-​selection procedures of party leaders (Hazan and Rahat 2010; Pilet and Cross 2014). If, in the past, it was more interesting and useful to concentrate on the power links between the different internal party factions in order to understand the organizational structures of parties (Panebianco 1988), today it has become necessary to analyse the ways of recruiting leaders (and the candidates for political office). Put differently, if in the past leaders were the expression of the party and its internal dynamics, today the party is the expression –​or the function –​of the leader. This is why it has become important to analyse the transformations in leadership selection processes that have occurred in the meantime. To this end, Figure  2.4 allows for the observation of these transformations over a medium-​long time period, from the beginning of the 1960s to today. More specifically, six main party leader selection processes are identified that can be listed in order of increasing inclusivity: 1) single leader; 2) party elite; 3) parliamentary group; 4) party delegates (or party congress/​convention); 5) closed primaries; 6) open primaries. As can be seen, the situation today is radically different with respect to the 1960s or 1970s. Initially, in fact, the main Italian parties would choose or confirm their own leaders (secretaries or presidents) through a particularly

Choosing party leaders in Italy  31 2010-2020

2

2000-2010 1990-2000

1

2

3

2

4

1

12 8

1980-1990

2

1

14 17

1970-1980

11 19

1960-1970 0%

4

4 24

10%

20%

Single leader

30%

40%

Party elite

50%

60%

70%

Parliamentary group

Closed primaries

80%

90%

100%

Party delegates

Open primaries

Figure 2.4 Party leader selection methods in Italy, by decade (1960–​2020).

limited body composed of the more restricted party leadership. Therefore, there was little variety in terms of the selection process; big or small, real mass or aspiring parties adopted a single method to select their respective party leaders. The choice of the leader was an elite affair, which involved a very small coterie of people. Since the 1970s and even more decisively in the decades that followed, the situation has become more dynamic. In particular, together with the party elite choosing the leader, another slightly more inclusive selection method emerged, which provides for the involvement of party delegates to congress elected by members or activists at a local level. This (partial) opening of internal decision-​making mechanisms is also a –​partly successful –​exploratory attempt by mass parties to respond to the demands of democracy and involvement of different social movements that were formed at the end of the 1960s. This situation relating to the method of selecting party leaders lasted until the beginning of the 1990s, but then it began to change abruptly with the disappearance of the old mass-​based parties and the emergence of new personal or highly personalized parties. While in many cases between the 1990s and the beginning of the 2000s, it was still the party delegates to congress that would select their leaders (around 60% of cases analysed), it is also true that new ways are emerging that are alien to the experiences of the past. The case of Forza Italia has already been discussed, where the party leader is essentially the product of self-​coronation and a purely individual decision (of a single individual). But Berlusconi’s party is not the only party that adheres to this method. The party founded by former prosecutor Antonio Di Pietro (Italy of Values), quickly catapulted into the political arena, also used a limited selection process entrusted to a single figure, that of its leader and founder.

32  Marco Valbruzzi However, the real change emerged above all from the end of the 2000s and, particularly, in the last decade, when the primaries open to supporters and voters became one of the most used leader selection methods. Naturally, this is mainly due to the PD, which since its creation in 2007, has included the instrument of primaries within its statute (Fasano and Seddone 2016; Vassallo and Passarelli 2016). But, as seen before, other parties –​such as Brothers of Italy –​have also resorted to open primaries to choose their leader. In recent years, a differentiation in methods for the selection of party leaders has been observed which, to a large extent, also depends on the diversity in organizational models adopted by individual parties. In addition, particularly since 2000, a slow process of greater inclusiveness of members/​sympathizers in internal decision-​making processes has been noted. It was in particular a process triggered by parties with a progressive ideological orientation which, however, recently seem to have ‘infected’ some conservative parties, which have managed to combine their traditional demand for strong leadership with the most widespread membership participation possible. This aspect can be observed in detail in Figure 2.5, which shows the frequency with which the main Italian parties have resorted to the different selection methods based on their ideological orientation. It is clear that the highest number of leaders selected through primaries can be found above all in the party family of socialist/​social democratic parties (to which the PD belongs). However, this instrument has also been adopted –​at least experimentally –​by conservative parties or by other actors (such as the M5S) with a less precise ideological position (Mosca and Tronconi 2019). Therefore, there is no longer a direct link between one particular method for choosing a leader and the ideology of a party. Increasingly inclusive selection processes of party leaders have become an increasingly common feature even among right-​ wing parties, including radical parties, which have succeeded in combining

M5S

1

Regionalist Conservative

1 2

2

1 3

Christian democrat

1

3

1

17

5

Liberal

16

Socialist

22

Communist 0%

1

3

1

4

9

10%

20%

Single leader

30%

40%

1

50%

60%

70%

Party elite Parliamentary group Closed primaries Open primaries

80%

90%

100%

Party delegates

Figure 2.5 Party leader selection method in Italy, by ideological orientation (1960–​2020).

Choosing party leaders in Italy  33 the personalization of the leadership with greater inclusiveness of internal decision-​making mechanisms. What can be observed so far is a profound transformation, linked above all to the instability of the Italian party system and the emergence of new parties characterized by very different organizational structures. These transformations have, on the one hand, set in motion a process of innovation that has affected, in particular, the recruitment mechanisms of candidates and leaders; and, on the other hand, an internal process of democratization in which primaries (open or closed) are the main channel of diffusion. It remains to be assessed, however, the extent to which the latter process of (presumed) internal democratization is real or artificial. In fact, in order to be effective, intra-​party democracy –​which is, in and of itself, a multi-​dimensional concept (Hazan and Rahat 2010) –​is based, not only on the degree of openness or inclusiveness of members/​supporters, but also on the degree of competitiveness in the selection process. If a wide number of members are called to choose (or ratify) just one candidate to the leadership, it is clear that in this case, the inclusiveness process represents just one facet, and perhaps not even the most important facet, of intra-​party democracy. For this reason, in addition to the variable of inclusiveness (or exclusiveness) of the selectorate in the leader selection process, it is necessary to take into consideration another factor that will allow for the observation and measurement of the degree of electoral competitiveness in the internal party contest. To this end, all the leadership selection methods of the Italian parties have been classified into six different categories according to the greater or lesser degree of competitiveness (see Table 2.3). Next, a clear cross-​time trend can be observed in terms of the two variables that have been used to define democracy within Italian parties. The image that emerges –​as shown in Figure 2.6 –​is very telling. In fact, the trajectories shown for the degree of inclusiveness of the selectorate and the electoral competitiveness are almost perfect mirror images of each other. From the post-​war period to today, the Italian party selectorate has grown continuously

Table 2.3  Operationalization of the variables ‘inclusiveness’ and ‘intra-​ party competitiveness’1 Inclusiveness

Intra-​party competitiveness

Category

Score

Margin between 1° and 2° candidate

Score

Single leader Party elite/​council Parliamentary group Party delegates Closed primaries Open primaries

0 1 2 3 4 5

100 (or only one candidate) 80-​99 60-​79 40-​59 20-​39 0-​19

0 1 2 3 4 5

34  Marco Valbruzzi 4.0 3.0 2.0 1.0 0.0

1940

1950

1960

1970

1980

Intra-party competitiveness

1990

2000

2010

Inclusiveness

Figure 2.6 Trend in intra-​party competitiveness and inclusiveness in Italian political parties, 1945–​2020.

and constantly over time, reaching its highest point in the last decade. On the contrary, the degree of competitiveness of party congresses or in leadership elections has progressively diminished, reaching its lowest point at the beginning of the twenty-​first century. Beyond the individual values, the data that emerge from this analysis shows a rather clear trend: while parties progressively enlarged their selectorate, opening internal channels of participation to members and supporters, the selection processes became less and less competitive. Therefore, it is as if there was a trade-​off between these two dimensions of intra-​party democracy in the sense that the maximization of one dimension seemed to bring about the reduction of the other. The most significant case, already mentioned above, is probably that of the election of Giorgia Meloni as leader of Brothers of Italy in 2014; she was elected by party members and supporters (about 250,000) in a competition that had one sole candidate. This was a coronation. Therefore, there was maximum inclusiveness with minimum (or even zero) competition. Hidden behind these long-​term trends, however, are significant differences between the individual parties that have formed (and some have disappeared) over the last seventy years. These differences can be seen in Figure  2.7, where each party is analysed in terms of the two aspects already mentioned above: ‘intra-​party competitiveness’ and ‘selectorate inclusiveness’. As can be seen, in the lower part of the graph, there were a number of parties created after the 1990s with varying degrees of inclusiveness in leadership choice, and they are characterized by a total lack of internal competition. Very often, these are personal parties that do not allow for the judgement or the questioning of the leader. Traditional mass parties such as the PCI (Italian Communist Party) belong to this category, which, despite widespread participation and broad involvement of activists, are based on the principle of ‘democratic

