Personnel Turnover and the Legitimacy of the EU (Palgrave Studies in European Union Politics) 3030600513, 9783030600518

Citation preview

PALGRAVE STUDIES IN EUROPEAN UNION POLITICS SERIES EDITORS: MICHELLE EGAN · NEILL NUGENT · WILLIAM E. PATERSON

Personnel Turnover and the Legitimacy of the EU

Edited by John A. Scherpereel

Palgrave Studies in European Union Politics

Series Editors Michelle Egan, American University, Washington, DC, USA Neill Nugent, Manchester Metropolitan University, Manchester, UK William E. Paterson, Aston University, Birmingham, UK

Following on the sustained success of the acclaimed European Union Series, which essentially publishes research-based textbooks, Palgrave Studies in European Union Politics publishes cutting edge research-driven monographs. The remit of the series is broadly defined, both in terms of subject and academic discipline. All topics of significance concerning the nature and operation of the European Union potentially fall within the scope of the series. The series is multidisciplinary to reflect the growing importance of the EU as a political, economic and social phenomenon. To submit a proposal, please contact Senior Editor Ambra Finotello [email protected]. This series is indexed by Scopus. Editorial Board: Laurie Buonanno (SUNY Buffalo State, USA) Kenneth Dyson (Cardiff University, UK) Brigid Laffan (European University Institute, Italy) Claudio Radaelli (University College London, UK) Mark Rhinard (Stockholm University, Sweden) Ariadna Ripoll Servent (University of Bamberg, Germany) Frank Schimmelfennig (ETH Zurich, Switzerland) Claudia Sternberg (University College London, UK) Nathalie Tocci (Istituto Affari Internazionali, Italy)

More information about this series at http://www.palgrave.com/gp/series/14629

John A. Scherpereel Editor

Personnel Turnover and the Legitimacy of the EU

Editor John A. Scherpereel Department of Political Science James Madison University Harrisonburg, VA, USA

ISSN 2662-5873 ISSN 2662-5881 (electronic) Palgrave Studies in European Union Politics ISBN 978-3-030-60051-8 ISBN 978-3-030-60052-5 (eBook) https://doi.org/10.1007/978-3-030-60052-5 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Acknowledgments

This volume’s contributors have been discussing turnover since the early 2010s, but the idea of producing a collection of turnover-focused analyses gained momentum in spring 2018. A number of eventual contributors discussed the prospect of an edited volume at the June 2018 meeting of the European Consortium for Political Research’s Standing Group on the EU in Paris. The project’s eventual framework became much tighter in the course of a workshop on “The Effects of Personnel Turnover in the EU” at James Madison University in July 2018. Special thanks to JMU’s College of Arts and Letters and Department of Political Science— and to Chris Arndt, Robert Aguirre, and Jon Keller—for supporting that workshop. In addition to building the project team, the JMU workshop helped contributors to define key concepts, discuss analytical approaches, talk through unexpected relationships among phenomena, and think more clearly about the ways that turnover relates to legitimacy. At subsequent conferences, contributors gained external feedback, brought each other up to speed, and talked through various concerns. Venues for such activities included meetings of the Società Italiana di Scienza Politica (Turin, September 2018), the European Union Studies Association (Denver, May 2019), the Midwest Political Science Association (April 2019), the Association for the Study of Nationalities (New York, May 2019), and the European Conference on Politics and Gender (Amsterdam, July 2019). While many colleagues provided helpful feedback at those conferences, I v

vi

ACKNOWLEDGMENTS

am particularly grateful to Edoardo Bressanelli, who served as an excellent discussant in Denver and was also willing—along with collaborators Christel Koop and Christine Reh—to share data. The COVID-19 pandemic has occurred during the project’s home stretch. The fact that the project has remained on schedule is a tribute to the dedication of the contributors—and to the various supports that they enjoy. When it comes to strong supports, I would like to thank Palgrave’s staff—particularly Ambra Finotello and Anne-Kathrin Birchley-Brun— who have checked in with just the right frequency and responded to questions with promptness and good cheer. In addition, the Senator Harry S. Byrd, Jr. Distinguished Professorship Program at JMU has supported the work on which the project is based. Finally, and most importantly, I would like to thank Melinda Adams, Catherine Scherpereel, and Maeve Scherpereel for their support and inspiration. Melinda has read countless drafts and provided many ideas. She has been particularly patient and flexible in the home stretch, which has played out, due to the pandemic, almost entirely on the home front.

Contents

Part I Introduction 1

Personnel Turnover and the Legitimacy of the European Union John A. Scherpereel

3

Part II Turnover and Input Legitimacy 2

3

Assessing the Impact of Membership Turnover on Constituent Views of the European Parliament William T. Daniel and Shawna K. Metzger Turnover, Turnout, and Input Legitimacy in the EU Nicholas Clark and John A. Scherpereel

Part III

35 59

Turnover and Throughput Legitimacy

4

Personnel Turnover and Legislative Efficiency in the EU Lauren K. Perez and John A. Scherpereel

5

Losing Women, Losing Power? Gender, Turnover, and EU Legislation Andrea S. Aldrich and Lauren K. Perez

85

107

vii

viii

6

CONTENTS

The Importance of Expertise: Political Careers, Personnel Turnover, and Throughput Legitimacy in the European Parliament Eugenio Salvati and Michelangelo Vercesi

Part IV 7

8

Turnover and Output Legitimacy

Making a Hasty Brexit? Turnover and Brexit Negotiations Jessica Adolino

169

Turnover, Conditionality, and Europeanization in the Western Balkans John Hulsey

193

Part V 9

137

Conclusion

Turnover, Legitimacy, and the EU Timescape John A. Scherpereel

Index

213

229

Notes on Contributors

Jessica Adolino is Professor of Political Science at James Madison University (USA). Her primary research interests lie in the areas of British politics, comparative public administration, and the participation of ethnic minorities in political life. She is the author, among others, of Ethnic Minorities, Electoral Politics and Political Integration in Britain. Andrea S. Aldrich is Lecturer in Political Science at Yale University (USA). Her research interests include political representation, gender, and comparative political institutions. Her work, which focuses on the relationship between internal party dynamics and legislative representation, has been published in a range of outlets including JCMS: Journal of Common Market Studies, Party Politics, and Politics & Gender. Nicholas Clark is Associate Professor of Political Science, Director of Public Policy, and Head of the Department of Political Science at Susquehanna University (USA). His areas of research interest include public opinion, public knowledge of the European Union, elections, and voting behavior. His recent work has appeared, among others, in European Union Politics, JCMS: Journal of Common Market Studies, and Journal of European Integration. William T. Daniel is Assistant Professor in Comparative Politics, School of Politics and International Relations at the University of Nottingham (UK). His areas of interest include political careers, political communication, institutional politics, and political parties. The author of Career ix

x

NOTES ON CONTRIBUTORS

Behaviour and the European Parliament: All Roads Lead Through Brussels?, his recent work has appeared in JCMS: Journal of Common Market Studies, European Union Politics, Journal of European Public Policy, Journal of European Integration, Party Politics, and elsewhere. John Hulsey is Associate Professor of Political Science at James Madison University (USA). His research focuses on ethnic politics, territorial politics, elections, political parties, and state-building. Recently, his work has been published in outlets including Communist and Post-Communist Studies, Democratization, International Peacekeeping, Nationalism and Ethnic Politics, and Regional & Federal Studies. Shawna K. Metzger is Assistant Professor, General Faculty, Department of Politics at the University of Virginia (USA). Her primary areas of interest include international security, interstate conflict, and political methodology. Her recent work has appeared in International Interactions, Journal of Conflict Resolution, Journal of Legislative Studies, Political Analysis, PS: Political Science & Politics, and elsewhere. Lauren K. Perez is Assistant Professor of Political Science at Francis Marion University (USA). Currently working on a book project that analyzes national parliamentary involvement in EU policy-making, her research focuses on comparative political institutions, executive-legislative relations, and political parties. Her recent work has been published in JCMS: Journal of Common Market Studies and Journal of European Public Policy. Eugenio Salvati is Postdoctoral Researcher at the Department of Political and Social Sciences of the University of Pavia (Italy). His primary areas of interest include EU politics, euroskepticism, local welfare, and social policy. His recent work has been published in venues including European Politics and Society, The International Spectator, Italian Political Science, Journal of Comparative Politics, and Modern Italy. John A. Scherpereel is Professor of Political Science at James Madison University (USA). The author of Governing the Czech Republic and Slovakia: Between State Socialism and the European Union, his research focuses on executive politics, territorial politics, and political representation. His work has been published in European Journal of International Relations, European Politics and Society, JCMS: Journal of Common

NOTES ON CONTRIBUTORS

xi

Market Studies, Journal of European Integration, Journal of European Public Policy, West European Politics, and elsewhere. Michelangelo Vercesi is Research Associate at the Center for the Study of Democracy/Institute of Political Science of the Leuphana University of Lüneburg (Germany). His research interests lie in the areas of parliaments, political elites, political parties, and women and politics. His recent work has been published in outlets including European Politics and Society, Journal of Modern Italian Studies, Parliamentary Affairs, Politics & Policy, Regional & Federal Studies, and Representation.

List of Figures

Fig. Fig. Fig. Fig.

1.1 1.2 1.3 2.1

Fig. Fig. Fig. Fig. Fig.

2.2 2.3 2.4 4.1 4.2

Fig. 5.1 Fig. 5.2 Fig. 5.3 Fig. 5.4 Fig. Fig. Fig. Fig.

5.5 5.6 6.1 6.2

Fig. 6.3

Legitimacy in the European Union Personnel Turnover’s Possible Effects on Legitimacy Descriptive Representation of Women in EU Institutions Predicted Probabilities: Respondent-Country MEP Dropout First Differences: Respondent-Country MEP Dropout Predicted Probabilities: Other Countries’ MEP Dropout First Differences: Other Countries’ MEP Dropout Quantity of Turnover in EU Institutions Over Time Cumulative Hazard Functions for Cox Regression of Legislative Duration Total Number of Ministers in the Council of the EU Number of Men and Women Ministers in the Council of the EU, 2009–2018 Ratio of Female Turnover to Male Turnover in the Council Ratio of Female Turnover to Male Turnover in the Council by Configuration Predicted Probabilities for Approval at First Reading Predicted Probabilities for Signature Delegation and Accountability in the EP MEPs by Turnover, Gender, Experience, and Role, 2009–2019 The Relationship between Experience in the EP and the Number of Reports

7 9 23 49 50 51 52 96 98 116 117 122 123 126 126 145 150 152

xiii

xiv

LIST OF FIGURES

Fig. 6.4 Fig. 7.1 Fig. 8.1

Distribution of Rapporteurs by EP Experience and Performance in Office Timeline of Resignations from May Government, June 2018 to April 2019 Worldwide Governance Indicators for Western Balkan Countries, 1995–2018

154 180 196

List of Tables

Table 1.1 Table 2.1 Table 2.2 Table Table Table Table

3.1 3.2 3.3 4.1

Table 4.2 Table 5.1 Table 5.2 Table 5.3 Table 6.1 Table 6.2

Personnel Turnover in EU Institutions Determinants of “Should the EP Play a More Important Role?” By EU Era, Determinants of “Should the EP Play a More Important Role?” MEP Turnover and Voter Turnout: Descriptive Statistics Predictors of EP Election Turnout Gender Progress and EP Election Turnout Over Time Personnel Turnover and Legislative Efficiency, Descriptive Statistics Cox Regression of Legislative Duration Annual Percentage Rates of Turnover by Gender and Council Configuration Turnover Ratio and Legislative Outcomes Number of Ministers Turned Over and Legislative Outcomes Distribution of Committee Chairs and Rapporteurs by Experience in the EP Type of Experience at the Sub-European Level and Performance in Office

22 46 48 70 72 76 95 97 121 125 127 151 155

xv

PART I

Introduction

CHAPTER 1

Personnel Turnover and the Legitimacy of the European Union John A. Scherpereel

The European Union (EU) is one of the world’s most complex political systems. Seven institutions operate at the center of the EU, and individuals rotate into and out of the institutions with great frequency. The EU institutions act alongside an array of EU agencies and advisory bodies. They cooperate (and occasionally conflict) with national and sub-national institutions in the EU’s 27 member-states. EU institutions carry out diverse legislative, executive, judicial, and financial control functions. The institutions are powerful: the decisions they make profoundly affect the distribution of resources and opportunities within and outside of Europe.

Electronic supplementary material The online version of this chapter (https://doi.org/10.1007/978-3-030-60052-5_1) contains supplementary material, which is available to authorized users. J. A. Scherpereel (B) Department of Political Science, MSC 7705, James Madison University, Harrisonburg, VA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. A. Scherpereel (eds.), Personnel Turnover and the Legitimacy of the EU, Palgrave Studies in European Union Politics, https://doi.org/10.1007/978-3-030-60052-5_1

3

4

J. A. SCHERPEREEL

Complexity, dynamism, and power: this combination of attributes places questions of legitimacy at the heart of political and scholarly debates about EU institutions and EU politics. Citizens know that the EU matters. But relatively few citizens identify strongly with EU institutions. They find it difficult to hold incumbents accountable. They feel poorly represented, physically and/or culturally distant from the corridors of power. European politicians, for their part, are often more interested in national and/or sub-national institutions than they are in EU institutions. All of these phenomena weaken the EU’s legitimacy. They make it harder for the EU to live up to its values and to improve the lives and livelihoods of hundreds of millions of people. This book asks how personnel turnover affects the EU’s legitimacy. In EU institutions, there are many comings and goings. New people enter and exit the system in different ways and at different moments. They bring particular experiences, outlooks, and working styles to their positions. Theoretically, these personnel dynamics might strengthen the EU’s legitimacy. They might infuse EU institutions with fresh ideas and assure that diverse viewpoints are incorporated into decisions. But the comings and goings might also weaken the EU’s legitimacy. They might work against the accumulation of expertise, the development of effective networks, and the timeliness of public responses to pressing problems. Which is it? How does personnel turnover affect the legitimacy of the EU? This chapter establishes a framework through which the latter question can be addressed. After discussing legitimacy in greater detail, the chapter presents three sets of broad propositions about the ways that personnel turnover might affect EU legitimacy. It distinguishes among three aspects of turnover and provides an empirical overview of the turnover that characterizes the institutions. The volume’s subsequent chapters use the framework established here to illuminate the ways that turnover does and does not affect the legitimacy of the EU.

Theorizing EU Legitimacy The volume builds from Max Weber’s classical account of legitimacy, which has theoretically distinctive attitudinal and behavioral dimensions. “The legitimacy of a system of domination,” Weber writes, “may be treated sociologically only as the probability that to a relevant degree the appropriate attitudes will exist, and the corresponding practical conduct

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

5

will ensue” (Weber 1978: 214). A governing system has attitudinal legitimacy when institutional incumbents and “ordinary citizens” view the system as visible and relevant, and when they regard the rules that the system produces as worthy of being obeyed. A governing system has behavioral legitimacy when the individuals in the system comport their actions to comply with the system’s rules, rhythms, procedures, and underlying values. The EU is a rational-legal system based on voluntary consent among member-states that endorse a particular set of values (e.g., the values presented in Article 2 of the Treaty on European Union: human dignity, freedom, democracy, equality, the rule of law, and respect for human rights, including the rights of minorities). In such a system, behavioral legitimacy involves citizen actions that prospectively and/or retrospectively recognize the relevance of EU law and some degree of commitment to and participation in both event-based and quotidian aspects of communal life (e.g., participating in elections, reading news on EU developments, discussing issues of community concern, signing petitions). Since the early 1990s, scholars have published a broad corpus of work on EU legitimacy (see, for example, Banchoff and Smith 2005; Beetham and Lord 2014; Kohler-Koch and Rittberger 2007; Luccarelli et al. 2011; Moravcsik 2002; Nicolaïdis and Howse 2001; Schmitt and Thomassen 1999). This literature is rich and multivocal; it generally focuses on issues of institutional design, citizen participation, and/or European identity. Its diverse participants engage, to one extent or another, with both attitudinal and behavioral dimensions of legitimacy. With regard to institutional design, scholars of EU legitimacy have scrutinized the Treaty-based balance of power among institutions and debated whether this balance conflicts with, reinforces, or perhaps somehow transforms ideals of liberal democracy. With regard to citizen participation, they have focused on citizens’ attitudes about the EU (as expressed in public opinion polls), debates about European/national/subnational political issues (and the relative prominence of the three kinds of debates), and (non-)participation of citizens in EP elections. And with regard to European identity, they have debated the extent to which legitimate democratic governance is possible in a context where system-wide (European) identities are under-developed and/or complexly layered with national, sub-national, and non-territorial identities.

6

J. A. SCHERPEREEL

An approach that prioritizes institutional dynamics —as opposed to static issues of institutional design—is less well-represented in this otherwise broad literature. How might the comings and goings mentioned above affect EU legitimacy? Might activities that transpire within and among EU institutions help to (de)legitimize the EU, regardless of the system’s institutional architecture? Theoretical purchase on these questions can be gained by investigating the ways that personnel turnover might affect three widely discussed facets—input, throughput , and output —of legitimacy. The input and output facets have a particularly venerable history (Scharpf 1970). Working in the shadow of Weber, Scharpf suggests that democratic systems can be legitimated via reference to their participatory process (input legitimacy) and/or the ways that the rules that they promulgate affect the ranks of the governed (output legitimacy). Further discussing legitimacy’s facets, Schmidt (2013: 4) suggests that democratic systems have input legitimacy when the “participatory quality of the process leading to laws and rules as ensured by. … ‘majoritarian’ institutions of electoral representation” is high. Input legitimacy is high when the system assures government by the people. To gauge the extent to which the EU (or any other political system) enjoys input legitimacy, scholars examine both attitudinal and behavioral phenomena. Attitudinally, for example, they study whether citizens know about the EU; how citizens evaluate EU institutions and actors; and how institutional incumbents see their work. Behaviorally, they examine (among others) voter turnout and voting behavior; parties’ and actors’ campaign behaviors; and mass communications on EU issues. Studies of input legitimacy often make use of data on public opinion and elections. In terms of the schematic presented in Fig. 1.1, studies of input legitimacy tend to focus on level ‘A’ in hopes of gaining insight into the quality/strength/durability of the arrow linking levels ‘A’ and ‘B’. Output legitimacy occurs, again according to Schmidt (2013: 4), when the “problem-solving quality of laws and rules” is high. Scholars of output legitimacy are concerned with the extent to which the policies that emerge from the EU’s decision and negotiation processes successfully address public concerns and/or solve public problems. They are concerned, in other words, with whether government is truly for the people. Studies of output legitimacy can prioritize attitudes and/or behaviors. Attitudinally, studies investigate the extent to which citizens appreciate the outputs of the policy process—do they approve of EU policies, for example,

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

7

B

EU insƟtuƟons ECB

European Council President MS1

A

MS2

Public

Council of the EU Presidency MS1

MS2

European Commission

European Parliament

President Commissioner1

ECJ

ECA

President

Commissioner2

CommiƩee1

Policy outputs

CommiƩee2

C

Fig. 1.1 Legitimacy in the European Union (Source Perez and Scherpereel (2017: 1159) and own elaboration)

that promote free movement of people or address inter-regional developmental gaps? Behaviorally, studies of output legitimacy have sought to understand, for example, whether national administrative structures can adequately implement laws passed at the European level so that EU outputs can be “felt” at national and sub-national levels. Studies of output legitimacy tend to focus on public policies (depicted in area ‘C’ of Fig. 1.1) and are ultimately interested in the strength and durability of the arrow linking ‘C’ and ‘A’. Despite numerous theoretical and empirical applications (Bekkers and Edwards 2007; Iusmen and Boswell 2017; Risse and Kleine 2007), throughput legitimacy remains the least-known facet of legitimacy. According again to Schmidt, throughput legitimacy involves analysis of the black box of EU decision-making. It: … focuses on the quality of the governance processes of the EU as contributing to a different kind of normative legitimacy from both the performance-oriented legitimacy of output and the participation-oriented legitimacy of input. Throughput is process-oriented, and based on the interactions—institutional and constructive—of all actors engaged in EU governance. The point here is the quality of the interaction among actors engaged in the EU decision-making process. (2013: 6–7)

8

J. A. SCHERPEREEL

Analyses of throughput legitimacy scrutinize the extent to which institutional processes reflect core EU values. Are decision-making processes truly open to substantive participation by a diverse set of actors? Are such processes transparent? Do interactions within the Brussels bubble tilt power towards non-majoritarian institutions? To what extent might such power balances violate community values and expectations? In terms of Fig. 1.1, analyses of throughput legitimacy focus on level ‘B’. Empirically, they tend to focus on behaviors more than attitudes and on institutional incumbents more than “ordinary citizens.” If studies of input legitimacy devote primary attention to ‘A’ and studies of throughput legitimacy focus mostly on ‘B’, the goal for both kinds of study is to understand quality/strength/durability of the arrow linking ‘B’ and ‘A’. Legitimacy, then, has three distinctive facets, and this volume is divided into parts that focus, respectively, on input (Chapters 2 and 3), throughput (Chapters 4–6), and output (Chapters 7–8) legitimacy. Theoretically, personnel turnover—the phenomenon to which we turn below—may affect different facets of legitimacy differently. Turnover may, for example, increase input legitimacy while simultaneously undermining throughput and/or output legitimacy.

The Promise and Perils of Turnover in EU Institutions Propositions about the ways that turnover affects legitimacy are diverse and widespread—they can be found in contemporary political narratives and in the research literature. This section presents three general sets of sets of propositions about the relationship between turnover and legitimacy. The first two—the populist critique and the feminist critique— come predominantly from contemporary politics. Both of these critiques are comprehensive: they suggest that turnover can redeem EU politics by promoting input, throughput, and output legitimacy. The third set of propositions, which comes from academic studies of turnover, is generally less comprehensive: it focuses predominantly on throughput and output legitimacy and directs attention to the ways that turnover might imperil legitimacy.

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

9

The Populist Critique of EU Politics The populist critique of EU politics began gathering momentum in the 1990s (Berezin 2009; Betz 1994; Mudde 2004; Rooduijn and Akkerman 2017; Taggart 2004). This critique is comprehensive; it suggests that the question marks in Fig. 1.2 can all be replaced by the word “enhances.” Populism, following Mudde (2004: 543), is “a thin-centered ideology [e.g., an ideology that can be combined with a wide variety of other ideologies] that considers society to be ultimately separated into two homogeneous and antagonistic groups, ‘the pure people’ versus ‘the corrupt elite,’ and which argues that politics should be an expression of the volonté générale (general will) of the people.” Questions of personnel—of who staffs public institutions and of whose interests incumbents protect—thus lie at the core of populist politics. European populists posit a massive split between incumbent power holders (“the elite”) and the true and incorruptible people; in general, this people is conceived in national rather than European terms. In questions of quantity, populists defend wholesale replacements. They echo

Fig. 1.2 Personnel Turnover’s Possible Effects on Legitimacy

10

J. A. SCHERPEREEL

the calls of American strategist Steve Bannon,1 who propelled Donald Trump to office on a message of “draining the swamp” and identified the “deconstruction of the administrative state” as a key governing priority. The turnover that populists defend, therefore, is a sweeping one: it involves the extensive replacement of corrupt elites with currently marginalized representatives of the people. European populists are also unabashedly majoritarian, perceiving institutional checks as obstacles to the volonté générale. EU populists tend to be opportunistic when it comes to institutional templates but often defend mechanisms of direct democracy. The latter is the lodestar, to take two examples, of Italy’s Five Star Movement and Czechia’s Freedom and Direct Democracy party. They tend to mistrust institutions that curb majority rule or threaten governing efficiency (e.g., courts, central banks, the press, universities). For populist critics, the path to a more legitimate system involves the rejection of the elite, the election to office of representatives of the un(der-)represented majority, and, if necessary, direct democratic end runs around non-majoritarian institutions. Often, contemporary populist claims are intertwined with ethnic exclusivity; populist politicians are suspicious of (ethnic) minority subterfuge purportedly carried out via control of public and private organizations. While the nature and style of European populist appeals vary across EU member-states, the underlying view of personnel—the juxtaposition of corrupt elite with virtuous people, the notion that redemption begins with replacement—has been a constant. In recent decades, strands of European populism have melded with Euroskepticism. Both ideologies incorporate bureaucrat bashing and mistrust of geographically distant gray eminences. Efforts to promote cooperation among various populist and Euroskeptic political parties have often failed. The European Parliament, for example, featured multiple groups containing populist parties in its eighth term (2014–2019), and cohesion both within and between these groups was limited. In the ninth European Parliament (2019– 2024), the two populist/Euroskeptic groups have been reduced to one, but the resulting group (Identity and Democracy) excludes a number of highly visible populist parties. In addition, efforts to promote common cause with populist parties that have traditionally allied with other party groups (e.g., Hungary’s Fidesz, Poland’s Law and Justice) have so far not materialized. Obstacles to cohesion notwithstanding, the run-up to the 2019 EP elections featured a particularly visible set of events featuring populist

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

11

leaders and the notion that a populist victory was necessary not to bury Europe but to redeem it. At these events, and in their national campaigns, politicians channeled the defiant tone of Nigel Farage’s well-known repudiation of Herman Van Rompuy after his selection as European Council president in 2009: I don’t want to be rude. But, really, you have the charisma of a damp rag and the appearance of a low-grade bank clerk. And the question I want to ask is: ‘Who are you?’ I’d never heard of you. Nobody in Europe had ever heard of you … I can speak on behalf of the majority of British people in saying that we don’t know you, we don’t want you, and the sooner you are put out to grass, the better. (Press Association 2010)

There were strong echoes of Farage, for example, in Italian League leader Matteo Salvini’s speech at a pre-election rally in Milan in May 2019. There, Salvini shared the stage with populist party leaders from other EU members, including France’s Marine Le Pen (who had kicked off her EP electoral campaign behind a podium declaring Le pouvoir au peuple), the Netherlands’ Geert Wilders, and Germany’s Jörg Meuthen. Salvini presented the populist vision—disdain for disconnected elites, support for replacement of elites with representatives of “the people,” and promise that the people’s victory would result in Europe’s redemption: A Europe of common sense is gathered here in Milan to free the continent from the illegal occupation orchestrated in Brussels. .. The dream of Europe’s founding fathers has been betrayed by the Merkels, the Macrons, the Soroses, and the Junckers who built a Europe based on finance and uncontrolled migration. (Gostoli 2019)

Three days before voting in Italy began, the incipient Independence and Democracy grouping posted a tweet that emphasized the opportunity to displace the entrenched elite with representatives of the people: “You [voters] can change Europe’s destiny and say ‘no’ to this technocratic EU that despises peoples.”2 The path to the redemption that populists seek—to a Europe that is more aligned with popular values and identities (input legitimacy), works better (throughput legitimacy), and produces policies that are more aligned with the general will (output legitimacy)—is personnel turnover.

12

J. A. SCHERPEREEL

The Feminist Critique of EU Politics As Salvini’s supporters flocked to the Piazza del Duomo for the May 2019 event, they passed through a city gate overseen by a statue that had been spontaneously outfitted with a pink scarf (Gostoli 2019): they encountered a feminist counter-rally. This confrontation exposed the antipathy that generally separates populists and feminists. While the two sets of critics disagree on many issues, though, they share at least one basic grievance—they are both uncomfortable with contemporary EU personnel dynamics. They also share one comprehensive hope—that new patterns of turnover will promote all three facets of legitimacy. Like populist critics, feminist critics propose that Fig. 1.2’s question marks can all be replaced by the word “enhances.” Rather than putting hopes in the replacement of “elites” with “the true people,” though, feminist critics prioritize considerations of gender equality and gender balance. Like the populist critique, the feminist critique has gained momentum since the 1990s. Issues of gender equality have figured into the European integration process since the European Economic Community’s (EEC’s) first days. Article 119 of the Treaty Establishing the EEC institutionalized the principle of equal pay for equal work for men and women. Generations of feminist critics, however, have discussed the narrowness of this framework. The literature on gender and EU politics (Abels and Mushaben 2012; Hoskyns 1996; Jacquot 2014; Kantola 2010; MacRae 2010) traces the expansion and limitations of subsequent efforts to expand gender equality in the EU. One dimension of these efforts has been the promotion of women’s political representation in EU institutions.3 Feminists who have prioritized representation generally build from Hanna Pitkin’s (1967) distinction between descriptive representation (e.g., the extent to which an institution’s incumbents resemble the political community they govern) and substantive representation (e.g., the extent to which incumbents promote the interests of a particular group) and/or Anne Phillips’s (1995) work on the politics of presence (e.g., the implications of the representation of societal groups within a governing institution). Their work identifies the gendered nature and consequences of EU institutions and advocates for institutional steps to promote gender balance and foreground gender concerns. Approaches that promote gender balance and/or parity have not been universally endorsed within feminist circles. In recent years, for example,

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

13

diverse strains of broadly feminist research have pivoted away from a focus on “critical mass” and stressed the importance of “critical actors” (Childs and Krook 2009). Other feminists (e.g., Sawer and Turner 2016) have highlighted that gendered institutional barriers continue to exist even after glass ceilings or representational thresholds have been broken. And still others (Kantola and Nousiainen 2009; Verloo 2006) have emphasized the ways that intersectionality complicates campaigns for gender balance; spikes in women’s representation, they suggest, do not necessarily result in the representation of all women’s interests. In addition, some feminist critics have problematized the ways that balance efforts risk essentializing women and/or institutionalizing an unjust gender binary (Lépinard 2007). Despite the richness and multivocality of the feminist academic space, calls for gender balance and parity have resonated widely in the political realm. An important early milestone in European collective action for equal representation in political decision-making positions was the Athens Declaration (1992). Signed by a range of contemporary and former women executives and parliamentarians, the declaration decried the “profound inequality in all public and political decision-making authorities and bodies at every level—local, regional, national, and European” and called on authorities to adopt measures to promote gender balance (Athens Declaration 1992). In the ensuing decades, “velvet triangles” (including femocrats in EU institutions; civil societal organizations; and experts from universities and think tanks) have supported the ideas of balance and/or parity in both EU and national institutions (Woodward 2004). Their activity developed in tandem with the expansion of gender quotas in a range of countries and political parties and responded to the perceived vagueness of previous commitments to women’s institutional empowerment. The movement produced a number of minor breakthroughs—the European Commission’s definition of “balance” as pertaining when women and men both control no less than 40% of an institution’s seats, for example, and the establishment of a Commission database on the representation of women and men across a range of specific institutions (national parliaments, cabinets, EU institutions, public administration, and the judiciary).4 The feminist critique of EU personnel dynamics has recently gained momentum. The activities of a range of actors—politicians, lobbyists, journalists, academics, and others—has encouraged regular public scrutiny of the quality of representation, especially in the EP and the

14

J. A. SCHERPEREEL

European Council. With regard to the latter, in a 2008 interview, commissioner Margot Wallström lamented that the EU was governed by an “inner circle of male decision-makers” who “agree behind closed doors on whom to nominate to EU top jobs.” Wallström bemoaned the fact that “old men choose old men, as always” (quoted in Kantola 2010: 50). This situation has begun to change, however, pushed forward by a number of overlapping developments, including the establishment of parity cabinets in certain member-states, the activities of long-established pan-European lobbies (e.g., the European Women’s Lobby) and newer organizations (e.g., the Gender Five Plus think tank), and journalistic efforts like Politico.eu’s The XX Factor. The 2019 iteration of the “top jobs process” suggested that their efforts had borne some fruit. In the aftermath of the 2019 EP elections, heads of state and government bargained over whom to nominate/appoint to the Commission presidency, the European Council presidency, the EU foreign policy chief position, and the European Central Bank presidency. The 2019 top jobs process was still largely played out behind closed doors. It still involved disproportionate shares of “old [white] men,” who collectively ignored a Spitzenkandidat process intended to shrink the distance between citizens and officeholders. All the same, the process resulted in a significant gendered breakthrough: the male-dominated European Council split the top jobs evenly between women and men. What is more, once the European Council nominated Ursula von der Leyen to the Commission presidency, she plead for support by MEPs by trumpeting her commitment to gender balance: I will ensure full gender equality in my College of Commissioners. If Member States do not propose enough female Commissioners, I will not hesitate to ask for new names. Since 1958 there have been 183 Commissioners. Only 35 were women. That is less than 20%. We represent half of our population. We want our fair share. (European Commission 2019)

Feminist critics, then, are not solely focused on bringing new incumbents into institutions; they are primarily interested in diversifying the backgrounds of incumbents and promoting gender equality. They defend the notion of boosting representation for a range of reasons—because increasing women’s representation promotes symbolic representation (e.g., by providing role models to prospective women office holders and changing broad cultural assumptions about “what a leader looks like”),

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

15

symbolizes justice (a key democratic value), promotes substantive representation, assures that people with a diverse range of experiences are heard, increases public trust in political institutions, and/or reinforces systemic legitimacy.5 In addition, a number of studies suggest that women bring distinctive experiences and styles to institutions and increase the quality of rule-making processes (Bratton and Haynie 1999; Volden et al. 2013). How, specifically, might the redemption through gender balance that feminist critics have defended be achieved? Theoretically, there are a number of potential paths—(a) the retention of women incumbents; (b) maintenance of the number of available seats, with disproportionate numbers of men exiting and disproportionate numbers of women entering; and/or (c) an expansion in the number of available positions, with a large share of new positions going to women. In general, the EU’s feminist critics have focused on steps that could be taken, within the current institutional framework, to increase both the supply of and demand for women incumbents. In a 2019 policy paper, for example, Gender Five Plus advocated for five reforms designed to promote gender balance: gender quotas, multi-member electoral systems with high district magnitudes, gender progressive party policies (e.g., party quotas), steps to promote women aspirants, and fair media representation (Irigoien 2019). Overall, the feminist critique suggests that turnover will help to legitimize the EU system. It will improve systemic inputs, help institutions to function more effectively, and produce stronger and more inclusive policies. Academic Propositions on Turnover and Legitimacy in the EU Despite their differences, both populist and feminist critics suggest that turnover can enhance all three facets of legitimacy. Academic studies of turnover have generally been less comprehensive. They have focused, to date, more on throughput and output legitimacy than on input legitimacy. They have generally focused on institutional incumbents and have rarely analyzed mass attitudes and behaviors. In general, their work suggests that while turnover can, indeed, help to legitimize systems, turnover also carries significant risks. Academic analysts have discussed turnover as a “Goldilocks dilemma”—quantities of turnover that are too low and too high, they suggest, can both undermine legitimacy. The fictional Goldilocks can tell which bowl of porridge is “just right.”

16

J. A. SCHERPEREEL

Scholars of turnover, though, have not identified any particular quantity, mechanism, or quality of turnover as “just right.” Rather, they have suggested that turnover that promotes legitimacy in one context may undermine it in another. The question marks above the arrows running from turnover to throughput and output legitimacy in Fig. 1.2, they suggest, can sometimes be replaced with “enhances” and other times be replaced with “imperils.” In some ways, academic studies of turnover have echoed populist and feminist critiques: if a democratic system does not experience enough turnover, they suggest, incumbents will become isolated from public preferences. Citizens will lose control over system operations (Katz 1997) and may lose faith in the system and its outputs. In addition, low levels of turnover tend to reinforce the marginalization of women, ethnic minorities, and other groups that have historically faced discrimination (Norris and Lovenduski 1995: 197). Personnel stagnation deepens historical injustices. For these reasons, some level of turnover is necessary to avoid the “three As”: arrogance, apathy, and atrophy (Atkinson and Docherty 1992; Matland and Studlar 2004). Political scientists have also argued, though, that institutions beset by excessively high levels of turnover imperil throughput and output legitimacy. Scherpereel and Perez (2015) discuss three interrelated perils. The peril of amateurism arises from the complex nature of modern governance and threatens throughput legitimacy. If incumbents lack the technical and political skills necessary to carry out demanding tasks, the risk of their domination by (undemocratic and/or unresponsive) actors elsewhere in a political system increases. This peril materialized, according to Kernell (1977), in the antebellum United States, where members of the House of Representatives from northern states (whose delegations turned over with great frequency) were dominated by House members from southern states (whose delegations were more stable). Perez and Scherpereel (2017) have also argued that this peril manifests in today’s EU, where turnover among ministers empowers unelected bureaucrats in the Council system. The peril of disunity also threatens throughput legitimacy by undermining solidarity and mutual understanding among incumbents. New incumbents may have different priorities and/or operating styles than old incumbents. The misunderstandings that emerge between freshmen and upperclassmen may undercut institutional performance and productivity. Finally, the peril of poor-quality policy threatens output legitimacy: even if new incumbents enter institutions with grand

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

17

plans and new ideas, their inexperience increases their risk of making errors, of passing measures that agents cannot execute, and/or of empowering agents in ways that violate democratic norms. The three perils of high turnover threaten systemic legitimacy. In sum, academic studies of turnover are attentive to the Goldilocks dilemma and caution that turnover can have significant negative effects, particularly on legitimacy’s throughput and output facets.

Conceptualizing Personnel Turnover in the European Union Before considering turnover patterns in EU institutions, it is important to provide a tighter conceptualization of the volume’s key explanandum. Turnover occurs when new individuals come to inhabit a political institution. Institutions, following Hall (1986: 98), include “the formal rules, compliance procedures, and standard operating practices that structure the relationship between individuals in various units of the polity and economy.” The volume’s authors note that the entry of new personnel can produce changes in informal norms and ways of doing things. In general, though, we begin our analyses by observing the turnover that takes place within the EU’s formal organizations. Legally, the EU distinguishes among “institutions,” “agencies,” and “advisory bodies.” Article 13 of the Treaty on European Union establishes the European Parliament (EP) , the European Council, the Council (of the European Union), the European Commission, the Court of Justice of the European Union, the European Central Bank, and the Court of Auditors as EU institutions. These seven organizations are generally larger, more powerful, and/or more internally complex than both agencies (e.g., the European Environment Agency) and advisory bodies (e.g., the Committee of the Regions). Internally, each EU institution has its own architecture. The EP, for example, contains various committees (e.g., Internal Market & Consumer Protection), political groups (e.g., European People’s Party), and political bodies (e.g., Conference of Presidents), while the Council of the EU meets in different configurations (e.g., Competitiveness) and relies heavily upon a committee of permanent representatives and various working groups. A study of turnover in the EU could examine personnel dynamics within any of the seven treaty-established institutions or their respective component parts. Below, this chapter presents summary descriptive statistics on

18

J. A. SCHERPEREEL

all seven institutions, and the book’s empirical chapters focus predominantly on the three institutions that matter most for day-to-day EU operations—the EP, the Council, and the European Commission. Turnover is related to the concept of elite circulation, an idea that plays a central role in the political sociology of Vilfredo Pareto (Pareto 1935, 1976, 1980). Pareto (1848–1923) devoted his career to studying social (in)stability and saw the circulation of elites—the replacement of one elite group with another—as the key indicator of regime change. In the current volume, we prefer “personnel turnover” to “elite circulation” for two reasons—one theoretical, the other based on contemporary empirical realities. First, there is an ambiguity in Pareto’s work about the “elite” label itself: should the label be attached to individuals or to positions ? Pareto sometimes stressed the inherent inequality of human beings and the uneven distribution of intellectual, leadership, and other capacities. He suggested that elites naturally emerge within particular social classes. At other times, Pareto implied that history involves the successive replacement of one governing class with another. This replacement implies a shift in the control of elite positions. The notion of personnel turnover employed in this volume attempts to avoid the individual vs. position ambiguity by keeping the focus squarely on positions. The second reason we prefer “personnel turnover” to “elite circulation” derives from trends in contemporary EU politics. The populist critics discussed above would certainly not self-identify as elites. In addition, there is room to question whether (all) positions in EU institutions deserve the “elite” label. Many EU citizens are unaware of the specific positions that comprise the EU’s governing system. Levels of voter turnout in European Parliament elections progressively declined between 1979 and 2014 before ticking upward in 2019. Incumbents operate in a union whose raison d’être has been brought into question in recent years. Given these contextual realities, are these (all) truly elite positions? Might that label be more appropriate for positions that are more universally vested with status? The personnel turnover concept attempts to avoid these issues by avoiding the language of “elites.” A comprehensive consideration of turnover’s possible effects on the three facets of legitimacy must likewise consider three aspects of turnover—quantity, mechanism, and quality. Quantity is a relatively straightforward phenomenon: to determine how much turnover takes place in an institution, one must determine whether new individuals come to control the institution’s seats. Consider the case of the European

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

19

Commission. On December 1, 2019, Ursula von der Leyen’s 27-member college of commissioners took office. Von der Leyen’s college replaced the college of Jean-Claude Juncker. To determine the quantity of turnover that occurred in this transition, we would count the number of members that carried on between the two commissions (8), subtract that number from the total number of members in the newly seated college (27−8 = 19), and divide the resulting number by the total number of members in the new college (19/27 = 0.7037). The quantity of turnover in this transition (70.37%) was quite high. In addition to questions of quantity, studies of turnover should also attend to the mechanism through which turnover takes place. By mechanism, we mean the way that incumbents gain and maintain their seats. Analysts of turnover have traditionally focused on turnover in legislatures, which generally (but not always) takes place via election. Scholars have occasionally differentiated between across-session turnover (sometimes called “between-session turnover”), which occurs across a period of time punctuated by a general election, and within-session turnover (sometimes called “intra-session turnover”), which occurs when sitting legislators vacate their seats during a session and are replaced by new incumbents. When we attend to the full range of EU legislative, executive, judicial, and financial oversight institutions, we encounter at least two additional mechanisms of turnover. The first is turnover by appointment. The ministers who participate in Council negotiations, for example, are generally appointed to their national ministerial positions by their respective heads of state and government: they gain their posts in a European legislative institution by virtue of their national executive position. A prime minister may replace the minister of finance who represented her country at the February 5 meeting of the economics and finance configuration with a new minister who represents the same member-state at the March 5 meeting of the same configuration. Within the European Parliament (whose members are generally directly elected), appointment is an additional mechanism of turnover: within-term appointments occasionally occur. In addition, group leaders appoint sitting colleagues to particular EP committees. European commissioners, some exiting members of the European Parliament (MEPs), and members of the Court of Justice, Court of Auditors, and European Central Bank are also replaced by appointment. In general, appointment introduces at least one intermediary between EU citizens and the individuals who exercise power.

20

J. A. SCHERPEREEL

The second non-electoral mechanism of turnover is rotation. To understand this mechanism, again consider the Council. Since its early days, the Council has depended on a six-month rotating presidency. Council presidents wield a certain amount of power (Brandsma and Hoppe 2020; Farrell and Héritier 2004; Häge and Naurin 2013). Historically, turnover by rotation has also been present in other domains of EU politics—in center-right/center-left gentleman’s agreements that have determined the EP’s presidency and the sequence of Commission presidents, for example, or in the small state/big state rotation that traditionally governed the appointment of the Commission president. In addition to questions of quantity and mechanism, studies of turnover should also consider the aspect of quality. By quality, we mean the sociological characteristics and life experiences that individuals bring to institutions. How diverse are incumbents? Do they represent (dis)similar gender, class, ethnic, religious, professional, marital, sexual identity, and/or other backgrounds? Many of the EU’s critics are exasperated by a sense that no matter how much turnover an institution experiences, and no matter how incumbents legally ascend to their positions, individuals with similar identities, experiences, and ways of seeing the world seem perpetually to occupy positions of power.

Turnover in EU Institutions A summary of the quantity of turnover in EU institutions helps to contextualize the propositions discussed above and provides a sense of the empirical trends that the volume’s contributors examine in subsequent chapters. Before presenting institution-specific summary data, it is instructive to recall the distinction between majoritarian institutions (MIs) and non-majoritarian institutions (NMIs) in the EU (Majone 1994; Pollack 1997; Thatcher and Stone-Sweet 2002). MIs’ incumbents claim positions as a result of elections. While directly elected legislatures and presidencies are the clearest examples of MIs, national governments—which may (e.g., the United Kingdom) or may not (e.g., Italy) be staffed by politicians with direct electoral mandates—are also typically considered majoritarian. NMIs, following Thatcher and Stone Sweet’s (2002: 2) definition, are “governmental entities that (a) possess and exercise some grant of specialized public authority, separate from that of other institutions, but (b) are neither directly elected by the people, nor directly managed by elected officials.”

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

21

NMIs are particularly common targets of populist critics. Still, democratic politicians establish NMIs for reasons that have been discussed at length in the literature (Thatcher and Stone Sweet 2002: 4–9). NMIs can resolve commitment problems, overcome information asymmetries, enhance rule-making efficiency, and promote blame avoidance. Politicians can also establish NMIs because doing so satisfies external expectations or fits with prevailing notions of appropriate behavior. While different NMIs are vested with different amounts of discretion, and while the eventual operations of NMIs may diverge from the visions of the politicians that establish them, all NMIs are designed to be at least partially isolated from political control. Institutionally, there are a number of common ways of achieving and symbolically representing such isolation. The terms of NMI incumbents, for example, may be longer than and/or deliberately asynchronous with those of MI incumbents, and NMIs may be geographically distant from MIs. Three of the EU’s four NMIs—the Court of Justice (Luxembourg), the European Court of Auditors (Luxembourg), and the European Central Bank (Frankfurt)—are headquartered outside of Brussels, and the Commission (the only EU NMI located in Brussels) is a frequent target of criticism. Table 1.1 and Fig. 1.3 present summary statistics on turnover in EU institutions over time. Table 1.1 shows that, on average, the quantity of turnover in the EU’s MIs is higher than the quantity of turnover in NMIs. The EP’s annual average turnover rate is 12.86%. In the EP, turnover spikes in the wake of elections (which take place once every five years), and the observed annual minimum of 1.47% turnover is quite low. Still, the institution can experience significant levels of within-session turnover (Daniel and Metzger 2018). While the mechanism for replacing withinterm departures varies by member-state, recourse is generally made to candidate lists from the previous general election or to pre-set lists of substitutes. On average, it takes the EP 17.85 years to reach “extinction,” which here refers to the average number of years it takes for (nearly) all incumbents to exit an institution. Of the 3695 unique individuals who served as MEPs between 1979 (the year of the EP’s first direct elections) and the end of the EP’s eighth term (July 1, 2019), 982 (26.58%) were women. As Fig. 1.3 shows, the percentage of serving MEPs who are women has grown from 16.6% after the 1979 elections to 40.4% after the 2019 elections.6

Legislative Executive Executive Executive Fin. control

Judicial

MI

MI

NMI

NMI

NMI

NMI

1952–2018

1978–2019

1999–2019

1958–2019

1975–2019

1999–2019

1979–2019

Time span

11.67

12.18

13.59

15.88

21.85

36.25

12.86

14.14

16.07

8.34

25.96

12.92

8.82

18.97

Std. Dev.

0.00

0.00

0.00

0.00

0.00

10.36

1.47

Min

Annual personnel turnover (%) Mean

52.00

53.33

33.33

88.00

54.55

61.85

59.59

Max

7.34 (109)

26.58 (3695) 27.85 (3592) 6.25 (208) 18.51 (189) 6.41 (78) 12.26 (136)

11.65

13.23

12.20

11.00

10.94

6.73

17.85

Total % women, Average years full time span to personnel (N) extinction*

Annualized turnover measures reflect observations on January 1 of each year *Personnel extinction occurs when 100% of the individuals in office at a particular moment have exited the institution. For the EP and the grand plenary of the Council, the long-term survival of even a single individual can keep the institution from reaching extinction. In the Council in the observed time span, for example, a solitary long-serving Belgian minister (Didier Reynders) kept every Council grand plenary between January 1, 2000 and January 1, 2019 from reaching extinction. Thus, for the EP and the Council, a threshold of 95% is used

Legislative

MI

European Parliament Council grand plenary European Council European Commission European Central Bank European Court of Auditors European Court of Justice

Function

MI or NMI?

Personnel Turnover in EU Institutions

Institution

Table 1.1

22 J. A. SCHERPEREEL

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

Council grand plenary

ECA

ECB

ECJ

EP

0 10 20 30 40

% women

0 10 20 30 40

Commission

23

1960

1980

2000

2020

1960

1980

2000

2020

0 10 20 30 40

European Council

1960

1980

2000

2020

year

Fig. 1.3 Descriptive Representation of Women in EU Institutions

The Council of the European Union brings together relevant ministers from the governments of the member-states: when a contentious issue related to agriculture needs to be decided, for example, the 27 ministers responsible for agriculture convene in the Council’s Agriculture and Fisheries configuration to take the decision. The European Council, meanwhile, brings together the EU’s heads of state and government, the president of the European Commission, and, since 2009, the President of the European Council, who coordinates the institution’s work. The European Council is an executive body that determines the Union’s longterm political direction. While both bodies can take decisions on certain matters by qualified majority—as part of the EU’s ordinary legislative procedure, this threshold is particularly relevant in the Council of the EU—both institutions have traditionally preferred to work by consensus (Häge 2013; Heisenberg 2005). The Council and the European Council are relatively unusual in terms of their mechanisms of entry. Those mechanisms derive from EU-specific rules and the constitutional requirements and (generally unsynchronized)

24

J. A. SCHERPEREEL

political dynamics of the member-states. Angela Merkel first attended the European Council in December 2005, for example, after her party’s national electoral victory and after assembling a coalition government that took office in November 2005. Emmanuel Macron, for his part, first attended the European Council in June 2017, after winning a direct election and formally taking office in May 2017. Charles Michel, as an example of a third pathway to the European Council, had been present at multiple European Council summits during his years as prime minister of Belgium (2014–2019). His first summit as European Council president took place in December 2019, shortly after he assumed the president’s role. Table 1.1 shows that the Council is characterized by particularly high levels of turnover. Between 1999 and 2019, the Council’s annual average turnover rate was 36.25%; it took the Council 6.73 years, on average, to reach extinction.7 The annualized average percentage of women ministers in the Council was 27.85%. This figure grew modestly from 26.39% in 1999 to 33.63% in 2019. The European Council, for its part, has had an average annual turnover rate of 21.85% since its establishment. Its average extinction rate has been 12.92 years, and, of the 208 unique individuals (including European Commission presidents and European Council presidents) who have served in the European Council since 1975, only 6.25% have been women. Between January 1975 and January 2019, women never controlled more than 13.33% of European Council seats; that maximum was reached in 2013, when four women were serving as head of government. The four remaining institutions—the European Commission, the European Central Bank (ECB), the European Court of Auditors (ECA), and the European Court of Justice (ECJ)—are NMIs that are all sheltered, to some extent, from political winds. Term structures are one component of this shelter. The mechanism of appointing commissioners involves a number of steps. Following each EP election, the European Council nominates a candidate for Commission president by qualified majority.8 Once the EP has voiced its confidence in the nominee (via at least 50% + 1 of all members), the governments of all memberstates besides the member-state from which the incoming president hails work with the president-elect to propose a commissioner. The president-elect allocates portfolios among the nominees. EP committees vet each nominee, and, at the conclusion of the EP hearings, the EP either supports or rejects the entire college. The Commission is

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

25

an executive institution that wields significant agenda-setting and policy implementation powers. As Table 1.1 shows, the average annual turnover rate among commissioners is 15.88%. It takes eleven years, on average, for the Commission to reach extinction. Between its establishment as the Commission of the EEC in 1958 and the conclusion of the Juncker Commission’s term (2019), 18.51% of commissioners were women, and women’s share of Commission seats has increased over time. Turnover in the Commission spikes in the aftermath of EP elections and levels out in periods between elections; departing commissioners are replaced through a process involving member-state nomination and EP vetting and approval. The governing council of the European Central Bank (ECB) brings together the president, five additional members of the executive board, and the governors of the national central banks of all countries that comprise the euro area. The president, vice-president, and four other executive board members are appointed by a qualified majority of member-states, who make appointment decisions after consulting the EP and the current ECB governing council. The president and five other executive board members are appointed to non-renewable eight-year terms. Members of the initial class of executive board members (1998) received terms of variable length (e.g., each year from 2001 forward, one executive board member was slated to exit the executive board), a feature intended to promote stability among subsequent executive board members. While all euro area states must commit to the idea of central bank independence, the precise mechanisms of institutionalizing the insulation of governors vary from country to country. In most countries, governors serve relatively long terms: Ennser-Jedenastik’s (2014) study of term durations across thirty European democracies from 1945 to 2012 finds a median term duration of just over six years. Since the ECB’s establishment, governing council members have turned over at an average annualized rate of 13.59%. The ECB remains one of the most gender imbalanced of all EU institutions—of the 78 governing council members between 1999 and 2019, only five (6.41%) were women. Women’s representation on the governing council hit a statistical peak of 8.33% in January 2019 and averaged only 5.19% over the course of the 1999–2019 span. The European Court of Auditors and the European Court of Justice play very different roles in EU governance. The ECA is relatively weak and understudied; the ECJ has played a critical integrative role and has

26

J. A. SCHERPEREEL

received significant academic attention. In terms of personnel dynamics, though, the two NMIs have relatively similar designs and dynamics. Both institutions currently receive one member from each memberstate. Members of the ECA are nominated by each member government and appointed by the Council, which acts by a qualified majority after consulting the EP. ECJ judges, for their part, are appointed by common accord of the member-states, following a consultation with an expert panel governed by Article 255 of the Treaty on the Functioning of the EU. Members of both institutions serve six-year renewable terms, and terms are staggered. As Table 1.1 makes clear, average annual turnover rates in the ECA and the ECJ are relatively low. And, as Fig. 1.3 shows, both institutions have traditionally been dominated by men; neither has had a woman president, and, while levels of women’s representation have edged slightly higher over time, the historical maximums for the ECA and the ECJ are 21.43% (2019) and 18.52% (2011), respectively. Of the 136 individuals who served the ECA between 1978 and 2019, thirteen (12.26%) were women. And of the 109 ECJ judges between 1952 and 2018, only eight (7.34%) were women. In terms of the broader quality of members, the ECB, ECA, and ECJ are professionally homogeneous: ECB governing council members tend to have significant experience on national high courts; ECA auditors generally enter the organization with national-level experience in public-sector institutions of financial control; and ECJ judges go to Luxembourg with significant experience on national and/or international court benches.

Perspectives on Turnover’s Effects The authors of the volume’s subsequent chapters work within the framework presented in this chapter and throw light on the ways that personnel turnover does and does not affect the EU’s legitimacy. Part II of Personnel Turnover (Chapters 2 and 3) examines the relationship between turnover and input legitimacy. Chapter 2 (Daniel and Metzger) investigates how turnover among MEPs may influence citizens’ attitudes toward the EP. Chapter 3 (Clark and Scherpereel) analyzes the extent to which turnover alters citizens’ behaviors by focusing on turnout in EP elections.

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

27

The volume’s third part (Chapters 4–6) investigates whether turnover affects throughput legitimacy. Chapter 4 (Perez and Scherpereel) analyzes the extent to which quantities of turnover in the Commission, Council, and Parliament affect the behaviors of incumbents and the efficiency of the legislative process. Chapter 5 (Aldrich and Perez) focuses on both the quantity and quality of turnover, examining whether the exit of women from the Council decreases the probability of EU legislative success. Chapter 6 (Salvati and Vercesi) also examines both quantity and quality, focusing on the ways that MEPs’ political biographies and past experiences affect who leads the EP and who succeeds as an EP leader. Chapters 7 and 8 form the fourth part of the volume. These chapters examine whether the quantity of turnover affects output legitimacy. They look, specifically, at the ways that turnover affects the policies that link the EU to external and exiting states. Such policies are particularly critical, given the increasing politicization of membership issues and citizens’ eagerness for the EU to play a strong role in global politics. Chapter 7 (Adolino) focuses on Brexit negotiations, and Chapter 8 (Hulsey) on negotiations with a potential EU candidate (Bosnia-Herzegovina). Both authors assess the ways that turnover among key personnel affects policy outcomes. While the authors reach different conclusions, the chapters collectively suggest that personnel turnover has significant effects on the legitimacy of the EU, that wholescale replacements are likely to undermine EU legitimacy, and that turnover’s gendered effects are noticeable. The volume’s concluding part (Chapter 9) develops these themes, locates personnel turnover within the broader EU timescape (Goetz and MeyerSahling 2012), and identifies potentially fruitful avenues of future work on turnover and legitimacy in the EU context.

Notes 1. In early 2019, following his departure from the White House, Bannon teamed with Belgian lawyer Mischaël Modrikamen to support “The Movement,” a club of European populist politicians. Preliminary press analysis (e.g., De La Baume and Sciorilli Borrelli 2019) suggests that while Bannon’s message and technique resonated on the other side of the Atlantic, the extent of meaningful cooperation was limited. 2. Tweet available at https://twitter.com/IDParty_/status/113151559956 8429059. Accessed May 28, 2020.

28

J. A. SCHERPEREEL

3. Kantola (2010: 3–4), for example, suggests that feminist attention to the interaction between “institutions, processes, actors and discourses” is one of three “interlocking aspects” related to the EU’s role in shaping gender relations. The other aspects are (a) EU gender policy (e.g., policies on family policy, policies and violence), gender mainstreaming, and the gendered nature of EU policies (e.g., trade policy, agricultural policy) in general; and (b) Europeanization and the extent of diversity that pertains across EU member-states, even after the effectuation of soft or hard EU laws. 4. For additional details, see Kantola (2010: 54–56) and Van der Vleuten (2012). 5. This list of rationales for gender balance draws from Kantola (2010), who in turn recalls justifications offered by Phillips (1998), Dovi (2007), and Mansbridge (1999). 6. The post-2019 figure (40.4%) is based on a total of 746 MEPs (302 women, 446 men); at the time of observation (July 2, 2019) data were missing on five seats. 7. In order to give a sense of the flux that occurs at the peak of the Council system, Table 1.1 and Fig. 1.1 report turnover, extinction, and gender figures for the Council as a “grand plenary,” e.g., as a congeries of all national cabinet members participating in the Council system at a particular moment in time. In reality, the Council never meets in grand plenary. For analyses that examine configuration-by-configuration turnover rates, see Scherpereel and Perez (2015) and Aldrich and Perez (this volume). 8. In 2014, this nomination followed the so-called Spitzenkandidat system: government leaders nominated Juncker, who represented the plurality party coming out of the election and whom the EPP had chosen, before the election, as its preferred candidate for the role of Commission president. In 2019, despite many MEPs’ continued commitment to the Spitzenkandidat process, government leaders nominated Ursula von der Leyen, who had not featured in the EPP’s 2019 electoral campaign.

References Abels, Gabriele, and Joyce Marie Mushaben (eds.). 2012. Gendering the European Union: New Approaches to Old Democratic Deficits. Houndsmills: Palgrave Macmillan. Athens Declaration. (1992). Available at https://eurogender.eige.europa.eu/ posts/athens-declaration-1992. Accessed May 28, 2020. Atkinson, Michael M., and David C. Docherty. 1992. Moving Right Along: The Roots of Amateurism in the Canadian House of Commons. Canadian Journal of Political Science 25 (2): 295–318.

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

29

Banchoff, Thomas, and Mitchell Smith (eds.). 2005. Legitimacy and the European Union: The Contested Polity. London: Routledge. Beetham, David, and Christopher Lord. 2014. Legitimacy and the European Union. London: Routledge. Bekkers, Victor, and Arthur Edwards. 2007. Legitimacy and Democracy: A Conceptual Framework for Assessing Governance Practices. In Governance and the Democratic Deficit: Assessing the Democratic Legitimacy of Governance Practices, ed. Victor Bekkers, Geske Dijkstra, Arthur Edwards, and Menno Fenger, 35–60. London: Routledge. Berezin, Mabel. 2009. Illiberal Politics in Neoliberal Times: Culture, Security and Populism in the New Europe. Cambridge: Cambridge University Press. Betz, Hans-Georg. 1994. Radical Right-wing Populism in Western Europe. New York: Springer. Bratton, Kathleen A., and Kerry L. Haynie. 1999. Agenda Setting and Legislative Success in State Legislatures: The Effects of Gender and Race. The Journal of Politics 61 (3): 658–679. Brandsma, Gijs Jan and Alexander Hoppe. 2020. He Who Controls the Process Controls the Outcome? A Reappraisal of the Relais Actor Thesis. Journal of European Integration 0 (0). https://doi.org/10.1080/07036337.2020.175 3041. Childs, Sarah, and Mona Lena Krook. 2009. Analysing Women’s Substantive Representation: From Critical Mass to Critical Actors. Government and Opposition 44 (2): 125–145. Daniel, William T., and Shawna K. Metzger. 2018. Within or Between Jobs? Determinants of Membership Volatility in the European Parliament, 1979– 2014. The Journal of Legislative Studies 24 (1): 90–108. De La Baume, Maïa and Silvia Sciorilli Borrelli. 2019. Steve Bannon’s Stuttering European Adventure. Politico, April 19. Available at https://www.politico. eu/article/steve-bannon-european-parliament-the-movement-stuttering-eur opean-adventure/. Accessed May 28, 2020. Dovi, Suzanne. 2007. Theorizing Women’s Representation in the United States. Politics & Gender 3 (3): 297–319. Ennser-Jedenastik, Laurenz. 2014. Party Politics and the Survival of Central Bank Governors. European Journal of Political Research 53 (3): 500–519. European Commission. 2019. Opening Statement in the European Parliament by Ursula von der Leyen, Candidate for President of the European Commission. Accessed May 28, 2020 at https://ec.europa.eu/commission/presscorner/ detail/en/speech_19_4230. Farrell, Henry, and Adrienne Héritier. 2004. Interorganizational Negotiation and Intraorganizational Power in Shared Decision Making: Early Agreements under Codecision and Their Impact on the European Parliament and Council. Comparative Political Studies 37 (10): 1184–1212.

30

J. A. SCHERPEREEL

Goetz, Klaus H., and Jan-Hinrik Meyer-Sahling (eds.). 2012. The EU Timescape. London: Routledge. Gostoli, Ylenia. 2019. European Far-Right Populists Rally with Matteo Salvini in Milan. Al Jazeera, May 18. Accessed May 28, 2020 at https://www.aljazeera.com/news/2019/05/european-populists-rally-mat teo-salvini-milan-190518202914838.html. Häge, Frank M. 2013. Coalition Building and Consensus in the Council of the European Union. British Journal of Political Science 43 (3): 481–504. Häge, Frank M., and Daniel Naurin. 2013. The Effect of Codecision on Council Decision-making: Informalization, Politicization and Power. Journal of European Public Policy 20 (7): 953–971. Hall, Peter A. 1986. Governing the Economy: The Politics of State Intervention in Britain and France. New York: Oxford University Press. Heisenberg, Dorothee. 2005. The Institution of ‘Consensus’ in the European Union: Formal versus Informal Decision-making in the Council. European Journal of Political Research 44 (1): 65–90. Hibbing, John R. 1999. Legislative Careers: Why and How We Should Study Them. Legislative Studies Quarterly 24 (1): 149–171. Hoskyns, Catherine. 1996. Integrating Gender: Women, Law and Politics in the European Union. New York: Verso. Irigoien, Alazne. 2019. European Parliament’s Elections 2019: Towards Parity Democracy in Europe. Brussels: Gender Five Plus. Available at https:// docs.wixstatic.com/ugd/530efa_2a575f9fa3dc45afb68e4ffb516fef30.pdf. Accessed May 28, 2020. Iusmen, Ingi, and John Boswell. 2017. The Dilemmas of Pursuing ‘Throughput Legitimacy” Through Participatory Mechanisms. West European Politics 40 (2): 459–478. Jacquot, Sophie. 2014. Transformations in EU Gender Equality: From Emergence to Dismantling. Houndsmills: Palgrave Macmillan. Kantola, Johanna. 2010. Gender and the European Union. Houndsmills: Palgrave Macmillan. Kantola, Johanna, and Kevät Nousiainen. 2009. Institutionalizing Intersectionality in Europe: Introducing the Theme. International Feminist Journal of Politics 11 (4): 459–477. Katz, Richard S. 1997. Democracy and Elections. New York: Oxford University Press. Kernell, Samuel. 1977. Toward Understanding 19th Century Congressional Careers: Ambition, Competition, and Rotation. American Journal of Political Science 21 (4): 669–693. Kohler-Koch, Beate, and Berthold Rittberger (eds.). 2007. Debating the Democratic Legitimacy of the European Union. Lanham, MD: Rowman & Littlefield.

1

PERSONNEL TURNOVER AND THE LEGITIMACY …

31

Lépinard, Eléonore. 2007. The Contentious Subject of Feminism: Defining Women in France from the Second Wave to Parity. Signs: Journal of Women in Culture and Society 32 (2): 375–403. Luccarelli, Sonia, Furio Cerutti, and Vivien A. Schmidt (eds.). 2011. Debating Political Identity and Legitimacy in the European Union. London: Routledge. MacRae, Heather. 2010. The EU as a Gender Equal Polity: Myths and Realities. JCMS: Journal of Common Market Studies 48 (1): 155–174. Majone, Giandomenico. 1994. The Rise of the Regulatory State in Europe. West European Politics 17 (3): 77–101. Mansbridge, Jane. 1999. Should Blacks Represent Blacks and Women Represent Women? A Contingent ‘Yes’. The Journal of Politics 61 (3): 628–657. Matland, Richard E., and Donley T. Studlar. 2004. Determinants of Legislative Turnover: A Cross-National Analysis. British Journal of Political Science 34 (1): 87–108. Moravcsik, Andrew. 2002. Reassessing Legitimacy in the European Union. JCMS: Journal of Common Market Studies 40 (4): 603–624. Mudde, Cas. 2004. The Populist Zeitgeist. Government and Opposition 39 (4): 541–563. Nicolaïdis, Kalypso, and Robert Howse (eds.). 2001. The Federal Vision: Legitimacy and Levels of Governance in the US and the EU . Oxford: Oxford University Press. Norris, Pippa, and Joni Lovenduski. 1995. Political Recruitment: Gender, Race and Class in the British Government. New York: Cambridge University Press. Pareto, Vilfredo. 1980. Compendium of General Sociology. Abridged with the approval of the author by Giulio Farino from Pareto’s Trattato di Sociologia Generale. Minneapolis: University of Minnesota Press. Pareto, Vilfredo. 1976. Sociological Writings. Selected and Introduced by S.E. Finer. Totowa, NJ: Rowman & Littlefield. Pareto, Vilfredo. 1935. The Mind and Society: A Treatise on General Sociology. New York: Harcourt, Brace. Perez, Lauren K., and John A. Scherpereel. 2017. Vertical Intra-institutional Effects of Ministerial Turnover in the Council of the European Union. Journal of European Public Policy 24 (8): 1154–1171. Phillips, Anne. 1995. The Politics of Presence. Oxford: Oxford University Press. Pitkin, Hannah. 1967. The Concept of Representation. Berkeley: University of California Press. Pollack, Mark. 1997. Delegation, Agency, and Agenda Setting in the European Community. International Organization 51 (1): 99–134. Press Association. 2010. UKIP’s Nigel Farage Tells Von Rompuy: You Have the Charisma of a Damp Rag. The Guardian, February 25. Available at https://www.theguardian.com/world/2010/feb/25/nigel-farage-her man-van-rompuy-damp-rag. Accessed May 28, 2020.

32

J. A. SCHERPEREEL

Risse, Thomas, and Mareike Kleine. 2007. Assessing the Legitimacy of the EU’s Treaty Revision Methods. JCMS: Journal of Common Market Studies 45 (1): 69–80. Rooduijn, Matthijs, and Tjitske Akkerman. 2017. Flank Attacks: Populism and Left-Right Radicalism in Western Europe. Party Politics 23 (3): 193–204. Sawer, Marian, and Alicia Turner. 2016. Specialised Parliamentary Bodies: Their Role and Relevance to Women’s Movement Repertoire. Parliamentary Affairs 69 (4): 763–777. Scharpf, Fritz W. 1970. Demokratietheorie zwischen Utopie und Anpassung. Konstanz: Universitätsverlag. Scherpereel, John A. and Lauren K. Perez. 2015. Turnover in the Council of the European Union: What It Is and Why It Matters. JCMS: Journal of Common Market Studies 53 (3): 658–673. Schmidt, Vivien A. 2013. Democracy and Legitimacy in the European Union: Input, Output, and ‘Throughput’. Political Studies 61 (1): 2–22. Schmitt, Hermann, and Jacques Thomassen. 1999. Political Representation and Legitimacy in the European Union. Oxford: Oxford University Press. Taggart, Paul. 2004. Populism and Representative Politics in Contemporary Europe. Journal of Political Ideologies 9 (3): 269–288. Thatcher, Mark, and Alec Stone Sweet. 2002. Theory and Practice of Delegation to Non-Majoritarian Institutions. West European Politics 25 (1): 1–22. Van der Vleuten, Anna. 2012. Gendering the Institutions and Actors of the EU. In Gendering the European Union: New Approaches to Old Democratic Deficits, eds. Gabriele Abels and Joyce Marie Mushaben. Houndsmills: Palgrave Macmillan. Verloo, Mieke. 2006. Multiple Inequalities, Intersectionality and the European Union. European Journal of Women’s Studies 13 (3): 211–228. Volden, Craig, Alan E. Wiseman, and Dana E. Wittmer. 2013. When are Women More Effective Lawmakers than Men? American Journal of Political Science 57 (2): 326–341. Weber, Max. 1978. Economy and Society: An Outline of Interpretive Sociology. Berkeley, CA: University of California Press. Woodward, Alison E. 2004. Building Velvet Triangles: Gender and Informal Governance. In Informal Governance in the European Union, ed. Thomas Christiansen and Simona Piattoni, 76–93. Cheltenham: Edward Elgar.

PART II

Turnover and Input Legitimacy

CHAPTER 2

Assessing the Impact of Membership Turnover on Constituent Views of the European Parliament William T. Daniel and Shawna K. Metzger

The European Parliament (EP) represents a crucial piece of the European Union’s (EU) institutional commitments to representative democracy and is often cited as a key remedy for the perception of the EU democratic deficit. However, both its constituents and its members (MEPs)

Electronic Supplementary Material The online version of this chapter (https://doi.org/10.1007/978-3-030-60052-5_2) contains supplementary material, which is available to authorized users. W. T. Daniel (B) School of Politics and International Relations, University of Nottingham, Nottingham, UK e-mail: [email protected] S. K. Metzger Department of Politics, University of Virginia, Charlottesville, VA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. A. Scherpereel (eds.), Personnel Turnover and the Legitimacy of the EU, Palgrave Studies in European Union Politics, https://doi.org/10.1007/978-3-030-60052-5_2

35

36

W. T. DANIEL AND S. K. METZGER

have ascribed to the EP a status of “second order” importance at various moments in its history. While direct elections to the EP were introduced in 1979 in order to give EU citizens a direct voice in the selection of the legislature and therefore bolster its input legitimacy, voter turnout has underwhelmed, and a consistent “electoral connection” that could be seen as analogous to the representative links offered by national legislatures has yet to be fostered between MEPs and members of the voting public. In addition, EP personnel have long struggled to overcome the popular adage that EP service is used either as a “kindergarten, a hospital, or a retirement home”—taking in a mix of political novices, scandal-ridden politicians, and has-beens. This is particularly true for the case of MEP turnover, both within and between EP terms, as members routinely exit for national positions that might be seen as more prominent (e.g., Daniel 2016). Although the EP has undoubtedly increased its legislative power as a co-decider on most EU legislation since the effectuation of the Maastricht Treaty, it has yet to fully shake the stigma of Reif and Schmitt’s (1980) seminal “second order” labelling. However, it is less clear from existing scholarship just how MEP treatments of the body they serve may also condition the views of their constituents and thus negatively affect the attitudinal legitimacy ascribed to it. Indeed, MEP membership volatility has been consistently higher than in national legislatures. Whereas relatively high degrees of turnover between sessions of a legislature may not be any more troubling from a normative perspective than legislatures with exceptionally low replacement rates of their members, the EP further sets itself apart from other bodies insofar as it also has comparatively high rates of early exit. Between 1979 and 2014, an average of 13.1% of all sitting MEPs failed to complete their full, elected mandate (Daniel and Metzger 2018). Moreover, patterns of early exit are also highly correlated with national background. In every session of Parliament since 1979, at least one country delegation has seen more than a third of its members leave office early (ibid.). While scholars have already examined the causes and consequences of such high turnover rates for legislative behavior within the EP, what impact might high turnover have on the citizens that such MEPs are meant to represent? Naturally, early exit from the EP may also be conditioned by negative valuations of the EP by constituents. Therefore, we examine this question by using a cross-sectional analysis of EU citizen views of the EP that

2

ASSESSING THE IMPACT OF MEMBERSHIP TURNOVER …

37

incorporates various waves of survey data from the Eurobarometer over the course of the EP’s existence. Our results indicate that high degrees of MEP early exit from across the EU have indeed lowered citizen valuations, up to a point. We narrow this relationship to the pre-Maastricht Treaty European Community (EC) era, and further show that this trend has reversed and then disappeared entirely in recent years as the EP’s legislative power has grown. This indicates that EP personnel turnover may not be a key source of citizen attitudes around the legislature’s importance. This chapter is concerned with input legitimacy’s attitudinal dimension. Like Clark and Scherpereel in this volume’s next chapter, we seek to determine the extent to which turnover among MEPs affects EU citizens. While Clark and Scherpereel (2021) focus on citizens’ (voting) behaviors, however, we focus on citizens’ attitudes—specifically, on citizens’ valuations of the EP as an institution. We examine whether or not high quantities of turnover among a country’s MEPs during a legislative session lead citizens to think more negatively about the EP and its importance. Whereas turnover is oftentimes studied in terms of ‘acrosssession’ changes in a legislature’s composition, we view these early exits as sending a particularly strong signal about members’ esteem of the institution. If members view other opportunities as more important to them than finishing their elected mandate, what are their constituents to think about the value of their position—and thus the institution that is meant to represent them? We believe this has strong relevance for the input legitimacy that the EP is meant to provide. In the following section, we discuss existing research on EP elections and public opinion, before developing a theory for the negative effect of early exit on constituent views.

EP Elections, Citizen Views, and Personnel Turnover European elections were intended to provide democratic legitimacy and popular input to the EU policymaking process, but the existing literature has long maligned the electoral connection between MEPs and EU citizens as deficient, in comparison with national legislatures. EP elections are viewed as “second order” to national ones, misunderstood by voters, characterized by populist voices as being in contest with national sovereignty, and hobbled by a proportional election system that removes personal connections between members and their constituents.

38

W. T. DANIEL AND S. K. METZGER

Reif and Schmitt’s (1980) work on the “second-order elections” model continues to dominate scholarly assessments of European elections. Their theory articulates how EU citizens view the EP as less important than national parliaments and how EP elections accordingly become tools of national oppositions—whose voters are more willing to vote with their “hearts” than with their “heads.” Others have used their work to explore both low turnout rates in EP elections, as well as the disproportionate prevalence of populist, fringe, and extremist voices among MEPs (e.g., Daniel 2016; Hix and Marsh 2011; Hobolt and Wittrock 2011; Reif 1984; Reif and Schmitt 1980). More recent scholarship has even suggested that EP elections may be self-defeating, as disinterested citizens vote in campaigns that are poorly understood and actually enhance the presence of Euroskeptic politicians within the EU’s only directly elected institution (van der Brug and de Vreese 2016). A general lack of citizen knowledge and understanding of EU institutions further plagues EP elections. For example, Hobolt and Tilley (2014) discuss how the EU’s lack of a clearly defined government leads citizens to distrust and ascribe blame to its institutions, such as the EP. Similarly, works by Franklin and Hobolt (2011) and Beaudonnet and Franklin (2016, in van der Brug and de Vreese 2016) detail how EP elections actually lead to decreased participation in both future national and European elections and decrease “diffuse support” for the EU, as they lead citizens to focus on their lack of understanding of European integration. Conversely, Schmitt et al. (2015) demonstrate how knowledge of the Spitzenkandidat process—meant to link the EP elections with the formation of the new European Commission—had a positive effect on EP election turnout in 2014. And, on the aggregate, turnout rose even further in the most recent 2019 elections. The general feeling is therefore that European voters would love the EP and participate more fully in its elections, if only they knew what it was for. In a similar vein, others have viewed European elections in relation to the citizen valuations of their national governments. Hobolt and de Vries (2016) review support for the EU as a tension between theories that pit the “winners” against the “losers” of European integration, in contrast with studies that highlight the role of European identity on fears of lost national sovereignty. As relates to support for the EP, Winzen et al. (2015) find that national parliaments led by MPs who are more in favor of European integration are less likely to view the EP as a threat or a rival and are therefore less likely to call for increased oversight of

2

ASSESSING THE IMPACT OF MEMBERSHIP TURNOVER …

39

EU affairs by national MPs. Relatedly, Muñoz et al. (2011) find that while more trusting citizens are more likely to share their trust with both national and EU governments, higher levels of trust in one’s national government across the national population correspond with lower levels of trust in the EP. In other words, not only must citizens more fully understand the purpose of the EP in order to support it, but they must also be influenced by national institutions and actors that view the EU as a useful partner to national governments, as opposed to a challenger or an exclusive rival. This is particularly important in an age of increased Euroskepticism, where populist voices oftentimes inhabit the EP itself. Finally, other studies of EP elections have pointed to foundational flaws in the institutional structure of European elections that remove the ability of MEPs to form an “electoral connection” with their constituents. Farrell and Scully (2007) find that the EP loses out on support from citizens, due in part to many member-states employing closed-list forms of proportional representation (PR) that privilege political parties. Similarly, Hix and Hagemann (2009) call for a reform of EU electoral systems to favor citizen representation via the creation of smaller electoral districts and more open lists. Their “natural experiment” approach evaluates the shift of the UK’s European elections from single-member district plurality (SMDP) elections—as in Westminster—to list-style PR voting; they conclude that the corresponding drop in turnout harms citizen connectedness to MEPs. And Obholzer and Daniel (2016) demonstrate the importance of citizen representation via their assessment of electoral district size and MEP Twitter campaigns during the 2014 European elections. In sum, a pessimistic view of the literature might suggest that Europeans are destined to misunderstand and mistrust the European Parliament, due to its particular institutional complexities, its connection to national governments, and its electoral process. Still, despite the existence of a range of factors encouraging citizens to mistrust the EP, citizen attitudes toward the institution vary significantly across countries and over time. To what extent might turnover affect such variation? The Effect of MEP Turnover on Citizen Support While the bulk of the literature on EP elections has looked at citizen views and knowledge of Parliament, as well as the institutional nature of European elections, there is reason to suspect that the individual behavior of MEPs themselves may also impact support for the legislature.

40

W. T. DANIEL AND S. K. METZGER

As mentioned in the introductory section, one striking feature of MEP behavior is their volatile careers—with many representatives eschewing their elected mandates in favor of other positions in national politics or at the EU level. Could it be reasonable, therefore, to expect that high degrees of early exit from one’s career as MEP might negatively affect the views of one’s constituents about the importance of the job? The quantity and effects of member turnover have already been the focus of much study, both comparatively and within the EU. Matland and Studlar’s (2004) work refers to turnover as a classic “Goldilocks problem”: too little can hamper representative democracy, while too much can signal internal chaos and prevent the efficiencies of institutional memory. While their work does not come down on a “just right” amount of turnover, it does frame the broad contours of its causes and points to the role of national political parties and election systems as paramount for engendering healthy levels of membership change. Within the EU, most scholars seem to view turnover across sessions of the EP as a self-selecting problem. For instance, Whitaker (2014) assesses the role that too-high levels of turnover have on the ability of MEPs to signal a commitment to the legislature and accordingly lose out on intra-institutional positions of power. Similarly, Beauvallet-Hadded et al. (2016) view MEP turnover as decreasing among members who choose to stand for a second term in office; van Geffen (2016) also views these MEP “careerists” as having self-selected into more active roles within the legislature. However, others also point to national-level factors. Daniel and Metzger (2018) assess turnover from a national-institutional perspective, identifying degrees of national federalism, electoral systems, and election cycles as each contributing to high levels of early exit from the EP. This work takes the EP out of isolation and relates it more with the national constituencies that serve as the legislature’s selectorate. Scherpereel and Perez (2015) further connect turnover between the EP and Council of the European Union, suggesting that national political shifts that lead to changes in the Council’s composition may limit its inter-institutional power vis-à-vis the EP. On the whole, the literature suggests that MEPs are often in control of their careers, to the extent that they choose to leave early or not return to Parliament, and that these decisions may have broad-reaching impacts on the ability of politicians to effectively do their jobs.

2

ASSESSING THE IMPACT OF MEMBERSHIP TURNOVER …

41

So far, we have examined how citizen views of the EP have been negatively affected by a lack of citizen understanding of and enthusiasm for European integration, competing valuations with national governing systems, and the complex institutional characteristics of European elections. However, we have also seen that MEPs themselves play a role in the squandering of institutional power via early exit and high across-session turnover, and that national political institutions may further exacerbate such membership volatility. It is worth considering, however, whether these two problems relate. Are citizens views of the EP negatively affected by MEP turnover? Does a lack of stability from among the EP’s membership hurt more than just the ability of an MEP to get important legislative work done? We turn to this question in crafting our theory for MEP early exit and citizen views of the legislature. Theory and Hypotheses We view high levels of within-session turnover (i.e., early exits) as potentially harmful for citizen valuations of the EP. In terms of citizen attitudes, knowledge of politicians “jumping ship” from the EP for domestic-level office may leave voters pessimistic about the EP’s purpose. In other words, if politicians do not take their positions seriously, then why should voters? High turnover rates may therefore be damaging to the input legitimacy that is needed for the EP to serve as a balm to representative democracy in the EU. More practically, however, we know that high turnover rates affect the ability of politicians to do their jobs—hobbling the institutional seniority and expertise needed to have a strong impact on internal legislative processes. This may also harm citizen valuations of the throughput legitimacy of the institution, as high turnover rates lead MEPs to be viewed as unproductive or ineffective in their job as policy makers. However, other processes may condition citizens’ views of the EP. As the EP has moved from an unelected talk shop to a directly elected veto player, its membership has expanded to more than 700 fulltime and professionalized politicians. MEP careers have also lengthened, and studies of a new “European class” of legislators have enhanced the perspective of MEPs as important players in both the EU policymaking process and within national political systems. Nevertheless, the EP remains an interesting laboratory of study, insofar as the singular body has been comprised of 28 (now 27) differing member-states—each with

42

W. T. DANIEL AND S. K. METZGER

its own national political culture, institutions, and parties. This suggests that a cross-sectional study of MEP turnover, over time, is necessary to answer our question. Between 1979 and 2014, some national delegations saw 100% of their MEPs complete a five-year mandate, while others saw more than half leave their positions early. It is reasonable to expect that this pronounced degree of intra-institutional heterogeneity, which we know to be heavily conditioned on a national basis, may also affect the views of MEP constituents. Might heightened levels of MEP early exit detract from citizen valuations of the EP? We believe that it will. More specifically, we expect that citizens from countries in which their MEPs are more likely to leave their mandate early will view the EP more negatively. We assume that such early exit signals to constituents a lack of importance for the job of MEP, as well as reduces the ability of MEPs to form an electoral connection with their constituents. In keeping with the extant literature on EP support, we further assume that high rates of MEP turnover will lead to lower awareness of and knowledge about the EU system. This lack of understanding of the body may lead to pronouncements of its unimportance. In other words, we posit the following: H1a: High rates of MEP early exit from one’s own national delegation will negatively affect one’s views of the EP. However, it may be that a weak electoral connection means that citizens are unaware of the specific internal dynamics of their national delegations, but do recognize the general propensity for MEPs to exit early from within the system. For instance, the well-known tactic of political parties to use high-profile national politicians to lead electoral lists that never intend to actually take up their seat and serve in the EP may lead voters to think negatively about the legislature. Similarly, ongoing scandals, whereby MEPs use their office to either enrich themselves or to funnel money to national party organizations—never doing any real work at the EP level—may also lead citizens to view the legislature as generally unimportant or even harmful to the European project. We anticipate in this way that high turnover rates will negatively affect constituent views of the EP, even when such turnover does not come from one’s own national delegation of MEPs. We therefore posit a related hypothesis: H1b: High rates of MEP early exit from other national delegations will negatively affect citizens’ views of the EP.

2

ASSESSING THE IMPACT OF MEMBERSHIP TURNOVER …

43

Alternative Explanations It may be that other variables lead EU citizens to view the Parliament negatively, across both time and space. These variables, which are mostly hinted at in the above literature review, could range from national-level traits—such as the national electoral calendar and political system—to individual-level characteristics of voters—such as general knowledge of and support for the EU. We examine these competing explanations in the following section, after introducing our data sources and modelling approach.