3

Choosing party leaders in Italy  35

2

PRI

MSI DC

PDS/DS

PD

PSI PPI

1

Intra-party competitiveness

PSDI

LN RC UDC

M5S

PCI

0

FI

0

IDV

1

SC

2

DL

3 Inclusiveness

AN

FDI

4

5

Figure 2.7 Intra-​ party competitiveness and inclusiveness of the Italian political parties, 1945–​2020.

centralism’ and, as a consequence, an (almost total) lack of competition for national leadership (Pasquino 1983). It is interesting to note that a ‘digital party’ such as the M5S, created only a decade ago, has similar characteristics to the PCI; low internal competition, but a heightened role of members in the leader selection processes. It is no coincidence that, when talking about the M5S, some scholars have spoken about a new form of ‘cybercratic centralism’ (Calise and Musella 2019; De Rosa 2014; Mosca 2015), where the control of the party by a central elite is guaranteed by the management of the online participation platforms. In line with the description of digital parties given by Paolo Gerbaudo (2019), the data outlined above also shows the existence of a ‘hyperleadership’ within the M5S, i.e. a leadership structure that is strongly concentrated around a single individual, supported by a so-​called ‘superbase’, or, rather, a base of online activists (‘nettivists’), who often limit themselves to (digitally) reacting to or supporting choices already made by the party electorate. Of all the parties taken into consideration here, it is the PD that has the most ‘eccentric’ and different profile compared to the others. Indeed, in this case, the extensive inclusiveness of the leader selection processes –​with primaries potentially open to the whole of the electorate –​is accompanied by a relatively high level of internal competition (and division). It is significant, from this point of view, that, with regards to the competitiveness of internal

36  Marco Valbruzzi

3

consultations, the PD is in line with the ‘traditional’ party principles of the First Republic (with the exception of the PCI), where the internal dynamic was characterized by a plurality of factions or currents competing with one another. For example, in the case of the DC –​and the PSI to a lesser extent –​ the leadership structure was defined as a ‘competitive oligarchy’ where single groups competed and distributed control of the main internal and external roles of the party. Considering the greater openness of the decision-​making processes observed in the PD, in this case, one could speak of ‘competitive polyarchy’; a situation which, as seen in the second section, has often led to a strong internal division and a frequent and frenetic change of leaders. To sum up, the profound transformations that have affected the Italian party system over the last thirty years have also changed the selection processes of party leaders which had consolidated from the end of the Second World War. The most significant changes can be observed above all with the two main aspects that give substance to intra-​party democracy: the inclusiveness of the selectorate and the competition in internal consultations. From this particular perspective, the relevant fact that emerges from the analysis of Italy is that the two aspects do not seem to have moved in the same direction but, rather, proceed along different and opposite paths. Indeed, when observing the transformations of leadership selection processes decade by decade, as shown in Figure 2.8, the progressive increase of power entrusted to members or supporters has moved at the same rate as the gradual decrease of

1950

2

1960

1980

1990

1

1970

2010

2000

0

Intra-party competitiveness

1940

0

1

2 Inclusiveness

3

4

Figure 2.8 Intra-​ party competitiveness and inclusiveness of the Italian political parties, by decade.

Choosing party leaders in Italy  37 intra-​party competitiveness. In this sense, the figure reported for the last two decades (2000s and 2010s) is significant. During this time, the highest levels of inclusiveness were accompanied with the lowest levels of competitiveness ever recorded in the post-​war period. Therefore, from the data examined in this section a particular ‘Italian model’ seems to emerge in terms of the process of personalization of politics in comparison with other countries. The strengthening and greater importance of party leaders have certainly been accompanied with a demand for greater openness and inclusiveness in decision-​making processes from the grassroots that is often fulfilled. At the same time, however, while members and supporters have a greater chance of making their voices heard within the party dynamics, their real capacity to influence clear and truly competitive choices is very limited, and sometimes totally absent. This is, therefore, the compromise that seems to have emerged in Italy to contain the ever-​increasing push towards personalization:  more extensive participation, that is often a façade, without real competition. A personalization that is praised from below but conditioned from above. A situation in which it is actually the ‘selected’ who select their own ‘selectors’. But what are the consequences on the political, social, and demographic characteristics of the party leaders? This will be addressed in the next section.

2.4  Demographics characteristics As seen in the previous sections, the transformations in the organizational models of Italian parties have had significant consequences on how leaders are chosen and, as a result, on their survival in office. So far, however, nothing has been said about the consequences that these changes can have (if any) on the characteristics of individual leaders, both in terms of sociodemographic as well as political profiles. It is therefore more useful to examine these aspects in more detail, especially to see if and to what extent different ways of selecting leadership have changed the profile of Italian party leaders. The first aspect that needs to be taken into consideration is personal, sociodemographic data. In Figure 2.9, all the leaders of Italian parties (see the list of 97 leaders analysed in Table A.1 in Appendix) are classified according to age group at the time of their first appointment as party leader. From this point of view, the changes in selection processes do not appear to have had any significant consequences on the age structure of the party leaders. On the contrary, many might have hypothesized that there would have been a positive correlation between introducing more inclusive instruments for choosing leaders and the generational turnover process, but the accessible data does not confirm this hypothesis. It is particularly significant, in this regard, that, when looking at the last decade (2010–​20), there were two party leaders over the age of 70 who came into office (both in Scelta Civica –​Civic Choice –​the party founded by former Prime Minister and current Senator for life Mario Monti). Even the average age of leaders taking office for the first time has remained

38  Marco Valbruzzi 2010-2020

1

3

2

2000-2010

4

3

4

1990-2000

2

1980-1990 1970-1980

4

5 2

7

2

1

2

5

2 3

5 2

0%

4

2

2

1945-1950

3

2

10%

20%

20-30 yrs.

1

6

4

1950-1960

5

4

1960-1970

2

3

30%

30-40 yrs.

1

40%

40-50 yrs.

50%

2

60%

70%

50-60 yrs.

80%

60-70 yrs.

1 1

90%

100%

70-80 yrs.

Figure 2.9 Age of the Italian party leaders at the time of the first selection (by class age), 1945–​2020.

2010-2020

5

2000-2010

2

1990-2000

1945-1950 0%

12

3

1970-1980

1950-1960

4

5

1980-1990

1960-1970

9

10

2

9

1

1

1

7

1

9

2

10%

7

20%

30%

Primary

40%

50%

Secondary

60%

70%

80%

90%

100%

Tertiary

Figure 2.10 Italian party leaders by level of education, 1945–​2020.

practically unchanged. In the first twenty years after the war, party leaders were, on average, 48.4 year old, and over the last twenty years (2000–​20) the average age is 48.9. Other than a few isolated exceptions, party leadership has remained firmly in the 40s–​50s range and does not appear to be linked to different selection processes. A similar argument holds when considering the level of education of leaders over time. As Figure 2.10 shows, from the post-​war period to today, there have been some changes in the level of education, but many of these changes, particularly over the last few years, do not appear to be in line with expectations. It is true that, at least for the 1960s, there were party leaders with

Choosing party leaders in Italy  39 only primary education (elementary or middle school), but over the following decade the number of leaders with secondary school diplomas increased, while the share of leaders with a university degree decreased. Between the 1970s and 1980s, university graduate leaders accounted for 80% of the total, while over the last twenty years, this percentage has reduced to just over 60%. In a certain sense, this fact makes party leadership slightly more representative with respect to the Italian population (where only 20% of citizens have a university diploma, see ISTAT 2020 report), but it certainly bucks the trend with respect to the increase in the level of education of citizens observed over the last 50 years. In the Italian case, there is no point taking the gender variable into consideration but for one reason that will be quickly explained. In fact, of all the party leaders considered in this chapter, there exists only one case of female leadership  –​the already mentioned Giorgia Meloni. Therefore, even from this perspective, the introduction of more inclusive mechanisms for selecting leaders has not affected the strongly imbalanced situation, in terms of gender, of Italian parties and has left the entry barriers for women intending to climb to the top of the party largely unchanged. Moving on now to the data analysis of the professional profiles of individual leaders (see Table  2.4) there are some elements that deserve to be analysed in detail. Firstly, contrary to our expectations, the share of professional politicians who have become party leaders has not decreased over the last decades but, particularly over the last twenty years, has significantly increased again, reaching 50% between 2010 and 2020. In fact, of the 14 party leaders who took office during this period, one in two falls under the category of ‘pure’ or professional politician.

Table 2.4 Occupation of the Italian party leaders, by decade (%), 1945–​2020

Fully paid politician Self-​employed and freelancer Trade unionist Entrepreneur Manager, prosecutor, professor Teacher, employee Blue-​collar worker N.

1945-​ 1950-​ 1960-​ 1970-​ 1950 1960 1970 1980

1980-​ 1990-​ 2000-​ 2010-​ 1990 2000 2010 2020

40.0

25.0

33.3

41.7

53.3

29.4

33.3

50.0

40.0

33.3

33.3

25.0

13.3

5.9

16.7

21.4

20.0 0 0

0 0 41.7

0 0 22.2

0 0 25.0

0 6.7 20.0

23.5 5.9 29.4

16.7 0 33.3

7.1 7.1 14.3

0

0

11.2

8.3

6.7

0

0

0

0

0

0

0

0

5.9

0

0

10

12

9

12

15

6

14

17

40  Marco Valbruzzi At the same time, while the number of professional politicians increased, the share of leaders who are magistrates or university professors decreased, especially over the last few years (to 14.3% in the last decade), as well as employees or teachers (now completely obsolete) and, to a lesser extent, trade unionists. It is significant to note that from the 1980s onwards there has been an increase in the share of entrepreneurs (in the economic sphere) who decided to ‘climb the ladder of politics’ and make their own way in an environment where internal ‘scalability’ was practically zero. Obviously, Silvio Berlusconi is the prime example of this, but he was then followed by other political entrepreneurs. It is, however, difficult to evaluate the overall impact of the different selection methods on transformations in the professional characteristics of leaders. The most significant aspect remains that most methods are linked to the growth of the share of leaders being professional politicians, particularly in a period of politics that is characterized by anti-​political populist rhetoric which, put simply, has exalted the image of civil society against the political class (Tarchi 2018). Despite the progressive opening of internal party participation channels, with the adoption of primaries and direct membership ballots, professional politicians have continued to maintain control of the key resource of the internal organization and, therefore, decisively influence the choice of the leader. To summarize, it might be useful to examine the different characteristics of Italian party leaders based on the different ways that they are selected. Table  2.5 analyses three different variables  –​time in office, the average age when selected, and political seniority (as MPs) when taking office –​in relation to the six selection methods mentioned in the previous sections. We can see that where the duration of the mandate is concerned, the relationship with greater or lesser inclusiveness of the selectorate is anything but linear. It is true that the longest periods in office (15.7 years on average) apply to cases of self-​selection, which is where the leader reappoints him or herself. However, it should also be noted that, even in cases of closed primaries, the length of Table 2.5 Features of Italian party leaders by selection method, 1945–​2020

Single leader Closed primaries Party delegates Open primaries Party elite Parliamentary group Mean

Length in office (in years)

Age at time of first selection

National political seniority as MP (in years)*

15.7 7.0 6.1 4.3 5.0 1.0 5.5

50.3 37.5 50.4 46.3 53.3 66.0 52.3

1.0 4.0 8.9 8.0 10.1 21.0 9.3

Note: * = at time of first selection.

Choosing party leaders in Italy  41 time of leaders in office (on average 7 years) is higher than all the other selection processes. Therefore, beyond the purely personal parties, it would appear that the survival in office of party leaders is not determined by the degree of inclusiveness the selectorate. Things are very different, however, when analysing the age of leaders when they first take office. As outlined above, the link with the different selection methods of leaders is very weak with regards to age, but the data in Table  2.5 does, however, show a distinctive trend. In fact, primaries  –​as a specific selection method –​are associated with a lower average age of selected leaders: 37.5 years for closed primaries and 46.3 years for open primaries. This instrument has effectively allowed relatively young leaders (such as Matteo Renzi for the PD, Luigi Di Maio for the M5S or Giorgia Meloni for Brothers of Italy) to reach the top of the party against previous cohorts of leaders that tend to be older. Finally, looking at the data on political seniority of party leaders in Italy, the greater or lesser degree of inclusiveness of the selectorate appears to have a limited impact. The lowest level (one single year of seniority, on average, before taking office) is recorded with regard to leaders selected without any competition, where the power of selection is in the hands of one person (the same leader). As already outlined above, this is particularly the case for personal parties, often created in opposition to traditional parties (and politicians). They are therefore created with the input of ‘newcomers’ or outsiders who are not (yet) part of the state institutions and with little to no prior political experience. If, however, the cases of these personal parties are excluded –​an anomaly in comparative terms, but absolutely the norm in Italy  –​a weak link between inclusiveness of the selectorate and political seniority of leaders can be observed. Indeed, if the average time in office of national parliament for party leaders is always less than 9  years in the case of primaries or the selection amongst delegates to congress, this number goes beyond 10 years of political seniority for leaders selected with less inclusive selection mechanisms (the party elite/​council or parliamentary group). In sum, even if the evidence of the impact of leader selection methods on the sociodemographic and political characteristics is anything but conclusive, there is a trend that seems to emerge more or less clearly. It is actually the leaders who take office thanks to more inclusive and open selection methods integrating party members or sympathizers who bring about greater renewal of the political class, both in generational terms as well as political seniority. Even if the available data is still insufficient to prove this hypothesis, the first signs seem to indicate a link between selectorate inclusiveness and the potential turnover of the political class.

2.5  Conclusions From the 1990s the Italian party system has undergone profound and sudden changes such as justifying the distinction between first-​and second-​party

42  Marco Valbruzzi systems (Pasquino and Valbruzzi 2015). These were not just changes in form or in appearance, but rather, transformations in the organizational structures of the parties themselves, as well as in their (post-​)ideological positions. Many authors have focused on the systematic effects of these transformations, starting with the consequences an increasingly de-​structured party system can have on the functioning of the political system and a representative democracy (Chiaramonte and Emanuele 2017). However, much less effort has been devoted to investigating the effects on the very functioning of parties in the performance of their essential functions. This chapter has analysed the consequences of party transformations on the crucial activity in the life of a party, namely the selection of a leader. What has changed under this perspective in the Italian context? And what consequences have these changes had on political leadership? The main trend that emerges from the analysis has a dual nature. On the one hand, particularly when observed from below, i.e. from the perspective of members and voters, we can detect a progressive opening of parties towards society and, more specifically, an increasingly bigger space for supporters when making key party decisions, starting with the selection of a leader. The broadening of the selectorate is not just an Italian trend nor a European trend. However, Italy finds itself in the driving seat of this institutional innovation process that, starting within progressive parties, has ended up spreading to centre-​ right parties, including the more radical forces. However, the fact that members or voters are increasingly making their voices heard in parties does not automatically mean that they have acquired more power or that their opinions are decisive (Katz and Mair 1995; Ignazi 2017). Here emerges the second ‘façade’ of the transformations observed in Italy with regards to the leader selection mechanisms and particularly with regards to the degree of competitiveness in internal consultations. Indeed, Italian parties have progressively broadened their selectorate, but, at the same time, have also strongly restricted the menu of options available to their members or supporters. There are obviously moments or episodes where the contest has been intense and unpredictable in its outcome (as was the case for a few leadership votes in the Democratic Party), but this is the exception that proves, rather than disproves, the rule. This particular combination of opening-​up of and elite control over such processes or, to be more specific, high inclusiveness and low competitiveness, has allowed Italian parties to coexist in a context characterized by increasing personalization of party and political organizations. Save for rare cases of successful personal parties (as is the case for Berlusconi’s party), the increase in personalization required a broader support base that went beyond that of traditional local activists. Primary elections or other similar instruments of involvement of voters in internal party processes have served as an effective weapon for leaders seeking legitimacy. Broadening the potential foundation of their own supporters, aspiring party leaders have not only managed to broaden their support but have often achieved the goal of overcoming

Choosing party leaders in Italy  43 or silencing internal opposition. Paradoxically, primaries, thought of as a tool for empowerment for members/​voters have ended up assigning more power and more resources to party leaders, often to the detriment of the organizational structure or the so-​called middle-​level elite (Katz 2018). In a certain sense, it is a kind of personalization that can be defined as ‘bottom-​ up’ because it require can the base in order to strengthen the top (unlike traditional personal parties that typically operate using a top-​down approach). However, beyond the continuation of further personalization of politics, the result is always the same:  a weakening of parties as organizational structures supporting groups, identities, and collective action (Ignazi 2020). Nevertheless, when an organization becomes little more than the shadow of a single leader, it is legitimate to begin to wonder whether these are still political parties.