Research Design and Analysis In order to test our hypothesis about MEP early exit and citizen views of the EP, we begin with Daniel and Metzger’s (2018) individual-level data on all MEPs that served during the first seven sessions of Parliament (1979–2014) and combine this information with data on citizens’ views of the EP from Eurobarometer (EB) surveys. Our unit of analysis is the individual survey respondent. We use the Mannheim Cumulative Trend Dataset (Schmitt and Scholz 2009) for our Eurobarometer data, which compiles and harmonizes Eurobarometer questions across survey waves from 1970 to 2002. With the survey questions we use, our main estimation sample can only include information on 15 EU countries, spanning from 1983 to 20001 ; however, we ran separate robustness checks using similar questions from additional waves and countries. Our estimation N for this baseline dataset is 260,577.2 In later models, we divide the baseline dataset by historical era and expand our observational period through 2011, using additional data from Clark and Rohrschneider (2019). Our main dependent variable is EPIMPF, a respondent’s personal feeling about whether the EP should play a more important role. The question appeared on 28 Eurobarometer waves that enter our sample, which reflects by far the broadest coverage of any individual question regarding the EP’s value.3 The question’s wording for the bulk of our included EBs is, “Would you personally prefer the European Parliament played a more important or a less important part than it does now?” This wording remains fairly constant across all the waves, with only minor changes.4 The question’s responses are on a 3-point scale, recoded such

44

W. T. DANIEL AND S. K. METZGER

that: (1) = less important, (2) = about the same, and (3) = more important. We assume respondents who would like to see the EP increase in its importance are more likely to be supportive of the institution, generally speaking. Our primary analysis includes any individual who provided a valid question response for EPIMPF, and omits any individuals who answered “don’t know” or for whom the question was not applicable (collectively, DK/NAs).5 We use an ordered probit model for our primary analyses with robust standard errors, clustered by respondent-country. We also include country fixed effects, which capture any time-invariant factors for each country. Further, we include fixed effects for each EP session. Our main results are unweighted, but we report weighted results for reference in the online appendix (DuMouchel and Duncan 1983; Solon et al. 2015).6 Our main independent variable is the amount of within-session turnover among the respondent’s MEPs within the past year. We generate the variable using Daniel and Metzger’s (2018) within-session turnover dataset, which has information on how long each MEP stayed in the EP within and across sessions. We begin by determining the earliest date each EB survey was deployed in a country, and treat this as the earliest day the respondent could have taken the survey. We then use Daniel and Metzger’s dataset to count the number of MEPs from the respondent’s country who had dropped out of the European Parliament in the year preceding the earliest possible survey date.7 Finally, we normalize this turnover count by dividing the count by a weighted average of the number of seats allotted to the respondent’s country in the EP sessions occurring within the prior-year window. For instance, if 45% of the preceding year’s days occurred during EP6 and 55% of the preceding year’s days occurred during EP7, the denominator would be 0.45*(country’s # of seats in EP6) + 0.55*(country’s # of seats in EP7). For countries that were not EU members in one of the preceding year’s EP sessions, we reallocate the entire weight to the session in which the country was an EU member. The final variable ranges from 0 to 50% in our sample.8 Finally, we include a number of respondent-level and country-level control variables. For respondents, we control for their gender (=1 if female, = 0 if male), age (and age2 , both in years), and education level.9,10 At the country level, we control for time since last national election,11 time since the last EP election (in days), voting age population turnout in the last EP election (%), and how long the country has been

2

ASSESSING THE IMPACT OF MEMBERSHIP TURNOVER …

45

a full EU member (in years).12 Our online appendix includes descriptive statistics for all the variables mentioned above. Table 2.1 contains the main results from our ordered probit models, with the relevant rows for hypothesis 1a shaded dark gray. Positively signed coefficients mean that higher covariate values increase the probability of the respondent answering “very important” to EPIMPF.13 We begin with Model 1, a pooled model containing all EB surveys between 1983 and 2000. Consistent with H1a, we see that country within-session turnover has a negative and statistically significant effect. Respondents from countries with higher within-session MEP turnover rates are less likely to believe the EP should play a more important role in the European Union. This effect persists when we include an additional variable for the number of within-session dropouts within the past year in all other EU countries (Model 2).14 Model 2 also serves as a test of H1b (rows with light-gray shading). Initially, we see no support for H1b, but the non-result is a product of a non-linear relationship between other countries’ within-session turnover and respondents’ views about the EP. Once we include a squared term for other countries’ within-session turnover, we find an inverse-U relationship (Model 3). The non-linearity means for certain values of other countries’ within-session turnover, more turnover decreases the probability of a respondent believing the EP should play a more important role, consistent with H1b. However, for other of the covariate’s values, the relationship is flipped—higher levels of other countries’ within-session MEP turnover increases the probability of respondents believing the EP should play a more important role. Model 3’s results continue to hold if we remove any control variables with two-tailed p-values greater than 0.10 (Model 4). While our analysis so far offers some support for our hypotheses, our pooled sample may also be masking important era-specific effects. In other words, the relationship between within-session turnover and respondents’ beliefs about the EP may vary at differing moments in the broader context of the EU’s historical evolution. To investigate, we further split our sample into the three distinct eras, starting from the EP’s shift to an elected body in 1979: (1) the European Community period (August 8, 1979—October 31, 1993); (2) the EU’s pre-enlargement phase, once the Maastricht Treaty granted co-decision to the EP (November 1, 1993— April 30, 2004); and (3) the EU’s post-enlargement phase, following its near-doubling of member-states (since May 1, 2004). We use Clark and

46

W. T. DANIEL AND S. K. METZGER

Table 2.1 Determinants of “Should the EP Play a More Important Role?” Country WS turnover

Model 1

Model 2

Model 3

Model 4

–0.003*** (0.001)

–0.003*** (0.001) –0.007 (0.007)

0.010*** (0.003) –0.000*** (0.000) –0.110*** (0.014) 0.165** (0.072) 0.085*** (0.010) 0.242*** (0.029) 0.002 (0.002) 0.000 (0.000) –0.000 (0.000) –0.025*** (0.009) –1.709*** (0.219) –0.800*** (0.236)

0.010*** (0.003) –0.000*** (0.000) –0.110*** (0.014) 0.168** (0.071) 0.085*** (0.010) 0.242*** (0.029) 0.002 (0.002) 0.000 (0.000) –0.000 (0.000) –0.020** (0.009) –1.666*** (0.190) –0.756*** (0.211)

–0.003** (0.001) 0.050** (0.025) –0.009** (0.004) 0.010*** (0.003) –0.000*** (0.000) –0.110*** (0.014) 0.164** (0.072) 0.085*** (0.010) 0.243*** (0.029) 0.002 (0.002) 0.000 (0.000) –0.000 (0.000) –0.024*** (0.008) –1.625*** (0.193) –0.716*** (0.217)

–0.003** (0.001) 0.049** (0.025) –0.009** (0.004) 0.010*** (0.003) –0.000*** (0.000) –0.110*** (0.014) 0.164** (0.073) 0.085*** (0.010) 0.243*** (0.028)

–0.023*** (0.008) –1.927*** (0.248) –1.017*** (0.273)

260577 –226342.20 452712.40 452858.99

260577 –226339.93 452709.86 452866.92

260577 –226318.91 452665.83 452812.42

260577 –226343.44 452714.89 452861.48

Other countries’ WS turnover Other countries’ WS turnover2 Age Age2 Female Education: DK/NA Education: 16–19 Education: 20+ Turnout, last EP elect. Days since last EP elect. % since last nat’l elect. Length of EU m’ship τ1 τ2 τ3 N Log–likelihood AIC BIC

Model 5 DV = EPIMP1

0.002 (0.002) 0.044** (0.020) –0.006* (0.003) –0.001 (0.002) –0.000 (0.000) 0.071*** (0.022) 0.058 (0.069) 0.010 (0.023) –0.095** (0.038) 0.007** (0.003) 0.000*** (0.000) –0.000 (0.000) 0.008 (0.005) –0.409* (0.211) 0.782*** (0.198) 2.284*** (0.207) 237520 –263688.48 527404.97 527550.26

(Notes: * = p ≤ 0.10, ** = p ≤ 0.05, *** = p ≤ 0.01, two-tailed for all variables. WS = withinsession. Robust standard errors clustered on country reported in parentheses. Country fixed effect and wave fixed effect estimates omitted to save space. Omitted category for education: ≤ 15 years. DV = epimpf1 for all models except Model 5, where DV = epimpf1 (see discussion in note 17))

2

ASSESSING THE IMPACT OF MEMBERSHIP TURNOVER …

47

Rohrschneider’s (2019) data for information on this third era, as our principal dataset only goes through 2000. To preview our findings: we find support for both of our hypotheses about within-session turnover (H1a, H1b), but only during the European Community era (e.g., 1979– 1993). We find the opposite of what we posit for both hypotheses during the EU’s pre-enlargement era (1993–2004), and we find no support for either hypothesis during the EU’s post-enlargement era (2004 forward). Our era-specific models (Table 2.2) reveal distinct differences in the relationship between within-session MEP turnover and respondents’ EP beliefs. We now find non-linear relationships for both withinsession turnover variables when we disaggregate by era. We generate our predicted probabilities analytically using the era-specific models, varying within-session dropout’s values, as usual. However, for all the other covariates, we use our dataset’s observed values (“observed value approach”), instead of the typical convention of setting everything else to its mean or median (“average case approach”) (Hanmer and Kalkan 2013). The predicted probabilities still represent the same quantity, but their meaning is now different: it is the average predicted probability across all our dataset’s respondents, for specific values of within-session dropout. Figure 2.1 displays the predicted probabilities for respondent-country MEP within-session dropout’s 5th, 25th, 50th, 75th, and 95th percentiles within each era. The three columns correspond to EPIMPF’s three possible answers, while the three rows correspond to the three EU eras. For respondent-country within-session dropout to have an effect, the difference between two covariate profiles’ predicted probabilities must be statistically distinguishable from zero, indicated in Fig. 2.2 by a difference’s confidence interval (CI) excluding zero.15 Figure 2.1 shows that respondent-country within-session turnover exhibits a different effect within each EU era. Starting with the EC era (“pre-Maastricht” row), our results are as expected: as a country’s level of within-session turnover increases, respondents from that country are less likely to believe the EP should play a more important role (Fig. 2.1, upper-right panel). For instance, holding all other variables at their observed values, a respondent from a country with no within-session MEP turnover in the past year has a 63.8% chance, on average, of believing the EP should play a more important role,

48

W. T. DANIEL AND S. K. METZGER

Table 2.2 By EU Era, Determinants of “Should the EP Play a More Important Role?” Without Survey Weights Model 1 Country WS turnover Country WS turnover2 Other countries’ WS turnover Other countries’ WS turnover2 Age Age2 Female Education: DK/NA Education: 16-19 Education: 20+ Turnout, last EP elect. Days since last EP elect. % since last nat’l elect. Length of EU m’ship τ1 τ2 N Log-likelihood AIC BIC

Model 2

Model 3

With Survey Weights Model 6a

Model 7a

Pre-Maastricht

Post-Maastricht Post-Enlargement Pre-Maastricht

Post-Maastricht

(08AUG79–31OCT93)

(01NOV93–30APR04)

(01NOV93–30APR04)

-0.0078** (0.0031) 0.0001 (0.0001) -0.0699** (0.0284) 0.0090** (0.0038) 0.0117*** (0.0035) -0.0001*** (0.0000) -0.1305*** (0.0177) 0.0026 (0.0702) 0.0973*** (0.0117) 0.2524*** (0.0360) 0.0011 (0.0022) 0.0000*** (0.0000) -0.0001 (0.0003) -0.0265*** (0.0071) -1.8871*** (0.1897) -0.9666*** (0.1768) 170243 -141607.04 283236.07 283346.57

0.0107*** (0.0036) -0.0004*** (0.0001) 0.2990*** (0.0714) -0.0688*** (0.0195) 0.0072*** (0.0027) -0.0001** (0.0000) -0.0785*** (0.0152) 0.2527*** (0.0915) 0.0668*** (0.0132) 0.2155*** (0.0302) 0.0034* (0.0020) -0.0000 (0.0000) -0.0017*** (0.0004) -0.0030 (0.0149) -0.8964** (0.4124) 0.0027 (0.4263) 90334 -84236.89 168501.78 168633.54

(01MAY04–07NOV11)

-0.0010 (0.0040) 0.0002** (0.0001) 0.6983* (0.3722) -0.0706* (0.0386) 0.0087*** (0.0017) -0.0001*** (0.0000) -0.0815*** (0.0135) 0.0727 (0.0807) 0.1009*** (0.0169) 0.1694*** (0.0238) 0.0036 (0.0027) 0.0002** (0.0001) 0.0002 (0.0005) -0.0037 (0.0321) 5.8403** (2.8936) 6.2839** (2.8919) 81051 -77598.09 155228.19 155377.03

(08AUG79–31OCT93)

-0.0060** (0.0027) 0.0001 (0.0001) -0.0666*** (0.0256) 0.0082** (0.0035) 0.0120*** (0.0032) -0.0001*** (0.0000) -0.1331*** (0.0150) -0.0567 (0.0772) 0.0940*** (0.0091) 0.2461*** (0.0378) 0.0010 (0.0019) 0.0000*** (0.0000) -0.0001 (0.0003) -0.0265*** (0.0068) -1.8708*** (0.1749) -0.9444*** (0.1590) 170243 -140419.83 280863.66 280984.20

0.0089** (0.0039) -0.0003*** (0.0001) 0.3126*** (0.0686) -0.0726*** (0.0190) 0.0077*** (0.0027) -0.0001*** (0.0000) -0.0761*** (0.0153) 0.2465*** (0.0880) 0.0667*** (0.0162) 0.2163*** (0.0346) 0.0035* (0.0021) -0.0000 (0.0000) -0.0017*** (0.0004) -0.0031 (0.0142) -0.9566** (0.4172) -0.0550 (0.4321) 90334 -82753.79 165535.57 165667.33

(Notes: * = p ≤ 0.10, ** = p ≤ 0.05, *** = p ≤ 0.01, two-tailed for all variables. WS = withinsession. Robust standard errors clustered on country reported in parentheses. Country fixed effect and wave fixed effect estimates omitted to save space. Omitted category for education: ≤ 15 years. DV = EPIMPF. Model 8 uses data from Clark and Rohrschneider (2019))

2

ASSESSING THE IMPACT OF MEMBERSHIP TURNOVER …

Pr(About the Same) .3 .2 .15

.55 .1 4.9

0 13

4.9 1.2

13

.65

13

.2

.6

.25

.3 .25

.15

.2

.55 .1

Post−Maastricht

.15 .1 0

0

12

0

12

.65

.3

.1

.6

.1

.55

.15

.15

.2

.2

.25

.25

.3

12 2.3

2.3

2.3

(01MAY04−07NOV11)

0

4.9 1.2

.3

1.2

(01NOV93−30APR04)

.6

.25

.3 .25 .2

Pre−Maastricht

.15

(08AUG79−31OCT93)

.1 0

Post−Enlargement

Pr(More Important) .65

Pr(Less Important)

49

0

20 7.1

0

20

0

7.1

20 7.1

Resp−country’s MEP WS Dropout, Past Year (%)

Fig. 2.1 Predicted Probabilities: Respondent-Country MEP Dropout (Notes WS = within-session. Probabilities generated using Table 2.2’s unweighted era models for 5th, 25th, 50th, 75th, and 95th era-specific percentiles. 95% confidence intervals. Only four probabilities displayed because x’s 5th and 25th percentiles are equal; fewer than four lines occur in some panels for the same reason. Note the same y-scale within columns but different scales across columns)

pre-Maastricht. Increasing the amount of respondent-country withinsession MEP turnover to 1.2% lowers the respondents’ average probability of holding “should be more important” beliefs to 63.5%. This 0.3 percentage point decrease is statistically distinguishable from zero (Fig. 2.2, upper-right panel, first estimate), as are all the decreases in Fig. 2.1’s upper-right panel (Fig. 2.2, upper-right panel). Once the Treaty of Maastricht goes into effect, our evidence no longer supports H1a. In the post-Maastricht era (1993–2004), more respondent-country within-session MEP turnover increases the probability of a respondent believing the EP should play a more important role (Fig. 2.1, middle-right panel), and these increases are all statistically distinguishable from zero (Fig. 2.2, middle-right panel). After the

50

W. T. DANIEL AND S. K. METZGER

ΔPr(Less Important)

−.025 −.02 −.015 −.01 −.005 0

.015 .005 0 1.2

13

1.2

4.9

13

.04

4.9

.03 .02

−.005

.01

−.01

0

−.015

2.3

2.3 12

12

7.1 20

.04 .02 0 −.02

−.04

−.02

0

−.008 −.006−.004−.002 0

.002

.02

12

(01MAY04−07NOV11)

1.2

0

0 Post−Maastricht

−.02 −.015 −.01 −.005

(01NOV93−30APR04)

13 4.9

2.3

Post−Enlargement

ΔPr(More Important)

.01

.01 .005

Pre−Maastricht

0

(08AUG79−31OCT93)

.015

ΔPr(About the Same)

7.1

7.1 20

20

Resp−country’s MEP WS Dropout, Past Year (%)

Fig. 2.2 First Differences: Respondent-Country MEP Dropout (Notes WS = within-session. Plotted quantity: (labelled percentile’s predicted probability)(previous percentile’s predicted probability). Probabilities generated using Table 2.2’s unweighted era models for 5th, 25th, 50th, 75th, and 95th era-specific percentiles. 95% confidence intervals. Fewer than four first differences appear because some of x’s percentiles are equal, within eras. Note the different y-scales within and across columns, to maximize readability)

2004 EU enlargement, the relationship becomes statistically indistinguishable from zero (Fig. 2.2, bottom-right panel)—there appears to be little connection between respondent-country within-session turnover and respondents’ beliefs about the role that the EP should play. Given the EP’s increased importance after co-decision was codified by the Maastricht Treaty, it may be that citizen feelings towards the EP’s importance were already at a high-enough baseline and were therefore less sensitive to turnover rates. On the whole, then, our results suggest turnover in a respondent’s own MEP delegation within an EP session does lead respondents to believe the EP should play a less important role, but only pre-Maastricht.

2

51

ASSESSING THE IMPACT OF MEMBERSHIP TURNOVER …

We also run a second set of predicted probabilities in which we vary other countries’ within-session MEP turnover rate to assess H1b, computing the same era-specific percentile values as before (Fig. 2.3).16 Figure 2.3 shows that other countries’ within-session turnover rates affect respondents’ personal beliefs about whether the EP should play a more important role. Notably, like the respondent-country dropout results, the nature of this relationship varies across eras. For the pre-Maastricht era, our evidence is consistent with H1b: increasing other countries’ within-session dropout decreases the probability of a “should play a more important role” response. However, this relationship only holds for lower levels of turnover, with only the 5th–25th and 25th–50th differences being statistically different from zero (Fig. 2.4, upper-right panel). Pr(Less Important)

Pr(About the Same) .25

.6

.5

.15

.4

.2

.5

.4 .3

.1

.1

.3

Pre−Maastricht

.2

(08AUG79−31OCT93)

.3

.6

.7

Pr(More Important)

1.4

3.2

1.4

5.3

3.2 2.3

4

5.3

1.4

4

3.2 2.3

5.3 4

.25

.6

.5

.15

.4

.2

.5

.4 .3

.1

.1

.3

Post−Maastricht

.2

(01NOV93−30APR04)

.3

.6

.7

2.3

1.7 1.3 2.3

.52

3.1

1.7 1.3 2.3

3.1

.52

1.7 1.3 2.3

3.1

.25

.6

.5

.15

.4

.2

.5

.4 .3

.1

.1

.3

(01MAY04−07NOV11)

.2

Post−Enlargement

.3

.7

.6

.52

2.1 2.5

5.1 5.8 5.7

2.1 2.5

5.1

5.8 5.7

2.1 2.5

5.1

5.8 5.7

Other Countries’ MEP WS Dropout, Past Year (%)

Fig. 2.3 Predicted Probabilities: Other Countries’ MEP Dropout (Notes WS = within-session. Probabilities generated using Table 2.2’s unweighted era models for 5th, 25th, 50th, 75th, and 95th era-specific percentiles. 95% confidence intervals. Note the different y-scales within and across columns, to maximize readability)

52

W. T. DANIEL AND S. K. METZGER

ΔPr(Less Important)

ΔPr(About the Same)

−.02 −.01

4 3.2

5.3

.05

0

.02

.01

.1

.02

.04

2.3 5.3

2.3

−.05 1.3

3.1

2.3 1.7

1.3 3.1

3.1

−.3

2.5 5.1

5.7 5.8

.1 0

−.2

−.1

−.03 −.02 −.01

.2

0

0

.1

2.3 1.7

.01

1.7

.3

1.3

0

−.03 −.02 −.01

0

Post−Maastricht

4 3.2

5.3

−.04 −.02

(01NOV93−30APR04)

.02 0

.005 0 −.005

2.3

4 3.2

(01MAY04−07NOV11)

.01

.01

.01 .005 0

Pre−Maastricht

(08AUG79−31OCT93)

−.01 −.005

2.3

Post−Enlargement

ΔPr(More Important)

2.5 5.1

5.7 5.8

2.5 5.1

5.7 5.8

Other Countries’ MEP WS Dropout, Past Year (%)

Fig. 2.4 First Differences: Other Countries’ MEP Dropout (Notes WS = within-session. Plotted quantity: (labelled percentile’s predicted probability)(previous percentile’s predicted probability). Probabilities generated using Table 2.2’s unweighted era models for 5th, 25th, 50th, 75th, and 95th era-specific percentiles. 95% confidence intervals. Note the different y-scale within and across columns, to maximize readability)

In the 1993–2004 era, we find the opposite of what H1b implies; this surprising finding is similar to the respondent country–turnover finding. Increases in dropouts among MEPs from other member states increases, rather than decreases, a respondent’s probability of believing the EP should play a more important role (Fig. 2.4, middle-right panel). Finally, in the post-2004 era, there is no detectable relationship between the two variables (Fig. 2.4, bottom-right panel), thus offering no support for H1b. In short: within-session MEP turnover in other EU countries will lead a respondent to believe the EP should play a less important role, but again, only prior to the Maastricht Treaty.17

2

ASSESSING THE IMPACT OF MEMBERSHIP TURNOVER …

53

Although not the focus of our investigation, a number of control variables behave as expected throughout the results. Corroborating discussions of the feminist critique from the introductory chapter, for instance, we find that female respondents are consistently more pessimistic about the EP across all three eras, even after controlling for respondent age and education level. This lends empirical heft to the statements made by Irigoien (2019, 25–27). On the other hand, our results also confirm that more educated voters have been the most supportive of the EP throughout history. This speaks to populist critiques of EU institutions, which oftentimes cater to less-educated voters with simplistic characterizations of political institutions. Although we do not test these variables in light of varied degrees of turnover, it is worth noting such relationships within the broader context of the EU’s input legitimacy.

Discussion and Conclusions Citizens’ attitudinal support for the EU broadly, and the EP specifically, is an important component of the EU’s input legitimacy. However, citizens may reasonably be expected to view the EP in a negative light, should MEPs themselves not take their jobs seriously and leave early from their elected mandates. Our initial analysis offers some support for the expectation that high levels of MEP early exit may dampen the attitudinal dimension of the EU’s input legitimacy. More specifically, we find signs of a statistically significant relationship between rates of early exit and the importance that citizens ascribe to the EP, when both the early exit rates of one’s national MEPs and other countries’ MEPs are taken into consideration. We find, however, that this effect is only apparent in the expected direction until the advent of co-decision, at which time higher rates of turnover actually leave citizens valuing the EP as even more important. The effect then disappears entirely after the 2004 enlargement period. What, then, does this say about the connection between EP personnel turnover and citizen attitudes towards the legislature? And what, more broadly, does it tell us about the EU’s input legitimacy? Having only been elected since 1979, the EP was still a relatively new—and comparatively powerless—institution throughout the 1980s. This may explain a general view that the EP should be more important, particularly among those citizens who perceived membership turnover as lower. Naturally, the opposite appears to be true for those citizens exposed to high levels of turnover—why shift more powers to a new

54

W. T. DANIEL AND S. K. METZGER

legislature unable to keep its members? Unlike in present times, a much larger proportion of MEPs during the pre-Maastricht period were indeed yesteryear’s political retirees, as well as unfamiliar faces, and indeed even some national MPs who continued to serve dual mandates in both national legislatures and the EP. It would therefore be reasonable for these citizens to take a pessimistic attitude towards a legislative body whose representatives may not have seemed fully committed to the demands of the job. On the other hand, since the expansion of EP powers to include codecision, we found that high turnover may actually have had a positive effect on citizen views of the EP. Perhaps this also coincides with an era in which EU politicians were beginning to build their own political “class” and voters saw a need for new, specialist faces that were not just angling for a national position? Another way of saying this is that if politicians were continuing to use the EP as a springboard to national office (cf. Daniel 2015) and thereby leaving it early, even as its institutional capacity had grown, voters might continue to view national legislatures as the most important and fail to support further increases in the EP’s level of power. Finally, the lack of effect for turnover on citizen views, post-2004, could be seen as a sign of the EU’s maturation. Nearly three decades of co-decision on, the EP may now just be seen as important “enough” and therefore our dependent variable less sensitive to a wide range of inputs that might once have swayed citizen opinions about the EP. Regardless, this chapter’s analysis provides no empirical support for the populist notion that turnover will bolster input legitimacy. Whatever the case, the EP remains an interesting example for the study of turnover’s effect on citizen views. Similar to other chapters in this volume, those effects are not always obvious or straightforward. However, so long as the EP is seen as both of secondary importance to the national parliaments of the EU, as well as the crucial balm of input legitimacy for a Union plagued by the narrative of democratic deficits, then its personnel turnover is worthy of consideration. If EP personnel turnover does not matter much for the attitudinal dimension of input legitimacy, the volume’s subsequent chapters will investigate the extent to which it matters for the behavioral dimension of input legitimacy, for throughput legitimacy, and for output legitimacy.

2

ASSESSING THE IMPACT OF MEMBERSHIP TURNOVER …

55

Notes 1. As we mention later, we eventually split our sample into different EU eras, using Clark and Rohrschneider’s (2019) data for the period 2004–2011. 2. We display the specific breakdown of countries based on our included survey waves, with the values representing the number of respondents from that country in a particular year, in the online appendix. 3. EB19, EB20, EB21, EB22, EB23, EB24, EB25, EB26, EB27, EB28, EB29, EB30, EB31, EB33, EB34.0, EB35.0, EB36, EB37.0, EB38.0, EB39.0, EB42, EB43.1, EB44.1, EB47.2, EB48.0, EB49, EB52.0, EB53. 4. “Would you, personally, prefer that the EP played…,” “Would you personally like the EP to play a more or a less important role than it does now?”. 5. Different countries coded DK and NA in different ways in different waves (e.g., as two separate values for some waves, but a single value in others). To address this, DK and NA are merged into a single value in the Mannheim dataset. 6. The Mannheim dataset lacks the specific survey structure information we would need to make further design-related adjustments. 7. Technically, the end point of our year-long windows is the day before the earliest possible survey date. 8. For descriptive information on within-session turnover broken down by country and EP session, see Daniel and Metzger (2018, Appendix A). 9. Education level is a categorical variable, recording the age at which respondents “finished” (for some waves) or “stopped” (in other waves) their full-time education. There are four categories: (1) < 15 years old; (2) 16–19 years old, (3) 20 + years old, or (4) DK/NA. For respondents still in school, we use their age at the time of the survey. We use < 15 years as our omitted category and include the other three categories as regressors. 10. We considered including other respondent-level variables, such as political knowledge, interest, efficacy, and/or participation; degree of Euroskepticism; and the respondent’s community size (urban/suburban/rural). However, these questions do not appear consistently across the same EB waves as EPIMPF. We lose at least 75% of our observations if we include at least one of these variables in our regressions. Further, some of the questions (political participation, political interest) do not ever appear in the same EB wave as EPIMPF; we lose 100% of our observations. We nonetheless check for robustness in our findings by including the variables we can in separate models for applicable EB waves; results remain consistent. 11. Formally, we normalize this variable. It appears as the percent of days that have passed in the country’s current constitutional inter-election period (CIEP). The CIEP for all countries in our sample is either 4 or 5 years.

56

W. T. DANIEL AND S. K. METZGER

12.

13.

14.

15.

16.

17.

A value of 10% in a country with a 5-year CIEP means 6 months have passed since the last national election. Similar to note 10, there were additional country-level variables we wished to include, but could not without losing observations (e.g., average number of voting age population per MEP). Ordered probit coefficients indicate the direction of x’s relationship with the probability of observing the scale’s first or last categories, similar to basic probit. However, the coefficients do not have a similarly easy meaning for the probability of observing the scale’s middle category (or categories) because of how ordered probit models are constructed (Greene 2003: 738–739). This variable is also a percent, with the denominator computed using the same weighted average logic as the country-specific turnover variable. The weights are multiplied by the number of seats allocated in the relevant EP session(s) to all countries other than the respondent’s. We opted to assess differences in adjacent covariate profiles only (e.g., 25th vs. 50th percentile), but could have also assessed others. See Austin and Hux (2002) for why we must use first differences to gauge statistical significance instead of overlapping predicted probability CIs. This variable exhibited a quadratic relationship with respondents’ beliefs about the EP’s importance and was statistically significant in Model 3’s pooled results. This pattern continues to hold for Table 2.2’s era-specific models. We also estimated a model with a slightly different dependent variable: EPIMPF1, which addresses whether a respondent thinks the EP plays an important role—different than our dependent variable, which asks should the EP play a more important role. Interestingly, neither type of within-session dropout has any effect on respondents’ belief about the role that is played by the EP. The implication, then, is that within-session dropout does not affect respondents’ EP-related views about what is the EP’s importance, but only about what should the EP’s importance be, relative to now. These results are featured in Table 2.1, Model 5.

References Austin, Peter C., and Janet E. Hux. 2002. A Brief Note on Overlapping Confidence Intervals. Journal of Vascular Surgery 36 (1): 194–195. Beauvallet-Haddad, Willy, Sébastien. Michon, Victor Lepaux, and Céline Monicolle. 2016. The Changing Composition of the European Parliament: MEPs from 1979 to 2014. French Politics 14 (1): 101–125. Brug, Wouter, and van de and Claes H. de Vreese. 2016. (Un)intended Consequences of EU Parliamentary Elections. Oxford: Oxford University Press.

2

ASSESSING THE IMPACT OF MEMBERSHIP TURNOVER …

57

Clark, Nick, and John A. Scherpereel. 2021. Turnover, Turnout, and Input Legitimacy in the EU. In Personnel Turnover and the Legitimacy of the EU , ed. John A. Scherpereel, London: Palgrave. Clark, Nicholas J., and Robert Rohrschneider. 2019. The Relationship between National Identity and European Union Evaluations, 1993–2017. European Union Politics 20 (3): 384–405. Daniel, William T. 2016. First-Order Contests for Second-Order Parties? Differentiated Candidate Nomination Strategies in European Parliament Elections. Journal of European Integration 38 (7): 807–822. Daniel, William T., and Shawna K. Metzger. 2018. Within or Between Jobs? Determinants of Membership Volatility in the European Parliament, 1979– 2014. Journal of Legislative Studies 24 (1): 90–108. DuMouchel, William H., and Greg J. Duncan. 1983. Using Sample Survey Weights in Multiple Regression Analyses of Stratified Samples. Journal of the American Statistical Association 78 (383): 535–543. Farrell, David M., and Roger Scully. 2007. Representing Europe’s Citizens?: Electoral Institutions and the Failure of Parliamentary Representation. Oxford: Oxford University Press. Greene, William H. 2003. Econometric Analysis, 5th ed. Upper Saddle River, NJ: Prentice Hall. Hanmer, Michael J., and Kerem Ozan Kalkan. 2013. Behind the Curve: Clarifying the Best Approach to Calculating Predicted Probabilities and Marginal Effects from Limited Dependent Variable Models. American Journal of Political Science 57 (1): 263–277. Hix, Simon, and Sara Hagemann. 2009. Could Changing the Electoral Rules Fix European Parliament Elections ? Politique Européenne 28: 37–52. Hix, Simon, and Michael Marsh. 2011. Second-Order Effects plus Pan-European Political Swings: An Analysis of European Parliament Elections across Time. Electoral Studies 30 (1): 4–15. Franklin, Mark N., and Sara B. Hobolt. 2011. The Legacy of Lethargy: How Elections to the European Parliament Depress Turnout. Electoral Studies 30 (1): 67–76. Hobolt, Sara B., and James Tilley. 2014. Blaming Europe?: Responsibility Without Accountability in the European Union. Oxford: Oxford University Press. Hobolt, Sara B., and Catherine E. de Vries. 2016. Public Support for European Integration. Annual Review of Political Science 19 (1): 413–432. Hobolt, Sara B., and Jill Wittrock. 2011. The Second-Order Election Model Revisited: An Experimental Test of Vote Choices in European Parliament Elections. Electoral Studies 30 (1): 29–40. Irigoien, Alazne. 2019. European Parliament’s Elections 2019: Towards Parity Democracy in Europe. Brussels: Gender Five Plus. Available at https://

58

W. T. DANIEL AND S. K. METZGER

docs.wixstatic.com/ugd/530efa_2a575f9fa3dc45afb68e4ffb516fef30.pdf. Accessed May 28, 2020. Matland, Richard, and Donley Studlar. 2004. Determinants of Legislative Turnover: A Cross-National Analysis. British Journal of Political Science 34 (1): 87–108. Muñoz, Jordi, Mariano Torcal, and Eduard Bonet. 2011. Institutional Trust and Multilevel Government in the European Union: Congruence or Compensation. European Union Politics 12 (4): 551–574. Obholzer, Lukas, and William T. Daniel. 2016. An Online Electoral Connection? How Electoral Systems Condition Representatives’ Social Media Use. European Union Politics 17 (3): 387–407. Reif, Karlheinz. 1984. National Electoral Cycles and European Elections 1979 and 1984. Electoral Studies 3 (3): 244–255. Reif, Karlheinz, and Hermann Schmitt. 1980. Nine Second-order National Elections: A Conceptual Framework for the Analysis of European Election Results. European Journal of Political Research 8 (1): 3–44. Scherpereel, John A., and Lauren K. Perez. 2015. Turnover in the Council of the European Union: What It Is and Why It Matters. JCMS: Journal of Common Market Studies 53 (3): 658–673. Schmitt, Hermann, Sara Hobolt, and Sebastian Adrian Popa. 2015. Does Personalization Increase Turnout? Spitzenkandidaten in the 2014 European Parliament Elections. European Union Politics 16 (3): 347–368. Schmitt, Herman, and Evi Scholz. 2009. Mannheim Eurobarometer Trend Dataset 1970–2002 (ZA3521). Mannheimer Zentrum fur Europaische Sozialforschung (MZES) and Leibniz Institute for the Social Sciences (GESIS, former “Zentrum für Umfragen, Methoden und Analysen” and “Zentralarchiv für Empirische Sozialforschung”). Available at https://www. gesis.org/eurobarometer-data-service/search-data-access/eb-trends-trendfiles/mannheim-eb-trend-file/. Accessed July 10, 2020. Solon, Gary, Steven J. Haider, and Jeffrey M. Wooldridge. 2015. What Are We Weighting For? Journal of Human Resources 50 (2): 301–316. van Geffen, Robert. 2016. Impact of Career Paths on MEPs’ Activities. JCMS: Journal of Common Market Studies 54 (4): 1017–1032. Whitaker, Richard. 2014. Tenure, Turnover and Careers in the European Parliament: MEPs as Policy-Seekers. Journal of European Public Policy 21 (10): 1–19. Winzen, Thomas, Christilla Roederer-Rynning, and Frank Schimmelfennig. 2015. Parliamentary Co-Evolution: National Parliamentary Reactions to the Empowerment of the European Parliament. Journal of European Public Policy 22 (1): 75–93.

CHAPTER 3

Turnover, Turnout, and Input Legitimacy in the EU Nicholas Clark and John A. Scherpereel

Legitimacy, as discussed by Scherpereel in this volume’s introduction, has both attitudinal and behavioral dimensions. Legitimacy requires the population to see the governing system and its rules as relevant and worthy of respect (the attitudinal dimension) and to behave in accord with the governing system’s rules, procedures, and values (the behavioral dimension). Daniel and Metzger (this volume) suggest that turnover does not have a particularly robust effect on the attitudinal dimension of input legitimacy. Might turnover have a stronger/more temporally robust effect on the behavioral dimension of input legitimacy?

N. Clark (B) Department of Political Science, Susquehanna University, Selinsgrove, PA, USA e-mail: [email protected] J. A. Scherpereel Department of Political Science, James Madison University, Harrisonburg, VA, USA © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. A. Scherpereel (eds.), Personnel Turnover and the Legitimacy of the EU, Palgrave Studies in European Union Politics, https://doi.org/10.1007/978-3-030-60052-5_3

59

60

N. CLARK AND J. A. SCHERPEREEL

This chapter examines that question. It seeks to determine whether personnel turnover among MEPs influences participation in EU elections. We begin by situating our study within the canonical literature on comparative voter turnout (Gray and Caul 2000; Jackman 1987; Powell 1986) and the more specific literature on turnout in EP elections (Bhatti and Hansen 2012; Clark and Rohrschneider 2009; de Vries et al. 2011; Fiorino et al. 2019; Flickinger and Studlar 2007; Franklin 2001; Franklin and Hobolt 2011, 2015; Hobolt and Spoon 2012; Mattila 2003; Schmitt et al. 2015; Studlar et al. 2003; Weber 2011). We suggest that dominant explanations of turnout and the prevailing theory of EP voting behavior (second-order elections theory) have devoted little attention to the ways that legislatures’ composition and personnel dynamics might affect citizens’ electoral engagement. In the case of the EP, the lack of attention is striking, given the centrality of concerns about institutional composition and dynamics in populist and feminist critiques of EU politics. We begin the chapter by considering why composition and dynamics might affect EP voter turnout. We present hypotheses about turnout that derive from the populist and feminist logics. After discussing our data and methods, we present results of models that seek to explain national-level variation in EP elections between 1984 and 2019. We find no support for hypotheses that derive from the populist critique and limited support hypotheses that derive from the feminist critique. Our evidence suggests that voters turn out in greater numbers when national delegations of MEPs become more gender-balanced.

Personnel Turnover and Voter Turnout Voter turnout is a widely studied phenomenon in comparative politics. Scholars of turnout have taken analytically distinctive approaches to the topic. They have used different units of analysis, for example, and treated time in distinct ways. Regarding units of analysis, aggregate or macro approaches study turnout at the level of the nation-state or other aggregated political unit. Macro studies seek to determine, for example, why turnout rates are higher in one country than another. Individual- or micro approaches, on the other hand, ask why certain individuals (e.g., people belonging to different age groups) are more or less likely than others to vote. In recent years, scholars (e.g., Fiorino et al. 2019) have used multi-level models that incorporate both aggregate and voter-based characteristics. With regard to treatments of time, some scholars use (pooled)

3

TURNOVER, TURNOUT, AND INPUT LEGITIMACY IN THE EU

61

cross-sectional snapshots, while others prioritize temporal dynamics. The latter group of scholars seeks to understand, for example, why turnout in a particular country is higher at one moment than at some other moment (see Gray and Caul 2000). In this chapter, we take an aggregate approach and rely on pooled cross-sections. We seek to contribute to a long series of studies on EP turnout (Fiorino et al. 2019; Franklin 2001; Franklin and Hobolt 2011, 2015; Mattila 2003; Studlar et al. 2003; Flickinger and Studlar 2007) and to determine whether and how EP personnel composition and dynamics might help to explain variation in member-state level turnout in EP elections. For reasons discussed below, we focus on EP elections from 1984 through 2019. We focus exclusively on elections that take place on the EP’s five-year schedule. We exclude, for example, off-cycle elections that take place in acceding states (e.g., Greece 1981; Croatia 2013) and empty-seat by-elections that took place in the United Kingdom during the period (1979–1999) when UK MEPs were elected in single-member districts. While our study speaks most directly to the stream of work on EP voter turnout, it also addresses the broader turnout literature. Canonical cross-sectional studies of turnout in national parliamentary elections (Jackman 1987; Powell 1986) have shown that institutional variables— e.g., laws requiring citizens to vote or constitutional structures (like unicameralism) that increase the stakes of elections—go a long way toward explaining cross-country variation. As subsequent generations of scholars have noted, though, countries whose institutional characteristics remain constant over time still often experience significant variation in turnout from one election to the next. Many established democracies, for example, have experienced decades-long declines in voter turnout even though their institutional systems have experienced little change. Declines in voter turnout over time can be driven, for example, by declines in rates of group mobilization (Gray and Caul 2000). Canonical micro-level studies, for their part, show that particular voter-level characteristics (e.g., income, education, political interest, and, at least to some extent, age) are positively correlated with higher turnout. The more targeted literature on turnout in EP elections has focused on explaining the effects of both macro conditions (such as the timing of the election) and micro conditions (such as knowledge about the EU) on voting behavior. EP elections contain analytically attractive quasi-experimental features. Elections are held over similar sets of days

62

N. CLARK AND J. A. SCHERPEREEL

once every five years, but electoral institutions—compulsory voting laws, minimum voting ages, days of the week on which voting takes place, openness of ballots—vary from member-state to member-state. In addition, citizens of member-states have different structural positions vis-à-vis the European polity. Some live in states that host EU institutions, for example. Some reside in countries that are net recipients from the EU’s budget. Some countries’ citizens trust the EU more than others. In addition, a range of domestic political factors—whether an election is a country’s first EP election, the amount of time that has elapsed since the country’s most recent general election, whether the EP election shares a ballot with a national/regional/local election and/or a referendum—vary from member-state to member-state. Within the EP turnout literature, there is a fair amount of consensus about the causal relevance of certain variables. At the macro level, for example, studies tend to agree that compulsory voting, weekend voting, EU identity and/or trust, simultaneity of EP elections and national elections, EP elections that fall late in national electoral cycles, older demographic age profiles, and higher levels of turnout in general elections are all associated with higher levels of turnout at EP elections. These insights into the correlates of EP turnout generally emerge from scholars’ collective grappling with similar empirical and theoretical concerns. Empirically, scholars of EP turnout generally contend with one or more of three observations. First, levels of turnout to EP elections are generally lower than levels of turnout to national general elections. Of the 147 EP-level turnout values that we examine in this chapter, for example, only seven (4.8%) featured turnout rates that were higher than the country’s most recent national general election. Second, levels of turnout in EP elections have always varied quite widely from one member-state to the next. For the on-cycle elections between 1979 and 2019, for example, the smallest annual standard deviation in the percentage of the national population that voted was 16.27 (2019); the largest was 20.61 (2004). Third, EU-wide election turnout has been on a rather steady downward trajectory over time, even as the competencies of the EP have increased. Before 2019, EU-wide turnout levels had fallen in every successive oncycle election, and the upward tick between 2014 (EU-wide turnout = 43.28%) and 2019 (EU-wide turnout = 48.82%) was noticeable but relatively modest. Theoretically, studies of turnout and voting behavior in EP elections generally begin with a nod toward Reif and Schmitt’s (1980) theory of

3

TURNOVER, TURNOUT, AND INPUT LEGITIMACY IN THE EU

63

second-order elections (SOE). SOE theory suggests that the results of EP elections are of little direct consequence to the electorate. It implies that citizens see EP elections as less important than national general elections, that EP electoral campaigns are filtered through national lenses and often revolve around issues of national (as opposed European) concern, and that voters often see EP votes more as a means of signaling to national politicians than of assuring that their interests are represented and/or their identities recognized in the course of EP business. Over the years, various extensions of the SOE logic have reinforced the idea that citizens are unlikely to vote in EP elections. Van der Eijk et al. (1996), for example, stress that EP elections have traditionally had no impact on the selection of the EU executive. While the Spitzenkandidat process—the process that seeks to link the selection of the European Commission president to EP election results—may have affected the applicability of this logic in the 2014 and 2019 elections, the European Council’s rejection of the European People’s Party’s Spitzenkandidat in the Commission formation process in 2019 may reactivate it in future elections. In addition, as Moravscik (2002) has argued, the EU has historically been most active in lower-salience policy areas (e.g., central banking, technical administration), while national and/or sub-national governments play larger roles in higher-salience issues (e.g., health care, education, law and order, pensions, and most taxation). These exigencies reinforce a sense that EP elections are lower-stakes affairs than national general elections. In general, SOE theory provides a stronger and more direct explanation of the first empirical observation mentioned above (the fact that turnout in EP elections is lower than turnout in national general elections) than it does to the second and third observations (the fact that EP election turnout varies widely across contexts; the fact that Union-wide EP election turnout has declined over time). It is rather striking, especially given the critiques spotlighted in chapter one of the current volume, that the existing literature—both the canonical/comparative and the EP-focused—has not focused on the ways that a chamber’s composition and personnel dynamics might affect citizens’ behavioral engagement with the chamber. By composition, we mean, simply, who serves as a legislator. And by personnel dynamics, we mean the quantity, mechanism, and quality of legislators entering and exiting the EP over time. For the case of the EP, a wide stream of research— tracing at least to Kreppel (2002) and extending at least to Salvati and Vercesi (this volume)—examines the ways that the institution has

64

N. CLARK AND J. A. SCHERPEREEL

developed, internally, over time. There has been little attempt, though, to examine whether composition and personnel dynamics affect voter turnout. Issues of composition and dynamics figure prominently, after all, in both populist and feminist critiques of EU politics. According to both sets of critics, composition and dynamics should affect the behavioral dimension of input legitimacy. Consider the bigger-picture implications of focusing exclusively upon electoral institutions, members’ structural positions, domestic political considerations, and citizens’ individual-level characteristics. Collectively, studies that spotlight these factors and/or are foregrounded in SOE theory suggest that the composition and activities of the representatives whom citizens elect are largely irrelevant for determining whether citizens exercise voting privileges.1 This suggestion clearly counters the claims of political parties, who are perpetually eager to convince voters that their representatives matter and that ceding control of parliamentary power to other parties would have negative consequences. It also counters the claims and implications of the EU’s populist and feminist critics. Populists, for example, identify a chasm between a decadent, out-of-touch, self-interested, self-perpetuating elite and a pure national people (Mudde 2007). According to populist critics, to increase the legitimacy of the EU—to kick start a process that activates certain behaviors including turning out to vote—one must replace the incumbent elite with office-holders who are more in touch with the general will. Feminist critics, for their part, decry the gendered compositions of EU institutions. They suggest that the EU will gain legitimacy—which, again, involves higher turnout—once institutions come to more accurately resemble the populations they govern. Thus, the logics that undergird both populist and feminist critiques are quite distinctive from the analytical foci of canonical studies and the most prevalent theory explaining low turnout in EP elections. Existing treatments suggest that institutional, structural, domestic political, and individual-level characteristics affect variation in turnout and that, insofar as EP elections are second-order phenomena, citizens pay little attention to EP composition and dynamics. Populist and feminist critiques, on the other hand, suggest that citizens are quite attuned to personnel composition and dynamics. They suggest that legitimacy—and, by extension, higher turnout—will come if and only if elites are replaced (for populist critics) and greater gender balance is achieved (for feminist critics).