Note 1 Although they are of a categorical/​ordinal nature, I  treat these two variables as continuous, ranging from 0 to 5. Treating these variables as continuous is common when ordinal variables have at least five categories (Bollen and Barb 1981; Johnson and Creech 1983).

References Albertazzi, D., Giovannini, A., and Seddone, A. (2016). ‘No regionalism please, we are Leghisti!’ The transformation of the Italian Lega Nord under the leadership of Matteo Salvini. Regional & Federal Studies, 28(5): 645–​671. Bakker, R., Hooghe, L., Jolly, S., Marks, G., Polk, J., Rovny, J., Steenbergen, M., and Vachudova. M. A. (2020). 2019 Chapel Hill Expert Survey.Version 2019.1. Available on chesdata.eu. Chapel Hill, NC: University of North Carolina, Chapel Hill. Biorcio, R. (1998). La rivincita del Nord. La Lega dalla contestazione al governo. Roma-​Bari: Laterza. Bittner, A. (2011). Platform or Personality?: The Role of Party Leaders in Elections. Oxford: Oxford University Press. Bollen, K. A. and Barb, K. H. (1981). Pearson’s R and coarsely categorized measures. American Sociological Review, 46(2): 232–​239. Bordignon, F. (2013). Il partito del capo. Da Berlusconi a Renzi. Santarcangelo di Romagna: Maggioli Editore. Bordignon, F. and Ceccarini, L. (2013). Five stars and a cricket. Beppe Grillo shakes Italian politics. South European Society and Politics, 18(4): 427–​449. Bordignon, F. and Ceccarini, L. (2019). 5 stars. 5  years. 5 (broken) taboos. In L. Ceccarini and J. L. Newell (eds), The Italian General Election of 2018:  Italy in Uncharted Territory, Basingstoke: Palgrave, pp. 139–​163. Calise, M. (2010). Il partito personale. I due corpi del leader. Roma-​Bari: Laterza. Calise, M. and Musella, F. (2019). Il Principe digitale. Roma-​Bari: Laterza. Chiaramonte, A. and Emanuele, V. (2017). Party system volatility, regeneration and de-​institutionalization in Western Europe (1945–​2015).Party Politics, 23(4): 376–​388.

44  Marco Valbruzzi Cross, W. P. and Pilet, J. B. (eds) (2016). The Politics of Party Leadership: A Cross-​ National Perspective. Oxford: Oxford University Press. Cross, W. P., and Katz, R. S. (eds) (2013). The Challenges of Intra-​Party Democracy. Oxford: Oxford University Press. De Rosa, R. (2014). Cittadini digitali. L’agire politico al tempo dei social media. Rimini, Maggioli. Diamanti, I. (2016). A un mese dal voto in vantaggio il No, il Pd si compatta intorno al premier. La Repubblica, 30 October. Fasano, L. M. and Seddone, A. (2016). Selecting the leader, Italian style. Contemporary Italian Politics, 8(1): 83–​102. Gallagher, M. and Marsh, M. (eds) (1988). Candidate Selection in Comparative Perspective: The Secret Garden of Politics. London: Sage. Gerbaudo, P. (2019). The Digital Party: Political Organisation and Online Democracy. London: Pluto Press. Hazan, R. Y. and Rahat, G. (2010). Democracy within Parties:  Candidate Selection Methods and Their Political Consequences. Oxford: Oxford University Press. Ignazi, P. (2014). Vent’anni dopo. La parabola del berlusconismo. Bologna: Il Mulino. Ignazi, P. (2017). Party and Democracy. The Uneven Road to Party Legitimacy. New York, Oxford: Oxford University Press. Ignazi, P. (2020). Party change and citizens’ retrotopia.Contemporary Italian Politics, 12(4).Johnson, D. R. and Creech, J. C. (1983). Ordinal measures in multiple indicator models:  a simulation study of categorization error. American Sociological Review, 48(3): 398–​407. Karvonen, L. 2010. The Personalization of Politics:  A Study of Parliamentary Democracies. Colchester: ECPR Press. Katz, R. S. (2018). Personalization, party government and democracy. In W. P. Cross, R. S. Katz, and S. Pruysers (eds), The Personalization of Democratic Politics and the Challenge for Political Parties, Oxford: Oxford University Press, pp. 215–​232. Katz, R. S. and Mair, P. (1995). Changing models of party organization and party democracy:  the emergence of the cartel party. Party Politics, 1(1):  5–​28.Meijers, M. and Zaslove, A. (2020). Populism and Political Parties Expert Survey 2018 (POPPA). https://​doi.org/​10.7910/​DVN/​8NE​L7B, Harvard Dataverse. Mosca, L. (2015). Problemi e limiti del modello organizzativo ‘cybercratico’ nell’esperienza del Movimento 5 Stelle. Ragion Pratica, 13(1): 37–​52. Mosca, L. and Tronconi, F. (2019). Beyond left and right:  the eclectic populism of the Five Star Movement. West European Politics, 42(6):  1258–​1283.Musella, F. (2015). Personal leaders and party change: Italy in comparative perspective. Italian Political Science Review, 45(3): 227–​247. Musella, F. (2018). Political Leaders beyond Party Politics. London: Palgrave. Panebianco, A. (1988). Political Parties: Organization and Power. Cambridge: Cambridge University Press. Pasquino, G. (1983). Programmatic renewal, and much more:  From the PCI to the PDS. West European Politics, 16(1): 156–​173. Pasquino, G. (2014). Italy:  The triumph of personalist parties. Politics & Policy, 42(4): 548–​566. Pasquino, G. (2019). Italian Democracy. How It Works. London: Routledge. Pasquino, G. and Valbruzzi, M. (2015). A Changing Republic. Politics and Democracy in Italy. Novi Ligure: Epoké.

Choosing party leaders in Italy  45 Pasquino, G. and Valbruzzi, M. (2017). The Italian Democratic Party, its nature and its Secretary. Revista Española de Ciencia Política, 44(3): 275–​299. Passarelli, G. and Tuorto, D. (2012). Lega & Padania Storie e luoghi delle camicie verdi. Bologna: Il Mulino. Passarelli, G. and Tuorto, D. (2018). La Lega di Salvini. Estrema destra di governo. Bologna: Il Mulino. Pennings, P. and Hazan, R. Y. (2001). Democratizing candidate selection: Causes and consequences. Party Politics, 7(3): 267–​275. Pilet, J. B., and Cross, W. (eds) (2014). The Selection of Political Party Leaders in Contemporary Parliamentary Democracies:  A Comparative Study. New  York: Routledge. Poguntke, T., and Webb, P. (eds) (2005). The Presidentialization of Politics:  A Comparative Study of Modern Democracies. Oxford: Oxford University Press. Polk, J., Rovny, J., Bakker, R., Edwards, E., Hooghe, L., Jolly, S., Koedam, J., Kostelka, F., Marks, G., Schumacher, G., Steenbergen, M., Vachudova, M. A., and Zilovic, M. (2017). Explaining the salience of anti-​elitism and reducing political corruption for political parties in Europe with the 2014 Chapel Hill Expert Survey data. Research & Politics, 4(1): 1–​9. Rahat, G. and Kenig, O. (2018). From Party Politics to Personalized Politics?: Party Change and Political Personalization in Democracies. Oxford:  Oxford University Press. Rombi, S. and Serricchio, F. (eds) (2020). L’elezione di Zingaretti. La rivincita del partito?. Novi Ligure: Epoké. Sandri, G., Seddone, A., and Venturino, F. (eds) (2015). Party Primaries in Comparative Perspective. Farnham: Ashgate. Sartori, G. (1976). Parties and Party Systems. A Framework for Analysis. Cambridge: Cambridge University Press. Tarchi, M. (2018). Italia populista. Dal qualunquismo a Beppe Grillo. Bologna:  Il Mulino. Tronconi, F. (ed.) (2015). Beppe Grillo’s Five Star Movement:  Organisation, Communication and Ideology. Farnham, Ashgate. Tronconi, F. (2019). The Italian Five Star Movement during the crisis: Towards normalisation? South European Society and Politics, 23(1): 163–​180. Valbruzzi, M. (2005). Primarie. Partecipazione e leadership. Bologna:  Bononia University Press. Vassallo, S. and Passarelli, G. (2016). Centre-​left prime ministerial primaries in Italy: The laboratory of the ‘open party’ model. Contemporary Italian Politics, 8(1): 12–​23.