3

TURNOVER, TURNOUT, AND INPUT LEGITIMACY IN THE EU

65

To what extent might legitimizing sequences foreseen by populist and feminist critics actually affect EP voter turnout? We include both frameworks in these analyses, since each is founded on the public’s perceptions of the EP’s composition and personal dynamics. Here, we present four hypotheses that derive from the assumption that European citizens are, indeed, attentive to EP composition and dynamics. The first two hypotheses test the populist claim that citizens will vest the EU with more legitimacy—and, thus, will be more likely to vote in EP elections—when there is more compositional disruption and dynamism in the ranks of a member-state’s delegation of MEPs. H1: The greater the quantity of turnover a member-state’s EP delegation experiences at one EP election, the higher the level of turnout the member-state will experience at the subsequent EP election. H2: The greater the quantity of turnover a member-state’s EP delegation experiences within an EP session, the higher the level of turnout the member-state will experience at the subsequent EP election. H1, which focuses on the composition of a member-state’s EP delegation, involves a logic of collective political efficacy (Lee 2010; Lubell et al. 2007; Seligson 1980). Collective (or group) efficacy, involves “the perception of whether a collective actor to which an individual belongs is capable of achieving desired outcomes” (Lee 2010: 393). While the collective efficacy concept has generally been used in studies of social movements, the populist critique implies that it may also pertain with regard to a national electorate: when members of a national electorate see that voters can disrupt the perpetuation of rule by incumbents, they will be more likely to seize future opportunities to disrupt incumbents’ rule. On the populist vision, two separate mechanisms may undergird the association between within-term turnover and turnout that H2 predicts. On one hand, within-term turnover may result in the same kind of disruption that electorally induced incumbent departure involves. In this case, voter turnout would reflect citizens’ approbation of the exit of sitting elites. On the other hand, within-term turnover is a particularly opaque process. As noted above, the UK was exceptional in the 1979–1999 era in that it held by-elections for MEPs who had died or been disqualified. In practice, EP rules allow for significant variation in the ways that memberstates fill vacated seats. Some member-states, for example, fill such seats by appointing the “first excluded” candidate from the same party list as

66

N. CLARK AND J. A. SCHERPEREEL

the exiting MEP. Others allocate such seats to pre-designated substitutes. Regardless of the national-level processes in place, such within-session replacements are often made long after the latest EP election. If parties make controversial and/or contestable substitutions, critics may become angry and fold the episode into their general critique of subversion by insiders. Seats that exist to amplify the voice of the people, according to the latter logic, are being allocated on the basis of, at best, obsolete aggregations of citizen interests or, at worst, elite subterfuge. Regardless, citizens would be more likely to turn out in the future EP election, as the election might promote a restoration of the people’s voice. The third and fourth hypotheses test the feminist claim that citizens will vest the EU with more legitimacy—and, thus, will be more likely to vote in EP elections—when there is more compositional diversity and gender balance in a member-state’s delegation of MEPs. Like H1, H3 posits a relationship between across-session turnover and voter turnout. Like H2, H4 posits a relationship between within-session turnover and voter turnout. H3: The greater the increase in women’s representation a member-state’s EP delegation experiences at one election, the higher the level of turnout the member-state will experience at the subsequent EP election. H4: The more gender-balanced a member-state’s EP delegation becomes over the course of an EP session, the higher the level of turnout the member-state will experience at the subsequent EP election. According to the logic of the feminist critique, behavioral legitimacy will respond to gender progress: voter turnout will rise when EU institutions are more reflective of societal gender balances. We might expect this legitimacy boost to be particularly pronounced in the case of the EP, since it is the Union’s only directly elected institution. Like H1, H3 is undergirded by a logic of collective political efficacy. It assumes, in this case, that when members of a national electorate see that their actions increase the descriptive representation of the EU’s directly elected legislature, they will become more likely to vote in subsequent elections. Like H2, H4 examines within-term turnover. It assumes, in line with the feminist critique, that within-term movement toward gender balance promotes a sense of responsiveness and inclusion that can induce more citizens to vote. Even if women’s representation is low among the set of national MEPs that are seated immediately after an election, voters may welcome attempts by empowered selectors to promote gender balance when seat vacancies materialize.

3

TURNOVER, TURNOUT, AND INPUT LEGITIMACY IN THE EU

67

Data and Research Design To test the hypotheses presented above, we examine national-level voter turnout data from all on-cycle EP elections from 1984 through 2019 (e.g., 1984, 1989, 1994, 1999, 2004, 2009, 2014, 2019; N = 147). We exclude results from the 1979 elections since our critical independent variables involve MEP composition and dynamics in the legislative session preceding each on-cycle election. Our dependent variable—the percentage of a member-state’s eligible voters that exercised the right to vote—come from the European Parliament (2020). Data for our independent variables are calculated on the basis of Daniel’s (2015) MEP career database, supplemented by EP data (from europarl.eu) on MEP composition and dynamics. Across-session turnover measures how much change in the personnel composition of a country’s MEP delegation occurred at the previous EP election. Specifically, it compares the composition of a country’s MEP delegation on the first day of the first post-election EP plenary and the final day of the EP plenary session that preceded the previous election. An example helps to illustrate how this variable is constructed. Consider the EP election that was held in Ireland in 1989. Ireland’s 1989 Across-session turnover value is determined by comparing the composition of Ireland’s EP delegation on the first day of the EP’s 1984–1989 session with the composition of Ireland’s EP delegation on the final day of the EP’s 1979–1984 session. In this example, eight of the 15 Irish MEPs who were part of Ireland’s EP delegation on July 24, 1984 were not part of Ireland’s EP delegation on May 25, 1984. Thus, Ireland’s Across-session turnover value for 1989 is 8/15 = 53.33%.2 Within-session turnover is constructed in a similar fashion. For Ireland 1989, it is calculated by dividing the number of Irish MEPs at the end of the 1984–1989 session who did not have seats at the beginning of the session (2) by the total number of MEPs in the Irish delegation at the end of the 1984–1989 session (15); thus, 2/15 = 13.33%. Across-session gender progress records the extent to which voters have used the ballot box in previous elections to boost women’s descriptive representation. Specifically, it examines the difference between women’s representation on the first days of the two most proximate EP sessions. Again, an example is illustrative. Consider Ireland’s Across-session gender progress value for 1989. This value (=0) is determined by subtracting the percentage of members of the Irish EP delegation that were women on July 24, 1984 (13.33%) from the percentage of members of the Irish EP

68

N. CLARK AND J. A. SCHERPEREEL

delegation that were women on July 11, 1979 (13.33%). This illustration demonstrates that Across-session gender progress can take on a positive, negative, or zero value. Positive values denote progress toward gender balance; negative values denote regress from gender balance. The values of Within-session gender progress can also be positive, negative, or zero. The latter variable measures how much more or less gender-balanced a member-state’s EP delegation becomes over the course of the legislative session preceding an election. We calculate it by subtracting the percentage of women in a member-state’s delegation at the first post-election plenary from the percentage of women in the same member’s delegation at the final plenary of the same session. Here, consider France 1989. On the first day of the first plenary after the 1984 EP elections (again, July 24, 1984), 22.09% of France’s MEPs were women. On the final day of the final plenary before the 1989 EP elections (May 26, 1989), 20.73% of France’s MEPs were women. Thus, in the 1984–1989 EP session, France’s EP delegation became slightly less gender-balanced. Its Within-session gender progress score for 1989 is 20.73–22.09 = - 1.36. Our multivariate models control for factors that previous macro-level studies of EP turnout have found significant. We include controls related to electoral institutions, member-state positions within the EU, and domestic political factors. Institutions Studies consistently find that countries with compulsory voting have higher turnout levels. Thus, we construct a dichotomous Compulsory measure (0 = no compulsory voting; 1 = compulsory voting); our data come from International IDEA (2020) and the European Parliament Research Service (EPRS 2014, 2019). Because existing research also suggests that weekend voting boosts turnout, we construct Weekend, which codes whether polls are open on Saturday and/or Sunday (0 = no weekend voting; 1 = weekend voting). Data for Weekend come from the EPRS (2014, 2019), national elections authorities, and Wikipedia. Position Within the EU We also control for a number of factors related to countries’ positions in the EU political economy. Institution host takes a value of 0 if a country

3

TURNOVER, TURNOUT, AND INPUT LEGITIMACY IN THE EU

69

does not host one or more of the EU institutions mentioned in Article 13 TEU and 1 otherwise; following the literature (e.g., Flickinger and Studlar 2007; Mattila 2003; Studlar et al. 2003), we expect host states to have higher levels of turnout. EU net recipient incorporates data from Begg and Grimwade (1998) and Commission budget reports on the net budgetary balance for each EU member-state in the year prior to the relevant EP election. Countries are coded 0 if they are not net budgetary recipients and 1 otherwise; the intuition (which is borne out in the analyses of Mattila 2003) is that citizens of net-recipient countries turn out in greater numbers. Our third EU positionality control is EU identity. Our indicator comes from nationally aggregated results of the standard Eurobarometer question that inquires about a respondent’s sense of European identity. The indicator’s values represent the percentage of a national public that identifies Europe as part of their identity in the year prior to the election. Higher values should correspond with greater turnout. Previous research (Studlar et al. 2003; Fiorino et al. 2019) suggests that citizens in countries with higher aggregate levels of European identity are more likely to vote than citizens in countries with lower levels of European identity. The final measure here is whether an election took place in 2019. The aggregated voter turnout in EP elections declined in each successive election between 1979 and 2014. The 2019 election represents a deviation with the first ever increase in turnout across Europe. As such, we use EP election 2019 to control for that election year in the analyses. Domestic Political Factors Existing research consistently finds that the location of an EP election within a country’s national electoral cycle matters (Flickinger and Studlar 2007; Franklin 2001; Mattila 2003). Citizens tend to turn out at lower levels when EP elections are held soon after national general elections, while elections held in the run up to general elections provide indications of possible future general election results, thus driving turnout upward; similarly, EP elections that are held at the same time as national elections tend to have higher rates of turnout. Our Electoral cycle variable—calculated on the basis of data from national election authorities, the European Elections Database (EED, http://eed.nsd.uib.no), and Wikipedia—calculates the number of days elapsed between the date of the most recent national general election and the EP election. Finally, as one means of controlling for country-specific effects, and because existing

70

N. CLARK AND J. A. SCHERPEREEL

studies find a clear positive association between national turnout levels and EP turnout levels (Flickinger and Studlar 2007; Studlar et al. 2003), we control for General election turnout. Our measure of turnout in the most recent national general election comes from International IDEA (2020). To test our hypotheses, we rely primarily on linear regression models using STATA. In all models, the standard errors are clustered by country to account for any country-based variation not accounted for in the model parameters.

Results We begin by presenting descriptive statistics, in Table 3.1, for each of the measures that we eventually include in our models. In reviewing Table 3.1, we first note that the mean national-level turnout in EP elections (51.0%) is indeed much lower than the mean turnout in national general elections (72.7%). Additionally, memberstates’ delegations tend to experience more turnover across sessions than they do within sessions: an average of 50.0% of a state’s delegation turns over across sessions vs. 16.4% within a session. These findings suggest that the EP experiences a high level of turnover. On average, half of the members of a country’s EP delegation are new entrants after an election, and nearly one-sixth of MEPs exit the chamber in the interval between Table 3.1 MEP Turnover and Voter Turnout: Descriptive Statistics

EP turnout Across-session turnover Within-session turnover Across-session gender progress Within-session gender progress Compulsory Weekend Institution host EU net recipient EU identity EP election 2019 Electoral cycle General election turnout

Number

Mean

Std. Dev

Minimum

Maximum

147 110 147 110 147 147 147 147 137 125 147 147 147

51.0 50.0 16.4 3.3 0.8 0.2 0.8 0.2 0.5 1.7 0.2 798.2 72.7

19.3 14.7 15.7 11.8 7.8 0.4 0.4 0.4 0.5 0.2 0.4 447.0 13.3

13.1 16.7 0.0 −33.3 −30.0 0.0 0.0 0.0 0.0 1.4 0.0 37.0 37.8

92.1 100.0 75.0 66.7 50.0 1.0 1.0 1.0 1.0 2.3 1.0 1834.0 94.6

3

TURNOVER, TURNOUT, AND INPUT LEGITIMACY IN THE EU

71

elections. Both of the gender-sensitive measures have quite wide value ranges. Across-session gender progress has a minimum value of –33.3 and a maximum value of 66.7, and Within-session gender progress ranges from –30.0 to 50.0. Both variables’ mean values (3.3 and 0.8, respectively) are relatively low, however. On average, elections lead to modest gains in women’s representation within member-state delegations, and withinsession replacements do not (again, on average) effect significant net gains for women. Many of the measures included in Table 3.1 are dichotomous. The low means on Compulsory, Institution host, and 2019 EP election suggests that few member-states meet these conditions. In other words, elections are not compulsory in most member-states, few member-states host an EU institution, and most of the elections in the dataset did not take place in 2019. The high mean on the weekend measure shows that many EU countries hold EP elections on the weekend. The mean on EU net recipient (0.5) indicates a fairly even split between net contributor and net beneficiary member-states. Table 3.2 presents the results of our multivariate models. The first model in the table focuses on the relationship posited in H1. The results do not support the hypothesis: turnover in one EP election is not associated with turnout in the subsequent EP election. Given the low levels of knowledge about the EU amongst the public (Karp et al. 2003; Clark 2014; Schmitt et al. 2015), most EP voters may simply not be aware of levels of turnover in the EP. We ran the model with fewer measures to ensure that the standard errors were not affected by one or more of the controls. The turnover measure does not achieve significance in different variations of the model. The level of turnover across EP elections does not appear to prompt more individuals to cast a ballot. Elsewhere in model (1), the compulsory measure is significant, suggesting that turnout is higher in those few countries that require the public to vote. The electoral cycle measure is also positively associated with turnout, meaning that EP election turnout is higher the further one gets from a previous national general election. In addition, higher turnout in the last national election is correlated with higher turnout in an EP election, likely capturing a greater tendency to vote in some countries. The most recent EP election (May 2019) is also correlated with a higher turnout rate, which is not surprising given that EU-wide turnout

72

N. CLARK AND J. A. SCHERPEREEL

Table 3.2 Predictors of EP Election Turnout (1) Across-session turnover

Weekend Institution host EU net recipient EU identity EP election 2019 Electoral cycle General election turnout Constant N R squared

(3)

(4)

0.01 (0.09)

Within-session turnover Across-session gender progress Within-session gender progress Compulsory

(2)

0.05 (0.06) −0.01 (0.08)

16.22*** (3.84) 2.23 (5.80) 8.53 (5.55) 2.61 (3.66) 7.85 (10.36) 4.27** (1.75) 0.004*** (0.001) 0.72*** (0.14) −28.45 (18.74) 109 0.72

15.32*** (4.09) 0.90 (5.79) 9.75 (5.90) 3.66 (3.58) 9.23 (10.20) 4.89*** (1.62) 0.004*** (0.001) 0.74*** (0.14) −32.34** (15.96) 125 0.72

16.03*** (3.80) 2.67 (6.13) 8.26 (6.25) 2.27 (3.73) 5.96 (10.47) 4.96*** (1.65) 0.004*** (0.001) 0.76*** (0.13) −27.80* (15.83) 101 0.72

0.21** (0.08) 15.93*** (3.89) 1.08 (5.65) 9.41 (5.86) 3.41 (3.46) 9.18 (10.17) 5.65*** (1.60) 0.003** (0.001) 0.74*** (0.14) −31.21** (15.41) 125 0.73

Notes * = p ≤ 0.10, ** = p ≤ 0.05, *** = p ≤ 0.01. The observations are EP elections in each individual EU member-state. The dependent variable is national-level voter turnout at on-cycle EP elections

levels increased in that election for the first time in the history of EP elections. Model (1)’s relatively high R squared value (0.72) suggests that this model accounts for much of the variation in turnout across the EU. We investigate H2 by replacing Model (1)’s across-session turnover measure with the measure of within-session turnover. The results of that model, which appear in column (2) of Table 3.2, largely parallel the results of model (1). The measure of within-session turnover is also not associated with higher levels of voter turnout. This finding suggests that any turnover due to replacement of departing MEPs does not motivate

3

TURNOVER, TURNOUT, AND INPUT LEGITIMACY IN THE EU

73

greater turnout. As noted above, within-session turnover is a particularly opaque process, and the method by which MEPs are replaced varies across member-states. Here again, we suspect that most EU citizens are not following EP politics closely enough to know when an MEP from their national delegation has been replaced. The remaining variables in model (2) have the same effects as in model (1). Next, we consider the possibilities suggested by feminist critics that gains in the descriptive representation of women will instill the EU with more input legitimacy and prompt higher levels of voter turnout. Column (3) presents the results of a model that includes Across-session gender progress, which measures the extent to which a member-state’s EP delegation became more or less gender-balanced as a result of the most recent EP election. Here, too, the predicted relationship is not observed: there is no significant association, across the broad 1984–2019 temporal span, between across-session gender progress and voter turnout. Elsewhere in the model, results largely mirror those observed in models (1) and (2), and model (3)’s R squared value (0.72) remains similar to the R squared values in models (1) and (2). Finally, the model presented in column (4) of Table 3.2 examines the influence of within-session gender progress on national-level turnout in EP elections. Recall, in this context, that Within-session gender progress summarizes the extent to which a member-state’s EP delegation becomes more (positive value) or less (negative value) gender-balanced over the course of the EP session that precedes an election. According to the logic sketched above, movement toward gender balance should motivate more people to perceive the EP as a legitimate representation of the people and, thus, to participate in EP elections. The results of model (4) support the hypothesis: each percentage point increase in the difference between women’s seat shares at a session’s final plenary and a session’s first plenary is associated with a 0.21 percentage point increase in voter turnout at the subsequent election. Beyond this finding, which supports H4, model (4)’s results are effectively identical to those of model (3). Like Daniel and Metzger (Chapter 2 of this volume), we recognize that the effects that appear in pooled models may cover up significant era-specific effects. We are particularly alert, in this regard, to potential era-specific effects of our measures of gender progress. While women’s descriptive representation has concerned many citizens for decades, the issue’s profile has risen significantly in the wake of the UN’s Conferences on Women (Krook and True 2012). For this reason, and given that the

74

N. CLARK AND J. A. SCHERPEREEL

only new significant finding to come out of the models presented in Table 3.2 related to Within-session gender progress, we have also constructed models based on samples split by era. Table 3.3 presents the results of these models. In Table 3.3, we examine the effects of our two indicators of gender progress in “earlier elections”—those in 1984, 1989, 1994, and 1999—and “later elections”—those in 2004, 2009, 2014, and 2019. As in the larger pooled models, the results provide some sense that gendered turnover patterns may affect turnout. While across-session gender progress does not register as significant in the full pooled model (model (3) of Table 3.2), it has a small effect on turnout in the subset of more recent elections. There (e.g., in model (3) of Table 3.3), an increase of one percentage point in across-session gender progress is associated Table 3.3 Gender Progress and EP Election Turnout Over Time Earlier elections (1) Across-session gender progress Within-session gender progress Compulsory Weekend Institution host EU net recipient EU identity Electoral cycle General election turnout Constant N R squared

Later elections (2)

−0.02 (0.10)

12.90** (5.45) 2.48 (6.98) 11.40 (7.05) 0.26 (5.04) 1.76 (15.48) 0.002 (0.002) 0.69*** (0.20) −11.67 (20.39) 67 0.64

(3)

(4)

0.28* (0.14) 0.12 (0.08) 12.00** (5.57) 1.56 (6.70) 12.61** (6.29) 0.99 (4.63) 2.16 (12.66) 0.001 (0.002) 0.72*** (0.20) −13.28 (17.70) 83 0.66

22.30*** (4.62) 3.65 (5.25) 1.72 (5.20) 7.44* (3.08) 20.46* (10.79) 0.007** (0.002) 0.61** (0.21) −45.90** (18.58) 34 0.86

0.32 (0.22) 22.07*** (3.31) 1.44 (4.18) 2.69 (4.62) 8.51*** (2.64) 23.41** (10.06) 0.006** (0.002) 0.59*** (0.18) −47.65** (19.80) 42 0.85

Notes * = p ≤ 0.10, ** = p ≤ 0.05, *** = p ≤ 0.01. The observations are EP elections in each individual EU member-state. The dependent variable is national-level voter turnout at on-cycle EP elections. Earlier elections are elections in 1984–1999. Later elections are elections in 2004–2019.

3

TURNOVER, TURNOUT, AND INPUT LEGITIMACY IN THE EU

75

with a 0.28 percentage point rise in voter turnout. On the other hand, Within-session gender progress, whose coefficient is positive and significant in model (4) of Table 3.2, does not have a statistically significant association within turnout in either of the time subsets investigated in Table 3.3. Overall, the models presented in Tables 3.2 and 3.3 suggest that gendered patterns of turnover are associated with EP voter turnout. The precise nature of the relationship and the ways that gender progress matters across time, however, require additional scrutiny in future studies.

Conclusions and Future Research Trajectories Accumulated research suggests that electoral institutions, structural positions, domestic political concerns, and citizens’ individual-level characteristics all play important roles in affecting voter turnout and, thereby, input legitimacy. In this chapter, we have suggested that the personnel composition and dynamics of legislative delegations also affect input legitimacy. Our empirical analyses provide no support for hypotheses inspired by the populist critique—across-session turnover and within-session turnover are not associated with levels of turnout to EP elections. Our tests of hypotheses inspired by the feminist critique receive some support. In the large pooled model, within-session gender progress is associated with higher voter turnout. And in the models split between earlier and later elections, across-session gender progress is associated with higher turnout. Overall, then, our analysis does not confirm the expectations of the populist critique and offers some support for the feminist critique. In terms of the populist critique, we have found no evidence for the notion that European voters derive a sense of collective efficacy from “kicking the bums out”: there is no relationship between the quantity of personnel turnover that a member-state’s delegation of MEPs experiences at one election and the extent to which voters in that country turn out in the next election. The existence of such a relationship presumes a relatively strong awareness of the EP’s personnel dimension; future efforts might examine the extent to which greater knowledge of the EU and/or the EP condition citizens’ propensity to vote. It is possible that the critique may hold for more knowledgeable voters. Outside of the EP context, future researchers might examine the extent to which personnel composition and dynamics affect turnout in other legislative

76

N. CLARK AND J. A. SCHERPEREEL

elections. Citizens may be more attentive—and more electorally responsive—to personnel issues in the national (and/or sub-national) legislatures with which they are more familiar. In addition, this chapter’s analysis has dealt strictly with questions involving the quantity of turnover. Future research might inquire into whether the quality of turnover might affect voting behavior. As chapter one stresses, populist critics sometimes stress issues of quantity (e.g., “expel the incumbents”) and other times stress issues of quality (e.g., “even if the mainstream parties nominate new people, those ‘new’ people will be no different than the old ones; they will carry the same baggage—professional, educational, political, otherwise—to office”). Future research might investigate whether, say, the replacement of an old European People’s Party (EPP) MEP via the election of a new EPP MEP might have a different effect than replacing, say, an EPP MEP with a newly elected MEP who represents a populist party and/or has a dramatically different set of pre-office experiences. We have also found no evidence that within-session volatility affects input legitimacy. Above, we presented two effectively opposite logics that might drive such an effect—one based on approbation, the other on disgust. Future micro-level research (whether quantitative or qualitative in approach) might also explore the extent to which approbation and/or disgust are relevant factors in individual voters’ decisions about whether/how to vote. The support for H3 (in the time-split model) and H4 (in the broader pooled model) suggests that citizens respond positively—they turn out in greater numbers—when the EP becomes more gender-balanced. These findings, together with those of Aldrich and Perez (Chapter 5 of this volume), suggest several potential benefits to greater gender balance in the EU. This chapter suggests that such representation improves input legitimacy by prompting higher turnout in EP elections. And, as Aldrich and Perez demonstrate, higher levels of women’s representation also improve throughput legitimacy by increasing the consensus-building capacity of the Council. In other words, greater gender balance generates a greater public response to the EU and improves the ability of EU institutions to govern on the issues of the day. Moreover, the salience of gender balance for the voting public suggests potential tactics for successfully competing in EP elections. Politicians, EP and Commission officials, political parties, and activists have pursued various strategies designed to increase voter turnout and promote citizen

3

TURNOVER, TURNOUT, AND INPUT LEGITIMACY IN THE EU

77

buy-in. This chapter’s analysis suggests that gender balance considerations should factor into these strategies. Future micro-oriented research might examine the extent to which citizens’ gender and other individual attributes affect how gendered turnover patterns affect the likelihood to vote. Such research, if conducted in the EP context, could complement broadly similar research completed in quite different geographic and institutional settings (e.g., Bhalotra et al. 2013; Broockman 2014; Wolbrecht and Campbell 2017). Future research might also explore the possible effects of critical actors in the EP. When critical actors use their EP positions to promote gender equality, are more feminist candidates likely to emerge? Are more voters likely to turn out when MEPs advocate for gender equality? These findings have implications for the bodies of research on EU voting behavior and on the effects of turnover. In regard to the former, the literature on turnout and voting behavior in EP elections would benefit from greater regard for macro-level predictors. Early analyses of EP voting behavior focused largely on country-level effects, such as the timing of EP elections within the national electoral cycle (Reif and Schmitt 1980). The behavioral turn within this literature has focused more on individual-level attributes. This chapter indicates that unique aspects of national electoral competition, such as changes in the composition and dynamics of MEPs, might also shape the choices of the electorate. Of note, we are not arguing against the validity or importance of individual-level analyses, but rather that such analyses might be strengthened through the inclusion of additional macro-level explanations. The literature on turnover offers mixed cues as to the potential effects of high turnover. On one hand, high turnover corresponds with less aggregate experience in government on the part of elected representatives, which could diminish the public’s confidence in governing institutions. On the other hand, one should expect and even hope for some degree of turnover in a democratic system; such change is evidence that the system is working. The question, then, is whether turnover will increase or decrease the public’s confidence in their governing institutions. This chapter does not directly examine the public’s confidence in their governing institutions, but one might surmise—as done by the EU’s populist critics—that higher turnout reflects certain perceptions of those institutions. Our results offer little indication as to whether increasing turnout is motivated by higher or lower public confidence in the EP.

78

N. CLARK AND J. A. SCHERPEREEL

Having established that these macro-level explanations might help us to better understand national turnout, additional individual-level analyses might permit a better understanding of the motivations shaping voting behavior. First, future efforts should model the interactive effects of knowledge about the EU and different types of turnover on EP voting behavior. Higher levels of turnover may have a stronger effect on more knowledgeable and attentive individuals, particularly types of turnover that require a greater degree of attention from the public. Second, individual-level analyses may provide a better understanding of the motivations shaping public responses to higher turnover. The Eurobarometer, for instance, includes questions about the public’s level of confidence or trust in different EU institutions. That survey data might be leveraged to understand why exactly higher turnover leads to greater turnout.

Notes 1. A number studies of individual-level determinants of EP turnout show that some voters do use EP elections as an opportunity to address concerns about the EU (and not just the national government). Defections from a governing party to an opposition party in EP elections are more likely to occur, for example, if government supporters perceive of themselves as more Euroskeptic than the government (Hobolt et al. 2009) or disapprove of the government’s performance on EU issues (Clark and Rohrschneider 2009). In addition, the broader information environment influences the extent to which voters consider the EU when making vote choices. EP voters are more motivated by EU concerns when and where EP political campaigns, the media, and the national political environment provide more information about EU affairs (de Vries et al. 2011; Hobolt et al. 2009; Hobolt and Spoon 2012; Weber 2011). These insights complicate the most reductive variants of second-order theory. While they show ways that the EU can matter in the EP turnout equation, though, their focus is still slightly different than the one we pursue here. We focus, again, on issues related to MEP composition and dynamics, on the relatively narrow issue of the EP’s personnel dimension. 2. Because Across-session turnover and Across-session gender balance require comparison of delegation compositions across two previous legislative sessions, it is not possible to calculate values for these variables for (a) observations in 1984 or (b) the first observations of acceding member-states (e.g., Portugal 1989, Spain 1989).

3

TURNOVER, TURNOUT, AND INPUT LEGITIMACY IN THE EU

79

References Begg, Iain, and Nigel Grimwade. 1998. Paying for Europe. Sheffield: Sheffield Academic Press. Bhalotra, Sonia, Irma Clots-Figueras, and Lakshmi Iyer. 2013. Path-Breakers: How Does Women’s Political Participation Respond to Electoral Success? Bonn: IZA Discussion Paper No. 7771. Available at https://papers.ssrn.com/ sol3/papers.cfm?abstract_id=2363272. Accessed June 12, 2020. Bhatti, Yosef, and Kasper M. Hansen. 2012. The Effect of Generation and Age on Turnout to the European Parliament: How Turnout Will Continue to Decline in the Future. Electoral Studies 31 (2): 262–272. Broockman, David E. 2014. Do Female Politicians Empower Women to Vote or Run for Office? A Regression Discontinuity Approach. Electoral Studies 34: 190–204. Clark, Nicholas. 2014. Explaining Low Turnout in European Elections: The Role of Issue Salience and Institutional Perceptions in Elections to the European Parliament. Journal of European Integration 36 (4): 339–356. Clark, Nick, and Robert Rohrschneider. 2009. Second-Order Elections Versus First-Order Thinking: How Voters Perceive the Representation Process in a Multi-Layered System of Governance. Journal of European Integration 31 (5): 645–664. Daniel, William T. 2015. Career Behaviour and the European Parliament. Oxford: Oxford University Press. Daniel, William T., and Shawna K. Metzger. 2018. Within or Between Jobs? Determinants of Membership Volatility in the European Parliament, 1979– 2014. The Journal of Legislative Studies 24 (1): 90–108. de Vries, Catherine E., Wouter van der Brug, Marcel H. van Egmond, and Cees van der Eijk. 2011. Individual and Contextual Variation in EU Issue Voting: The Role of Political Information. Electoral Studies 30 (1): 16–28. European Parliament. 2020. Turnout by Country (%). Available at https:// europarl.europa.eu/election-results-2019/en/turnout/. Accessed June 12, 2020. European Parliament Research Service. 2014. 2014 European Elections: National Rules. Available at https://www.europarl.europa.eu/EPRS/InfoGr aphic-2014-European-elections-national-rules-FINAL.pdf. Accessed June 12, 2020. European Parliament Research Service. 2019. 2019 European Elections: National Rules. Available at https://www.europarl.europa.eu/spain/barcel ona/resource/static/files/import/normes_nacionals/eprs-national-rules-en. pdf. Accessed June 12, 2020. Fiorino, Nadia, Nicola Pontarollo, and Roberto Ricciuti. 2019. Supranational, National and Local Dimensions of Voter Turnout in European Parliament Elections. JCMS: Journal of Common Market Studies 57 (4): 877–893.

80

N. CLARK AND J. A. SCHERPEREEL

Flickinger, Richard S., and Donley T. Studlar. 2007. One Europe, Many Electorates? Comparative Political Studies 40 (4): 383–404. Franklin, Mark N. 2001. How Structural Factors Cause Turnout Variations at European Parliament Elections. European Union Politics 2 (3): 309–328. Franklin, Mark N., and Sara B. Hobolt. 2011. The Legacy of Lethargy: How Elections to the European Parliament Depress Turnout. Electoral Studies 30 (1): 67–76. Franklin, Mark N., and Sara B. Hobolt. 2015. European Elections and the European Voter. In European Union: Power and Policy-Making, 4th ed., ed. Jeremy Richardson and Sonia Mazey, 399–418. New York: Routledge. Gray, Mark, and Miki Caul. 2000. Declining Voter Turnout in Advanced Industrial Democracies, 1950 to 1997. Comparative Political Studies 33 (9): 1091–1122. Hobolt, Sara B., and Jae-Jae Spoon. 2012. Motivating the European Voter: Parties, Issues and Campaigns in European Parliament Elections. European Journal of Political Research 51 (6): 701–727. Hobolt, Sara B., Jae-Jae Spoon, and James Tilley. 2009. A Vote Against Europe? Explaining Defection at the 1999 and 2004 European Parliament Elections. British Journal of Political Science 39 (1): 93–115. International IDEA.. 2020. Compulsory Voting. Available at: https://www.idea. int/data-tools/data/voter-turnout/compulsory-voting. Accessed June 12, 2020. Jackman, Robert W. 1987. Political Institutions and Voter Turnout in the Industrial Democracies. American Political Science Review 81 (2): 405–423. Karp, Jeffrey A., Susan A. Banducci, and Shaun Bowler. 2003. To Know It is to Love It? Satisfaction with Democracy in the European Union. Comparative Political Studies 36 (3): 271–292. Kreppel, Amie. 2002. The European Parliament and Supranational Party System: A Study in Institutional Development. Cambridge University Press. Krook, Mona Lena, and Jacqui True. 2012. Rethinking the Life Cycles of International Norms: The United Nations and the Global Promotion of Gender Equality. European Journal of International Relations 18 (1): 103–127. Lee, Francis L.F. 2010. The Perceptual Bases of Collective Efficacy and Protest Participation: The Case of Pro-Democracy Protests in Hong Kong. International Journal of Public Opinion Research 22 (3): 392–411. Lubell, Mark, Sammy Zahran, and Arnold Vedlitz. 2007. Collective Action and Citizen Responses to Global Warming. Political Behavior 29 (3): 391–413. Mattila, Mikko. 2003. Why Bother? Determinants of Turnout in the European Elections. Electoral Studies 22 (3): 449–468. Moravscik, Andrew. In Defence of the ‘Democratic Deficit’: Reassessing Legitimacy in the European Union. Journal of Common Market Studies 40 (4): 603–624.

3

TURNOVER, TURNOUT, AND INPUT LEGITIMACY IN THE EU

81

Mudde, Cas. 2007. Populist Radical Right Parties in Europe. Cambridge: Cambridge University Press. Powell, G. Bingham. 1986. American Voter Turnout in Comparative Perspective. American Political Science Review 80 (1): 17–43. Reif, Karlheinz, and Hermann Schmitt. 1980. Nine Second-Order National Elections: A Conceptual Framework for the Analysis of European Election Results. European Journal of Political Research 8 (1): 3–44. Seligson, Mitchell A. 1980. Trust, Efficacy and Modes of Political Participation: A Study of Costa Rican Peasants. British Journal of Political Science 10 (1): 75–98. Schmitt, Hermann, Sara B. Hobolt, and Sebastian A. Popa. 2015. Does Personalization Increase Turnout? Spitzenkandidaten in the 2014 European Parliament Elections. European Union Politics 16 (3): 347–368. Studlar, Donley, Richard S. Flickinger, and Stephen Bennett. 2003. Turnout in European Parliament Elections: Towards a European-Centred Model. British Elections & Parties Review 13 (1): 195–209. van der Eijk, Cees, Mark N. Franklin, and Michael Marsh. 1996. What Voters Teach Us about Europe-wide Elections: What Europe-wide Elections Teach Us about Voters. Electoral Studies 15 (2): 149–166. Weber, Till. 2011. Exit, Voice, and Cyclicality: A Micrologic of Midterm Effects in European Parliament Elections. American Journal of Political Science 55 (4): 907–922. Wolbrecht, Christina, and David E. Campbell. 2017. Role Models Revisited: Youth, Novelty, and the Impact of Female Candidates. Politics, Groups, and Identities 5 (3): 418–434.

PART III

Turnover and Throughput Legitimacy

CHAPTER 4

Personnel Turnover and Legislative Efficiency in the EU Lauren K. Perez and John A. Scherpereel

While chapters two and three have investigated the relationship between turnover and input legitimacy, chapters four, five, and six focus on the connections between turnover and throughput legitimacy. To what extent, the three chapters’ authors ask, do the comings and goings of

Electronic supplementary material The online version of this chapter (https://doi.org/10.1007/978-3-030-60052-5_4) contains supplementary material, which is available to authorized users. L. K. Perez (B) Department of Political Science, Francis Marion University, Florence, SC, USA e-mail: [email protected] J. A. Scherpereel Department of Political Science, James Madison University, Harrisonburg, VA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. A. Scherpereel (eds.), Personnel Turnover and the Legitimacy of the EU, Palgrave Studies in European Union Politics, https://doi.org/10.1007/978-3-030-60052-5_4

85

86

L. K. PEREZ AND J. A. SCHERPEREEL

politicians in the Commission, the Council of the EU, and the European Parliament affect the operations of the EU’s institutional machinery? Does turnover affect the quality and efficiency of the EU’s political system? Does it help, hurt, or otherwise affect the system’s capacity to tackle public problems and improve citizens’ lives? Academics, populist critics, and feminist critics have all suggested that it is important for the individuals who exercise power in a democracy to turn over with some regularity. Without turnover, power holders are likely to be aloof, to develop attitudes of impunity, and to dismiss pressing issues of concern. New entrants to office may have tighter connections with citizens. They may have a particularly strong sense of the public mood, and they may bring new experiences, fresh approaches, and energy to office. Yes, turnover has serious risks (Atkinson and Docherty 1992; Francis and Baker 1986; Hyneman 1938; Rosenthal 1992). But turnover can also promote popular representation and help to legitimate systems of social control. These virtues notwithstanding, this chapter suggests that personnel turnover can threaten another democratic virtue—efficient decisionmaking. Efficiency, as discussed in the volume’s first chapter, is a critical component of throughput legitimacy. The failure to address public problems expeditiously can undermine systemic legitimacy. Inefficient decision processes can lead to a situation where today’s policy outputs address yesterday’s public’s concerns. This risk of temporal misfit is doubly dangerous: a problem that requires public intervention persists, deepens, or morphs as the decision process bogs down, and the inappropriateness of the ultimate institutional response weakens public respect and support for democratic institutions and processes. Citizens in inefficient systems may come to question the usefulness of public officeholders. Such processes increase the likelihood that non-democratic actors will ignore or weaken publicly unpopular institutions. To what extent might turnover in EU institutions affect efficiency and feed into such processes? This chapter attempts to throw light on this question. It feeds into a stream of research related to the duration of the EU legislative process (Chalmers 2014; Golub 1999, 2007; Klüver and Sagarzazu 2013; Hertz and Leuffen 2011; König 2008; Rasmussen and Toshkov 2013; Schulz and König 2000; Sloot and Verschuren 1990). Scholars of EU legislative efficiency have used different data and employed a range of analytical methods, but they have produced quite a robust set of findings. They have shown, for example, that qualified

4

PERSONNEL TURNOVER AND LEGISLATIVE EFFICIENCY IN THE EU

87

majority voting in the Council—an institutional step explicitly designed to decrease legislative lordeur and increase decision-making speed (Sloot and Verschuren 1990)—does, in fact, catalyze quicker decision making. They have also shown that certain reforms designed to address the EU’s democratic deficit—the empowerment of the European Parliament, the use of directives versus blunter legal instruments, and the implementation of public consultations—have increased the amount of time it takes to reach decisions. While the latter reforms have sought to increase the quality of democracy, they have increased the risk of slow policy responses. Personnel turnover is not an institutional reform like EP empowerment. It is, rather, a structural feature of (EU) politics. We expect turnover, though, to have a similar effect as EP empowerment and other efforts to address the democratic deficit. In the first portion of this chapter, we develop a rationale for this expectation. We suggest that new entrants to all three of the EU’s main “day-to-day” institutions— the Commission, the Council, and the EP—face steep learning curves. We suggest that these learning curves increase uncertainty and promote a systemic ritardando. We stress that new entrants’ learning curves are particularly steep in the EU context, due to the EU system’s complexity and frequently shifting institutional patterns. We portray the relationship between turnover and decision-making speed as a governance tradeoff (Dahl 1994). We begin the empirical section of the chapter by discussing our data and method. We present descriptive statistics on turnover in the Commission, Council, and EP over time and test the associations between turnover and legislative duration using a Cox survival model. Focusing on proposals introduced between 2000 and 2018, we find that turnover in the Commission and the Council are associated with longer legislative processes, while turnover in the Parliament may actually speed up the legislative process. After presenting our results, we discuss the findings’ implications and present possible avenues for further research.

Personnel Turnover: A Brake on the Efficiency of the EU’s Legislative Process? Recent work on personnel trends in the EU institutions suggests that turnover has important, and occasionally unexpected, effects. Turnover among national ministers, for example, weakens the Council vis-à-vis the EP and the Commission (Scherpereel and Perez 2015) and weakens

88

L. K. PEREZ AND J. A. SCHERPEREEL

ministers vis-à-vis the lower-level executive actors that staff the Council machinery (Perez and Scherpereel 2017). Turnover among at least some of those lower-level executive actors, however, may actually increase member-states’ bargaining success and strengthen the Council vis-à-vis the other institutions (Scherpereel et al. 2018). The chapter pursues an intuition that turnover may also affect the duration of the legislative process and undermine the EU’s throughput legitimacy. This intuition derives from the fact that the EU is a complex and dynamic organism that governs disparate policy sectors. No two actors that rotate into the EU system will be similarly (un)prepared: one new actor may be more familiar with her new position’s context, requirements, and opportunities than some other new actor. We assume, however, that the adjustment will be significant for all new incumbents. The EU’s institutional arrangements are notoriously intricate. Successful negotiation of the landscape requires familiarity with rules and processes that actors who have often established their political careers and instincts at sub-national and national levels do not innately possess. The EU systemic framework is also unusually dynamic, at least when compared to most national contexts. The EU’s treaties have been overhauled five times over the last three-anda-half decades, and significant institutional improvisations—trilogues, for example, or the open method of coordination—have occurred “beneath” this shifting institutional framework. The steepness of new incumbents’ learning curve also derives from variation in the universe of governance regimes that comprise the EU’s policy space. Even politicians who are deeply familiar with one EU policy domain (e.g., the internal market) may lack the tools to thrive, immediately, in other, even broadly similar domains (e.g., trade, competition). Of course, no two policy settings are identical in any contemporary liberal democracy. In the EU, however, decision-making rules, alliance configurations, and patterns of state-society relations are particularly variable from one policy area to the next and from one year to another. In addition, the institutional decisions that help to structure policy domains—decisions, for example, about the policy responsibilities of Commission vice-presidents or the number and remit of Commission DGs and EP committees—change quite frequently. As a result, an actor who has become comfortable in one policy sector has a substantial amount of knowledge to gain when she transitions to a new policy sector. At the individual level, we suggest that these characteristics decelerate the pace of activity. Actors rotating into positions with steep learning

4

PERSONNEL TURNOVER AND LEGISLATIVE EFFICIENCY IN THE EU

89

curves are reluctant to commit to files under consideration for fear that doing so will have reputational costs, monetary costs, and/or unintended consequences. These costs outweigh any benefits to be gained by prematurely supporting a live dossier. At the systemic level, we suggest that episodes of turnover increase uncertainty and that this uncertainty produces a ritardando effect. To demonstrate this point, consider the process that might unfold in the Council after a right-leaning minister from a pivotal member-state replaces a left-leaning minister from the same member-state. Other actors in the Commission, Council, and EP may have clues about the likely action orientations of the new minister. They may have encountered, for example, relevant election manifestos, public speeches/interviews, and/or government programs. But these clues are likely to be imperfect. Even in cases of “intra-partisan turnover”— turnover, that is, where one right-leaning minister is replaced by a minister from the same party—systemic actors will not know the action orientations of new incumbents until they take on their new roles. Once again, the effect is to prolong decision making: it takes time for new actors and their deputies to establish positions and for interlocutors to gain certainty about new incumbents’ preferences and attributes. We assume that adjustment costs will manifest and systemic uncertainty will increase regardless of which institution—Commission, Council, or EP—experiences turnover. Consistent with the argument just advanced, we assume that lateral moves in the Commission are equivalent to initial entries. Thus, for example, we assume that a commissioner who has overseen agriculture and moves to trade faces similar hurdles and induces similar uncertainties when compared to a commissioner who enters the Commission on the heels of a stint in national government or the private sector. These considerations support the following hypothesis: H1: As levels of turnover in the Commission, Council, and EP increase, the time that it takes to reach EU decisions will increase.