3  Playing with fire? The organizational consequences of party primaries in Spain Oscar Barberà and Juan Rodríguez-​Teruel

3.1  Introduction Since the late 1960s, more and more parties have opened their leadership and candidate selection methods to all members and, eventually, to supporters and voters (Scarrow and Kittilson 2003; Caul Kittilson 2006; Kenig 2008, 2009; Krowel 2012). This is not, however, a universal trend (Cross and Blais 2012; Pilet and Cross 2014; Cross and Pilet 2015). Spanish parties, for example, were able to resist more participatory and democratic practices until the late 2000s (e.g. Méndez et al. 2004; Verge 2007; Pérez-​Moneo 2012; Barberà et al. 2014; Barbera, Lisi, and Rodríguez-​Teruel 2015). As stated in the introduction of this edited volume, democratic innovations, and party primaries in particular have had a relevant impact on the way parties operate and, hence, prompted several research avenues (see Chapter  1 in this volume). Some scholars have tried to link this phenomenon with others such as the long-​term transformation of intra-​party power relations and changing party models (e.g. Katz and Mair 1995, 2009; Scarrow 1999; Bolleyer 2009). Others have also noted their connections with other rather complex phenomena such as the personalization of politics or its growing professionalization (e.g. Poguntke and Webb 2005; Rahat and Kenig 2018). Finally, a third group of studies has been focusing on the internal or external consequences of these processes (e.g. LeDuc 2001; Pennings and Hazan 2001; Kenig 2008, 2009; Hazan and Rahat 2010; Pilet and Cross 2014). The aim of this chapter is to develop a theoretical framework to empirically assess the organizational consequences of introducing party primaries, and then apply it to properly understand the main organizational impacts in Spain’s mainstream parties. More specifically, our main research question focuses on the impact that the degree of inclusiveness of the selectorates and other internal regulations of primary elections have on the elites and leaders’ power within the party. Hence, the next section is devoted to build the theoretical framework. The following one will describe the main formal features of party primaries in Spain. An assessment of party primaries’ effects on several Spanish case studies will be presented and discussed according to the

Party primaries in Spain  47 theoretical framework. The chapter will close with a short overview of our results and some concluding remarks.

3.2  The organizational consequences of introducing party primaries on the party elites’ strength The study of the political consequences of leadership and candidate selection methods is a promising yet underdeveloped field. It seeks to establish the internal or external political effects produced by candidate or leadership selection methods generally defined by the degree of inclusiveness of the selectorate, the most relevant feature of any selection procedure (Hazan and Rahat 2010: 35; Kenig et al. 2015).1 Until recently, much of this literature focused on the analysis of US candidate selection processes and was mainly concerned with the study of their external effects, namely on the election results. This approach has substantially changed over the last two decades. A  new and emerging (European) literature has focused mainly in parliamentary democracies, expanded its attention to party leaders, and inquired into their internal effects (e.g. Ware 2002; Hazan and Rahat 2010; Sandri et al. 2015). This new research avenue has received a growing amount of attention in recent years. Lead by Rahat and Hazan’s seminal works, the literature has identified the potential effects that the selectorate might have  –​at both the candidates or party leaders’ level –​on several dimensions of party internal life, such as in terms of participation, representation, competition, and responsiveness patterns (Rahat et al. 2008; Hazan and Rahat 2010; Kenig 2009; Pilet and Cross 2014; Cross and Pilet 2015). At an organizational level, the literature has mostly focused on analysing the potential impacts of highly inclusive selection methods on party cohesion.2 Two main approaches have been working on opposite directions. The first one, that might be labelled as party decline thesis, is claiming that there is a negative relationship between the inclusiveness of the selectorate and the party cohesion (e.g Ranney 1972; Penning and Hazan 2001; Hopkin 2001). The rationale underpinning that approach is that more inclusive selectorates transform parties towards a candidate-​ centred responsiveness model, which weakens cohesion and that leads to electoral defeats or membership decline. The cartel party model (Mair 1994; Katz and Mair 1995, 2009) has embraced the opposite direction of the party decline thesis: party primaries do not promote conflicts and weaken cohesion. Instead, they strengthen the powers of the party leadership and, in a broader sense, the party in public office. The well-​known arguments are that the introduction of highly inclusive procedures engenders the concentration of decision-​making powers and the empowerment of the party leadership at the expense of the intermediate party elites and activists. This provides the former with the much-​needed autonomy to compromise with other party leaders across party lines. In addition, stratarchically organized parties (Carty 2004) might be seen as well as

48  Oscar Barberà & Juan Rodríguez-Teruel an extension of the cartel party model. This is because they provide both autonomy for the party leadership at the national level and autonomy for their regional branches. The result is then, the strengthening of the party leadership at the national level and the vertical party cohesion between the centre and the periphery (Hazan and Rahat 2010: 154). Both the party decline and the cartel party approaches assume that party primaries do produce internal consequences in terms of party cohesion. However, a third approach denying their strong impact could also be adopted. Borrowing from Hirschman (1991), the futility thesis might claim that the consequences of party primaries on both dimensions are irrelevant or hardly change the previous (in)stability of the party. One of the main arguments underpinning this hypothesis states, on the one hand, that highly centralized and cohesive parties might be willing to implement party primaries as a purely marketing strategy and, hence, set up quite strict formal requirements to control the processes, avoid conflicts, and preserve party unity. Party primaries could have, for example, high candidacy requirements to discourage potential candidates and either led to ‘cover-​up’, manipulated elections or to ‘coronations’ (i.e. elections with just one contending candidate). On the other hand, in highly factionalized parties, primaries can be understood as the last resort to solve already existing conflicts (Ramiro and Verge 2013). In this context, high levels of competitiveness and negativity should be expected. They, in turn, would hardly lead to strengthening the power of the party leadership or calming down the internal divisions. The main assumption behind the futility thesis is that the previous organizational background of the party is a relevant control variable that has to be taken into account in order to fully understand the consequences of party primaries. Highly divided parties with primaries should expect to have very different consequences than highly united ones. Unfortunately, this is something that the literature has not properly addressed. This is indeed even more important when considering the consequences of their introduction for the first time, because they set an important precedent for the organization (see Astudillo and Detterbeck 2020). The literature bridging both phenomena has not clearly pointed out so far the path dependency effects involved in this relationship. At the candidates and party leaders’ level, the consequences of intra-​party elections might be analytically understood as a one-​off phenomenon. However, when considering the party as a whole, the impacts go far beyond that. This is especially relevant when considering the effects of party primaries being held for the very first time. Primaries become then a critical turning point for the organizational evolution of the party and set important precedents for the future. In this respect, the degree of institutionalization of primaries should be considered a relevant variable shaping future outcomes, as this has extensively discussed in the introduction of this edited volume (see Sandri and Seddone 2021, Chapter 1 in this volume).

Party primaries in Spain  49 DIRECT OUTCOME

INDIRECT OUTCOMES

SEQUENCE 1

PARTY COHESION BACKGROUND

PARTY ELITES’ STRENGTH

Minor or non-existent Previous conflicts

Strong party leader / candidate

(End to Previous conflicts)

NO New conflicts

FUTURE USE OF PRIMARIES

Regular use of primaries

SEQUENCE 2

FIRST PRIMARIES

Major Previous conflicts

Weak party Leader / candidate

(Previous conflicts Remain)

Delayed use of primaries

New conflicts

Figure 3.1 Ideal types of sequences linking party primaries with party elites’ strength and other indirect outcomes. Source: Authors’ own.