Data and Methods Our main independent variables are the annualized turnover rates for each of the three institutions involved in everyday EU decision-making: the Commission, the Council, and the EP. To calculate the quantity of Commission turnover, we begin by assembling an original dataset of all commissioners, including portfolio names and dates of entry and exit.

90

L. K. PEREZ AND J. A. SCHERPEREEL

This dataset covers the entire lifespan of the Commission, from January 1958 to July 2019. Because our dependent variable data (discussed below) span from 2000 through 2018, our analysis involves scrutiny of commissioners that served from January 1999 through December 2018.1 For all months within that time span, we determine whether each commissioner was leading the same portfolio one year before the date of observation. For example, for our January 2000 observation, we determine whether a commissioner in office on January 1, 2000 had or had not been serving in the same role on January 1, 1999. In a small number of cases, this determination requires a judgment as to whether the portfolio managed at t + 1 (e.g., January 1, 2000) was substantially equivalent to the portfolio managed at t (e.g., January 1, 1999). In such cases, if the name of the portfolio at t + 1 contains the same brief as the portfolio at t (e.g., portfoliot = Competition; portfoliot +1 = Competition and regional policy), we code the commissioner as retained. If not, we code the commissioner as having turned over. After making these individual-level determinations, we calculate Commission turnover, which is the percentage of commissioners serving in a role at t + 1 that had not been serving in that role at t. For Council turnover, we follow a similar process, analyzing national cabinet ministers who served between January 1999 and December 2018. Our raw data come from Lars Sonntag’s Politica database (Sonntag 2020). Our analysis of the Council focuses on the “Council grand plenary”—the set of cabinet members whose portfolios involve policy sectors covered by the Council.2 As is the case for Commission turnover, Council turnover includes one observation each month. That observation measures the percentage of ministers in the grand plenary at t + 1 (e.g., January 1, 2000) that were not in the grand plenary at t (e.g., January 1, 1999). We construct EP turnover in a similar way. For this measure, we update Daniel and Metzger’s (2018) data to include information on all MEPs serving between July 1979 and July 2019. Our outcome of interest is the duration of the legislative process, which we operationalize as the number of months that elapse from the first to the final step in the codecision procedure. Codecision, which originated with the Treaty on European Union, requires the Commission, the Council, and the EP to play major roles: the Commission drafts legislative texts, and the EP and the Council must agree to identical versions of a draft in order for the proposal to become law. The procedure’s usage has expanded since the early 1990s, and the Lisbon Treaty

4

PERSONNEL TURNOVER AND LEGISLATIVE EFFICIENCY IN THE EU

91

officially designated codecision the EU’s ordinary legislative procedure. When compared to other EU legislative procedures, codecision involves more veto players (Tsebelis and Yatanagas 2002) and requires intense inter-institutional negotiations. While numerous studies have shown that codecision files take longer to conclude than files subject to alternative legislative procedures (Golub 1999, 2007; Golub and Steunenberg 2007; Klüver and Sagarzazu 2013; König 2007; Schulz and König 2000), there is still significant variation in the duration of specific codecision dossiers. In the sample of completed codecision files discussed below, for example, the average length of the legislative process was 26.63 months. The minimum value was two months, the maximum was 182 months, and the standard deviation was 21.92 months. In addition, a significant number of files introduced during our period of observation remained “legally alive” without having reached conclusion. The longest duration among this set of right-censored observations was 203 months. We begin by extracting files subject to codecision from the EU’s EurLex database. In all, we monitor the progress of all codecision files introduced between 2000 and 2018 (inclusive). Our dataset includes information about whether each legislative file reached certain stages in the legislative process, as well as the first and last dates of those events. In most cases, the first step is the adoption of the proposal by the Commission, and the final step is signature by the Council and the EP. We also code withdrawal by the Commission as a decision that concludes the legislative process (for an analysis of the politics of proposal withdrawal under codecision, see Reh et al. 2020). To account for factors that previous studies have found to affect the duration of the legislative process, we include a number of controls in our models. First, we code for the type of legal act under consideration. We expect the decision-making process to take longer for a Directive than for either of the other two other types of legal instrument we study— Decisions (which are binding in their entirety but are generally directed to a subset of member-state/s) and Regulations (which are binding in their entirety and have direct effect in all member-states). The idea that directives will take longer to conclude than regulations and (especially) decisions was originally discussed by Schulz and König (2000) and subsequently confirmed by others (Klüver and Sagarzazu 2013; Rasmussen and Toshkov 2011). Theoretically, the difference in duration may derive from the fact that directives, because they require national actors to adjust national laws, incline states to be less flexible when negotiating

92

L. K. PEREZ AND J. A. SCHERPEREEL

them, “either because they lack domestic legislative majorities or because of opposition from domestic interest groups” (Schulz and König 2000: 658). In addition to legal instrument, we control for the number of memberstates in the EU at the moment of observation (# of member-states ). There is no guarantee that the entry of additional members (via enlargement) will increase either preference polarization (e.g., how much ideological distance separates the extreme actors on any preference scale) or preference heterogeneity (e.g., how evenly preferences are distributed across such a scale). The most careful treatment of EU enlargement’s effects on decision-making speed, however (Hertz and Leuffen 2011), finds that enlargement lengthens decision-making duration. Häge (2012: 23) provides one theoretical argument for why this might be the case: as the number of member-states expands, “a larger number of states need to come together in order to form a blocking minority.” We also control for the legislative backlog that the EU faces in a given month. Sloot and Verschuren (1990) were the first to consider this idea, which they called “policy pressure.” They suggested that the Council is aware of how much outstanding legislation there is and faces pressure from the Commission, the EP, and even the Court of Justice to clear out its backlog; as time has developed and the EP has gained power, the Parliament itself may now also feel pressure to perform. Most national legislatures have rules that limit the shelf life of submitted draft acts; among other things, such “discontinuity” (Van Schagen 1997) can limit legislative backlog. In the EU context, however, there is no such principle, even if recent commissions have tended to engage in “cleaning up” exercises upon taking office (Reh et al. 2020). Although their dependent variable is legislative volume (not legislative duration), Junge et al.’s (2015) suggestion that the Commission expands its activities in strategic response to gridlock in the other institutions is potentially relevant in this context: actors (at least those in the Council and the EP) may sense that failure to conclude dossiers may wind up empowering other, potentially rival institutions (e.g., the Commission). Regardless of the precise mechanism, numerous studies of duration (Golub 1999, 2007; Golub and Steunenberg 2007; Klüver and Sagarzazu 2013) find that larger backlogs in the EU decision pipeline decrease decision-making duration. To operationalize Backlog, we use a count of the number of open files (of all kinds) in a given month, which ranges in our data from a minimum of 285 to a maximum of 961. We combine our data with Häge’s (2011) EUPOL

4

PERSONNEL TURNOVER AND LEGISLATIVE EFFICIENCY IN THE EU

93

data in order to ensure adequate coverage of files that began before our data but were still open when our data coverage begins in 2000. We also control for whether a proposed act is New legislation (“1”) or seeks to amend existing policy (“0”). When undertaking a new legislative endeavor, the Commission may be particularly uncertain about the preferences of decision makers. In such a situation, it may be more likely to make a mistake in the placement of its original proposal. Actors in the Council and the EP may, in turn, have more difficulty reaching agreement. We code this variable by determining whether the title of the proposal includes words such as “amending,” “correcting,” “replacing,” “revising,” “consolidating,” “codifying,” “repealing,” and “modifying.” Next, we include three institutional controls that are expected to slow the legislative process. Legally, codecision can involve up to three readings in the EP and the Council. If the institutions cannot agree to a proposal at first reading, the dossier moves to a second reading; and if they cannot negotiate an agreement at second reading, the proposal moves to a third reading. We expect proposals that reach a Second reading to take longer to conclude than those that are concluded at first reading and proposals that reach a Third reading to take longer than those concluded at both first and second reading. Similarly, legislative acts initiated under particular treaty articles require decision makers to solicit the Opinions of EU advisory bodies like the Committee of the Regions and the Economic and Social Committee. We expect acts that require this step (“1”) to take longer to conclude than those that do not (“0”). Finally, we include two measures to control for preference divergence in the Council. Previous studies have measured preference divergence in a variety of ways. In early work, for example, Schulz and König (2000) made assumptions about heterogeneity based on a whether a proposal belonged to the functional “core” of the EU (e.g., internal market, agriculture, competition, and trade) or more peripheral areas; they assumed preferences to diverge less in core sectors than in peripheral ones. König later (2008) used manifestos to calculate policy sector-specific estimates of member-state preferences. This approach became even more finely grained with the completion of the EULIS project (König and Luig 2012). Others have used broader left–right scores to estimate preferences (Crombez and Hix 2014; Klüver and Sagarzazu 2013). Intuition and previous research both suggest that when there is greater heterogeneity of preferences in the Council, it is harder to reach a conclusion, thus increasing the duration of the legislative process (Klüver and Sagarzazu

94

L. K. PEREZ AND J. A. SCHERPEREEL

2013). Thus, we include two controls for preference heterogeneity in the Council; both measures utilize data from the Chapel Hill Expert Survey (Bakker et al. 2015; Polk et al. 2017). Left–right preference divergence measures divergence on a left–right spectrum; it is calculated by taking the standard deviation of the party positions of all ministers in the Council grand plenary in a particular month. Natl-suprantl. preference divergence records the standard deviations of Council parties’ approaches to European integration in a given month. We expect higher values of both variables to be associated with longer legislative durations. Our independent variables and all of our controls vary over time, which implies that we cannot simply have one observation per legislative proposal. Instead, we observe each legislative proposal at each month, recording both the number of months elapsed so far and the values of each of the independent and control variables. We then evaluate whether that month was the one during which the legislation was finally concluded. We use the month as the time unit because our turnover data are measured at the month level and because, as noted above, the legislative process can take many months or even years to complete. To test our hypothesis, we use a semi-parametric Cox model. This kind of event history (survival) model allows us to include right-censored observations—in the present case, proposals that had not been concluded by our final observation date. We find that one of our independent variables of interest (EP turnover) and most of our controls do not have a consistent effect over time; they fail to meet the proportional hazards assumption. For this reason, we interact such variables with the natural log of time to estimate the non-proportional effects (Box-Steffensmeier and Zorn 2001). We provide the results of our test of the proportional hazards assumption in the online appendix and report coefficients for time-varying covariates (TVCs) in a separate column of our results table (Table 4.2).

Results Table 4.1, which presents descriptive statistics for all of our variables, shows that turnover is substantially higher in the Council than in the Commission or the Parliament. The Council’s average annualized monthly quantity of turnover (36.25%) is more than twice the Commission’s annual average quantity (17.28%) and almost three times higher than the EP’s annual quantity (13.39%). Figure 4.1, which graphs institutional turnover quantities over time, shows that turnover patterns in

4

95

PERSONNEL TURNOVER AND LEGISLATIVE EFFICIENCY IN THE EU

Table 4.1 Personnel Turnover and Legislative Efficiency, Descriptive Statistics

Monthly data

Legislative file data

Commission turnover (%) Council turnover (%) EP turnover (%) Left–right pref. divergence Natl-suprantl. pref. divergence Directive Regulation # of member-states Backlog New legislation Second reading Third reading Opinions Legislative duration (months)*

Number

Mean

Std. Dev

Minimum Maximum

241

17.28

27.76

0

88

241

36.25

8.82

10.36

61.85

241

13.39

19.89

1.07

55.13

241

1.58

0.13

1.31

1.80

241

1.10

0.27

0.73

1.78

1,852 1,852 1,852

0.33 0.57 25.52

0.47 0.50 4.44

0 0 15

1 1 28

1,852 1,852

613.19 0.46

212.16 0.50

285 0

961 1

1,852

0.22

0.42

0

1

1,852 1,852 1,591

0.04 0.74 26.63

0.20 0.44 21.92

0 0 2

1 1 182

Note *Duration statistics are listed for completed files only

the Commission and the Parliament are significantly more punctuated than turnover patterns in the Council. The Commission and the EP tend to experience periods of high turnover followed by periods of low or (in the case of the Commission) zero turnover. In contrast, the Council exhibits generally high levels of turnover. This aligns with the fact that commissioners and MEPs are appointed and elected (respectively) to clear, temporally bounded terms. While there is certainly turnover within those terms, most of the personnel flux occurs at the breaks between terms: quantities of within-session turnover are lower than quantities of across-session turnover. In contrast, the Council is made up of national ministers who serve according to national electoral calendars and the

L. K. PEREZ AND J. A. SCHERPEREEL

60 40 0

20

Turnover (%)

80

96

Jan00

Jan02

Jan04

Jan06

Jan08

Jan10

Jan12

Jan14

Jan16

Jan18

Jan20

Date Commission EP

Council

Fig. 4.1 Quantity of Turnover in EU Institutions Over Time

whims of heads of government, and the EU calendar has no clear effect on their tenure. These different types of turnover also have different effects on the duration of the legislative process. The results of our Cox Proportional Hazards model are presented in Table 4.2. For all variables, we present both coefficients (columns 1 and 2) and hazard ratios (columns 3 and 4). Variables with negative coefficients have hazard ratios between 0 and 1. In terms of interpretation, a negative coefficient means that increasing covariate values make legislative completion less likely. When the value of a covariate with a negative coefficient rises, the legislative process takes longer to complete. Thus, covariates with negative coefficients are associated with increases in legislative duration. In contrast, variables with positive coefficients have hazard ratios that are greater than 1. When the value of a variable with a positive coefficient increases, legislative completion becomes more likely. Thus, covariates with positive coefficients are associated with decreases in legislative duration.3

4

PERSONNEL TURNOVER AND LEGISLATIVE EFFICIENCY IN THE EU

97

Table 4.2 Cox Regression of Legislative Duration (1) Coefficients main Commission turnover (%) Council turnover (%) EP turnover (%) Directive Regulation # of member-states Backlog New legislation Second reading Third reading Opinions Left–right preference divergence Natl-suprantl. preference divergence Observations # of subjects # of failures X2 Log likelihood

−0.0044*** (0.0016) −0.0155*** (0.0039) 0.0254*** (0.0068) −2.1046*** (0.2738) −0.8453*** (0.0854) −0.0414*** (0.0086) 0.0025*** (0.0008) −1.3180*** (0.2424) −4.0536*** (0.3510) −0.4329*** (0.1330) −0.7445*** (0.2645) −2.3001** (0.9423)

(2) Coefficients TVC

−0.0073*** (0.0021) 0.3650*** (0.0856)

−0.0007*** (0.0002) 0.3735*** (0.0774) 1.1554*** (0.1061)

0.2613*** (0.0864) 0.5835* (0.2985)

−0.0609 (0.1962) 48,615 1852 1591 464.7 −10,166

(3) Hazard ratios main 0.9956*** (0.0016) 0.9846*** (0.0038) 1.0257*** (0.0069) 0.1219*** (0.0334) 0.4294*** (0.0367) 0.9594*** (0.0083) 1.0025*** (0.0008) 0.2677*** (0.0649) 0.0174*** (0.0061) 0.6486*** (0.0862) 0.4750*** (0.1257) 0.1002** (0.0945)

(4) Hazard ratios TVC

0.9927*** (0.0021) 1.4406*** (0.1233)

0.9993*** (0.0002) 1.4528*** (0.1124) 3.1752*** (0.3369)

1.2987*** (0.1122) 1.7923* (0.5350)

1.0628 (0.2086) 48,615 1852 1591 464.7 −10,166

48,615 1852 1591 464.7 −10,166

48,615 1852 1591 464.7 −10,166

Note * = p ≤ 0.10, ** = p ≤ 0.05, *** = p ≤ 0.01. Standard errors in parentheses. TVC variables interacted with time

In Table 4.2, we see that turnover in all three institutions is significantly associated with legislative duration. The results for Commission turnover and Council turnover confirm our expectation: higher quantities of turnover are associated with lower monthly likelihoods of file

98

L. K. PEREZ AND J. A. SCHERPEREEL

completion. As quantities of turnover in the Commission and the Council increase, the legislative process takes longer to complete. Attention to the model’s hazard ratios (Table 4.2, column 3) suggests that a one percentage point increase in Commission turnover is associated with a 0.44% decrease in the likelihood of completing the legislative process in a particular month. Similarly, a one percentage point increase in Council turnover is associated with a 1.54% decrease in the likelihood of completion. Figure 4.2, which shows the cumulative hazard function for turnover at the observed minimum and maximum values for each institution, helps to visualize these associations. The findings for the Commission and the Council fit with the idea that as turnover increases, experience decreases, uncertainty rises, and the legislative process becomes less efficient. In this sense, turnover in the Commission and the Council threaten EU throughput legitimacy. Council turnover

0

0

Cumulative Hazard 1 2 4 5 3

Cumulative Hazard 5 1 2 3 4

Commission turnover

0

25

50

75

100

125

0

25

50

75

100

125

Time (months)

Time (months)

Cumulative Hazard 1 2 3 4 5

EP turnover

0

Minimum turnover Maximum turnover

0

25

50 75 Time (months)

100

125

Fig. 4.2 Cumulative Hazard Functions for Cox Regression of Legislative Duration

4

PERSONNEL TURNOVER AND LEGISLATIVE EFFICIENCY IN THE EU

99

In contrast, and contrary to H1’s expectations, higher quantities of turnover in the European Parliament are associated with a speedier legislative process: a one percentage point increase in turnover among MEPs increases the likelihood of file completion in a month by 2.57%. As column (2) of Table 4.2 suggests, however, the effect of EP turnover on legislative duration decreases slightly as time moves forward, reaching zero by the thirty-third month of a legislative file’s life. What might explain the unexpected result for EP turnover? As a partial explanation, Salvati and Vercesi (chapter 6) find that those MEPs with considerable national experience are able to step into roles as EP rapporteurs without negative effects on efficiency, and it is only full political outsiders who see delays in their reports. It seems that national experience may be a better substitute within the EP than it is within the Council and the Commission. However, this would only help explain some of what we find (e.g., why we might see less of an effect of EP turnover on efficiency, but not why that effect would be positive). Further research into why turnover would increase efficiency is necessary. For the moment, however, we might propose three potential explanations. First, it may be that as more inexperienced MEPs enter the Parliament, they and/or their fellow parliamentarians (who are not yet certain about the issue orientations of their new colleagues) look to the Council (traditionally the stronger chamber) for guidance. Second, the electoral cycle may play a mediating role. MEPs may be particularly keen to push dossiers forward in periods following personnel punctuations. After elections, for example, they may want to convince citizens that they are delivering on their fresh electoral mandates. The third potential explanation relates to the fact that the EP tends to experience high levels of across-session turnover. Such disruptions cause shifts in the ideological positions of MEPs. Such shifts may open up areas of potential agreement that were blocked by veto players in the previous parliamentary session. In general, the controls included in Table 4.2 perform in line with expectations. Like other scholars, we find that regulations take longer to conclude than decisions and that directives take longer to conclude than both decisions and regulations. The inclusion of more member-states (e.g., enlargement), new legislation (as opposed to amending or consolidating acts), the addition of a second or third reading, opinions from other institutions like the Economic and Social Committee, and greater left–right preference divergence all extend the amount of time needed to conclude a dossier. For all of the variables whose effects vary with time,

100

L. K. PEREZ AND J. A. SCHERPEREEL

we find that the effect diminishes the longer a dossier is under consideration. We find, again in line with expectations, that legislative backlog increases the speed of the legislative process. National-supranational preference divergence, on the other hand, is not significantly associated with legislative duration.

Conclusions and Future Research Our empirical analysis has produced mixed findings: the intuition that higher levels of turnover will lengthen the EU’s legislative process has been supported in the case of the Commission and Council but not in the case of the Parliament. Turnover among commissioners and national ministers negatively affects throughput legitimacy. But turnover among MEPs is associated with a more efficient EU legislative process. When viewed from a distance, it is somewhat remarkable that the EU institutional system manages to reach any decisions at all. The European Commission, after all, is a rather “unnatural” institution. While the Spitzenkandidat process has sought to bring the Commission presidency more into line with the partisan configuration of the Parliament, the Parliament’s largest political group controls significantly less than a majority of the institution’s seats. The Commission president has limited control over who sits in the college of commissioners, and the politicians who nominate the commissioners have diverse reasons for putting forward particular nominees. The Council is one of the most protean legislative institutions in the world, and the executive-legislators who sit at its peak have national “day jobs” that pull them away from the business of governing a large, diverse, and complex community. The Parliament may be the most recognizable political object out of the three institutions studied in this chapter, but it faces its own significant obstacles to efficiency. The EP has over 700 MEPs, and its seats are shared by representatives of well over 150 national political parties. It is no mean achievement, given these obstacles, to reach decisions in any amount of time. But nor is it unreasonable—in a community founded on a commitment to liberal democratic values—to expect EU institutions to operate efficiently to address public problems and promote the common good. Previous research has suggested that there is a tradeoff between democracy and efficiency in EU decision-making. Steps that are designed to increase legitimacy—for example, the empowerment of the EP, the use of legal instruments that carve out key spaces for national

4

PERSONNEL TURNOVER AND LEGISLATIVE EFFICIENCY IN THE EU

101

decision-makers, and the consultation of societal stakeholders—come at the cost of a slower legislative process. Turnover is a structural feature of the EU system—not a deliberate institutional innovation. Turnover brings new voices into the system and can help to keep public institutions in tune with the public mood. Our results suggest, however, that turnover in the Commission and the Council slow down the legislative process. Slower legislative processes, in turn, may (further) undermine already low levels of trust in EU institutions and the integration process. We must take care, however, not to overstate the existence of a tradeoff between turnover and legislative efficiency. For the European Parliament, after all, we find no support for the idea that turnover undermines efficiency. Indeed, EP turnover is associated with faster decision making. Future research would do well to explore the reasons behind the retarding effect of Commission and Council turnover, on one hand, and the accelerating effect of EP turnover, on the other. Above, we have suggested that an explanation may involve the relative prominence of the various institutions. Codecision has been established for more than twenty-five years. It has been the EU’s ordinary legislative procedure for more than a decade. During the same periods, though, the leaders of national governments have emerged as particularly important steerers of Europe’s political course (Bickerton et al. 2015; Puetter 2014). It is possible, given these phenomena, that MEPs may (still) look toward the Council when they (MEPs) are adjusting to in-house institutional flux. It is also possible that post-election dynamics establish new coalitional opportunities and/or encourage a sense, among actors in all three institutions, that the people have spoken and that principals have a responsibility to act. In spite of our finding that both Commission and Council turnover are associated with longer legislative processes, we see our chapter’s overall findings as broadly consistent with other contributions (e.g., Hagemann and Franchino 2016; Novak and Hillebrandt 2020) that problematize simple tradeoff arguments. The notion that democracies face tradeoffs between desirable normative ends—e.g., transparency vs. efficiency, turnover vs. efficiency—is intuitive. Governance tradeoffs have been discussed in the broader political science literature (e.g., Dahl 1994) and in the EU studies community (e.g., Curtin and Meijer 2006; Leino 2017; Naurin 2017; Stasavage 2006). The current chapter suggests that the turnover/efficiency tradeoff does not apply across all three institutions: while turnover in two of the three institutions is associated with lower efficiency, turnover in the third institution is associated with greater

102

L. K. PEREZ AND J. A. SCHERPEREEL

efficiency. Additional research is necessary to identify the reasons behind the differences across institutions. The chapter’s analyses throw some light on the propositions suggested by the EU’s populist and feminist critics. As discussed in chapter 1, the populist critique suggests that personnel replacements can help to legitimize EU politics. Here, we have focused exclusively on questions involving the quantity of turnover, and we hesitate to make bolder claims without further considering turnover’s mechanism/s and (especially) quality. It remains to be seen, for example, whether the replacement of a set of center-right MEPs with a set of populist MEPs is effectively distinguishable from the replacement of one set of center-right MEPs with a second set of center-right MEPs. That proviso notwithstanding, our analysis does not suggest that legislative business will grind to a halt when new MEPs replace sitting MEPs. Indeed, such replacements are associated with an acceleration of the legislative process. This finding may encourage critics who are eager to replace “entrenched elites” with representatives of the pure people. Anyone eager to reach that conclusion, though, would do well to review Salvati and Vercesi’s analysis in this volume (chapter 6). The latter authors find that most new MEPs who serve as rapporteurs do not seem to slow the legislative process but that “pure outsiders”—politicians who lack national or EP experience—are associated with throughput delays. What is more, our findings on the Commission and the Council may deflate at least some populist critics’ hopes. Quantities of turnover in the latter institutions are associated with legislative lordeur. Were populists to gain a foothold in national governments and thereby replace Council and, potentially, Commission principals with new, more sympathetic allies, it would likely take the new incumbents some time to transform their mandate into policy. The finding that the quantity of Council turnover tends to delay decision making is particularly relevant in this regard, since populist critics may trust the Council—the bulwark of organic national interests, the “brake on Brussels”—over the “other-dominated” Commission and EP. But the longer decision-making processes that are associated with Council turnover may ultimately compound populist critics’ frustrations with democratic decision-making and the EU system. In terms of the feminist critique, our findings would seem to suggest caution, at least in regard to the Council and Commission. In these two institutions, it would seem that while higher turnover might allow more women to enter, turnover might also decrease efficiency. However, in

4

PERSONNEL TURNOVER AND LEGISLATIVE EFFICIENCY IN THE EU

103

chapter 5, Aldrich and Perez show that this is an example of when the quality of turnover matters in addition to the quantity of turnover. Their examination throws additional light on the ways that turnover affects throughput legitimacy.

Notes 1. Values for January 2000 require observation of the Commission’s personnel composition in January 1999; values for February 2000 require observation of the Commission on February 1999; etc. 2. Information about our process for determining whether a cabinet member is part of the Council grand plenary can be found in the chapter’s online appendix. 3. To use the language of survival analysis: negative coefficients mean that the hazard is decreasing and that there is a decreased likelihood of failure. In the present setting, “failure” is defined as the passage of the legislation. A negative coefficient means that a proposal has a lower likelihood of passing, thereby increasing duration. Conversely, positive coefficients increase the hazard or risk of failure. Covariates with positive coefficients increase the risk of passage, shortening the amount of time it takes to reach a final decision.

References Atkinson, Michael M. and David C. Docherty. 1992. Moving Right Along: The Roots of Amateurism in the Canadian House of Commons. Canadian Journal of Political Science/Revue Canadienne De Science Politique 25(2): 295–318. Bakker, Ryan, Catherine de Vries, Erica Edwards, Liesbet Hooghe, Seth Jolly, Gary Marks, Jonathan Polk, Jan Rovny, Marco Steenbergen, and Milada Anna Vachudova. 2015. Measuring Party Positions in Europe: The Chapel Hill Expert Survey Trend File, 1999–2010. Party Politics 21(1): 143–152. Bickerton, Christopher J., Dermot Hodson, and Uwe Puetter (eds.). 2015. The New Intergovernmentalism: States and Supranational Actors in the Post-Maastricht Era. Oxford: Oxford University Press. Box-Steffensmeier, Janet M., and Christopher J. W. Zorn. 2001. Duration Models and Proportional Hazards in Political Science. American Journal of Political Science 45 (4): 972–988. Chalmers, Adam William. 2014. In Over Their Heads: Public Consultation, Administrative Capacity and Legislative Duration in the European Union. European Union Politics 15(4): 595–613.

104

L. K. PEREZ AND J. A. SCHERPEREEL

Crombez, Christophe and Simon Hix. 2014. Legislative Activity and Gridlock in the European Union. British Journal of Political Science 45(3): 477–498. Curtin, Deirdre, and Albert Jacob Meijer. 2006. Does Transparency Strengthen Legitimacy? Information Polity 11(2): 109–122. Dahl, Robert. 1994. A Democratic Dilemma: System Effectiveness versus Citizen Participation. Political Science Quarterly 109(1): 23–34. Daniel, William T. and Shawna K. Metzger. 2018. Within or between Jobs? Determinants of Membership Volatility in the European Parliament, 1979– 2014. The Journal of Legislative Studies 24(1): 90–108. Francis, Wayne L. and John R. Baker. 1986. Why do US State Legislators Vacate their Seats? Legislative Studies Quarterly 11(1): 119–126. Golub, Jonathan. 2008. The Study of Decision-Making Speed in the European Union: Methods, Data and Theory. European Union Politics 9(1): 167–179. Golub, Jonathan. 2007. Survival Analysis and European Union Decision-Making. European Union Politics 8(2): 155–179. Golub, Jonathan. 1999. In the Shadow of the Vote? Decision Making in the European Community. International Organization 53(4): 733–764. Golub, Jonathan and Bernard Steunenberg. 2007. How Time Affects EU Decision-Making. European Union Politics 8(4): 555–566. Häge, Frank M. 2012. The Duration of Multilateral Negotiations in the Council of the European Union, Unpublished manuscript available at https:// www.cergu.gu.se/infoglueCalendar/digitalAssets/1776176368_BifogadFil_ FMHaege%202012%20Duration%20of%20Multilateral%20Negotiations%201 10817.pdf. Accessed July 17, 2020. Häge, Frank M. 2011. The European Union Policy-Making Dataset. European Union Politics 12 (3): 455–477. Hagemann, Sara, and Fabio Franchino. 2016. Transparency vs. Efficiency? A Study of Negotiations in the Council of the European Union. European Union Politics 17(3): 408–428. Hertz, Robin and Leuffen, Dirk. 2011. Too Big to Run? Analysing the Impact of Enlargement on the Speed of EU Decision-making. European Union Politics 12(2): 193–215. Hyneman, Charles S. 1938. Tenure and Turnover of Legislative Personnel. The Annals of the American Academy of Political and Social Science 195(1): 21– 31. Junge, Dirk, Thomas König, and Bernd Luig. 2015. Legislative Gridlock and Bureaucratic Politics in the European Union. British Journal of Political Science 45(4): 777–797. Klüver, Heike and Iñaki Sagarzazu. 2013. Ideological Congruency and Decisionmaking Speed: The Effect of Partisanship across European Union Institutions. European Union Politics 14(3): 388–407.

4

PERSONNEL TURNOVER AND LEGISLATIVE EFFICIENCY IN THE EU

105

König, Thomas. 2008. Analysing the Process of EU Legislative Decision-Making: To Make a Long Story Short . . . European Union Politics 9(1): 145–165. König, Thomas. 2007. Divergence or Convergence? From Ever-growing to Everslowing European Legislative Decision Making. European Journal of Political Research 46(3): 417–444. König, Thomas and Bernd Luig. 2012. Party Ideology and Legislative Agendas: Estimating Contextual Policy Positions for the Study of EU Decision-Making. European Union Politics 13(4): 604–625. Leino, Päivi. 2017. Secrecy, Efficiency, Transparency in EU Negotiations: Conflicting Paradigms? Politics and Governance 5(3): 6–15. Naurin, Daniel. 2017. The Puzzle of Transparency Reforms in the Council of the EU. Politics and Governance 5(3): 87–90. Novak, Stéphanie and Maarten Hillebrandt. 2020. Analysing the Trade-off between Transparency and Efficiency in the Council of the European Union. Journal of European Public Policy 27(1): 141–159. Perez, Lauren K. and John A. Scherpereel. 2017a. Vertical Intra-institutional Effects of Ministerial Turnover in the Council of the European Union. Journal of European Public Policy 24(8): 1154–1171. Polk, Jonathan, Jan Rovny, Ryan Bakker, Erica Edwards, Liesbet Hooghe, Seth Jolly, Jelle Koedam, Filip Kostelka, Gary Marks, Gijs Schumacher, Marco Steenbergen, Milada Anna Vachudova and Marko Zilovic. 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 (January–March): 1–9. Puetter, Uwe. 2014. The European Council and the Council: New Intergovernmentalism and Institutional Change. Oxford: Oxford University Press. Rasmussen, Anne and Dimiter Toshkov. 2013. The Effect of Stakeholder Involvement on Legislative Duration: Consultation of External Actors and Legislative Duration in the European Union. European Union Politics 14(3): 366–387. Rasmussen, Anne and Dimiter Toshkov. 2011. The Inter-institutional Division of Power and Time Allocation in the European Parliament. West European Politics 34(1): 71–96. Reh, Christine, Edoardo Bressanelli, and Christel Koop. 2020. Responsive Withdrawal? The Politics of EU Agenda-setting. Journal of European Public Policy 27(3): 419–438. Rosenthal, Alan. 1992. The Effects of Term Limits on Legislatures: A Comment. In Gerald Benjamin and Michael J. Malbin (eds.), Limiting Legislative Terms. Washington, DC: CQ Press, 205–208. Scherpereel, John A. and Lauren K. Perez. 2015. Turnover in the Council of the European Union: What It Is and Why It Matters. JCMS: Journal of Common Market Studies 53(3): 658–673.

106

L. K. PEREZ AND J. A. SCHERPEREEL

Scherpereel, John A., Lauren K. Perez, and Rachel Young. 2018. How Permanent are the Permanent Representatives? The Effects of Personnel Turnover in the Committee of Permanent Representatives. Paper presented at the biennial meeting of the European Consortium for Political Research’s Standing Group on the European Union, Paris. Schulz, Heiner and Thomas König. 2000. Institutional Reform and DecisionMaking Efficiency in the European Union. American Journal of Political Science 44(4): 653–666. Sloot, Thomas and Piet Verschuren. 1990. Decision-making Speed in the European Community. JCMS: Journal of Common Market Studies 29(1): 75–85. Sonntag, Lars. 2020. Politica, 2020. Available at http://www.kolumbus.fi/taglar sson/. Accessed June 25, 2020. Stasavage, David. 2006. Does Transparency Make a Difference? The Example of the European Council of Ministers. In Christopher Hood and David Heald (eds.), Transparency: The Key to Better Governance? Oxford: Oxford University Press, 165–179. Tsebelis, George and Xenophon Yataganas. 2002. Veto Players and Decisionmaking in the EU after Nice. JCMS: Journal of Common Market Studies 40(2): 283–307. Van Schagen, Jan. 1997. The Principle of Discontinuity and the Efficiency of the Legislative Process. Journal of Legislative Studies 3(4): 115–125. Wonka, Arndt. 2015. The European Commission. In Jeremy Richardson and Sonia Mazey (eds.), European Union: Power and Policy-making 4 ed. London: Routledge, 83–105.

CHAPTER 5

Losing Women, Losing Power? Gender, Turnover, and EU Legislation Andrea S. Aldrich and Lauren K. Perez

The inclusion of women in political institutions can have important effects on systemic legitimacy. Clark and Scherpereel’s contribution to this volume (chapter 3), suggests that the descriptive representation of women (e.g., the inclusion of women in leading political roles) can contribute to input legitimacy by encouraging increased political participation. Other research suggests that women’s inclusion in office can help to increase political knowledge and engagement among women (Barnes and Burchard 2013; Fridkin and Kenney 2014), to reduce stereotypes about women in politics (Beaman et al. 2009), and to provide role models for other women (Wolbrecht and Campbell 2007). Research also

A. S. Aldrich (B) Department of Political Science, Yale University, New Haven, CT, USA e-mail: [email protected] L. K. Perez Department of Political Science, Francis Marion University, Florence, SC, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. A. Scherpereel (eds.), Personnel Turnover and the Legitimacy of the EU, Palgrave Studies in European Union Politics, https://doi.org/10.1007/978-3-030-60052-5_5

107

108

A. S. ALDRICH AND L. K. PEREZ

suggests that women’s descriptive representation can affect output legitimacy. The more women ascend to high political office, the more likely it is that women’s views, needs, and opinions will be expressed in policy choices. Women often play distinctive roles in legislative committees and policy debates (Heath et al. 2005; Kerevel and Atkeson 2013). They are more likely to work in policy areas that affect citizens’ private lives, including social policy, education, and health care (Bratton and Haynie 1999; Davis 1997; Escobar-Lemmon and Taylor-Robinson 2009). In addition, more diverse and representative decision-making bodies are more likely to reflect the views and preferences of all relevant groups, to ensure that policy decisions are taken with diverse preferences and groupspecific outcomes in mind (Mansbridge 1999), and, thereby, to increase the quality of policy outputs. This chapter picks up where chapter four left off. Like chapter four, it is concerned with the relationship between personnel turnover and throughput legitimacy. Here, however, we pay particular attention to issues raised by the feminist critique of EU politics. Our study of the relationship between women’s representation (in EU institutions) and (the EU’s) throughput legitimacy pays particular attention to the ways that women and men work in the Council of the EU. Thus, we are interested in the quality of turnover – in whether the personnel who enter and leave the Council are male or female – in addition to the quantity of turnover. We draw from the literature that examines whether women bring specific skills and resources to decision-making processes (Anzia and Berry 2011; Volden et al. 2013). We begin by presenting an overview of the expected impact of women in legislative and executive processes, and we discuss why this is particularly relevant to decision-making in the Council. Then, we provide a more targeted discussion of women and turnover in the Council and an overview of empirical data on the status of female cabinet ministers in the Council system. We hypothesize that not all turnover in the Council is the same and that quality matters: because women tend to bring distinctive resources to decision-making tables, we expect the exit of women to have particularly deleterious effects on the EU’s throughput legitimacy. Our analysis supports the hypothesis: the exit of experienced women, we find, lengthens the EU’s legislative process and makes it less likely to have a successful conclusion, detracting from throughput legitimacy.

5

LOSING WOMEN, LOSING POWER? …

109

Women as Lawmakers When it comes to working in legislatures and pushing policy forward, previous research has shown that women often behave differently than their male counterparts. Research on policy preferences and bill sponsorship, for example, has found that women are more likely to initiate and support legislation that pertains directly to women and families (Bratton and Haynie 1999; Heath et al. 2005; Schwindt-Bayer 2006) and that women are more likely to serve on committees in areas like employment policy, environment, public health, culture, and education (Kantola 2010). Other scholars have focused on women’s impact on policy outcomes and concluded that some women are more effective than men at turning proposals into law (Bratton and Haynie 1999). Thomas (1991) shows that this is particularly true when it comes to bills on women’s issues and attributes this success to working with other women through legislative networks. Others have argued that women have a particular work style that can sometimes make them more effective legislators (Volden et al. 2013). Previous research examining women’s behavior in both legislative and executive politics has found that the working style of women is distinct from that of their male counterparts in several ways. First, women are said to be more collaborative than their male colleagues and to approach issues with more compassion and understanding (Jeydel and Taylor 2003). Such differences in approach can result in different policy solutions (Kathlene 1995). More broadly, this approach may make women more effective at legislating and in politics generally, promoting work beyond partisan lines and increasing focus on more pragmatic outcomes (Alvarez 2000). Research has even shown that this is true in the policy-making process of the European Union. In a study of organizational change in the Commission during the enlargement process, for example, Ban (2013) finds that the influx of women during enlargement had unexpected, yet positive, consequences for the Commission’s working culture. The largest differences between men and women involved their communication and management styles: women were described as more cooperative. Such a tendency to cooperation challenged the competitive norm that had existed in Commission decision making prior to the influx of women that came with enlargement. These characteristics can be especially helpful in policy-making scenarios that require consensus building. In fact, Volden et al. (2013)

110

A. S. ALDRICH AND L. K. PEREZ

find women to be the most effective when they are members of the minority party in (the US) congress. In this case, women are able to leverage the behavioral traits mentioned above in conjunction with their institutional position to guide more pieces of legislation from the proposal stage to the floor voting stage and into law. This is precisely why we expect women to be more successful in navigating through the EU’s legislative process, which requires cooperation and coordination across several institutions and with many diverse actors (see, for example, Chalmers 2014; Christensen 2001; Christiansen and Piattoni 2003; Golub 2007; Klüver and Sagarzazu 2013). In particular, the Council is an institution with “a distinctive ‘culture’ of collective decision-making” (Lewis 2010: 656) that operates largely on consensusbuilding and generally strives to reach unanimous decisions (Häge 2013; Hayes-Renshaw et al. 2006; Heisenberg 2005; Novak 2013; Warntjen 2010). Within the Council, the treaties allow for some decisions to be made via Qualified Majority Voting (QMV), while others require unanimity. Despite these rules, “one of the most basic and durable normative standards in the Council is to make collective decisions by consensus” (Lewis 2010: 656). Even when the treaties provide for QMV, the Council only actually votes around 10–20% of the time (Heisenberg 2005; Thomson 2011). This emphasis on cooperation and consensus-building suggests that women’s working styles may be particularly well-suited to the Council and they should be more effective at facilitating the sometimes difficult interactions that are necessary to pass new EU legislation. This also suggests that experienced women would be able to apply their skills and working style even more successfully, and that losing these women to turnover would therefore hinder decision-making and legislating in the Council. In research on the European Council, Tallberg (2008) also identifies several important individual-level factors that help ministers negotiate that could contribute to women’s effectiveness. He argues that negotiators with significant expertise, the ability to earn another’s trust, and the opportunity to leverage personal relationships tend to have an advantage over those who lack those attributes. In terms of expertise, previous research has shown that women who make it to high-level political positions are just as qualified as men, if not more qualified (Fulton 2012). Given the uphill battle that women often face to obtain high level positions of power, those that reach the top often have more experience and/or expertise than similarly situated men (Aldrich and Daniel 2019;

5

LOSING WOMEN, LOSING POWER? …

111

Besley et al. 2017). Scholars have also found that women are perceived as more trustworthy and honest and less likely to participate in corruption (Barnes and Beaulieu 2019; Funk et al. 2019), which may make them more trustworthy partners during negotiations. In addition, research has also shown how women have the ability to leverage relationships with other women in political institutions. Having a significant number of women in an institution can change the way the institution functions practically. If women are not struggling for respect of their basic rights in the workplace, they are freer to pursue their substantive policy goals (Dahlerup 2006). This means women’s presence in institutions can also change the way men work and behave when working with women. Using experimental evidence from the Council, Naurin et al. (2019) find evidence that negotiations between men and women are impacted by stereotypes but that this may also be useful for women. They find that when women behave stereotypically (“displaying emotions and vulnerability” [483]), men are more willing to compromise and support the women’s bargaining proposal. One additional important feature of Council negotiations is the role of the Council presidency, which rotates among the member-states every six months. On the one hand, this role is fairly limited, since the president is supposed to be a neutral actor, is limited to a short term, and has to work with what is already on the agenda and with what legislation the Commission has already proposed. However, researchers have nevertheless found that it is an important role (Schalk et al. 2007; Thomson 2008), mostly because the presidency country has a lot more information than the representatives from other member-states, both through conversations with the other EU institutions and with other member-state delegations to the Council, and can propose compromises (Schalk et al. 2007; Tallberg 2006, 2010; Thomson 2008; Warntjen 2008). Presidents engage in bilateral “confessionals” with other leaders to learn what policy components are most important to them and what they cannot accept. Then, from some set of potential compromises, they can propose one (likely the one closest to their own position) to the Council as a whole (Tallberg 2006). This role requires even more negotiation and cooperation than standard Council participation, and therefore seems like a prime opportunity for a feminine work style to be advantageous in helping find a compromise that everyone will accept. While we do not explicitly test whether female presidents help conclude legislation, if there are more women in the Council, there will be more female presidents.