Figure 3.1 suggests two ideal types of sequences linking the introduction of party primaries with some direct and indirect outcomes. The most direct outcome or consequence of party primaries is the future party leader or candidates’ strength. That leads to the indirect outcomes in terms of party cohesion such as the (possible) ending of previous conflicts or the emergence of new disagreements. Party elites’ strength and party cohesion then become the main factors influencing the most direct outcome of the sequences, namely the future use of party primaries (Figure 3.1). The main assumption behind the first ideal sequence is that first time party primaries strengthening a party leader or candidate’s power might serve to solving past conflicts and/​or avoiding the rise of new dissent will be greatly valued and then lead to their regular use and institutionalization in the future. On the other hand, the second ideal sequence suggests that first time party primaries could lead to the selection of weak party leaders and this might not contribute to end previous internal dissent or, eventually, lead to new conflicts. In these circumstances, primaries will then be perceived as a failure, and therefore cast away or irregularly implemented in the future, hence limiting or postponing their institutionalization. It is worth pointing out that in both sequences the existence of previous conflicts acts a control variable influencing the party cohesion and as an intervening variable influenced by the party leader or candidate’s strength.

50  Oscar Barberà & Juan Rodríguez-Teruel The sequences presented in Figure 3.1 do not include other related variables that might be shaping the primary elections’ effects. As suggested in the introduction of this edited volume, the level of institutionalization of party primaries on the party system could act as a strong incentive to promote their future, particularly if the party wants to be seen as democratic or responsive. On the other hand, the influence that first-​time party primaries might have on the power of the party elites might also be shaped by the party Statutes and other internal regulations, which in turn can give them the legitimacy to take decisions in order to promote party cohesion and unity. The rest of this chapter will try to discuss and assess, according to the suggested ideal sequences, which have been the main internal impacts derived from the introduction of party primaries in Spain. In this sense, the evidence will mainly focus on the consequences in terms of party cohesion and its subsequent use. Although a proper test of both ideal sequences should need more comparative research, the Spanish case study should be able to provide some preliminary results and allow further discussion of their value.

3.3  The introduction of party primaries in Spanish politics The introduction of party primaries in Spain came several decades after the restoration of democracy. In 1991 the Republican Left of Catalonia (ERC) changed the rules of its party congress by allowing all members to attend and vote and by banning vote delegation.3 In 1993 the United Left (IU) was the first party to embrace an OMOV contest as a candidate selection method.4 Since then, several mainstream parties begun testing with primary elections for either candidates or party leaders (Méndez et al. 2004: 196; Verge 2007: 321; Pérez-​Moneo 2012). Most of them were organized locally. The limited and fragmented information available on these local ballots makes them very hard to analyse (but see Ramiro 2016). That is why in the following pages only primary elections above the local level are considered. In addition, party primaries have also been embraced by almost all the main Spanish new statewide parties created after 2005 such as Ciudadanos (Citizens), Unión Progreso y Democracia (UPYD, Union Progress and Democracy), EQUO, Podemos (We can) or VOX. All these new statewide parties opted to organize party primaries since their inception as a way to differentiate themselves from the two biggest Spanish parties, the PP (Popular Party) and the PSOE (Spanish Workers’ Socialist Party). The same might be stated of new regional parties emerged by the mid or late 2000s (Barberà 2018; Barberà and Rodríguez-​ Teruel 2019). Due to their different nature, these parties are excluded from the analysis. Hence, these chapter is focusing on primaries for the main positions (top candidates and party leaders) in state-​wide and regional parties that successfully achieved representation in regional or national parliaments before 2005 (see Table A.2 in Appendix for full list and some features). Party (open or closed) primaries for top candidates and party leaders have been introduced quite unevenly (Barberà et  al. 2015). Table  3.1 shows the

Party primaries in Spain  51 Table 3.1 First time party primaries in mainstream Spanish parties National Level

Regional Level*

Top Candidate

PSOE 1998 ICV 2000 IU 2007 ERC 2011 ICV 2011 CHA 2015 BNG 2015

Party Leader

PSOE 2014 IU 2016 PP 2018

IU 1998 PSOE 1998 PSC 1999 EUiA 2002 ICV 2006 CHA 2010 BNG 2012 PAR 2014 ERC 1991 ICV 2000 BNG 2002 PP 2010 PSOE 2013 PSC 2014 IU 2016

Source: Author’s own based on press and official figures. Notes: *First time a regional party or the party branch of a statewide party introduced primaries. List of acronyms and party families (see Table A.2 in Appendix).

year when a Spanish mainstream party organized its first primary ballot. It also distinguishes between top candidates or party leaders and the national, regional, and EU level. All parties introducing the procedure in their party regulations but not organizing a ballot have been excluded. Parties that seldom used primaries to select some candidates but not their top candidate or their party board but not the party leader are also excluded. Finally, Table 3.1 do include the selection of party leaders or top candidates through one-​member-​ one-​vote assemblies (Kenig et al. 2015).5 ERC’s 1991 decision to allow One-​Member-​One-​Vote (OMOV) procedures in its assemblies was quite a pioneer move. Taking that exception aside, the first wave of party primaries was mainly limited to top candidate selection purposes. Most left parties (PSOE, PSC, IU, ICV, EUiA, BNG) decided to experiment with them at the regional level during the late 1990s and early 2000s. The Spanish Socialists (PSOE) also implemented them at the national level, while the Catalan Iniciativa per Catalunya-​Verds (ICV, Catalan Iniciative for Catalonia-​Greens) and the Galician Bloque Nacionalista Galego (BNG, Galician Nationalist Block) were the only ones to open these processes to the selection of their party leaders. The controversial effects that the PSOE experimented with the selection of its national top candidate slowed down their further expansion to new roles or parties for almost a decade (Boix 1998; Salazar 2000; Hopkin 2001). The second wave started in the early 2010s with the breakthrough of some new parties such as UPYD and, later on, Podemos or Ciudadanos. The main features of this new wave were the extension of party primaries to

52  Oscar Barberà & Juan Rodríguez-Teruel most positions and levels and its tentative adoption by centre right parties as the Partido Popular (PP, Popular Party) and the Partido Aragonés (PAR, Aragonese Party). At the beginning of 2016 most of the mainstream Spanish parties had experimented at least once with primaries to select either their top candidates or party leaders in at least one level. The main exceptions were some centre-​right regionalist parties such as Convergència Democratica de Catalunya (CDC, Democratic Convergence of Catalonia), the Partido Nacionalista Vasco (PNV, Basque Nationalist Party) or Coalición Canaria (CC, Canary Coalition). On the other hand, since the mid-​2010s several parties such as the regional ICV, IU, or the PSOE have started to extend party primaries at all possible levels. The regulation of party primaries differs by parties and territorial presence (Table  3.2). The early adopters (ERC, PSOE, IU, EUiA) kept candidate selections processes closed to their party members. The main exceptions were the Partit dels Socialistes de Catalunya (PSC, Catalan Socialists Party) that in 1999 selected its regional top candidate in a coronation open to all Catalan citizens, and ICV that opened up all processes to its supporters (registered but non-​paying fees).6 Since the mid 2010s some parties such as the PSOE, the PSC, IU, or Chunta Aragonesista (CHA, Aragonese Assembly) have extended the vote to their supporters and, eventually, citizens. On the other hand, all parties but ICV have kept closed primaries as their default party leadership selection method. The candidacy requirements are one of main criteria introduced by the mainstream party boards to control and restrict the internal competitiveness. Most parties have established membership as the minimum candidacy requirement for both top candidates and party leaders. However, the PSOE allowed no party members to contest in top candidate selection processes till 2014 (Salazar 2000:  147). In 2015 both the BNG and the CHA formed electoral coalitions and organized party primaries to select their candidates, allowing non-​party members to contest. The most restrictive feature is the required support for would-​be candidates. There are large differences between the Spanish parties. Some as Esquerra Unida i Alterantiva (EUiA, United Alternative Left) (20%), the PSC (10%), or even the PSOE (10%) have stipulated quite restrictive requirements for the would-​be top candidates. In other parties such as the PP (just 100 members for the party leader), ICV (50 members), ERC (5%) or IU (3% since 2014) the requirements are much lower or even inexistent. High supports have traditionally restricted the number of competitors, promoted ‘coronations’ (Kenig 2009) or avoided the elections. Avoiding the ballot when just one candidate stands is another key stipulation that has restricted the use of intra-​party democracy (Astudillo and Detterbeck 2020). This has been implemented to prevent the economic costs and potential legitimacy problems that ‘coronations’ might have for the would-​ be candidate (i.e. in case the turnout is very low). Most of the mainstream Spanish parties have opted for this mechanism while the new parties have stated the obligation to hold elections even for ‘coronations’. The downside

newgenrtpdf

Table 3.2 Spanish mainstream parties: Party primaries main features by 2015 Selectorate

BNG

M

CHA

M

ERC*

M

EUiA ICV IU

M M&S M+S

PAR

M+S

PP

M

PSC

M + C>16

PSOE M + C>18

Candidacy Member Support/​endorsement No National Candidate: None Regional Top Candidate: n.a. Yes Party Leader: n.a. No National Candidate: None Yes Regional Top Candidate: 150 members Yes Regional Top Candidate: 1/​3 Party Board or 25% of National Council or 10% members Party leader: 5% members Yes Top candidates: 20% members or 10% National Council Yes Top Candidates and Party Leaders: 50 members Yes National Top Candidate: Min. 3% Max. 15% members Regional Top Candidates: Between 3% and 5% members Regional Party Leaders: Between 5% and 10% members Party Leader: Between 5% and 10% members n.a. Top Candidates n.a. Yes Regional Party leader (Balearic I.): 90 members Party Leader: 100 members Yes Top candidates: Min. 10% Max 15% National Council or Min. 5% Max. 10% members Party leader: 10% members Yes National top Candidate: Min. 5% Max. 10% members Regional top Candidates: Min. 10% Max. 20% members Party leader: 5% members Regional Party Leaders: 10% members

Compulsory Election

Formula

Voting System

Yes

Plurality

M&S

Yes Not if 1 candidate Not if 1 candidate

n.a. Plurality Plurality

A M&S B B or P

Yes Not if 1 candidate Yes Not if 1 candidate

Plurality Plurality Plurality

B B or E B or P or E A or E

n.a.

Plurality

Yes

Plurality

Not if 1 candidate

Plurality (runoff)

B or P A M&A B

Not if 1 candidate

Plurality

B

Source: Author’s own from party regulations and the press. Notes: A: Party Assembly (centralized ballot boxes); B: Party Branches; C: Citizens; E: Electronic Voting; M: Members (decentralized ballot boxes); P: Postal voting; S: Supporters.

Party primaries in Spain  53

Party

54  Oscar Barberà & Juan Rodríguez-Teruel of such key stipulation is that the incumbent party leadership might try to exclude other candidates to avoid competition and, hence, maintaining the control of the nomination. The formal way to do it is, of course, by setting high candidacy requirements. The informal way might include all sorts of pressures to avoid not welcomed contenders. Since most of these processes just fill a unique position, all parties have used the plurality formula in their primary elections. Differences on the voting process are mostly due to the type of election. Some parties such as the BNG, ERC (till 2008) or the PP (in 2010) select their party leaders through OMOV ballots. Top candidates are, on the other hand, generally selected through decentralized ballots held on local and regional branches. Since the mid 2000 more and more parties have experimented with electronic voting methods. The early adopters such as the regional ICV used the technology as a way to complement the logistic problems of postal voting. By the mid-​2010s, the technology and internet penetration in Spain were ready to its widespread diffusion. Then, new parties such as Podemos decided to use internet remote internet voting as their default way to hold all types of voting. Since then, many other parties have started to allow members the option of online voting.

3.4  The organizational consequences of first-​time party primaries in Spain This section presents the main internal consequences derived from the introduction of party primaries either for both top candidates and party leaders at the national and regional level in contemporary Spain. To do so, Table 3.3 provides several indicators according to the main variables suggested in the theoretical framework. Our main dependent variable, the strength of the party leader or candidate elected through the first party primaries organized ever by the corresponding party is measured by the share that the winning nominee obtained in the internal elections. Table 3.3 also provides some contextual information on the introduction of party primaries in each party: the year in which they were held, the type of position being contested (party leader or candidate), and their territorial level (regional or national). Hence, for each party there might be up to four cases (two levels per two type of positions):  regional party leaders, regional top candidates, national party leaders and/​or national top candidates. Party cohesion, the first indirect outcome and second dependent variable, is measured qualitatively according to official information (i.e. the parties’ websites), news reported by Spain’s major media outlets (such as El País or El Mundo) and secondary literature (e.g. Argelaguet 2009; Verge 2007; Rodríguez-​Teruel et al. 2010; Pérez-​Moneo 2012; Barberà et al. 2014; Barberà, Lisi, and Rodríguez-​Teruel 2015; Gómez Reino 2011; Ramiro and Verge 2013). Party cohesion measures are divided into two dimensions: 1. Horizontal dynamics, which are related to conflicts between groups or individuals at the apex of each organizational layer of a party (e.g. Panebianco

newgenrtpdf

Table 3.3 The internal consequences of party primaries in mainstream Spanish parties: a first qualitative assessment Outocome of the first primaries

Indirect outcomes (I): cohesion post-​primaries

Year

Type of post

Share of winner

Ends Vertical conflicts

Ends Horizontal Conflicts

New Vertical conflicts

New Horizontal conflicts

Party splits

ICV PP

2000 2010

NC & RPL RPL

67 & 100 69,2

-​ YES

YES YES

NO NO

NO NO

NO NO

IU

2016

RPL

71,5

NO

2018

PL(NC)

37,0

NO

NO

YES (minor) NO

EUiA CHA

2002 2010

RC RC

63 80

-​ -​

YES (Major) YES (Minor) YES YES

NO

PP

YES (Major) -​

NO NO

NO NO

NO NO

ERC PAR

2011 2014

NC RC

65 58

-​ -​

YES YES

NO NO

NO NO

NO NO

PSC

2014

RPL

85

NO

YES

NO

NO

PSOE 1998

NC & RC

55,0 & 64*

-​

NO (Minor)

YES (Minor)

YES (Major)

YES (Major) NO

IU

1998

RC

70,5

-​

NO

2007

NC

62,5

NO (Minor)

YES (Major) YES (Major)

NO

IU

NO (Major) NO (Major)

NO

YES (minor)

Regular Regular (Not extended to RC) Regular (Extended to other RPL) Regular (Extended to RPL) Inexistent till 2008 Inexistent till 2019 (Not extended to PL) Inexistent till 2019 Irregular (not in 2019) (Not extended to PL) Irregular (Extended to other positions) Very occasional till 2014 (Not extended to PL) Inexistent till 2015 Inexistent till 2015 (continued)

Party primaries in Spain  55

Party

Indirect outcomes (II): Future use of primaries

newgenrtpdf

Outocome of the first primaries Party

Year

Type of post

Indirect outcomes (I): cohesion post-​primaries Share of winner

Ends Vertical conflicts

Ends Horizontal Conflicts

New Vertical conflicts

New Horizontal conflicts

Party splits

PSOE 2013/​4 RPL & PL

77 & 48,7

YES

1991

RPL

91,6

YES (Minor) NO

YES (Minor) NO

NO

ERC

NO (Minor) -​

BNG

2012

RC & RPL

53 & 54

-​

NO

NO

BNG

2002

RPL

73

-​

NO

NO

YES (Major) YES (Major) NO

PSC

1999

RC

99,2

-​

NO

NO

NO

BNG CHA

2015 2015

NC NC

60,3* n.a.

-​ -​

NO NO

NO NO

NO NO

NO (Major) NO (Minor) NO (Major) NO (Minor) -​ -​

Indirect outcomes (II): Future use of primaries

Regular (Extended to all positions) Regular (Not extended to RC) Regular (RC inexistent till 2020) Inexistent till 2012 Inexistent till 2014 (Not extended to PL) Inexistent till 2019 Inexistent till 2019

Source:  Authors’ own based in press coverage. *Mean value of all elections held. In grey the positive answers by dimension. Notes:  PL:  Party Leader; RPL: Regional Party Leader; NC: National Top Candidate; RC: Regional Top Candidate.