112

A. S. ALDRICH AND L. K. PEREZ

In contrast, there are a few other features of Council and EU decisionmaking that might limit the impact that female ministers can have and that are worth noting here. First, ministers are not independent actors, but are representatives of their member-states and need to act in accordance with instructions set by their governments and parliaments. We see no obvious reason why the constraints imposed by these instructions would have a differential impact by gender, but they do limit the free rein of all ministers. Second, much of the actual work and negotiating in the Council is done by its preparatory bodies. According to Häge (2008), only about half of all legislative dossiers are actually discussed by the ministers themselves, while the other half are negotiated by the preparatory bodies and the ministers simply vote on the final outcome. This reliance on the preparatory bodies is likely increased by a third important feature—the EU’s increasing use of trilogues. These tripartite meetings among representatives of the Commission, Council, and Parliament take place early in the codecision process and provide a venue for inter-institutional negotiations. Trilogues have further reduced the involvement of ministers, since members of the Council’s preparatory bodies (specifically from the presidency country) usually attend these meetings (Häge and Naurin 2013). If the ministers are not negotiating the dossiers themselves, then there is less possible impact for female ministers. However, it is possible that differing management styles, such as those found in the Commission (Ban 2013), could still mean that women (and turnover rates among women) have a greater impact. Finally, it is important to remember that an individual minister’s power may also come from the member-state and/or political party they represent. Again, we do not expect this to differ between women and men. These are all important features of the EU’s system that help put the potential role of female ministers in perspective. Overall, however, we expect the considerations discussed earlier in this section to survive these features’ potentially attenuating effects. It is also important to note that variation in electoral and party systems can affect the types of politicians that make it into ministerial positions and influence their behavior once in office. We know that different incentives exist for legislators in single-member electoral systems compared with proportional voting systems (Carey and Shugart 1995), but the effect of electoral systems on the career paths of women is still unclear. Single-member plurality or majority systems can make it more difficult for women to advance given the strong personalistic competition and the strength of incumbents (Schwindt-Bayer 2005), but such systems can

5

LOSING WOMEN, LOSING POWER? …

113

also allow women to highlight their unique strengths or personal ties to voters (Shugart 1994). The impact of proportional systems depends very much on the type of list system employed (open or closed), the presence or absence of quotas, and the electoral context (Aldrich 2018). Thus, general hypotheses about the effect of this type of institutional variation on the behavior of women in the Council are not immediately clear. In addition, a recent study of gender and cabinet ministers across seven different parliamentary and presidential systems found that these institutional factors are not that important in the nomination of ministers. Annesley et al. (2019) find no relationships between the selection of female cabinet ministers and political system type, party ideology, or the gender of the selectors. Instead, they conclude that variation in the number of ministers across countries is mostly due to the variation across political leaders when it comes to their qualifying criteria for choosing ministers. Thus, we do not expect any systematic effect of country or party-level factors to condition the behavior of women in the Council.

Women as Ministers In addition to a particular style of legislating, research has shown that women have traditionally tended to be overrepresented in ministries in “soft” policy areas and underrepresented in “hard” policy areas. Soft policy areas are those that pertain to education, social welfare, family, children, and culture, while hard policy areas deal with the more prestigious issues of finance, defense, security, and agriculture (Davis 1997; EscobarLemmon and Taylor-Robinson 2009). Some argue that the distribution of sexes across ministries is a reflection of traditional gender norms where men have been thought of as responsible for public affairs and women held responsibility over private, or familial, affairs. The first female ministers were appointed to portfolios closely linked to the private sphere, i.e. the “soft” areas (Davis 1997; Krook and O’Brien 2012), which may have reflected the natural outcome of the interests and experience of women in parliament and politics (Heath et al. 2005; Studlar and Moncrief 1999). The relegation of women to these softer positions, regardless of the cause, has consequences for behavior within the ministerial positions and career trajectories. In addition to being considered softer policy areas, these portfolios often carry less prestige and prominence than positions in defense, economics, or finance, meaning the work within the ministry is not highly visible and is rarely a stepping stone to higher office (Krook

114

A. S. ALDRICH AND L. K. PEREZ

and O’Brien 2012). When it comes to crafting laws and making policy, this can have both positive and negative consequences. Regardless of the causes or consequences of gendered portfolio distributions, we do see evidence of these patterns in the Council’s preparatory bodies. A study using data from 2007 highlights how both ministerial appointments in member-state cabinets and positions as permanent representatives in the Council were still very male-dominated. Women totaled only 3.7% of permanent ambassadors in Coreper II, which deals with the “hard” policy areas, whereas they made up 22.2% of Coreper I, which deals with the “soft” policy areas (Kantola 2010). Thus, while we expect there to be less women overall, we expect even fewer women in policy configurations of the Council that deal with economics, foreign affairs, and business-related policy. This difference in descriptive representation should induce differences in the style and type of policy being made in the Council and the overall effectiveness of the Council configurations in the legislative process.

Women and Turnover Turnover occurs in all long-term institutions: the same people cannot serve an institution forever. However, if quantities of turnover are too low or too high, problems can ensue. When the quantity of turnover is too low, politicians risk losing touch with citizens and may lose interest in doing the work necessary to serve them (Atkinson and Docherty 1992; Katz 1997; Matland and Studlar 2004). Low levels of turnover can also keep rising groups in society, such as women or minorities, from entering political institutions (Norris and Lovenduski 1995). This may be a problem in terms of the national pipelines that affect who gets into the Council. In order to get the experience to become a national minister, women often need to have served in parliament first and subsequently risen through the ranks. Thus, low turnover in national parliaments over the last fifteen to twenty years could be impacting the number of women serving in the Council today. However, within the Council itself, the problem tends to be that turnover is quite high (Scherpereel and Perez 2015). Our data show that an average of 41.2% of all ministers in the Council turned over within a year between 2008 and 2018. High turnover causes problems because the new entrants are less experienced, and it will take new entrants time to work up to full effectiveness. Atkinson and Docherty (1992)

5

LOSING WOMEN, LOSING POWER? …

115

argue that it can take novice politicians three or four years to become fully effective. Novices struggle because they likely do not have as much policy knowledge or as great an understanding of how new policy will fit with existing legislation, they do not have the personal connections with other policy-makers or the knowledge of the preferences that might help them in negotiations, and they are likely to be less familiar with the way their institution and counterpart institutions function (Francis and Baker 1986). The way that we would expect high female turnover to interact with our argument about women being more effective legislators, especially in a highly consensual institution like the Council, is that female turnover should have an even more detrimental effect than male turnover. When women rotate out of the Council, the Council loses legislators that, given the literature discussed above, are particularly effective actors. If women are leaving after only short periods, we would expect high female turnover to have a negative impact on the policy-making process. These women would not be able to gain as much experience as if they stayed in the institution for longer periods, and therefore would not be as likely to reach their full potential. On the other hand, if they have been there for longer and have been able to gain experience, the effect of their loss will likely be even greater. Additionally, some of the added value that women contribute comes through their ability to negotiate and collaborate. Both of these attributes are helped by having strong relationships with counterparts. This added value may increase over an individual woman’s tenure, as she builds and strengthens these relationships.

Women in the Council of Ministers Patterns of women’s participation in the Council of Ministers mirror more familiar patterns observed in Europe’s national parliaments and cabinets. Between 2009 and 2018, women accounted for 27.45% of total ministers in the Council. There were slightly fewer women in ministerial positions prior to 2014. The low ebb in this period occurred in 2012, when women made up only 25.76% of minsters. Since 2014, women’s representation in the Council has ticked upward: the percentage of female ministers rose from 26.48% in 2013 to 28.19% in 2015 and 30.49% in 2017. However, as Fig. 5.1 shows, this trend has largely been driven by a drop in the number of male ministers participating in Council business: in absolute terms, the numbers of female ministers in the Council

A. S. ALDRICH AND L. K. PEREZ

0

100

200

300

400

500

600

700

800

900

116

2009

2012 Men per year

Year Women per year

2015

2018 Ministers per year

Fig. 5.1 Total Number of Ministers in the Council of the EU

has remained largely stable or even declined.1 In 2014, we see a drop in the total number of ministers serving within a given year. Between January and December of 2013, nearly 900 different ministers participated in/oversaw Council business; this fact suggests that the year was characterized by a high quantity of turnover. The total number of ministers serving was lower in 2014, and, by 2015, the number had dropped to about 400. Ministers seem to have been serving in their positions for longer stretches of time in 2015. These changes in the rates of turnover appeared to have a positive effect on the share of women. The smaller gap between men and women in absolute terms may be a sign that women have had higher staying power in more recent years. While the growth in the share of women is overall positive, there is a significant amount of variance across countries and Council configurations. At the country level, Hungary had the fewest women in ministerial positions between 2009 and 2018: only 8.27% of the Hungarian ministers who participated in Council business over the course of the

5

LOSING WOMEN, LOSING POWER? …

117

decade were women. Over the same stretch of time, several countries— Austria, Denmark, France and Finland—were near parity. Croatia, Cyprus, Czechia, Estonia, Greece, Ireland, Lithuania, Malta, Poland, Slovakia, and the United Kingdom were significantly below average; less than 20% of the ministers they sent to the Council were women. The majority of the latter countries are new member-states from Eastern Europe, suggesting that these may be selecting especially low numbers of women when compared to their western counterparts. Finally, Sweden was the overall leader over the course of the observed time span: 56.18% of Swedish ministers in the Council system were women, making it the only country where women constitute the majority of ministers in the Council. At the level of the Council’s sectoral configurations, gender patterns also conform to the literature’s expectations. Figure 5.2 displays the distribution of men and women in each configuration from 2009 through 2018. This figure highlights a number of important patterns. First, women continue to be underrepresented in the “hard” configurations— Foreign Affairs, Economic and Financial Affairs and, but to a lesser extent, ECOFIN

JHA

EPSCO

COMPET

TTE

AGRIFISH

ENVI

EYCS

0

50 100 150

0

50 100 150

0

50 100 150

FAC

2009

2012

2015

2018 2009

2012

2015

2018 2009

2012

2015

2018

Year Men per configuration per year

Women per configuration per year

Fig. 5.2 Number of Men and Women Ministers in the Council of the EU, 2009–2018

118

A. S. ALDRICH AND L. K. PEREZ

Justice & Home Affairs. In Foreign Affairs, there are almost 3.5 times as many men as women in 2009, and this difference increases to almost 5 times in 2013. While the ratio of men to women in Economic and Financial Affairs is not as stark, the absolute number of women remains very low throughout the time period. In no configuration do we see more women than men, but the smallest differences can be found in (a) Environment and (b) Agriculture and Fisheries. Previous research (Escobar-Lemmon and Taylor-Robinson 2005) has shown that these areas are more gender-neutral. Overall, Fig. 5.2 suggests that while there is still room for improvement in terms of women’s representation in European cabinets, the existing patterns are similar to those found in the literature. Given that gendered patterns of representation exist in the Council, we expect turnover to have differential effects in both those areas where women have more representation and in areas where women turn over in the positions at both higher and lower rates than men. When women are turning over at high rates, they are likely being replaced by men or inexperienced women, and the added benefit of an experienced woman’s legislative style is lost. This consideration supports the chapter’s main hypothesis: H1: Higher quantities of women’s turnover will lower the probability of consensus and success in the EU’s legislative process.

Women and Turnover in the Council: Empirical Design In order to test the relationships between gender, turnover, and policy success, we calculate three different measures of turnover. First, we calculate a series of percentages that quantifies the share of individuals who turn over within each configuration, the share of men within a configuration that turn over, and the share of women within a configuration that turn over.2 To capture turnover, we borrow from the measures used in Perez and Scherpereel (2017), so that turnover is measured annually between a month in year t and the same month of t + 1. For example, the quantity of turnover among ministers in a configuration on April 1, 2017 is calculated by comparing the set of ministers on that date to the set of ministers that populated the same configuration on April 1, 2016. If a minister no longer appears in the April 1, 2017 set, she is considered to have exited the position, turning the position over to a new

5

LOSING WOMEN, LOSING POWER? …

119

minister. Turnover has been calculated in this manner from January 2009 through April 2018. In order to determine the percentage share of ministers turning over in any given month, we calculate the share of ministers that have exited as a portion of all ministers in the configuration. For example, the Employment, Social, Health, and Consumer Affairs configuration had 62 ministers in November 2010. Thirty-one of those ministers had been replaced by November 1, 2011. Thus, the rate of turnover was 50% for November 2011. The share of each gender is calculated similarly; we calculate the number of replaced women at time t + 1 as a share of all women present at time t. Our second measure of turnover captures the total number of ministers who exit a configuration over a period of time. We incorporate this measure in addition to the percentage measure because the low numbers of women that participate in some Council configurations can cause drastic percentage-point swings even if only one or two women exit their position. For example, a single woman leaving a configuration could cause a turnover rate of one hundred percent if she were the only female minister. In a more gender-balanced configuration, the departure of one woman might produce a turnover rate of three percent. For our third measure of turnover, we calculate the exit rate among women relative to the exit rate of men. Because some configurations contain very few women to begin with, the departure of just one or two women ministers can push the percentage rate of turnover quite high for women. In fact, the data contain a few observations where the quantity of turnover is greater than 80% for a given configuration in a given month. In order to avoid biasing the results of our analysis in favor of our hypothesis, we also construct a variable that measures turnover among women relative to turnover among men. This ratio is calculated as follows: N umber o f women tur ning over N umber o f men tur ning over When the ratio is equal to 1, the number of men and women leaving a configuration over the course of the previous year is equal. When the ratio is greater than one, the configuration has lost more women than men in the previous year, and when the ratio is less than one, more men have exited than women. The following paragraphs describe patterns in turnover using these three measures.

120

A. S. ALDRICH AND L. K. PEREZ

First, the raw turnover rates allow us to identify average patterns in personnel change and show that differences exist across configurations and across genders. If we examine the average number of days each minister stays in their position, we see this varies greatly by time, configuration, and gender. In the entire dataset, the average number of days served in a position is about 357, but the variance is large with a standard deviation of 344 days. When we examine patterns over time, we see that, in the early years of the dataset (between 2009 and 2011), ministers were serving an average of 514 days. Between 2012 and 2014, average tenure in position had fallen to 318 days, and, between 2015 and 2018, had fallen further to 235 days, perhaps reflecting more volatility in European politics writ large in recent years. Over the entire sample, women and men are almost equal with a mean of about 361 and 364 days, respectively, and there are no significant differences across genders in the time periods described above. However, if we look at rates of service of ministers across configurations, we see greater variation by gender. With the exception of Foreign Affairs, women in “hard” configurations tend to serve shorter terms than men. Women, on average, spend 316 days in Economic and Finance positions and 345 days in Justice & Home Affairs positions. The respective averages for men are 357 and 360. Women also tend to serve shorter terms than men in Competitiveness, Transport, Telecom & Energy, and Environment configurations. In contrast, women serve longer terms in “softer” configurations like Employment, Social Policy, Health & Consumer Affairs, and Education, Youth, Culture & Sport. Examining the average rates of turnover as a percentage of the total gender by configuration by year also shows that some difference exists across genders. Table 5.1 displays average rates of turnover for each configuration for a select number of years. This table shows that turnover percentage rates vary by configuration but have been trending downward in the later years of the data. Across both genders, there is a lot of fluctuation by year in turnover rates. On average, though, women turn over at higher percentage rates than men in configurations where their numbers are lower and at lower percentage rates in configurations where their numbers are higher. Figure 5.3 displays a cross-configuration monthly average of this ratio from January 2009 through December 2017. The figure shows this ratio ranges from zero to about 2.4 but that a large amount of variation exists from month to month. The mean of the ratio is about 1.02 with

5

LOSING WOMEN, LOSING POWER? …

121

Table 5.1 Annual Percentage Rates of Turnover by Gender and Council Configuration FAC ECOFIN JHA EPSCO Women 2009 34.3 2011 28.5 2013 69.2 2015 44.6 2017 31.7 Mean 45.4 Std. 27.4 Dev Men 2009 32.6 2011 40.5 2013 71.7 2015 46.2 2017 34.6 Mean 54.1 Std. 24.3 Dev Total Std. Dev

52.9 24.4

COMPET

TTE AGRIFISH ENVI EYCS

0.0 27.8 57.1 83.3 33.3 55.6 36.8

33.0 52.7 73.7 58.2 24.9 60.2 27.8

40.1 46.4 61.2 40.9 38.6 48.9 26.7

12.7 67.4 53.3 85.7 52.1 63.5 35.0

35.0 31.0 72.7 56.5 56.1 60.9 27.1

29.1 26.2 60.0 64.3 49.3 53.3 28.5

15.3 45.1 73.3 40.0 38.9 50.1 28.1

27.3 42.3 66.7 47.1 37.5 49.3 26.0

38.1 33.9 78.3 46.1 43.0 53.5 25.7

47.9 39.1 74.1 43.1 35.1 52.1 23.6

53.3 30.4 74.8 46.0 41.1 55.2 23.5

49.3 40.6 83.3 43.1 37.1 27.0 23.3

54.7 20.3 81.8 47.2 18.0 54.8 23.6

52.9 39.6 72.0 43.5 22.2 53.4 25.5

41.1 47.9 72.0 59.8 16.3 58.8 25.7

42.5 27.2 75.2 53.6 22.2 54.0 25.9

53.9 25.3

54.3 23.6

53.0 23.6

58.9 22.0

56.9 22.6

54.1 24.6

56.6 23.7

52.9 24.7

Notes Configuration abbreviations: FAC: Foreign Affairs ECOFIN: Economic and Financial Affairs JHA: Justice and Home Affairs EPSCO: Employment, Social Policy, Health, and Consumer Affairs COMPET: Competitiveness TTE: Transport, Telecom, and Energy AGRIFISH: Agricultures and Fisheries ENVI: Environment EYCS: Education, Youth, Culture, and Sport Means are calculated with data from 2009–2017.

a standard deviation of 0.34, the maximum of the ratio is about 1.5 (December 2017), and the minimum is about 0.04 (January 2016). While a mean around 1 may suggest that men and women are exiting at similar rates overall, the graph clearly shows that these trends are time-sensitive. Through 2014, the trend suggests that, on average, more men than women are exiting configurations. In the 2014–2017 period, however,

122

A. S. ALDRICH AND L. K. PEREZ

1.5 1 .5 0

Ratio of Female Turnover to Male Turnover

2

Monthly Turnover Ratio

Jan2009

Jan2011

Jan2013 Date

Jan2015

Jan2017

Fig. 5.3 Ratio of Female Turnover to Male Turnover in the Council

this trend appears to reverse, with more women, on average, exiting than men. However, averaging across configurations may obscure even more important variation across policy areas. Thus, in addition to looking at the aggregate ratio, Fig. 5.4 contains the ratio of turnover by configuration across time. This figure shows that the lowest ratio of turnover between women and men is in the Economic and Financial Affairs configuration (mean = 0.08). This configuration experiences the highest turnover rates of men relative to women, but, as Fig. 5.2 has shown, this configuration also has a very low number of women in general. In contrast, this turnover ratio appears highest in the Employment, Social, Health & Consumer Affairs configuration (EPSCO) (mean = 0.65), the Environment configuration (mean = 0.4), and the Education, Youth, Culture & Sport configuration (EYCS) (mean = 0.55). In order to examine how these dynamics affect the policy-making process, we choose to focus on three possible steps in codecision, which is the EU’s ordinary legislative procedure. Under codecision, the European Parliament and the Council can go through up to three rounds of amendments in order to reach agreement. Therefore, the first possible outcome we consider is whether or not the Council accepts the European Parliament’s position at first reading. This outcome means that there

LOSING WOMEN, LOSING POWER? …

ECOFIN

JHA

EPSCO

COMPET

TTE

AGRIFISH

ENVI

EYCS

0 .5 1 1.5 2

0 .5 1 1.5 2

FAC

123

0 .5 1 1.5 2

Ratio of Female Turnover to Male Turnover

5

Jan2009 Jan2011 Jan2013 Jan2015 Jan2017

Jan2009 Jan2011 Jan2013 Jan2015 Jan2017

Jan2009 Jan2011 Jan2013 Jan2015 Jan2017

Date

Fig. 5.4 Ratio of Female Turnover to Male Turnover in the Council by Configuration

is a consensus both within the Council and between the Council and the European Parliament. In this case, the legislative act will be adopted. We believe that having and retaining women in the Council should make this outcome more likely. However, if the Council offers further amendments to the EP’s position, then the legislation goes to a second reading, which is our second dependent variable. If a draft legislative instrument makes it this far, it means the EP and the Council did not agree during the first stage. If women are better able to find consensus and usher legislation through the policy-making process, then legislation should be less likely to need a second reading when there are more women involved. Finally, if the EP and the Council reach an agreement at any stage, the final step is to sign the legislation. Our third dependent variable codes for whether the legislation was ever signed. Again, we think that women are particularly likely to foster agreement and increase the likelihood of this outcome. Since our dependent variables are all dummy variables that take

124

A. S. ALDRICH AND L. K. PEREZ

a value of 1 if the legislation reaches that stage and a 0 if it does not, we use logistic regression. We use Häge’s (2011) EUPOL dataset, which was constructed by pulling information on the various stages of the legislative process from the EU’s online legislative database, Eur-Lex. From this, we retain all legislation decided under codecision that started after 2008 and ended before 2018. Because most EUPOL data stops in 2013 and ends entirely in 2014, we focus on legislation passed between 2008 and 2014. We match these data with our turnover data using the date of the final Council action, so that our measure of turnover indicates the quantity of turnover in the final year of the Council’s involvement with the proposal. We also matched by configuration based on which configuration deals with the legislation last. This generates a total of 261 observations. We develop two sets of models. The first set uses the ratio of female turnover to male turnover as the critical independent variable. The second set uses the number of female ministers that exited the configuration of interest and the number of male ministers that exited the configuration of interest. In all models, we control for the number of female ministers per configuration, which we expect should help consensus-building and therefore make approval of the EP’s position at first reading and signature more likely, while making it less likely that legislation needs to go to a second reading. We also include the number of male ministers per configuration. Next, we include two variables for the type of legislation. These variables control for whether a legislative proposal is a directive or a regulation, leaving decisions as the reference category. Finally, we include a variable for the duration of the legislative process, which measures the number of days between the first act in the legislative process and the last. We include clustered standard errors by configuration.

Results The results support our expectations that losing women makes it harder to conclude the legislative process, both in terms of the Council approving the Parliament’s position at first reading and in terms of getting legislation signed. Additionally, having more women on the Council helps to reach these terminal points, providing further evidence that women’s presence promotes throughput legitimacy. We find some evidence of these relationships using the ratio of female turnover to male turnover, while we find

5

LOSING WOMEN, LOSING POWER? …

125

stronger evidence when we look separately at the number of female and male ministers that exit each configuration. The results for the ratio of female turnover to male turnover are presented in Table 5.2. As the ratio of female exits to male exits increases, the Council becomes less likely to approve legislation at first reading and the likelihood that legislation will be signed decreases. However, this relationship is only significant at the 0.1 level (p = 0.056 for approval at first reading, and p = 0.076 for signatures). When the ratio is at its minimum (when female turnover is very low compared to male turnover), we see that the predicted probability of both approval at first reading (see Fig. 5.5b) and signature (see Fig. 5.6b) are about 0.7.3 If we consider the non-zero turnover ratio minimum of 0.0625, which occurs when at least one woman has exited, the predicted likelihood of a signature is slightly higher (0.7295) than the predicted likelihood of approval at first reading (0.6803). The predicted probability of signature remains slightly higher Table 5.2 Turnover Ratio and Legislative Outcomes Turnover ratio Female ministers Male ministers Directive Regulation Duration Constant Observations Pseudo R squared Wald χ2 (7) Prob > χ2

(1) Approval at 1st reading

(2) Second reading

(3) Signature

−2.019* (1.057) 0.078* (0.045) −0.042 (0.032) −0.188 (0.694) −0.006 (0.642) −0.000 (0.001) 1.512 (1.122) 261 0.050

−3.505 (3.265) 0.065 (0.095) −0.006 (0.050) −5.807*** (0.581) −3.921*** (0.837) 0.014*** (0.004) −7.071*** (2.589) 261 0.643

−1.870* (1.053) 0.081* (0.044) −0.034 (0.033) −1.059 (0.693) −0.840 (0.682) 0.002*** (0.000) 1.087 (1.213) 261 0.088

5.595 0.470

14035.241 0.000

57.925 1.186e−10

Notes Robust standard errors in parentheses. *** denotes significance at the 0.01 level; ** at the 0.05 level; and * at the 0.1 level

126

1

1

A. S. ALDRICH AND L. K. PEREZ

female

.8 .6 .4

0

0

.2

Pr(Approval at fist reading)

.8 .6 .4 .2

Pr(Approval at fist reading)

male

0

5

10 (a) Total Turnover

15

0

20

.2

.4 .6 .8 1 (b) Gender Ratio of Turnover

1.2

1.4

1.2

1.4

1

female

1

Fig. 5.5 Predicted Probabilities for Approval at First Reading

0

0

.2

.2

Pr(Signature) .4 .6

Pr(Signature) .4 .6

.8

.8

male

0

5

10 (a) Total Turnover

15

20

0

.2

.4

.6

.8

1

(b) Gender Ratio of Turnover

Fig. 5.6 Predicted Probabilities for Signature

than that of approval at first reading across turnover ratios. As the ratio increases to its maximum (1.44)—when women, that is, are exiting their positions at almost 1.5 times the rate of men—the predicted probability of a signature drops to 0.1532, and the probability of approval at first reading drops to 0.1156. A very similar relationship occurs when we look at the raw number of female ministers who exit a configuration (Table 5.3). As more women leave the Council, the likelihoods of approval at first reading and signature both decrease. When zero women leave the configuration, the predicted probability of approval at first reading is 0.8514; when the observed maximum of fourteen women exit, the probability drops to 0.0341 (see Fig. 5.5a). Similarly, when no women leave, the results suggest that that

5

LOSING WOMEN, LOSING POWER? …

Table 5.3 Number of Ministers Turned Over and Legislative Outcomes Male ministers turned over Female ministers turned over Female ministers Male ministers Directive Regulation Duration Constant Observations Pseudo R squared Wald χ2 (7) Prob > χ2

127

(1) Approval at 1st reading

(2) Second reading

(3) Signature

−0.157*** (0.047) −0.364*** (0.111)

0.242** (0.109) −0.293 (0.401)

−0.189** (0.076) −0.393*** (0.091)

0.170*** (0.053) 0.019 (0.032) 0.035 (0.652) 0.046 (0.555) −0.001 (0.001) 1.737* (0.962) 261 0.103

0.062 (0.143) −0.056 (0.060) −6.148*** (0.595) −3.947*** (0.938) 0.014*** (0.004) −8.712*** (2.684) 261 0.659

0.192*** (0.047) 0.036 (0.044) −0.854 (0.734) −0.806 (0.596) 0.002*** (0.001) 1.453 (0.914) 261 0.157

54.435 1.93e−09

34185.329 0.000

201.684 5.048e−10

Notes Robust standard errors in parentheses. *** denotes significance at the 0.01 level; ** at the 0.05 level; and * at the 0.1 level

legislation will have a 0.8857 probability of getting signed. This figure drops to only 0.0306 when fourteen women leave (see Fig. 5.6a). We do not observe a case when no men exit, but at the observed minimum of five, the predicted probability of approval at first reading is 0.8022, and the likelihood of signature is 0.8604. When fourteen men leave, the predicted probability of both outcomes is much higher than it is when fourteen women leave: it is 0.4969 for approval at first reading and 0.5305 for signature. In fact, even at the observed maximum for men, the probability of both outcomes is not as low as it is when fourteen women leave. When twenty-three men turn over, the chance of approval at first reading is 0.1939, and the chance of signature is 0.1716. In contrast, only about nine women have to leave for the predicted probabilities to be that low.

128

A. S. ALDRICH AND L. K. PEREZ

Having more women participating in a configuration also increases the likelihood that the Council will approve the EP’s position at first reading and that the legislation will be signed, thereby providing further evidence that higher levels of women’s representation facilitate progress toward the two different conclusion points in the legislative process. This finding holds across the models that use turnover ratio (significant at the 0.1 level) and those that include the raw number of male and female ministers that exit (significant at the 0.001 level). In the models with the turnover ratio, when there is only one female minister, we predict that there will be a 0.3629 probability of approval at first reading and a 0.3931 probability of signature. However, when there are thirty women (the maximum in our data), these probabilities rise to 0.8452 and 0.8729, respectively. The change in probabilities is even greater in the models that use the raw number of male and female ministers that exit a configuration at a given time. Here, when there is one female minister, the predicted probabilities are 0.1919 for approval at first reading and 0.1855 for signature. In contrast, when we have thirty women, we predict a 0.9702 chance of approval at first reading and a 0.9837 chance of signature. The third outcome that we consider—whether legislation makes it to the second reading—is not as well predicted by the quantity of turnover or female representation in the Council. The only turnover variable that is significant is the number of male ministers that exit, and this relationship is positive. Since legislation only makes it to a second reading if there was no agreement at first reading, legislation that makes it to this stage has taken longer than much other legislation. In our data, less than eight percent of legislation makes it to this stage. As such, since we expect male turnover to also add inefficiencies and complications to the legislative process, it makes sense that such turnover might make it more likely that legislation goes to a second reading. When only five men exit a configuration, the chance that legislation goes to a second reading is predicted to be quite low at only 0.0002, but when twenty-three men leave, the predicted chance goes up to 0.0152. The control variables were much more relevant in the model predicting whether legislation goes to a second reading than for either of the other models. In models using both turnover measures, directives and regulations are both significantly less likely to go to a second reading than decisions, our reference category. We also find that the longer a piece of legislation has been under consideration, the more likely it is to go to

5

LOSING WOMEN, LOSING POWER? …

129

a second reading, which makes sense since this process takes longer. A longer duration of consideration also makes it more likely to get signed.

Conclusions and Future Research Our gendered analysis of the Council shows that raw numbers and “portfolio patterns” are similar to those that have been observed in executives more broadly and in the Council’s committee of permanent representatives. During our period of observation, less than a third of ministers serving in the Council—only 27.45%—were women. However, we do find some evidence that the number of women has increased in recent years, reaching 30.49% in 2017. We also find that women are more likely to be present in some of the “softer” policy areas, particularly the Environment configuration and the Agriculture & Fisheries configuration. In contrast, the fewest women are present in the “hard” configurations of Foreign Affairs, Economic & Financial Affairs, and Justice & Home Affairs. Overall, the data support our expectations that having and retaining more women help to facilitate the legislative process—they help to bring the legislative process to a conclusion, both in terms of reaching agreement at first reading and of getting final legislation signed. In chapter 4 of this volume, Perez and Scherpereel find that turnover in the Council weakens throughput legitimacy: greater quantities of turnover are associated with a longer legislative process. Here, we find that the quality of turnover matters as well, that the relationship between turnover and throughput legitimacy has an important gendered dimension. Losing women decreases the Council’s ability to achieve its policy goals. When it loses female ministers, the Council becomes less likely to reach consensus and to bring the legislative process to a successful conclusion. When women are retained, by contrast, they are able to build up experience, relationships, and institutional knowledge. These resources allow them to maximize their consensus-building skills in ways that help the Council to reach internal and inter-institutional agreements. We observe this same effect when there are more women in each Council configuration. These findings suggest that high replacement (turnover) of women would work at cross-purposes for populist politicians. Populist critics seek to replace current elites with populist challengers; the replacement of any current elite, including a sitting woman, will be a populist victory. On the other hand, populists also tend to be displeased with the status quo, so

130

A. S. ALDRICH AND L. K. PEREZ

higher turnover, especially among women, would make it harder for them to pass new legislation and threaten the status quo. However, they might still prefer this outcome to continued policy-making by the existing elites. The implication of our results for the feminist critique is more clearly positive. Having more women and having them in the Council for longer periods of time (e.g., via lower rates of turnover among female legislators) promotes throughput legitimacy. When there are more women in the system and lower rates of turnover among women, institutions tend to find agreement more quickly and to bring the legislative process to a conclusion. We have proposed that women’s tendency to use collaboration is the mechanism that drives these effects, especially in such a consensus-driven institution as the Council. That mechanism, if accurate, bodes well for throughput legitimacy. It suggests that more voices and ideas may be considered during the policy-making process, rather than one group’s initial ideas simply triumphing. In addition, having and retaining more women may promote the attitudinal dimension of input legitimacy via descriptive representation. The female half of the population is likely to feel more represented by EU institutions when levels of women’s representation go up and remain high. Clark and Scherpereel (chapter 3 of this volume) find some evidence of descriptive representation’s effects on the behavioral dimension of input legitimacy; future efforts might study its effects on the attitudinal dimension in greater depth. Further research will also be necessary to uncover any associations between gendered turnover and output legitimacy—does having more women in EU institutions (and retaining them over time) lead to better quality policies, and especially to better substantive representation for women? To the extent that new policies are an improvement on the status quo, however, our results suggest that gaining and retaining female members of the Council increase the likelihood of passing legislation and doing so quickly, which would also contribute to the EU’s output legitimacy.

Notes 1. The number of ministers displayed in Fig. 5.1 and discussed in the text is higher than many textbook presentations of the Council would lead one to expect. If each member-state were represented by one minister in each Council configuration, the total number of ministers in an EU of (say) 27 members would be 27 times the total number of Council configurations.

5

LOSING WOMEN, LOSING POWER? …

131

In reality, the number of ministers participating in Council business varies significantly across countries and over time. One prime minister may split responsibility for education, youth, culture, and sport among four or more ministers; another may split responsibility for those sectors among one or two ministers. It is not unusual for two or more national ministers to be involved in sectors which, within the Council’s administrative order, fall under the umbrella of a single configuration. 2. In order to identify which ministers are present in a given month, we use data on ministers’ names from the CIA’s monthly “Chiefs of State and Cabinet Members of Foreign Governments” reports. Ministers are also counted as having turned over when they switch to a portfolio in a different configuration, since ministers in this situation will need to learn a new policy field and work with a new set of colleagues. 3. For all predicted probabilities, including in Figs. 5.5 and 5.6, the other variables are held at their means.

References Aldrich, Andrea S. 2018. National Political Parties and Career Paths to the European Parliament. JCMS: Journal of Common Market Studies 56(6): 1283–1304. Aldrich, Andrea S., and William T. Daniel. 2019. The Consequences of Quotas: Assessing the Effect of Varied Gender Quotas on Legislator Experience in the European Parliament. Politics & Gender. https://doi.org/10.1017/S17439 23X19000291. Alvarez, Lizette. 2000. Feminine Mystique Grows in Senate - The New York Times. New York Times December 7. Accessed June 8, 2020 at https://www. nytimes.com/2000/12/07/us/feminine-mystique-grows-in-senate.html. Annesley, Claire, Karen Beckwith, and Susan Franceschet. 2019. Cabinets, Ministers, and Gender. New York: Oxford University Press. Anzia, Sarah F., and Christopher R. Berry. 2011. The Jackie (and Jill) Robinson Effect: Why Do Congresswomen Outperform Congressmen? American Journal of Political Science 55 (3): 478–493. Atkinson, Michael M., and David C. Docherty. 1992. Moving Right Along: The Roots of Amateurism in the Canadian House of Commons. Canadian Journal of Political Science 25 (2): 295–318. Ban, Carolyn. 2013. Management and Culture in an Enlarged European Commission: From Diversity to Unity? London: Palgrave Macmillan. Barnes, Tiffany D., and Emily Beaulieu. 2019. Women Politicians, Institutions, and Perceptions of Corruption. Comparative Political Studies 52 (1): 134– 167.

132

A. S. ALDRICH AND L. K. PEREZ

Barnes, Tiffany D., and Stephanie M. Burchard. 2013. ‘Engendering’ Politics: The Impact of Descriptive Representation on Women’s Political Engagement in Sub-Saharan Africa. Comparative Political Studies 46 (7): 767–790. Beaman, Lori, Raghabendra Chattopadhyay, Esther Duflo, Rohini Pande, and Petia Topalova. 2009. Powerful Women: Does Exposure Reduce Bias? The Quarterly Journal of Economics 124 (4): 1497–1540. Besley, Timothy, Olle Folke, Torsten Persson, and Johanna Rickne. 2017. Gender Quotas and the Crisis of the Mediocre Man: Theory and Evidence from Sweden. American Economic Review 107 (8): 2204–2242. Bratton, Kathleen A., and Kerry L. Haynie. 1999. Agenda Setting and Legislative Success in State Legislatures: The Effects of Gender and Race. The Journal of Politics 61 (3): 658–679. Carey, John, and Matthew Soberg Shugart. 1995. Incentives to Cultivate a Personal Vote: A Rank Ordering of Electoral Formulas. Electoral Studies 14 (4): 417–439. Chalmers, Adam William. 2014. In Over Their Heads: Public Consultation, Administrative Capacity and Legislative Duration in the European Union. European Union Politics 15 (4): 595–613. Christiansen, Thomas. 2001. Intra-Institutional Politics and Inter-Institutional Relations in the EU: Towards Coherent Governance? Journal of European Public Policy 8 (5): 747–769. Christiansen, Thomas, and Simona Piattoni. 2003. Informal Governance in the European Union. Northampton, MA: Edward Elgar Publishing. Dahlerup, Drude. 2006. The Story of the Theory of Critical Mass. Politics & Gender 2 (4): 511–522. Davis, Rebecca Howard. 1997. Women and Power in Parliamentary Democracies: Cabinet Appointments in Western Europe, 1968–1992. Lincoln, NE: University of Nebraska Press. Escobar-Lemmon, Maria, and Michelle M. Taylor-Robinson. 2005. Women Ministers in Latin American Government: When, Where, and Why? American Journal of Political Science 49 (4): 829–844. Escobar-Lemmon, Maria, and Michelle M. Taylor-Robinson. 2009. Getting to the Top: Career Paths of Women in Latin American Cabinets. Political Research Quarterly 62 (4): 685–699. Francis, Wayne L., and John R. Baker. 1986. Why Do US State Legislators Vacate Their Seats? Legislative Studies Quarterly 11 (1): 119–126. Fridkin, Kim L., and Patrick J. Kenney. 2014. How the Gender of U.S. Senators Influences People’s Understanding and Engagement in Politics. The Journal of Politics 76 (4): 1017–1031. Fulton, Sarah A. 2012. Running Backwards and in High Heels: The Gendered Quality Gap and Incumbent Electoral Success. Political Research Quarterly 65 (2): 303–314.

5

LOSING WOMEN, LOSING POWER? …

133

Funk, Kendall D., Magda Hinojosa, and Jennifer M. Piscopo. 2019. Women to the Rescue: The Gendered Effects of Public Discontent on Legislative Nominations in Latin America. Party Politics: 1354068819856614. Golub, Jonathan. 2007. Survival Analysis and European Union Decision-Making. European Union Politics 8 (2): 155–179. Häge, Frank M. 2008. Who Decides in the Council of the European Union? JCMS: Journal of Common Market Studies 46(3): 533–558. Häge, Frank M. 2011. The European Union Policy-Making Dataset. European Union Politics 12 (3): 455–477. Häge, Frank M. 2013. Coalition Building and Consensus in the Council of the European Union. British Journal of Political Science 43 (3): 481–504. Häge, Frank M., and Daniel Naurin. 2013. The Effect of Codecision on Council Decision-Making: Informalization, Politicization and Power. Journal of European Public Policy 20 (7): 953–971. Hayes-Renshaw, Fiona, Wim Van Aken, and Helen Wallace. 2006. When and Why the EU Council of Ministers Votes Explicitly. JCMS: Journal of Common Market Studies 44(1): 161–194. Heath, Roseanna Michelle, Leslie A. Schwindt-Bayer, and Michelle M. TaylorRobinson. 2005. Women on the Sidelines: Women’s Representation on Committees in Latin American Legislatures. American Journal of Political Science 49 (2): 420–436. Heisenberg, Dorothee. 2005. The Institution of ‘Consensus’ in the European Union: Formal versus Informal Decision-Making in the Council. European Journal of Political Research 44 (1): 65–90. Jeydel, Alana, and Andrew J. Taylor. 2003. Are Women Legislators Less Effective? Evidence from the U.S. House in the 103rd-105th Congress. Political Research Quarterly 56 (1): 19–27. Kantola, Johanna. 2010. Gender and the European Union. Houndsmills: Palgrave Macmillan. Kathlene, Lyn. 1995. Alternative Views of Crime: Legislative Policymaking in Gendered Terms. The Journal of Politics 57 (3): 696–723. Katz, Richard S. 1997. Democracy and Elections. New York: Oxford University Press. Kerevel, Yann P., and Lonna Rae Atkenson. 2013. Explaining the Marginalization of Women in Legislative Institutions. The Journal of Politics 75 (4): 980–992. Klüver, Heike, and Iñaki. Sagarzazu. 2013. Ideological Congruency and Decision-Making Speed: The Effect of Partisanship across European Union Institutions. European Union Politics 14 (3): 388–407. Krook, Mona Lena, and Diana Z. O’Brien. 2012. All the President’s Men? The Appointment of Female Cabinet Ministers Worldwide. The Journal of Politics 74 (3): 840–855.

134

A. S. ALDRICH AND L. K. PEREZ

Lewis, Jeffrey. 2010. How Institutional Environments Facilitate Co-Operative Negotiation Styles in EU Decision-Making. Journal of European Public Policy 17 (5): 648–664. Mansbridge, Jane. 1999. Should Blacks Represent Blacks and Women Represent Women? A Contingent ‘Yes.’ The Journal of Politics 61 (3): 628–657. Matland, Richard S., and Donley T. Studlar. 2004. Determinants of Legislative Turnover: A Cross-National Analysis. British Journal of Political Science 34 (1): 87–108. Naurin, Daniel, Elin Naurin, and Amy Alexander. 2019. Gender Stereotyping and Chivalry in International Negotiations: A Survey Experiment in the Council of the European Union. International Organization 73 (2): 469–488. Norris, Pippa, and Joni Lovenduski. 1995. Political Recruitment: Gender, Race and Class in the British Parliament. New York: Cambridge University Press. Novak, Stephanie. 2013. The Silence of Ministers: Consensus and Blame Avoidance in the Council of the European Union. JCMS: Journal of Common Market Studies 51(6): 1091–1107. Perez, Lauren K., and John A. Scherpereel. 2017. Vertical Intra-Institutional Effects of Ministerial Turnover in the Council of the European Union. Journal of European Public Policy 24 (8): 1154–1171. Schalk, Jelmer, René Torenvlied, Jeroen Weesie, and Frans Stokman. 2007. The Power of the Presidency in EU Council Decision-Making. European Union Politics 8 (2): 229–250. Scherpereel, John, A. and Lauren K. Perez. 2015. Turnover in the Council of the European Union: What It Is and Why It Matters. JCMS: Journal of Common Market Studies 53(3): 658–673. Schwindt-Bayer, Leslie A. 2005. The Incumbency Disadvantage and Women’s Election to Legislative Office. Electoral Studies 24 (2): 227–244. Schwindt-Bayer, Leslie A. 2006. Still Supermadres? Gender and the Policy Priorities of Latin American Legislators. American Journal of Political Science 50 (3): 570–585. Shugart, Matthew. 1994. Minorites Represented and Unrepresented. In Electoral Systems in Comparative Perspective: The Impact on Women and Minorities, eds. Wilma Rule and Joseph Francis Zimmerman. Westport, CT: Westview Press. Studlar, Donley T., and Gary F. Moncrief. 1999. Women’s Work? The Distribution and Prestige of Portfolios in the Canadian Provinces. Governance 12 (4): 379–395. Tallberg, Jonas. 2006. Leadership and Negotiation in the European Union. Cambridge: Cambridge University Press. Tallberg, Jonas. 2008. Bargaining Power in the European Council. JCMS: Journal of Common Market Studies 46(3): 685–708. Tallberg, Jonas. 2010. The Power of the Chair: Formal Leadership in International Cooperation. International Studies Quarterly 54 (1): 241–265.