56  Oscar Barberà & Juan Rodríguez-Teruel

Table 3.3  Cont.

Party primaries in Spain  57 1988; Maor 2005). They mainly refer, in the first place, to conflicts between factions. On the other hand, horizontal dynamics also have to do with the conflicts between a newly elected top candidate/​party leader and the other incumbent holding the other position. 2. In multi-​ level systems such as Spain, vertical intra-​ party dynamics should also be taken into account. Vertical dynamics have to do with either top down or bottom up conflicts between individual elites or factions located at different levels (e.g. Fabre 2008; Swenden and Maddens 2008; Detterbeck 2013). Vertical dynamics involve conflicts between party leaders (or candidates) trying to push organizational changes upwards or downwards within the party. An additional measure of party cohesion is provided by the existence of party splits in the coming months after the party primaries. The second indirect outcome, the future use of primaries is also a qualitative variable relying on the same sources as the variables measuring party cohesion. This variable is measured though three indicators: (1) whether the party regularly held primary elections in the future, after the first one was organized; (2)  whether they were irregularly used at the end of each leadership or top candidate mandate, or postponed by a long period of time; (3)  whether the introduction of party primaries is just limited to one position or extended to other (or all) internal party mandates. It is important to state that this variable is to assess the practices of the parties, meaning the organization of actual ballots, not whether the legal framework was overall implemented. This distinction is important because, as already stated, many Spanish parties regularly start their legal procedures to hold party primaries, but these procedures are interrupted if a single would-​be candidate fulfils the candidacy requests. The would-​be candidate is then automatically nominated for the post. In order to provide a clear picture of the results, Table  3.3 is ordered following a hierarchy based on the potential response categories to the items measuring the two main indirect outcomes outlined in the theoretical framework (Figure 3.1): (1) Is the introduction of party primaries (for a given post and level) putting an end to previous cohesion problems? Or does it open up new cohesion problems for the party?; (2) Does the party hold regularly party primaries afterwards? The results can be grouped in three main blocks. The first group includes processes organized by some regional parties, IU and the PP. Half of them organized primaries for the first time to select candidates (EUiA, CHA, ERC, PAR), but others such as ICV (2000), IU (2016), and the PP (2010,2018) used them to select party leaders at the national or regional level. They generally had a clear winner getting around two-​thirds of the votes or more, the main exceptions being the PP (2018). That is due to the special two rounds contest for the party leadership selection process organized by the PP in 2018:  the party primary elections were conceived as a first round to select the two most voted candidates. The winner of the first round was a former vice-​president of the Spanish Government that achieved a slim victory over the second

58  Oscar Barberà & Juan Rodríguez-Teruel contender. In the party congress, the second and third contenders united their votes and got a clear majority (57.2%) over the former vice-​president. All the primary elections of the first block did have an impact in the organization because they were instrumental to ending previous internal conflicts. Most of the problems solved were cohesion disagreements at a horizontal level, although in some cases, such as the PP (2010), the PSC (2014), and IU (2016) also extended to the vertical dimension. Party primaries did not open up new horizontal or vertical new conflicts nor produced party splits (with IU’s being the minor exception). Most of the candidate selection processes did not led to the regular use of party primaries, being most of them experiments hardly extended to other positions. The party leadership selection processes had more influence, particularly on ICV and the PP (2018). The second group is mainly composed by left statewide parties such as the PSOE and IU, although there are also two cases from regional parties such as ERC (1991) and the BNG (2012). A great deal of these processes involved the selection of party leaders, although there are important exceptions such as the PSOE (1998), or IU (1998,2007). Interestingly, in most of them the winners got quite tight results, usually obtaining less than two-​thirds of the votes, although once again there are deviant cases such as the one of ERC in 1991 (92%) or the one of the PSOE in Galicia in 2013 (77%).7 In all of them the party primaries not only did not solve the previous cohesion problems (again, mostly involving horizontal conflicts), but sparked new conflicts or even major party splits, such as in ERC (1991) and the BNG (2012). As expected, parties did not extend the use of party primaries to other positions or introduced regularly in a long time. However, there are two important exceptions, the one of ERC (1991) and the BNG (2012) mostly related to the selection of party leaders. The last group might provide examples of the futility thesis mentioned in the theoretical framework. It involves some regional parties such as the BNG (2002, 2015), the PSC (1999) and the CHA. In the early cases, the PSC (1999) and the BNG (2002) the winners achieved landslide victories but were not able to finish previous cohesion problems. On the other hand, none of them opened up new conflicts or party splits. Therefore, these party primaries were most likely conceived as a one-​off experiment and did not seem to have any relevant impact on the party in the midterm. Both the BNG and the CHA conceived their primaries to select their candidates for the 2015 general elections as part of a strategy to form broader electoral coalitions and renew their image when competing against new leftist parties such as Podemos. In all cases (with the exception of the BNG 2002) party primaries were conceived more as a marketing strategy than a real contest.

3.5  Discussion Most of the academic literature analysing the internal impacts derived from the introduction of party primaries in Spain during the late 1990s concurred

Party primaries in Spain  59 with the party decline thesis. The destabilizing effects of the PSOE (1998) party primary on the party elites’ strength and cohesion casted a long shadow over these processes and, as expected, the future use of party primaries was postponed for quite some time (Boix1998; Salazar 2000; Hopkin 2001). The PSOE’s problems were also shared with IU, the other left statewide party, during the early 2000s (Ramiro and Verge 2013). As a result, primaries in both statewide parties were mainly taken as a last resort only to be held when two would-​be candidates achieved high candidacy requirements. This successfully limited the number of competitors able to compete and, hence, the number of contests (Méndez et al. 2004: 196ff; Verge 2007: 321ff; Rodríguez-​Teruel et al. 2010; Barberà et al. 2015). By the mid-​2010s the use of primaries started to become more widespread. Several new parties such as Podemos adopted them, which lead to a new wave of processes halfway between experiments and marketing strategies (Barberà 2018; Barberà and Rodríguez-​ Teruel 2019). The mainstream left statewide parties PSOE and IU organized party primaries to select their party leaderships at both the national and regional level, which had more relevance than previous processes. Table  3.4 presents a qualitative assessment of the relationship between the winner’s strength and party cohesion during the introduction of party primaries in Spanish parties. In line with the previous observation, most of the evidences from the two main left statewide parties (PSOE, IU) confirms the abovementioned decline thesis and points out their association with the results of the process: tight results proved unable to end old party conflicts and eventually opened up new conflicts or party splits. This is also true for the BNG (2012) assembly. However, in several examples, and that is probably why party primaries have been long considered as playing with fire, a clear victory wasn’t enough to strengthen internal party cohesion. That is the case of IU’s (1998) and the PSOE’s (2013) regional contests as well as the ones from two regional parties (PSC 2014; ERC 1991). The later are quite intriguing cases that might be explained by the resilience of the previous divides, the internal regulations dealing with the discipline or other internal affairs. On the other hand, Table 3.4 also presents strong evidences of the cartel party thesis where winners with high share of votes (over 2/​3 of the electorates) do are related to high levels of party cohesion. Interestingly, most of the examples come from regional parties (ICV, CHA, PAR) or have to do with the regional dimension of the PP (2010) and IU (2016). Only in two cases (EUiA 2002; ERC 2011) party cohesion was also achieved despite low winners’ results. The most interesting case here is the one from the PP (2018), that has been summarized in the previous section. The results from Table 3.4 can also be read otherwise: at the national level, first-​time party primaries produced very competitive contests that go hand in hand with weak party leaders, then resulting in lack of cohesion and the party decline arguments. At the regional level the introduction of party primaries seems to be less antagonistic and generally producing more strengthen party leaders that, eventually, leads to more internal cohesion. These findings

60  Oscar Barberà & Juan Rodríguez-Teruel Table 3.4 Winner’s strength and party cohesion in Spain: a qualitative assessment Winner’s strength

Party cohesion

Case

End to old conflicts

New conflicts

Confirming

Confirming Partially

High (> 66%)

YES

NO

EUiA 2002 ERC 2011

Low ( 66%)

2

Low NO (