5

LOSING WOMEN, LOSING POWER? …

135

Thomas, Sue. 1991. The Impact of Women on State Legislative Policies. The Journal of Politics 53 (4): 958–976. Thomson, Robert. 2008. The Council Presidency in the European Union: Responsibility with Power. JCMS: Journal of Common Market Studies 46(3): 593–617. Thomson, Robert. 2011. Resolving Controversy in the European Union: Inputs, Processes and Outputs in Legislative Decision-Making before and after Enlargement. Cambridge: Cambridge University Press. Volden, Craig, Alan E. Wiseman, and Dana E. Wittmer. 2013. When Are Women More Effective Lawmakers Than Men? American Journal of Political Science 57 (2): 326–341. Warntjen, Andreas. 2008. The Council Presidency: Power Broker or Burden? An Empirical Analysis. European Union Politics 9 (3): 315–338. Warntjen, Andreas. 2010. Between Bargaining and Deliberation: DecisionMaking in the Council of the European Union. Journal of European Public Policy 17 (5): 665–679. Wolbrecht, Christina, and David E. Campbell. 2007. Leading by Example: Female Members of Parliament as Political Role Models. American Journal of Political Science 51 (4): 921–939.

CHAPTER 6

The Importance of Expertise: Political Careers, Personnel Turnover, and Throughput Legitimacy in the European Parliament Eugenio Salvati and Michelangelo Vercesi

This chapter analyzes the relationship between personnel turnover, political experience, and throughput legitimacy in the European Parliament (EP). In particular, it argues that turnover affects the likelihood of being selected for EP leadership positions and, consequently, that turnover affects institutional performance. Since the first direct election in 1979—and, especially, since the Lisbon Treaty established codecision as the European Union’s (EU’s) ordinary legislative procedure—the European Parliament has acquired a pivotal

E. Salvati (B) Department of Economics and Management, University of Pavia, Pavia, Italy e-mail: [email protected] M. Vercesi Center for the Study of Democracy/Institute of Political Science, Leuphana University Lüneburg, Lüneburg, Germany e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. A. Scherpereel (eds.), Personnel Turnover and the Legitimacy of the EU, Palgrave Studies in European Union Politics, https://doi.org/10.1007/978-3-030-60052-5_6

137

138

E. SALVATI AND M. VERCESI

role in the EU’s political system. The EP’s greater decision-making power has increased the attractiveness of the EP as a career option for ambitious politicians (Whitaker 2014; Daniel 2015; Salvati 2019). The rise of a professional European political class and lower levels of legislative turnover have made the EP a more institutionalized body (Scarrow 1997; Verzichelli and Edinger 2005; Salvati 2016). The EP’s greater institutionalization has occurred within context of higher politicization of European integration (Salvati and Vercesi 2019). Career paths of MEPs (members of the European Parliament) before entering office can tell a lot about the legitimacy of the EU. First, it is important that European citizens can vote for candidates with some connection to their own constituency. This fact could make voters feel closer to their representatives. This feeling could help to address the EU’s democratic deficit. Second, previous political experience may affect the deliberative quality of the EU and the quality of EU policy outputs. Variations along these dimensions have substantial implications for the accountability and responsiveness of EU institutions (Scherpereel and Perez 2015). So far, scholars have mainly focused on explaining career models. They have mostly focused on the paths that MEPs take to the Parliament and the duration of their stints. The literature’s findings are mixed and limited: overall, Kaeding’s (2005: 100) statement that “puzzles in the internal working of the EP are far from being solved and understood” still holds. This chapter fills some remaining gaps. It theoretically connects MEPs’ political background, personnel turnover, and throughput legitimacy. We do not investigate whether MEPs’ personal attributes (e.g., nationality, party affiliation, policy position) affect policy making. Rather, in line with the volume’s general task, we explore whether the nature of the decision-making process within the EP can be understood as a function of MEPs’ political background and how MEPs’ political backgrounds affect throughput legitimacy. We address three questions. First: are parliamentary insiders more likely to attain EP leadership positions? Second: are insiders more successful in shepherding legislation through the policy-making process? Third: how does personnel turnover affect policy production? We argue that personnel turnover imperils EU legitimacy and that this effect is mediated by the type of previous political experience that newcomers bring to the parliament. We expect higher turnover to be conducive to lower throughput legitimacy, whose proxy is MEPs’ poor performance (measured as longer decision-making process). Our exploratory analysis deals with turnover’s quality aspect and is focused on the micro-individual

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

139

level of this phenomenon (François and Grossman 2015). This chapter contributes to the discussion about rules of self-government within the EP and the efficiency of the EU policy-making process. According to Costa’s (2019: 3–5) classification of topics in the EP literature, our investigation lies at the intersection between organizational studies (internal division of labor), legislative studies (MEP’s profiles and roles), and studies about EP activities (implementation of competences and influence within the assembly). We look at the relationships of delegation and accountability in the EP, in a way that reinterprets these three aspects as pieces of a causal chain running from personnel turnover to political legitimacy outcomes. Our framework is based on the well-known principal-agent model of political relations (e.g., Strøm 2003). The principal-agent model (“P-A model”) is an important heuristic tool that can account for delegations of authority within (and between) EU institutions. It can help analysts to understand why some actors execute their institutional tasks in a disciplined way rather than opportunistically. Moreover, it promotes explanations of why actors endowed with similar positional decision-making resources (i.e., political office, formal power) contribute to policy making in different ways (Delreux and Adriaensen 2017). In particular, scholars have stressed that, under some circumstances, previous professional experience can (or, at least, is expected to) be a powerful predictor of politicians’ performance in office (e.g., Alexiadou 2016; Costello and Thomson 2010; Obholzer et al. 2019). By focusing on committee chairs and rapporteurs, we first review, in the next section, the relevant state of the art. Subsequently, we present our principal-agent framework and derive four main propositions. Third, we present our data basis and our empirical analysis. Finally, we discuss the implications of our findings and present future research outlooks.

Pathways to Power and Policy Making in the EP Over the past years, a core group of specialized and long-serving MEPs has emerged (Salvati 2019). In addition to skilled personnel, the EP has developed a specialized internal division of labor, based on a strong committee system (Bowler and Farrell 1995; Mamadouh and Raunio 2003). The prestige and autonomy of EP committees encourage MEPs and party groups to seek committee leadership positions. These positions allow them to get a firmer grasp on decision-making. Committees

140

E. SALVATI AND M. VERCESI

help MEPs to overcome informational asymmetries. Much of the EU legislation decided via codecision is adopted in the committees via early agreement (Burns 2013), and individuals in committees play key roles in the early agreement process. According to its Rules of Procedure, the EP is organized into 20 standing committees,1 whose members are appointed in proportion to party groups’ levels of representation (McElroy 2006; Yordanova 2009). Party group leaders make nominations, which are ultimately approved by a final vote of the parliament. This process takes place at the beginning of each legislative session and is repeated 2.5 years into each EP term (e.g., at the midpoint of the session). Following the principle of proportional representation of party groups, every committee elects its chair and up to four vice chairs. This leadership selection reflects bargains among (a) party leaders and (b) national delegations. Committee chairs are influential politicians: they lead committee meetings, shape their committee’s agenda, and represent their committee inside and outside of the Parliament (Whitaker 2001). The second outstanding figure of the EP committee system is the rapporteur, an MEP who drafts committee reports on particular legislative proposals and follows proposals through the law-making process (Yoshinaka et al. 2010). Because EP committee positions are important, scholars have sought to identify the determinants of committee assignments. They have focused on (1) the influence (if any) of individual background and interests on committee assignments and (2) the role of parties as gate-keepers. They have generally concentrated on the factors that affect the likelihood that an MEP will be assigned to a committee, but they have focused less on the factors that affect the likelihood of appointment to the two leadership positions (chair, rapporteur) discussed above. Bowler and Farrell (1995) find no evidence that party seniority determines committee assignments; in their gate-keeping action, parties are more concerned with members’ occupational expertise. Mamadouh and Raunio (2003) similarly stress that MEPs aim to enter those committees that best cover their own field of expertise, and national parties tend to adapt to their motivations. Moreover, party loyalty is not an asset to be allocated to the desired committee (Yordanova 2009). However, committees are representative of the general ideological orientation of parties, and MEPs are not usually assigned where they can get the best pay-off for their electoral constituency (Whitaker 2001; McElroy 2006). Overall, the evidence suggests that specialization matters more than party

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

141

profiles: MEPs obtain the preferred position when they have previously been part of the same committee or have a consolidated experience in the EP (Whitaker 2019). Parties appoint committee members who will reduce transaction costs and make policy making more efficient (Krehbiel 1991). While this conclusion holds for committee appointments in general, it does not generally hold for committee chairs. For chairs, existing research suggests that parties’ priorities are central and that individual preferences are marginal (Whitaker 2019). MEPs who have served in EP leadership positions and/or had prominent political careers are more likely to be appointed to a chairship (Salvati 2019). A more complex configuration of factors seems to account for the selection of rapporteurs. Factors such as party affiliation, nationality, and policy preferences make the overall distribution “disproportional” when compared to the whole plenary (Kaeding 2004, 2005; Høyland 2006). More than committee chairs, rapporteur appointments are determined by a mixture of party preference and individual expertise. According to Yoshinaka et al. (2010: 477), this reflects the fact that the rapporteur has to be “a consensus builder in a setting that is multiparty with constant, ongoing negotiations among groups.” Thus, rapporteur selection is a delicate process, characterized by tough competition between party groups to get the assignment of the most important reports. Rapporteurs are chosen, based on a system of quota points that single party groups possess according to their size in parliament. Confronted with a specific issue to be discussed, each group bids for a report in the guise of auction and tries to win vis-à-vis competitors. Party groups can also follow a consensual decision-making process and agree on future reports. Moreover, it is worth noting that the proposal of a recognized expert in the relevant legislation field is usually pointwise cheaper (Corbett et al. 2007).2 A rapporteur is responsible for preparing and monitoring the evolution of a legislative text/draft from cradle to grave (Mamadouh and Raunio 2003).3 When the text is finally adopted by the EP, it is transmitted to the Council of Ministers as the parliament’s official position. Rapporteurs lead inter-institutional negotiations with the Council and Commission at different stages of the process (i.e., early agreements; trilogues; conciliation committee), and their work takes place in both formal and informal venues (Costello and Thomson 2010, 2011; Obholzer et al. 2019; Salvati 2019).

142

E. SALVATI AND M. VERCESI

The bulk of the literature on rapporteur assignments seeks to disentangle the relationship between rapporteurs and (national and European) parties (Kaeding 2004). Høyland (2006), for example, finds that MEPs from parties that are currently represented in the governments of member-states receive more reports than others. Yoshinaka et al. (2010) and Yordanova (2011) show that MEPs who are more loyal to their party group are more likely to be selected as rapporteur. Costello and Thomson (2011), for their part, find that MEPs with policy positions that are closer to the position of the median MEP of the whole parliament are more likely to receive rapporteurships. A second strand of research investigates rapporteurs’ influence on policy making and policy outputs. Studies in this tradition find that rapporteurs can use their informational advantage to control legislation, boost coordination, and reduce transaction costs (Costello and Thomson 2010, 2011; Salvati 2019; Thierse 2019). They find, furthermore, that individual traits affect bargaining capacity, especially when the rapporteur is backed (and constrained) by a parliamentary majority. Moreover, a rapporteur is more consequential when she represents the position of the median MEP and does not hold any leadership position within the EP and/or in the party group (Costello and Thomson 2010, 2011). In addition, two further conditions increase rapporteurs’ influence on decision making: either the EP has a consultative role or the dossier is still in the early agreement phase of the codecision procedure. This is because other MEPs are less involved in the “bargaining” process and/or suffer from information asymmetry (Costello and Thomson 2010; Ringe 2010; Roger and Winzen 2015). Rapporteurs’ specialization and the consequent policy-making quality are affected by personnel turnover. We know that seniority matters for the allocation of reports, endowing senior MEPs with more power (Daniel 2013). As Daniel (2013: 845) observes: “committee report allocation has become even more selective since the advent of codecision”; this can be interpreted as an indicator of the EP’s need to rely on expert members. Turnover is thus potentially dangerous: in this regard, Scherpereel and Perez (2015; also see Chapter 1 of this volume) have identified three sources of perils: (1) amateurism, due to politicians’ lack of skills; (2) policy’s poor quality, which results from policy-makers’ inexperience and technical inability; (3) disunity, which diminishes solidarity and coordination between the members of an organization. These three phenomena negatively affect the efficiency of the EP’s functioning (Häge and Ringe

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

143

2019). In case of higher turnover, party gate-keepers face a dilemma between concentrating responsibilities in the hands of a few skilled rapporteurs, on one hand, and risking longer decision-making processes and suboptimal outcomes, on the other. When policy making is controlled by a restricted number of MEPs, cooperative interrelationships in the EP are affected. Participation is reduced, and throughput legitimacy can potentially shrink (Häge and Ringe 2019: 230).

Linking Personnel Turnover to Institutional Performance Theoretical Framework Our theoretical framework is based on the principal-agent model (“P-A model”). The P-A model, while essentially empirical, has some features that “resemble elements that are found in political theory, where they form the building blocks for a legitimacy-related argument” (Brandsma and Adriaensen 2017: 37). For example, Piattoni (2013) concludes that the organization of principal-agent relationships affects legitimacy in the EU. She stresses that higher levels of deliberative quality in policy making processes increase the throughput legitimacy of particular EU institutions. The EU is a complex supra-national institutional setting, though, which is characterized by parallel chains of delegation. This fact makes accountability harder to institutionalize and can undercut the legitimacy of the EU as a whole (Kassim and Menon 2003: 133–135; Vaubel 2006). The basic theoretical argument is that principals select agents to fill political offices. Principals delegate political tasks to agents to overcome transaction costs, counterbalance the lack of information, and solve collective action problems. Principals expect agents to behave in line with principals’ will. Principals risk agency loss, however, which can result from adverse selection and/or moral hazard/agency rent (Dowding and Dumont 2015: 12–17; Lupia 2003). For these reasons, principals develop both ex-ante (to select the right agent) and ex-post (to “keep agents on track” and/or fire them) screening mechanisms to make agents accountable. Here, we seek to unravel the shortcuts that principals use to pick the best agents and minimize risks. We also seek to understand the ways that principals’ selections affect agents’ performance. When selecting agents, principals can follow distributive, informational, and/or partisan logics. The distributive approach suggests that principals

144

E. SALVATI AND M. VERCESI

allocate posts, based on agents’ desire to occupy positions that enable them to serve their constituents’ interests (Shepsle 1979). However, principals can also select agents whose expertise best fits with the policy sector of competence (Krehbiel 1991). Finally, from a partisan perspective, principals select agents when agents guarantee party reliability, party unity, and party control over policy making (Cox and McCubbins 1993). Party leaders encourage agents’ loyalty by gate-keeping future candidacies and affecting career promotions (Sieberer and Müller 2017). In the EP, the distributive logic especially applies to simple membership of collective bodies, such as parliamentary committees. In this case, the partisan logic appears weak (e.g., Yordanova 2009). As for single-incumbent positions such as committee chairs and vice-chairs, as well as group coordinators, the informational logic gains greater prominence (e.g., Daniel and Thierse 2017; Treib and Schlipphak 2019) together with partisan reasons. This is because agents selected to these positions have two responsibilities: (1) to coordinate intra-parliamentary decision-making efficiently, and (2) to act as a transmission belt between specialized committees and generalist party groups (Ringe 2010: 25). Party coordination is central even at the committee stage (Roger and Winzen 2015), and the policy preferences of prospective agents play a crucial role in guiding the selection of rapporteurs (Obholzer et al. 2019). Expectations4 Even when from a formal viewpoint they are not the prime principal, political parties remain actual “kingmakers” in the EP’s internal organization (Raunio 2006; Yordanova 2013). However, parties value professional experience differently for committee chairs and rapporteurs. Chairs and rapporteurs can both be seen as agents, but parties expect them to play different roles and provide different types of accountability. Parliamentary committees are, by definition, accommodating bodies. In committees, cooperation is the rule more than in the plenary (Curini and Zucchini 2015). Compromises between parties are requested, and the committee as a whole is in charge of selecting its own chair, who should be able to adjust different party and individual views. Because of chairs’ mediating and representative role, the MEPs that parties select for chairships are those who will be likely to “disempower” party conflicts and to manage committee works less divisively. Rapporteurs focus on shaping policies rather than coordinating activity. They “occupy an uncomfortable middle

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

145

ground” between committees and parties (Yoshinaka et al. 2010: 477) (Fig. 6.1). National parties decide about rapporteurs and are driven by the will to control very specific dossiers over the course of the legislative process. Because of rapporteurs’ centrality in the definition of EU legislation, national parties need extremely reliable and effective agents, who support national parties’ priorities. On one hand, rapporteurs need knowledge of the institutional dynamics (Daniel 2013; Hurka et al. 2015). On the other hand, they are vehicles through which parties impose their policy preferences. Parties mediate between committees and rapporteurs in the selection process through party coordinators. They check rapporteurs’ positions in advance, using individual party affiliations and national political profiles as proxies (e.g. Samuels and Shugart 2010). They play a greater role in delegating power to rapporteurs than to committee chairs. Therefore, our first expectation (institutional experience proposition) is that:

Fig. 6.1 Delegation and Accountability in the EP [2019: 243] and own elaboration)

(Source Obholzer et al.

146

E. SALVATI AND M. VERCESI

Proposition 1a:

previous knowledge of the EP is more important for the selection of committee chairs than it is for the selection of rapporteurs.

However, parties aim to maximize both party reliability and expertise of rapporteurs (while minimizing the screening costs) to be sure that the EP implements their ideal points when important decisions are at stake. Therefore: Proposition 1b:

parties delegate more frequently to an inner group of rapporteurs with greater knowledge of the EP relative to other rapporteurs.

Personnel turnover affects the process of selecting rapporteurs because it reduces the breadth of the pool of eligible persons with previous experience in the EP. So far, we have focused on the “delegation side” of the chain and the main reasons behind principals’ choices. If we look at the “accountability side,” rapporteurs are expected to provide competence. Party principals want tasks to be executed efficiently and successfully. If we consider a new entry in the EP as an indicator of personnel turnover and assume that previous experience enhances in-office performance (see Müller-Rommel et al. 2020), our second expectation (incumbency proposition) is that: Proposition 2:

incumbent MEPs perform better than incoming MEPs when selected as rapporteurs.

Both reelected and new entries in parliament can have already occupied a seat during legislative terms prior to the antecedent one (e.g., an MEP at t 2 could have been MEP at time t, time t 1 , or both). The longer one remains in an institution, the more likely she is to acquire institutional experience. Thus, our third expectation (longer experience proposition): Proposition 3:

MEPs who have spent more time in the EP prior to being selected as rapporteurs perform better.

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

147

Finally, we observe that those without previous experience in the EP do not all have the same institutional expertise. Some are full political outsiders, while others enter the EP with experience at the national and/or sub-national level. Since national experience is an important source of political capital to spend in inter-party bargaining at the EU level (Aldrich 2018), we hypothesize (national experience proposition) that: Proposition 4:

EP outsiders with previous political experience at the sub-European level perform better as rapporteurs than full outsiders.

Before testing our expectations, we discuss variable operationalization and present our evidentiary base.

Research Strategy Operationalization Our propositions contemplate four independent and three dependent variables. The former refer to political experience. We argue that principals select different profiles, based on the political office to be filled and the tasks that the agent is expected to fulfill. We operationalize previous experience as the tenure of specific institutional and/or party roles before being elected to the EP. In this regard, we proxy previous knowledge of the EP by determining whether a politician has (=1) or has not (=0) already been an MEP before the observed legislative term. Incumbency is straightforward and directly tackles the turnover issue: we ask whether a sitting MEP did (=1) or did not (=0) hold a seat in the previous session. Time spent in the EP records the number of previous EP legislative terms in which the MEP held a seat. Finally, we use an ordinal measure to operationalize experience at the sub-European level. Our ten-point ascending scale ranks the importance of (1) regional party leadership; (2) local MP/minister (for municipalities with more than 20,000 inhabitants); (3) mayor (for municipalities with more than 100,000 inhabitants); (4) regional MP; (5) regional minister; (6) regional chief executive; (7) member of the national party leadership; (8) national MP; (9) senior/junior minister; and (10) prime minister. The final value is the simple mean of the “national score” and the “regional score.” This

148

E. SALVATI AND M. VERCESI

approach allows us to “weigh” the national level more than the subnational one, while continuing to afford both levels due consideration. We focus on the most recent office held before being elected to the EP at national/sub-national level. When two or more positions overlap, we consider the most important one; where applicable, both parliamentary houses are considered for MPs. Our first dependent variable, leadership position in the EP, codes two things: whether an MEP was (=1) or was not (=0) appointed to a committee chairship, and whether an MEP was (=1) or was not (=0) appointed as a rapporteur. Second, we measure the frequency of selection for rapporteurships as the number of reports an MEP receives in one legislative term. Third, performance is considered as a proxy of throughput legitimacy. If we posit that throughput legitimacy can be conceptualized as efficiency, efficacy, or inclusiveness of decision-making (Schmidt 2013), the viable indicator of performance we choose is the length in days of the process to adopt a Commission’s legislative proposal in which a rapporteur is involved (we do not consider rejected and withdrawn proposals).5 Data Basis and Coverage Information about previous political experience and institutional roles are drawn from Political Careers and Parliamentary Roles of the Members of the European Parliament, 1979–2019 (Salvati 2019), an original dataset covering all MEPs in office for at least one month and one day, including replacements (the total number of observations is thus higher than the number of seats).6 Data come from MEPs’ personal websites, MEP’s profiles on the EP’s website (www.europarl.europa.eu), and the Vote Watch Europe databank (www.votewach.eu). Information about legislative proposals and the length of the decision-making process come from The Informal Politics of Codecision dataset (Bressanelli et al. 2014, 2016; Reh et al. 2020). Finally, we use the Legislative Observatory of the European Parliament (https://oeil.secure.europarl.europa.eu/oeil/ home/home.do) to link legislative proposals to the relevant rapporteurs. We analyze the allocation of committee chairships and rapporteurships for two legislative terms, from 2009 to 2019. This focus allows us to “fix” third conditions, which are a potential source of noise. Our observation period starts one legislative term after the 2004 EU enlargement and the 2007 entry of Bulgaria and Romania. Thus, MEPs from all “new”

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

149

member-states (except for Croatia, which joined the Union in 2013) had a chance to get previous experience in the EP. Moreover, the temporal compactness of the period limits the impact of possible long-term effects on the composition of the EU parliamentary class and the institutionalization of the EP (Salvati 2019). This makes appointments less likely to be affected by the organizational uncertainty and instability that are typical of underinstitutionalized bodies (Musella and Vercesi 2019). Finally, in both legislative terms, a plurality of MEPs were affiliated with the European People’s Party (EPP), and the second largest group was the Progressive Alliance of Socialists and Democrats (S&D). Together, these two groups occupied 61% (2009) and 54.8% (2014) of EP seats at the time of parliamentary inauguration. For information about the legislative process, we rely on the available data on the ordinary legislative procedure for the same time span, starting from the first meeting of the EP’s 7th EP session (July 14, 2009). EURLex codes the earliest procedure we analyze as COM(2009)0382; the latest procedure we cover is COM(2016)0731. Overall, we consider the political backgrounds of 817 MEPs for 2009– 2014 and 766 MEPs for 2014–2019 (total N = 1583), and we check the features of 637 legislative procedures. Figure 6.2 provides aggregate descriptive statistics about the distribution of MEPs by personnel turnover, institutional role in the EP, and gender. Turnover at the parliamentary level involved the majority of seats (55.8%). The overall value is in line with disaggregated numbers. For the 7th term (2009–2014), the level of turnover was 55.5%, and 17.6% of MEPs had no political experience before being elected. For EP 8, 56% of seats turned over, and 20% of MEPs were political outsiders. Women (35.7% of all MEPs) are underrepresented among MEPs, and most MEPs held at least one political office at the sub-European level before entering the Parliament. In the full sample, over 70% of MEPs received at least one report.

Empirical Analysis Institutional Experience Proposition According to our first proposition, knowledge of the functioning of the EP is more valued as a criterion for chair appointment than it is for rapporteurship appointment. We assess this expectation by comparing

150

E. SALVATI AND M. VERCESI

Fig. 6.2 MEPs by Turnover, Gender, Experience, and Role, 2009–2019 (Note N = 1583. Numbers above bars are percentages. Reelection refers only to the legislative term before the one at issue. A committee chair can be also selected as rapporteur)

the ratio of committee chairs and rapporteurs who held a seat in the EP before being elected for the observed legislative term. Moreover, we seek to understand whether there is a core group of rapporteurs who receive a particularly high number of reports over a legislative term. We expect these rapporteurs to be more similar to committee chairs in their level of EP experience, given the higher sensitivity of their task. For this purpose, we observe the background of rapporteurs by number of reports, checking differences’ robustness (Vercesi and Grimaldi 2019). In this regard, it is worth noting that our theoretical argument is not deterministic, but tendential. Findings should be hence understood as relative. For the comparison between committee chairs and rapporteurs, we exclude from calculations those rapporteurs who received only one report in one legislative term (442 individuals). This allows cancelling distortions due to “flash cases” whose lack of experience in the EP could be the result of a limited pool of agents (because of the high turnover rate in the parliament) rather than a deliberate principal’s choice. This means that we deal with 668 rapporteurs (60.2% of the total N ); only 166 of them (i.e., 24.9% of 668 and 15% of the total) received more than four reports in

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

151

one legislative term. Table 6.1 shows differences both by position and number of reports (in this latter case, all rapporteurs are counted). As expected, previous experience in the EP counts more in the selection of committee chairs. While 92% of chairs were MEPs at least once before being elected in the legislative terms under analysis, this is true for only 56% of rapporteurs. This relationship is straightforward and statistically significant at p < 0.01 (p = 0.000052). If we focus on rapporteurs and look at variation in the number of assigned reports, we observe fluctuations. However, the trend is in line with our expectation: MEPs with more reports more frequently have a parliamentary background at the EU level (see Fig. 6.3). The graph displays an increasing trend. Up to five reports, we see that four out of five times, rapporteurs with a background in the EP do not reach 60% (consistent with overall findings). However, from 6 reports Table 6.1 Distribution of Committee Chairs and Rapporteurs by Experience in the EP

Committee chairs Rapporteurs total Rapporteurs 1 report 2 reports 3 reports 4 reports 5 reports 6 reports 7 reports 8 reports 9 reports 10 reports 11 reports 12 reports 13–20 reports >20 reports

Previous experience in the EP

No previous experience in the EP

Total (absolute numbers)

91.7 (44) 55.8 (373)

8.3(8) 44.2 (295)

52 668

46.4 50.7 58.2 60.5 55.3 61.5 68.2 50.0 76.9 54.5 60.0 60.0 66.7 66.7

53.6 49.3 41.8 39.5 44.7 38.5 31.8 50.0 23.1 45.5 40.0 40.0 33.3 33.3

405 282 170 81 38 26 22 16 13 11 5 5 24 15

(188) (143) (99) (49) (21) (16) (15) (8) (10) (6) (3) (3) (16) (10)

(217) (139) (71) (32) (17) (10) (7) (8) (3) (5) (2) (2) (8) (5)

Note Absolute frequencies between brackets. “Rapporteurs total” does not include rapporteurs with only 1 report by legislative term. Reports’ numbers are calculated based on single legislative terms Source: Political Careers and Parliamentary Roles of the Members of the European Parliament, 1979– 2019 dataset (Salvati 2019), updated

152

E. SALVATI AND M. VERCESI

Fig. 6.3 The Relationship between Experience in the EP and the Number of Reports

upwards, only in two out of nine cases does the percentage remain below the threshold. Other factors are likely play a role in determining who becomes chair, who becomes rapporteur, and how many reports each MEP receives. Overall, though, we find evidence supporting our first expectation. Incumbency Proposition The second proposition suggests that personnel turnover matters for performance: rapporteurs who were in the EP before being elected should perform better than those who occupied other positions. We operationalize performance as the length of the decision-making process for single legislative proposals. However, the legislation discussed in the parliament can refer to more or less politically significant issues. In this regard, we assume that performance makes the difference when salient legislation is under consideration and, for this reason, focus only on the most salient proposals. Our measure of salience comes from The Informal Politics of Codecision dataset (Reh et al. 2020), whose authors use the average number of mentions of a certain procedure receives in English, German, French, and Italian press outlets as a proxy of salience. Although

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

153

the measurement does not fully capture the conceptual boundaries of salience, it seems a viable choice, since the most salient issues are more likely to appear in the public debate. In this regard, many proposals are never mentioned in the media. The average salience value for the 637 proposals is 0.755 (standard deviation = 2.932), and 0 is both the median and the mode of the distribution. For this reason, we choose a stricter notion of salience, focusing only on legislation with a value of ≥4. This threshold allows us to include proposals which, on average, have been mentioned at least once in each of the four languages taken into consideration. Using the ≥4 threshold, we retain 32 proposals (5% of the whole sample).7 Evidence suggests that incumbents do not perform better than newcomers, when important issues are discussed in the parliament. The average duration of procedures when incumbent MEPs are selected as rapporteurs is 705.8 days; the average duration for newcomers is 707.7, e.g., only two days more. This finding suggests that the simple turnover between two legislative terms does not account for performance. Other factors are probably more important. For example, previous experience—irrespective of when a politician acquires it—plays a greater role. Longer Experience Proposition Our third proposition foresees that the time previously spent in the EP counts, even if an MEP did not have a seat immediately before being elected to a term. In this regard, it is worth noting that in 55.2% of cases (16 MEPs out of 29), important reports were assigned to full newcomers. This indicates that, yes, principals value previous experience when it comes to selecting rapporteurs, but—at first sight—a report’s issue salience does not appear to be a conditional factor. The relationship between number of terms in the EP and performance—when checked against the salience of reports—is not significant. Although our conclusions must be cautious because of the low number of observations, the findings show that the two variables are weakly and insignificantly correlated (r = 0.039). However, one can think that experience does matter when working conditions get harder, while it does not make a difference when conditions are favorable. For example, previous knowledge of parliamentary dynamics could be an asset when the workload of the parliament or political polarization are higher and when unexpected external shocks

154

E. SALVATI AND M. VERCESI

100 90 80 70 60 50 40 30 20 10 0 Below average duraƟon

Above average duraƟon

MEPs with two or more previous terms MEPs with one previous term Newcomers

Fig. 6.4 Distribution of Rapporteurs by EP Experience and Performance in Office (Note Percentages on the vertical axis)

modify the agenda. If we take the average duration (708 days) as a benchmark and compare the background of rapporteurs below and above this number, interesting results emerge (see Fig. 6.4). The relevant separation is between the newcomers and others. While the two groups are distributed equally for shorter processes, the ratio of newcomers increases almost 10% (to 58.8%) when longer processes are observed. Overall, having been in the EP once or more does not seem to make a difference in terms of policy-making duration, but the distribution of newcomers is clearly skewed towards the “longer duration half” (63% of them are related to above average durations). This suggests that the “quantity” of previous experience in parliament is not an important determinant of performance unless an MEP is a parliamentary outsider. In the latter case, the absence of previous experience may slow down decision making. National Experience Proposition Our fourth proposition distinguishes EP outsiders, based on their political experience at the national level. In this regard, the evidence is in line

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

155

with our expectation. Six out of 16 MEPs with no previous experience in the EP (37.5%) were full outsiders, while the other 10 had previously acquired political experience below the EU level: full outsiders spent on average 807 days per report, while the rest of the group performed about five months better (645 days). Moreover, when we check in detail for the type of experience in national member-states (irrespective of previous service in the EP), we observe that a sub-European political background does matter (Table 6.2). MEPs who held a political position in their own country “save” about four months (670 days vs. 779 days). The gap is larger if we distinguish between MEPs with no experience below the EU level and MEPs with both types of sub-EU experiences: in this case, the difference reaches seven months of decision-making (i.e. 580 days vs. 779 days, months in bold in the table). A further indicator of the importance of a sub-European background is that it correlates negatively with the decision-making duration (when political experience is measured according to the aforementioned scaled operationalization). Again, our N is too small to draw statistically significant conclusions. However, we can observe a negative relationship between the two variables (r = −0.124). When we distinguish between durations that are below and above the overall average, the trend becomes clearer when duration is longer. While the negative relationship between national/sub-national experience and duration almost disappears with processes with a duration below the average (r = −0.012), for longer processes—which we have posited have more discriminatory power when Table 6.2 Type of Experience at the Sub-European Level and Performance in Office No sub-EU experience

Days Months

779.0 26.0

Note Values are on average

Sub-EU level experience Only national experience

Only sub-national experience

Both national and sub-national experience

Overall

750.3 25.0

645.3 21.5

579.8 19.3

670.2 22.3

156

E. SALVATI AND M. VERCESI

one wants to detect either efficient or inefficient performances—relatively experienced MEPs could provide faster decision-making (r = −0.204).

Discussion Overall, we have found that personnel turnover, appointments, and performance in the EP are linked. Although several explanatory variables are likely working behind the scenes, the evidence presented here is interesting in a number of respects. First, previous experience in the EP is less of an asset for rapporteur selection than it is for committee chair selection. For MEPs who aspire to receive a higher number of reports, though, the relevance of previous EP experience increases. Second, turnover does not, in general, affect rapporteurs’ performance on salient issues, when turnover is operationalized as referring only to the legislative term before the one under investigation. At the same time, previous EP experience— irrespective of the moment of acquisition—makes a difference when the decision process takes longer. While experienced MEPs perform better under this condition, experience is not significant when decision-making duration is below the average. Finally, we have seen that principals should consider agents’ regional and national experience if they are interested in shorter policy-making processes, as such experience correlates positively with decision-making efficiency. Again, this finding is especially strong when the decision process’s duration exceeds the average. Like Perez and Scherpereel (Chapter 4), our work concludes that turnover per se in the EP does not necessarily weaken throughput legitimacy. In this sense, our results can be understood as complementary suggestions that the relationship between personnel instability and institutional performance is structurally different in the EP, on one hand, than in the Council and the Commission, on the other. In the latter venues, indeed, turnover undermines throughput legitimacy, as shown by Chapter 4. That said, further data are necessary to corroborate our (preliminary) findings and for reaching a more conclusive picture. For example, we do not know yet how the sub-European political experience of ministers and commissioners can affect decision-making in the Council and the Commission. Moreover, our chapter does not address the gendered aspect of policy-making in the EP. As pointed out by Aldrich and Perez (Chapter 5), gender matters in negotiations; this might suggest that gender imbalance affects policy bargaining in the EP as well. If this held true, our analysis would be a strong basis to investigate further to

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

157

what extent sliding doors substituting women with men (or vice versa) can lead to scenarios that deviate from general trends. Even more important when it comes to evaluating the impact of turnover on legitimacy (in all its three facets) at the EU level is to consider the policy cycle in its entirety. Rather than focusing only on policy formulation and adoption, one should take into consideration all the involved actors in each stage of the process, both at the formal and informal level. How can we evaluate the impact of gender and/or personal expertise on informal bargaining? Do members of different EU institutions negotiate political outputs with different styles? Our chapter also contributes indirectly to populist and feminist debates about the EU, as they have been outlined in Chapter 1. Compared to the Council and the Commission, the EP is better equipped to cope with populist claims. In fact, the EP is the only EU institution whose members are directly chosen by citizens through popular vote. In theory, this provides voters with the chance to promote higher levels of personnel turnover and substitute elites with the preferred representatives. The fact that turnover does not seem to affect negatively institutional efficiency could be an argumentation in favor of populists, who argue that elite renewal improves the quality of the democratic process. However, our findings go against the populist claim when they show that turnover does not weaken performance, provided that new MEPs have acquired political experience prior to entering office. The relevance of experience to be selected as chair or rapporteur in the EP recalls technocratic ideas, according to which representation must be based on knowledge and responsibility rather than responsiveness, which is instead stressed by populists (Caramani 2017). In this regard, appointments to leadership positions in the EP are based on a mix between party representativeness and expertise-based criteria, which undermines the populist notion of representatives as ‘common people’. From a feminist viewpoint, we have seen (Fig. 6.2) that, in the period under consideration, women constituted only a minority of MEPs (36%). Although numbers can make one argue that a minimal “critical mass” has been reached, descriptive representation is still far from balance. Scholarship suggests that women in political institutions are more consensus-oriented and pursue compromise rather than conflict strategies (Krauss and Kroeber 2020). Building on this and in light of Chapter’s 5 evidence, our analysis provides arguments to support the idea that—all else equal—a more balanced distribution of seats between women and

158

E. SALVATI AND M. VERCESI

men could further improve throughput legitimacy and hence address some feminist critiques. A more efficient policy making process led by women is potentially beneficial in terms of women’s substantive representation, at least when the relationship between descriptive and substantive representation holds (Alexander et al. 2018). Nevertheless, significant differences in the size of women and men’s pools to pick up chairs and rapporteurs persist. An explanation is the alleged gendered character of the EP. It is possible that, in spite of the many efforts made to promote gender equality in the EU listed in Chapter 1, women remain less represented as a result of the reproduction of male-tailored selection procedures in the EP (Krook and Mackay 2011), with the possible repercussions that this phenomenon has on the quality of representation.

Conclusion With regard to the internal organization of the EP, our results provide a (partially) positive answer to our first question (are parliamentary insiders more likely to gain leadership positions?). The observed difference between committee chairs and rapporteurs can be explained not only by numerical variation in office availability, but also by the fact that MEPs serve two party principals—European party groups (EPGs) and national party delegations. On one hand, EPGs exert a gate-keeping function in the chair appointment processes. Chairs have the power to scrutinize the agenda in various policy fields and represent committees inside and outside the EP. In this regard, EPGs can define assignments according to their priorities. On the other hand, for rapporteur appointments, national party delegations have the upper hand. It is worth stressing that rapporteurs’ nationalities have an impact on final decisions on policy drafts, whereas it is controversial whether rapporteurs’ choices are affected at all by EPGs (Costello and Thomson 2010). Technical experience is less important, unless a rapporteur must manage many pieces of legislation in one legislative term. The answer to our second question (are insiders more successful in shepherding legislation through the policy-making process?) is negative. This means that the efficient management of the legislative process does not necessarily require long-standing MEPs, who are likely to have developed closer relationships with the EPG leaders. This has important implications. If newcomers can handle the most salient drafts, it is

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

159

possible for national parties to place trustworthy agents in sensitive positions, without worrying about possible negative effects of inexperience on the efficiency (and effectiveness) of the process. This implies, moreover, that national parties can also promote throughput legitimacy, due to their interest in safeguarding smooth management of legislative drafts. The most novel finding in this exploratory research is that newcomers can perform well. This finding holds, though, only when MEPs have acquired certain political skills at the national and sub-national levels. Political experience in a member-state is an important proxy of party reliability. National delegations can refer to it to distinguish between unqualified and qualified party agents, who can successfully lead the legislative draft through the perils of the EU legislative process. The length of previous service in the EP does not, overall, guarantee good policy performance; in fact, national political professionalization appears to be a better predictor of performance. With regard to our third question (how does personnel turnover affect policy production?), our findings suggest that—at least when salient issues are discussed—a high degree of turnover does not substantially undermine decision-making efficiency and effectiveness. When turnover injects professional politicians with solid sub-European experience into the EP, policy production works more quickly and successfully than when reports are assigned to MEPs with no experience in the national polity. As pointed out by van Geffen (2016: 1020–1021), MEPs who have already been MPs or ministers in their country are “expected to have closer connections with the national party […] and might be better positioned to take on certain rapporteurships which could enhance his ability to build a career in the EP compared with an EP careerist.” The institutionalization process and the strengthening of the EP sustain the access of skilled politicians and indirectly favor personnel turnover. This trend is attested to by the growing number of professional politicians who decide to “climb the ladder” to the EU level after developing a national political career (Salvati 2016, 2019; van Geffen 2016). There is a connection between MEPs’ political background and EP throughput legitimacy. For our purpose, it is worth noting however that the latter is not necessarily negatively affected by turnover; the negative effect only happens if inexperienced political outsiders enter the EP. One can conclude that MEPs can establish those coordination and negotiation skills that make them perform better as rapporteurs even when they are an MP or minister at a sub-European level.

160

E. SALVATI AND M. VERCESI

Overall, our findings suggest that the assignment of reports to newcomers who are national political professionals reduce the inherent perils of turnover (Scherpereel and Perez 2015). A more efficient legislative process can actually help improve throughput legitimacy and enhance the quality of EU democracy as a whole. In theory, this would also mean improving the effectiveness of a political system and reducing the consequent risk of citizens’ demobilization. In contrast, if newcomers are full outsiders, policy making slows down and the peril of amateurism pertains. It is this scenario that undermines throughput legitimacy (as argued in this book’s first chapter). In summary, this chapter contributes to the existing literature by disentangling the relationship between political experience/personnel turnover and the efficiency of EU policy making. Our N is limited, our analysis is mostly exploratory, and our results encourage further investigations in this direction. Future studies can further elaborate on turnover’s “quality” aspect and on its impact on EU legitimacy. One path is to look at MEPs’ background in a more fine-grained way and to link personal profiles to performance in more general terms. All reports (not only the most salient) should be taken into consideration. Moreover, the temporal focus of the analysis could be expanded, so that one can investigate longitudinal trends, which might strengthen our preliminary findings. Acknowledgements Preliminary drafts of this chapter were presented at the 2019 EUSA International Biennial Conference in Denver (May 9–11) and the 2019 ECPR General Conference in Wrocław (4–7 September). We thank all panels’ participants for their valuable suggestions and especially Edoardo Bressanelli, William Daniel, Matthew Kirby, Lauren Perez, and John Scherpereel for having read and commented on earlier versions. Thanks also to Pamela Pansardi for helping us with the access to the data on reports’ allocation in the 7th parliamentary term. Finally, we are grateful to Edoardo Bressanelli, Christel Koop, and Christine Reh for sharing with us updated and unpublished information from “The Informal Politics of Codecision” dataset.

Notes 1. The EP can also set up sub-committees and special temporary committees to deal with specific issues. 2. Sometimes, two co-rapporteurs can be chosen, and rotation is possible for recurring reports, for example on annual basis. 3. The draft can be amended both within the committee and in the plenary.

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

161

4. All our propositions work under an “all else equal” condition. 5. In case a rapporteur has more than one report, we calculate the average duration of different legislative proposals. 6. We have gathered new data about names and number of reports for rapporteurs for this chapter. 7. The proposal COM(2011)0785 is not referred to any rapporteur, but only to a shadow rapporteur (Marietje Schaake). We have opted to consider her profile for the analysis. When two rapporteurs were appointed (5 observations in our analysis), we counted two reports. Moreover, we looked at MEPs’ profiles when the proposal was made or, where not possible, when the rapporteur was appointed.

References Aldrich, Andrea S. 2018. National Political Parties and Career Paths to the European Parliament. Journal of Common Market Studies 56 (6): 1283–1304. Alexander, Amy C., Catherine Bolzendahl, and Farida Jalalzai. 2018. Introduction to Measuring Women’s Political Empowerment Across the Globe: Strategies, Challenges, and Future Research. In Measuring Women’s Political Empowerment Across the Globe: Strategies, Challenges and Future Research, ed. Amy C. Alexander, Catherine Bolzendahl, and Farida Jalalzai, 1–25. Basingstoke: Palgrave Macmillan. Alexiadou, Despina. 2016. Ideologues, Partisans, and Loyalists: Ministers and Policymaking in Parliamentary Cabinets. Oxford: Oxford University Press. Bowler, Shaun, and David M. Farrell. 1995. The Organizing of the European Parliament: Committees, Specialization and Co-ordination. British Journal of Political Science 25 (2): 219–243. Brandsma, Gijs J., and Johan Adriaensen. 2017. The Principal-Agent Model, Accountability and Democratic Legitimacy. In The Principal Agent Model and the European Union, ed. Tom Delreux and Johan Adriaensen, 35–54. Cham: Palgrave Macmillan. Bressanelli, Edoardo, Adrienne Héritier, Christel Koop, and Christine Reh. 2014. The Informal Politics of Codecision: Introducing a New Data Set on Early Agreements in the European Union. EUI Working Paper RSCAS 2014/64. Bressanelli, Edoardo, Christel Koop, and Christine Reh. 2016. The Impact of Informalisation: Early Agreements and Voting Cohesion in the European Parliament. European Union Politics 17 (1): 91–113. Burns, Charlotte. 2013. Consensus and Compromise Become Ordinary—But at What Cost? A Critical Analysis of the Impact of the Changing Norms of Codecision upon European Parliament. Journal of European Public Policy 20 (7): 988–1005.

162

E. SALVATI AND M. VERCESI

Caramani, Daniel. 2017. Will vs. Reason: The Populist and Technocratic Forms of Political Representation and Their Critique to Party Government. American Political Science Review 111 (1): 54–67. Corbett, Richard, Francis Jacobs, and Michael Shackleton. 2007. The European Parliament, 7th ed. London: John Harper. Costa, Olivier. 2019. Introduction: The Need for Further Research on the European Parliament. In The European Parliament in Times of EU Crisis. Dynamics and Transformations, ed. Olivier Costa, 1–21. Cham: Palgrave Macmillan. Costello, Rory, and Robert Thomson. 2010. The Policy Impact of Leadership in Committees: Rapporteurs’ Influence on the European Parliament’s Opinions. European Union Politics 11 (2): 219–240. Costello, Rory, and Robert Thomson. 2011. The Nexus of Bicameralism: Rapporteurs’ Impact on Decision Outcomes in the European Union. European Union Politics 12 (3): 337–357. Cox, Gary W., and Mathew D. McCubbins. 1993. Legislative Leviathan: Party Government in the House. Berkeley: University of California Press. Curini, Luigi, and Francesco Zucchini. 2015. Legislative Committees as Uncertainty Reduction Devices in Multiparty Parliamentary Democracies. West European Politics 38 (5): 1042–1061. Daniel, William T. 2013. When the Agent Knows Better than the Principal: The Effect of Education and Seniority on European Parliament Rapporteur Assignment. Journal of Common Market Studies 51 (5): 832–848. Daniel, William T. 2015. Career Behaviour and the European Parliament: All Roads Lead Through Brussels? Oxford: Oxford University Press. Daniel, William T., and Stefan Thierse. 2017. Individual Determinants for the Selection of Group Coordinators in the European Parliament. Journal of Common Market Studies 56 (4): 939–954. Delreux, Tom, and Johan Adriaensen. 2017. Introduction. Use and Limitations of the Principal-Agent Model in Studying the European Union. In The Principal Agent Model and the European Union, ed. Tom Delreux and Johan Adriaensen, 1–34. Cham: Palgrave Macmillan. Dowding, Keith, and Patrick Dumont. 2015. Agency Rent, Adverse Selection and Moral Hazard. In The Selection of Ministers Around the World, ed. Keith Dowding and Patrick Dumont, 1–24. London and New York: Routledge. François, Abel, and Emiliano Grossman. 2015. How to Define Legislative Turnover? The Incidence of Measures of Renewal and Levels of Analysis. The Journal of Legislative Studies 21 (4): 457–475. Häge, Frank M., and Nils Ringe. 2019. Rapporteur-Shadow Rapporteur Networks in the European Parliament: The Strength of Small Numbers. European Journal of Political Research 58 (1): 209–235. Høyland, Bjørn. 2006. Allocation of Codecision Reports in the Fifth European Parliament. European Union Politics 7 (1): 30–50.

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

163

Hurka, Steffen, Michael Kaeding, and Lukas Obholzer. 2015. Learning on the Job? EU Enlargement and the Assignment of (Shadow) Rapporteurships in the European Parliament. Journal of Common Market Studies 53 (6): 1230– 1247. Kaeding, Michael. 2004. Rapporteurship Allocation in the European Parliament. European Union Politics 5 (3): 353–371. Kaeding, Michael. 2005. The World of Committee Reports: Rapporteurship Assignment in the European Parliament. The Journal of Legislative Studies 11 (1): 82–104. Kassim, Hussein, and Anand Menon. 2003. The Principal-Agent Approach and the Study of the European Union: Promise Unfulfilled? Journal of European Public Policy 10 (1): 121–139. Krauss, Svenja, and Corinna Kroeber. 2020. How Women in the Executive Influence Government Stability. Journal of European Public Policy. https://doi. org/10.1080/13501763.2020.1773905. Krehbiel, Keith. 1991. Information and Legislative Organization. Ann Arbor: University of Michigan Press. Krook, Mona L., and Fiona Mackay, eds. 2011. Gender, Politics and Institutions. Towards a Feminist Institutionalism. Basingstoke: Palgrave Macmillan. Lupia, Arthur. 2003. Delegation and Its Perils. In Delegation and Accountability in Parliamentary Democracies, ed. Kaare Strøm, Wolfgang C. Müller, and Torbjörn Bergman, 33–54. Oxford: Oxford University Press. Mamadouh, Virginie, and Tapio Raunio. 2003. The Committee System: Powers, Appointments and Report Allocation. Journal of Common Market Studies 41 (2): 333–351. McElroy, Gail. 2006. Committee Representation in the European Parliament. European Union Politics 7 (1): 5–29. Müller-Rommel, Ferdinand, Corinna Kroeber, and Michelangelo Vercesi. 2020. Political Careers of Ministers and Prime Ministers. In The Oxford Handbook of Political Executives, ed. Rudy B. Andeweg, Robert Elgie, Ludger Helms, Juliet Kaarbo, and Ferdinand Müller-Rommel, 229–250. Oxford: Oxford University Press. Musella, Fortunato, and Michelangelo Vercesi. 2019. Definitions and Measures of Party Institutionalization in New Personal Politics: The Case of the 5 Star Movement. Zeitschrift für vergleichende politikwissenschaft 13 (2): 225–247. Obholzer, Lukas, Steffen Hurka, and Michael Kaeding. 2019. Party Group Coordinators and Rapporteurs: Discretion and Agency Loss Along the European Parliament’s Chain of Delegation. European Union Politics 20 (2): 239–360. Piattoni, Simona. 2013. Representation as Delegation: A Basis for EU Democracy? Journal of European Public Policy 20 (2): 224–242.

164

E. SALVATI AND M. VERCESI

Raunio, Tapio. 2006. Political Parties in the European Union. In The SAGE Handbook of European Union Politics, ed. Knud Erik Jørgensen, Mark Pollack, and Ben Rosamond, 247–262 London: Sage. Reh, Christine, Edoardo Bressanelli, and Christel Koop. 2020. Responsive Withdrawal? The Politics of EU Agenda-Setting. Journal of European Public Policy 27 (3): 419–438. Ringe, Nils. 2010. Who Decides, and How? Preferences, Uncertainty, and Policy Choice in the European Parliament. Oxford: Oxford University Press. Roger, Léa., and Thomas Winzen. 2015. Party Groups and Committee Negotiations in the European Parliament: Outside Attention and the Anticipation of Plenary Conflict. Journal of European Public Policy 22 (3): 391–408. Salvati, Eugenio. 2016. Towards a European Parliamentary Class? A Proposal for a Typology of the MEPs. Journal of Comparative Politics 9 (1): 59–74. Salvati, Eugenio. 2019. Il Parlamento Europeo tra crisi del processo di integrazione e politicizzazione dell’Unione Europea. Milano: Mondadori. Salvati, Eugenio, and Michelangelo Vercesi. 2019. Party Euroscepticism and the Conditions of Its Success: Conceptualization and Explanation. Quaderni Di Scienza Politica 26 (2): 301–330. Samuels, David J., and Matthew S. Shugart. 2010. Presidents, Parties, and Prime Ministers. How the Separation of Powers Affects Party Organization and Behavior. Cambridge: Cambridge University Press. Scarrow, Susan. 1997. Political Career Paths and the European Parliament. Legislative Studies Quarterly 22 (2): 253–263. Scherpereel, John A., and Lauren K. Perez. 2015. Turnover in the Council of the European Union: What It Is and Why It Matters. Journal of Common Market Studies 53 (3): 658–673. Schmidt, Vivien A. 2013. Democracy and Legitimacy in the European Union Revisited: Input, Output and ‘Throughput.’ Political Studies 61 (1): 2–22. Shepsle, Kenneth A. 1979. Institutional Arrangements and Equilibrium in Multidimensional Voting Models. American Journal of Political Science 23 (1): 27–59. Sieberer, Ulrich, and Wolfgang C. Müller. 2017. Aiming Higher: The Consequences of Progressive Ambition Among MPs in European Parliaments. European Political Science Review 9 (1): 27–50. Strøm, Kaare. 2003. Parliamentary Democracy and Delegation. In Delegation and Accountability in Parliamentary Democracies, ed. Kaare Strøm, Wolfgang C. Müller, and Torbjörn Bergman, 55–106. Oxford: Oxford University Press. Thierse, Stefan. 2019. Policy Entrepreneurship in the European Parliament: Reconsidering the Influence of Rapporteurs. Journal of European Public Policy 26 (2): 267–285.

6

THE IMPORTANCE OF EXPERTISE: POLITICAL CAREERS, PERSONNEL …

165

Treib, Oliver, and Bernd Schlipphak. 2019. Who Gets Committee Leadership Positions in the European Parliament? Evidence from the 2014 Selection Process. European Union Politics 20 (2): 219–238. van Geffen, Robert. 2016. Impact of Career Paths on MEPs’ Activities. Journal of Common Market Studies 54 (4): 1017–1032. Vaubel, Roland. 2006. Principal-Agent Problems in International Organizations. The Review of International Organizations 1 (2): 125–138. Vercesi, Michelangelo, and Selena Grimaldi. 2019. Using Large N Longitudinal Comparison to Explain Political Recruitment in Changing Democracies. SAGE Research Methods Cases—Politics & International Relations Part 2: 1–11. Verzichelli, Luca, and Michael Edinger. 2005. A Critical Juncture? The 2004 European Elections and the Making of a Supranational Elite. Journal of Legislative Studies 11 (2): 254–274. Whitaker, Richard. 2001. Party Control in a Committee-Based Legislature? The Case of the European Parliament. The Journal of Legislative Studies 7 (4): 63–88. Whitaker, Richard. 2014. Tenure, Turnover and Careers in the European Parliament: MEPs as Policy-Seekers. Journal of European Public Policy 21 (10): 1509–1527. Whitaker, Richard. 2019. A Case of ‘You Can Always Get What You Want’? Committee Assignments in the European Parliament. Parliamentary Affairs 72 (1): 162–181. Yordanova, Nikoleta. 2009. The Rationale Behind Committee Assignment in the European Parliament: Distributive, Informational and Partisan Perspectives. European Union Politics 10 (2): 253–280. Yordanova, Nikoleta. 2011. Inter-institutional Rules and Division of Power in the European Parliament: Allocation of Consultation and Co-decision Reports. West European Politics 34 (1): 97–121. Yordanova, Nikoleta. 2013. Organising the European Parliament: The Role of the Committees and their Legislative Influence. Colchester: ECPR Press. Yoshinaka, Antoine, Gail McElroy, and Shaun Bowler. 2010. The Appointment of Rapporteurs in the European Parliament. Legislative Studies Quarterly 35 (4): 457–486.

PART IV

Turnover and Output Legitimacy

CHAPTER 7

Making a Hasty Brexit? Turnover and Brexit Negotiations Jessica Adolino

By any measure, under Theresa May’s leadership, negotiating the United Kingdom’s (UK) exit from the European Union (EU) was a highly contentious, chaotic, and ultimately ineffective process. However, the May government’s failure to reach an agreement on an approach to leaving the EU—and the unprecedented level of political turmoil observed in the UK—was not preordained. In fact, the negotiation process led by May stood in marked contrast to the country’s prior approach to and success with EU internal negotiations. In reality, the UK side underperformed in the Brexit negotiations. The previous chapters in this volume considered personnel turnover and its effects on input and throughput legitimacy: this is the first of two chapters examining the dynamics between personnel turnover and output legitimacy. To what extent did personnel turnover affect the Brexit negotiations between the

J. Adolino (B) Department of Political Science, James Madison University, Harrisonburg, VA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. A. Scherpereel (eds.), Personnel Turnover and the Legitimacy of the EU, Palgrave Studies in European Union Politics, https://doi.org/10.1007/978-3-030-60052-5_7

169

170

J. ADOLINO

EU and the UK? And how might Brexit-related turnover affect output legitimacy? This chapter considers these questions by examining EU and UK personnel dynamics between 2017 and 2019. It argues that, while the structural features of “differentiated disintegration” clearly worked to the EU’s advantage in the Brexit negotiations (Schimmelfennig 2018), the May government’s failures were also the product of processes of personnel turnover, in particular a steady and unprecedented stream of ministerial resignations. The hazards commonly associated with personnel turnover often affect public attitudes towards policies and institutions. Citizens recognize that the decision-making inefficiencies personnel turnover may create often result in public problems failing to be addressed—either effectively or, indeed, at all. Thus, personnel turnover may have significant implications for questions of democratic legitimacy. Of particular relevance is the notion of output legitimacy, or legitimacy through performance. Here, the concern is whether the policy solutions that governments design and implement are effective in addressing the issues facing the people (Scharpf 2003). Output legitimacy is thought to increase when the needs of more stakeholders or individuals in society are met by a policy decision. That is, output legitimacy is strong where problem-solving quality—the effectiveness and performance—of laws and rules is judged to be high. Personnel turnover, especially when it involves replacement of political veterans with newcomers—can slow down decisions, weaken governing effectiveness, and ultimately undermine output legitimacy. Citizens in the EU and the UK both expect government officials to make policies that solve their problems. In the case of Brexit, this has meant, first and foremost, reaching agreement on the terms of the UK’s withdrawal. On the EU side, polls demonstrate that citizens want an EU that can project power on the world stage, so output legitimacy would be endangered if the EU negotiating team appeared weak by conceding to the UK’s demands. On the UK side, a majority of citizens looked to the government to deliver a clean break with the EU; a failure to successfully conclude the negotiations, then, would have negative implications for output legitimacy. The chapter proceeds by first examining the points of strength and weakness for the EU and UK sides as they entered into the Brexit negotiations. It continues with a discussion of the implications of high rates of personnel turnover for government performance. The chapter then documents patterns of personal turnover on the EU and UK sides. It concludes

7

MAKING A HASTY BREXIT? TURNOVER AND BREXIT NEGOTIATIONS

171

with reflections on the implications of personnel turnover for the Brexit negotiation process and for output legitimacy in both the UK and EU.

Anticipating the Brexit Negotiations On the eve of the Brexit negotiations the observed strengths and weaknesses of the EU and UK sides made it possible to envision at least three different ways in which the forthcoming talks would unfold. One, was that, as has often been the case in matters of foreign policy, the EU would struggle to maintain a unified or constant position, while the UK would fall into its traditional role as a formidable EU negotiator—and thus prove to have the upper hand. Another possibility was that both the EU and the UK would enter negotiations in full negotiating form— the EU as the highly effective international trade negotiator and the UK again leveraging its experience and aptitude in the EU negotiating arena—with a fight to the finish between two equally determined sides. The third scenario, which turned out to be the correct one, was rarely countenanced: that the EU would enter negotiations as the stronger side—unified, focused and resolute—and that it would face a UK team that was wholly unprepared and ill-suited for the task at hand. The British civil service has long been hailed as one of the best administrative services in the world, and, within the EU, the effectiveness of what was called a “Rolls Royce” British administration was widely recognized (Bulmer and Burch 2009). This view was based on the clarity of the positions that the British developed on particular issues, the efficiency with which the UK’s EU representatives could deliver their views, the coherence with which the UK’s stance was articulated by UK government ministers and officials in Brussels, and the thoroughness of their preparation. As a result, the UK was not marginalized at the EU level in terms of outcomes—rather, it was more often than not at the heart of EU decisionmaking (Kassim 2016). Thus, despite the enormity of the task, it was not unreasonable to expect in 2016 that the British were fairly well-situated to plan, negotiate, and implement a successful Brexit strategy. This is not to say that the EU entered this process at an absolute disadvantage vis-à-vis the UK; in fact, the EU had a long and successful track record with respect to external trade relations, with established institutional processes for such negotiations and deep expertise to draw upon. The EU’s intergovernmental nature highlights the necessity of speaking with one voice when negotiating with third parties and, typically, this

172

J. ADOLINO

imperative is met. Given its large size, the EU is usually the stronger party in international negotiations and is a “patient and inflexible negotiator,” tending to “toe the line” and offer countries “take it or leave it deals” (Maurice 2018; McConnell and Tormey 2019; Patel 2018). Trade negotiators from other countries have reported that EU negotiators often take a “relentless, dominant and uncompromising approach” (Patel 2018, 5). As an active participant in past EU trade negotiations, the UK certainly should have expected the EU to draw on this experience in the Brexit negotiations. Further, the unique circumstances presented by these negotiations also suggested at the outset that the EU entered these negotiations in a position of strength. More specifically, the EU was faced with a member-state seeking to “disintegrate” from the Union (for the UK, such “disintegration” was defined by Theresa May in 2017 as a “hard” Brexit, whereby the country would decouple entirely from the EU, thus fully restoring Britain’s economic and political sovereignty). As Schimmelfennig (2018, 1154) notes, the task at hand—the selective reduction of state’s level and scope of integration, or what he terms a process of “differentiated disintegration” —would necessarily involve a fundamental change in the constellation of institutional and material bargaining power between parties. Further, the requirements for such a process, as defined by Article 50 of the Treaty on European Union and the EU’s rules for treaty change would also work to the advantage of the EU in its role as a defender of the status quo. States seeking differentiated disintegration also would be in a situation of “asymmetric interdependence.” This situation would reflect the fact that the economic power of any single member-state is “small in comparison with the EU as a whole,” as well as the likelihood that the EU would remain such states’ most important foreign market (Schimmelfennig 2018, 1160). As such, the state seeking disintegration would have more to lose than the EU as a whole. In the midst of such asymmetry, the EU would be likely to have much greater negotiation leverage. Given these parameters, in seeking differentiated disintegration, states like the UK would be forced to “moderate their demands and make (asymmetrical) concessions to the EU in the course of negotiations” (Schimmelfennig 2018, 1169). One significant caveat should be noted to this characterization: while generally successful in external trade negotiations, the EU also is notorious for its poor performance and lack of unity with respect to foreign policy matters.1 EU foreign policy decisions are often reached by way

7

MAKING A HASTY BREXIT? TURNOVER AND BREXIT NEGOTIATIONS

173

of discussions framed, prepared, and undertaken as 27 distinct countries, with national, not EU, interests at the forefront of negotiations. For example, the Transatlantic Trade and Investment Partnership negotiations with the United States revealed distinct preferences among member-states and political groups, a fact that US negotiators clearly sought to exploit (Patel 2018, 8). Entering into the Brexit negotiation period, it was clear that the UK’s exit from the EU would have a differential impact across the EU-27, with some member-states anticipating costly economic repercussions. If a unified UK were able to leverage these distinct national interests across the EU bloc, it might gain some negotiating advantage. Thus, before negotiations began, there was considerable speculation that the UK might make some headway in advancing its preferred course of action during negotiations through direct diplomacy with national leaders, both individually and in regional groupings. If the Brexit negotiations were to unfold in a manner consistent with the EU’s previous foreign policymaking efforts, then, it was reasonable to envision some vulnerability on the EU’s part.

Personnel Turnover and Its Effects One variable overlooked in the assessment offered above is the possible effects of personnel turnover or, more specifically, how turnover among negotiating officials (especially ministers and their underlings) might affect parties to complex negotiations (such as those to be held to settle an EU member state’s unprecedented departure from the EU). The literature examining the effects of high rates of personnel turnover, whether the focus is on cabinets, legislatures, or bureaucracies, generally finds that rapid or frequent change can create a number of problems within governing institutions. In particular, three potentially negative outcomes are associated with high turnover: the perils of amateurism, poor-quality policy, and disunity (Scherpereel and Perez 2015). Increasing rates of turnover may reduce the level of experience and knowledge in political institutions, thereby weakening them, as newcomers prove less adept at navigating their new institutional setting, less knowledgeable about the rules of the game, and in command of fewer resources—including prior relationships with their peers, interpersonal trust, technical skills and institutional memory (see also Atkinson and Docherty 1992; Hibbing 1991; Matland and Studlar 2004; Niemi and Winsky 1987; Rosenthal 1974; Shin and Jackson 1979). Of particular relevance with respect to the

174

J. ADOLINO

impact amateurism may have on those in ministerial positions—and for the Brexit negotiations—is an argument that institutional and issue expertise is necessary for negotiations and coalition-building. Further, it is also helpful for ministers to have an institutional memory of past laws and relevant governing procedures. Frequent turnover, however, may mean that new recruits to ministerial positions may be lacking in all of the above (Francis and Baker 1986, 120; cited in Scherpereel and Perez 2015). High rates of turnover might further affect the quality of policy outputs since newer, and often less experienced, politicians (and ministers) will often lack the necessary substantive knowledge of the issues addressed by the legislature (or by their ministry) and also will often fail to have the resources required to translate ideas into laws and to ensure policy implementation. Officials who expect to serve in office for a only short period also may be less willing to invest in long-range policy projects and will almost certainly be less likely to take short-term risks to invest in longterm gains (Putnam 1976). One would expect this same pattern to hold true for government ministers, and, the more quickly they leave office, the more problematic for policy-making this is likely to be. A third hazard posed by high turnover is the potential for disunity. More specifically, the danger here is the presence of new, constantly changing members in the midst of longer-serving members—the experienced v. the neophytes—which may create difficult working conditions and perhaps even cleavages that will serve as significant impediments to meaningful work being accomplished (Atkinson and Docherty 1992; Scherpereel and Perez 2015). These divisions may be such that older members come to resent newly elected or newly appointed members, especially if the latter have dramatically different agendas or broader ambitions. We can certainly envision this same scenario in the midst of a cabinet, where, just as in a legislature, animosity between old and new actors can make it difficult to work together and to reach consensus— in some instances, these gaps can be so wide as to prevent any forward motion, bringing the body to a halt. Presumably, these qualities are even more detrimental for cabinet government, where consensus and collective responsibility are required. Finally, in the UK, civil servants are unlikely to make decisions in the absence of clear messaging and instructions from their ministers; a perennial strength of the UK’s bureaucratic system has been the clear continuity between ministers’ thinking and the work of their civil servants. To the extent that frequent turnover increases uncertainty for civil servants about

7

MAKING A HASTY BREXIT? TURNOVER AND BREXIT NEGOTIATIONS

175

their minister’s preferences, we might expect a negative change in bureaucratic behaviors (Perez and Scherpereel 2017, 1162). Further, because the British bureaucratic system generally has not been prone to rapid turnover at the top, we can anticipate that incumbent civil servants are less comfortable with frequent change. As such, it is reasonable to expect high rates of ministerial turnover might lead to inefficiencies in decision-making with respect to both timing and quality.

Personnel Turnover and Brexit Negotiations From July 2016 to July 2019, the circumstances surrounding the Brexit negotiations in the EU and the UK stood in marked contrast. Simply put, on the EU side, there was personnel continuity; this helped the EU to remain unified, to maintain a consistent approach to the negotiations, to gain and hold the upper hand, and to ultimately achieve negotiating success. On the UK side, there was rapid and substantial personnel turnover at the ministerial level; a rapid-fire series of resignations was a manifestation of a debilitating lack unity on Brexit matters that further politicized an already-fraught decision-making process, and, in the end, rendered the May government unable to act. The EU Side: Personnel Continuity and Inter-Institutional Coordination The EU side during 2016–2019 was notable for its focus, unity, and stability—features that created the circumstances for strong bargaining power. In this negotiation period, the EU had a steady, experienced, and unified team of negotiators; this team had strong leadership, political coherence, and clear organization. The EU side began with, and maintained, a commitment to the EU’s core principles and a clearly articulated set of negotiating objectives—”preserving the unity and integrity of the EU and its supranational rules and institutions” (Schimmelfennig 2018, 1167). By the time the UK invoked Article 50 in March 2017, beginning the formal two-year negotiating period, the European Council had developed a set of negotiating directives for the European Commission that made clear the central importance of European unity: “in these negotiations the Union will maintain its unity and act as one…” (European Council 2017). Over the course of the negotiations, the strength of the EU’s commitment to these tenets proved to be a powerful unifying factor

176

J. ADOLINO

and source of strength. Crucially, the Council’s directives also made it clear that the UK would negotiate only with the Commission Taskforce 50, led by the Union’s chief negotiator (Michel Barnier), eliminating the potential for May to try and leverage divergent national interests by engaging in bilateral negotiations with individual member-states. Further, the Commission was limited to discussing with the UK only those matters for which it had been given a mandate by the Council, which gave Barnier license to repudiate the UK’s demands. This whole-of-the EU, teambased approach, with well-defined institutional roles, and member-state unity around a set of clearly defined objectives, forced the UK to deal with a single, inflexible EU channel—a focused Commission negotiating team with a tightly constrained Council mandate (Patel 2018). EU cohesion was enhanced by a high level of preparation for the negotiations. In contrast to the UK, even before the June 2016 referendum was held, the EU was preparing for possible Brexit negotiations, a reflection of its natural inclination as a rules-based institution. Within the European Commission, work had begun to develop a joint technical response statement that would be issued by the heads of the European Council, the Council, the Commission, and the Parliament once the referendum was held. European Council President Donald Tusk also engaged in an all-out effort to communicate with national leaders about the need for speaking with one voice on the EU’s commitment to its core principles, regardless of the referendum result. On the day of the referendum, the European Council sent its first “lines to take” messaging to the EU-27, urging European unity to ensure a strong defense of the EU’s stability and integrity. The early drive for unity allowed the EU to hit the ground running following the referendum—defining the negotiation rules and procedures, clearly setting out its own objectives, and putting a strong team in place that was fully briefed on these plans. This mandate remained unchanged for the duration of the May negotiation period and was consistently referenced and adhered to by EU leaders—within and across EU institutions and in their communications with the UK. From experience, the EU knew the importance of remaining steadfast during negotiations, and this preparedness—and the resulting political coherence of its approach—put it in a strong position to control the negotiations from the outset. Compounding this advantage, Commission Taskforce 50, the European Commission team charged with primary responsibility for managing

7

MAKING A HASTY BREXIT? TURNOVER AND BREXIT NEGOTIATIONS

177

the EU’s Brexit response in this period, was notable for its experience, preparation, constancy, and strong leadership. Each of the team’s members had extensive EU diplomatic and policy-making resumes. They were clearly selected for their roles because of their negotiating prowess and technical expertise. In keeping with EU practice in international negotiations, this team, in close contact with the Council, had primary responsibility for the negotiations. The taskforce’s leader, Michel Barnier, was an ardent Europhile and a seasoned politician and negotiator. Having served as the EU’s Single Market Commissioner during the 2008 financial crisis, Barnier was a well-respected EU official who enjoyed the confidence of national leaders. As the Commission’s chief negotiator, Barnier was the key public figure in the Brexit negotiation process, a faithful messenger on the EU position, and a steady hand at the helm. He is credited in large part with the EU’s ability to maintain a united front throughout this period: “in coordination with (Didier) Seeuws at the Council and (Guy) Verhofstadt in the European Parliament…Barnier was…the pivot on which the continued unity of the 27 formed, managing differences in opinion and shaping the strategy of public expression and decision-making ahead of time” (Maurice 2018). Importantly, over the course the negotiations with the May government, the membership of the central EU team remained constant; thus, none of the hazards posed by personnel turnover were present. This personnel stability certainly enabled effective management of the negotiation process—both internally (within the EU) and externally (with the UK). The EU adhered to a robust inter-institutional informationsharing regimen, emphasizing complete transparency within and across EU institutions and member-states—this proved to be a key mechanism for both controlling the public narrative and maintaining unity among and within member-states. Critically, inter-institutional political accountability was high: Barnier reported to the European Council to keep member-states up-to-date at every stage of the negotiations, and relevant documents and updates were made available on the Commission’s sites. Barnier also regularly traveled to national capitals and met with the press to discuss the EU’s Brexit policy (Patel 2018; Maurice 2018). While the European Council was to be the guiding body and final decision-maker at the main negotiation stages, clearly defined roles and lines of responsibility, as well as an emphasis on transparency on the EU side, allowed the process to remain outside the political realm. Disagreements within the EU-27 were managed in a coordinated process between

178

J. ADOLINO

Didier Seeuws (the head of the European Council’s Taskforce on the UK), COREPER, and other institutional representatives (Maurice 2018). Thus, national leaders were not drawn directly into the talks, nor did they directly engage in negotiations with the UK, and Barnier’s team served as the only communicator with the UK’s negotiators. Key officials—within the Commission, Council, and Parliament—also proved to be effective gatekeepers with respect to the EU’s position throughout this period. Each of these further strengthened the EU’s projection of a unified front. When combined with the high level of personnel continuity, these features enabled EU negotiators to run rings around their UK counterparts. The UK Side: Personnel Turnover and Inter-Institutional Discord These strengths on the EU side meant that, from the outset, achieving any measure of success in forcing EU concessions and securing its preferred outcomes in the Brexit negotiations would require the UK negotiators to be at the top of their political and administrative game. As noted earlier, entering into the negotiations, the UK’s leaders and civil service had well-deserved reputations as skilled EU negotiators. Despite this past record of success, May and her Cabinet ministers, who effectively took on responsibility for overseeing and conducting the negotiations (with May essentially playing the role of chief negotiator), proved incapable of meeting the challenge at hand. The unity, coordination, openness, and consistency that defined the EU’s approach stands in marked contrast to the political divisions, poor planning, opacity, and incoherence that was observed on the UK side. In the lead-up to the June 2016 Brexit referendum, in anticipation of negotiating with the EU to secure a Brexit deal, the UK’s then-Justice Secretary Michael Gove declared that “the day after we vote to leave, we (will) hold all the cards and we can choose the path we want.” In July 2017, John Redwood, a Conservative Party MP, proclaimed that: “getting out of the EU can be quick and easy – the UK holds most of the cards” (Henley and Roberts 2018). As the Brexit negotiations unfolded in 2017–2019, it was obvious that the UK did not in fact “hold all the cards.” Instead, the May government proved unable to successfully play the hand it was dealt. Rather than their dictating the terms of engagement and agreement, over the course of the May’s premiership, the Brexit negotiations involved a series of successive British climbdowns in the face of EU unity, determination, and resolve; and, ultimately, she

7

MAKING A HASTY BREXIT? TURNOVER AND BREXIT NEGOTIATIONS

179

proved unable to deliver on the referendum result. Further, by March 2019, as the original Brexit deadline approached, the UK appeared to be in the midst of a political and constitutional crisis, raising concerns about democratic legitimacy. By almost any measure, the period that began in the immediate aftermath of the June 23, 2016 Brexit referendum and concluded with Prime Minister May stepping down in July 2019, was one of government chaos. The turmoil began with then-Prime Minister David Cameron’s resignation on June 24, 2016 and succession by May within days of the vote. Subsequently, May’s decision to call a snap election in 2017 and the resulting loss of the Conservatives’ parliamentary majority cast doubt on her leadership and further stirred up dissension in her party’s ranks at the Government and Parliamentary levels. At the Government level, a tendency toward dissent was at least in part a function of the composition of the Cabinet itself, which May deliberately constructed to encompass both sides of the Brexit divide. Beginning in 2016, and continuing after the 2017 election, the prime minister made the decision to balance her Cabinet between Leavers and Remainers (those who voted to Leave the EU in the referendum and those who voted to Remain).2 Included among these were three of the most prominent actors from the Leave campaign, each placed in a key Brexit post: Boris Johnson (Foreign Secretary), David Davis (Secretary of State for Exiting the European Union), and Liam Fox (International Trade Secretary). None of these officials was chosen because of his prior expertise or experience in the EU negotiating arena. Instead, May’s choices reflected two strategies: one, balancing Leavers and Remainers allowed May to demonstrate her commitment to delivering on Brexit (since May herself had voted Remain), and, two, it followed the time-tested logic of keeping one’s political enemies close. This proved to be a fateful decision, since it is clear that the close proximity of the two camps within the Cabinet was a formidable obstacle to building consensus on matters related to Brexit. Nonetheless, the Prime Minister continued to maintain this balance after the 2017 election, in her 2018 Cabinet reshuffle, and in appointments following resignations, which continued to complicate Cabinet dynamics and impede decisionmaking.3 The presence of soft and hardcore Brexiters, as well as ardent Remainers around the same table meant that Cabinet agreement on the approach to negotiations was elusive, as each faction competed for influence and rifts verged on the edge of warfare.

180

J. ADOLINO

The extent of these divisions was made manifest by the unprecedented number of ministers4 —from both senior and junior ranks—that quit the May government over Brexit-related policy disagreements. Between June 12, 2017 and April 3, 2019, the government witnessed 45 resignations, with high-profile secretaries of state and departmental ministers stepping down to return to the backbenches. Of these, 34 members of her government, including 9 serving in the Cabinet, departed over issues with some aspect of Brexit, ranging from dissatisfaction with the Prime Minister’s Withdrawal Agreement, to disagreements about the proper role of Parliament, to questions about the legitimacy of the entire Brexit process. All told, May lost more ministers, and at a more rapid pace, than any other prime minister in modern times.5 The Brexiters Exit Overwhelmingly, as Fig. 7.1 shows, it was supporters of a hard Brexit who made the decision to leave the Government on policy or political grounds. Of the 34 officials who resigned for policy or political disagreements in 2017–2019, 19 voted Leave in 2016, and, of these 19, 17 were members of the Conservative Party’s populist European Research Group (ERG). This group of nearly 90 backbench Conservative MPs, one of several private membership voting groups within the Tory parliamentary

Fig. 7.1 Timeline of Resignations from May Government, June 2018 to April 2019

7

MAKING A HASTY BREXIT? TURNOVER AND BREXIT NEGOTIATIONS

181

party, became the primary pro-Brexit group in the House of Commons after the 2016 EU referendum. Led by Jacob Rees-Moggs, the ERG was unwaveringly committed to the cleanest possible break with the EU and, as such, was the home for the “hardest” Brexiters in the party. Its efforts were the key driver behind May’s record 230 vote majority defeat on her Brexit deal on January 15, 2019 and at all subsequent Brexit junctures in the Commons they worked to obstruct the Prime Minister’s plans. Every resignation by an ERG member was based on a feeling that the PM’s deal would keep the UK too close to the EU. Not coincidentally, the vast majority of the Brexiters’ resignations corresponded to key markers on the withdrawal process timeline. As Fig. 7.1 indicates, the Brexiters’ exodus generally came in waves: (1) in the aftermath of the long-awaited revelation of May’s blueprint plans for Brexit at a full Cabinet meeting at Chequers in July 2018; (2) the tabling of the Withdrawal Agreement in November 2018; and (3) in concert with “meaningful votes” on the Agreement in January and March 2019. For example, following the Cabinet meeting at Chequers, two Cabinet level and three junior ministers resigned in a 24-hour period. All told, there were nine resignations by Leave-supporting ministers between July 8, 2018 and July 16, 2018. The resignations of Davis and Boris Johnson— two of the three original hard Brexiters in the Cabinet—were viewed as a heavy blow to the Prime Minister’s authority. At this point, Brexiter outrage over the Chequers Plan and May’s secretive and cloistered approach to her cabinet came into in full view. Once May tabled her Withdrawal Agreement in November 2018, the floodgates of ministerial resignations re-opened with even greater force, with seven Leave resignations on a single day, November 15, 2018, including two by Secretaries of State. Subsequent resignations by hard Brexiters around the first “meaningful vote” on the Withdrawal Agreement on January 15, 2019 and the vote to delay Brexit on April 3, 2019 continued the exodus in opposition to May’s deal. The return of the hard Brexiters—in particular, David Davis and Dominic Raab, both Secretaries of State for DEXEU, and Boris Johnson, the Foreign Secretary—to the parliamentary backbenches marked the end of May’s strategy of keeping her enemies close. The resignation over policy disagreement by some of the Conservative Party’s leading members, not to mention the most important members of her Cabinet on Brexit matters, was a tangible articulation of the outrage among committed Brexiters in the ERG and on the backbenches more broadly.

182

J. ADOLINO

Their return to the House of Commons provided additional leadership to and strengthened the resolve of both groups and further inflamed the discord between the Government and the House of Commons. In addition, their resignations arguably inspired Leave-supporting junior ministers to follow their lead and quit the Government. Upon leaving office, Davis, Raab, and Johnson engaged in very public attacks on the Prime Minister and her plan, and they continued to orchestrate a behindthe-scenes as well as a public assault on the Government. Two of the three, Raab and Johnson, also worked tirelessly to position themselves for the prime ministerial post once May stepped down (as she had promised to do once an agreement was approved), with Johnson ultimately proving successful in that regard. Even Remainers Leave Ministerial resignations by more centrist Conservative Party members who supported Remain followed a different pattern. This smaller group of ministers’ Brexit-related exits were more often based on quarrels with the Government’s approach to the Brexit decision-making process itself, rather than with the substance of the Prime Minister’s deal (although dissatisfaction with aspects of the substantive content of the Government’s Brexit options might also be in evidence). These departures came, for example, from those who disagreed with the Government over its management of the parliamentary process or who were advocating for May to support a second referendum. The latter, in particular, involved members of the small Conservatives for a People’s Vote group (PVG). The exceptions to this overall pattern were two ministers who had supported the Remain side in 2016, but transferred their allegiance to the Leave side and joined the ERG once joining the Government—like the other Leavers, these ministers exited in disputes over the content May’s Brexit plan. The first Remain supporter to leave the May government was Lord George Bridges, a pro-EU peer and minister at the Department for Exiting the European Union (DEXEU) in June 2017. He had served the Government since 2016 and in his role at DEXEU was charged with pushing Brexit legislation through the House of Lords, as well as working out the fine details of the Great Repeal Bill (2016). Bridges departed after a falling out with May on the lack of consultation between No. 10 and DEXEU; his departure was viewed as an early indication of the turmoil within DEXEU.

7

MAKING A HASTY BREXIT? TURNOVER AND BREXIT NEGOTIATIONS

183

Some Remain supporters who were affiliated with the PVG group resigned airing concerns about the Withdrawal Agreement, but the precipitating events for their resignations were matters of procedure. In July 2018, Defense Minister Gutto Bebb quit his frontbench role in the government to vote against an amendment to Brexit customs legislation tabled by the ERG’s leader, Jacob Rees-Mogg. Bebb said he felt duty-bound to oppose the amendment because he could not support the Prime Minister’s willingness to accept a series of backbench “wrecking amendments” to the blueprint offered at Chequers. In Bebb’s view, the amendments were a procedural maneuver intended to publicly undermine May’s authority to craft a Brexit deal. The Transport Minister Jo Johnson, another PVG member and committed Remainer, stepped down on November 10, 2018, as the debate over the Withdrawal Agreement was heating up. In his resignation letter, he told the Prime Minister: It is now my intention to vote against this Withdrawal Agreement. I reject this false choice between the PM’s deal and “no deal” chaos. On this most crucial of questions, I believe it is entirely right to go back to the people and ask them to confirm their decision to leave the EU and, if they choose to do that, to give them the final say on whether we leave with the Prime Minister’s deal or without it.

In his resignation letter, Johnson resignation made clear his intention to ardently campaign for a second referendum because of his discomfort with how the Brexit decision-making process had unfolded since 2017. In a similar vein, and also in November 2018, Sam Gyimah, the Universities Minister and a PVG member, resigned in protest after the Withdrawal Agreement had been revealed and, like Jo Johnson, actively campaigned in his resignation letter for a second referendum. Other Remain supporting ministers left the government in protest over strictly procedural matters. None of these officials were members of the PVG, and their decisions to leave their posts were focused on very specific votes in the Commons. For example, the resignation of Phillip Lee, Minister for Justice, in early June 2018 happened right before a key vote in Parliament about whether to give the House of Commons a role in approving the final Brexit outcome. In his departure letter, Lee wrote: The main reason for my taking this decision now is the Brexit process and the Government’s wish to limit Parliament’s role in contributing to

184

J. ADOLINO

the final outcome in a vote that takes place today For me resigning is a last resort—not something I want to do but something I feel I must do because, for me, such a serious principle is being breached that I would find it hard to live with myself afterwards if I let it pass.

Alberto Costa, the Parliamentary Private Secretary at the Scotland Office, resigned (February 27, 2019) to table an amendment in opposition to the Government’s position to protect EU citizens in Britain if the UK left the bloc without a deal, while Sarah Newton, the Minister of State for Disabled People, Work and Health, defied the Conservative Party whip (March 13, 2019) to vote in favor of an amended government motion that called for a no-deal Brexit to be ruled out in all circumstances, not just at the end of March 2019. In a similar vein, junior ministers Brine, Harrington and Burt resigned from the Government to vote to give Parliament control of the Brexit process. All quit their posts in April 2019 to back a cross-party amendment to allow a series of indicative votes on possible solutions to the Brexit impasse. Notably, Harrington had been one of May’s earliest supporters, serving as Treasurer for her 2016 leadership campaign, so his resignation announcement, in which he accused the Prime Minister of “playing roulette” with the lives of the British people suggested that May had lost control over not just the parliamentary process, but her party. Brine was also a 2016 May loyalist. Burt said he “opted to defy the whip for the country’s sake” and that parliament should seek other options without the instruction of party whips and the government and “should adopt any feasible outcome on its own.” Taken together, such words and actions were a clear rejection of the Government’s authority and the Prime Minister’s approach to decision-making, while the mass resignation of individuals who had previously been ultra-loyal to the Prime Minister strongly suggested a crisis of confidence within her party. Only two Remain supporters—Ben Bradley, a Conservative Party ViceChairman (July 10, 2018) and Shaleish Vara (November 15, 2018), the Northern Ireland minister, resigned over strictly substantive objections. Here, though, although each voted Remain in the 2016 referendum, they had made their ardent support for a hard Brexit clear once they joined the May government, as demonstrated by their membership in the European Research Group. In that sense, their motivations for resignation were closely aligned with those of the Leave supporters. Bradley’s position in this regard was quite clear:

7

MAKING A HASTY BREXIT? TURNOVER AND BREXIT NEGOTIATIONS

185

I admit that I voted to Remain in that ballot. What has swayed me over the last two years to fully back the Brexit vision is the immense opportunities that are available from global trade, and for the ability for Britain to be an outward looking nation in control of our own destiny once again. I fear that this agreement at Chequers damages those opportunities, that being tied to EU regulations, and the EU tying our hands when seeking to make new trade agreements, will be the worst of all worlds.

Vara was the first minister to quit the government after the draft deal was revealed in November, noting that the result of the referendum was “decisive” and that the Government “must deliver.” Vara indicated that he could not support the Withdrawal Agreement because “it leaves the UK in a halfway house with no time limit on when we will finally be a sovereign nation.” Location Matters The likelihood of success in the Brexit negotiations also was affected by the patterns observed in the distribution of resignations across ministries and ministerial ranks. Across ministries, the highest number of resignations were from ministers in the Department for Exiting the European Union (DEXEU). Of greatest significance here is the fact that, in less than three years, three different people served in the DEXEU Secretary of State position—David Davis, Dominic Raab, and Stephen Barclay. Further, between June 2017 and April 2019, the ministry lost seven officials overall, including these top positions, to resignations over policy disagreements with the government. By March 2019, only one junior DEXEU minister had been in post since the Department was formed in 2016. Such a high rate of turnover created a notable level of instability—as well as loss of expertise—in the department that had primary responsibility for coordinating Brexit. Looking across the entire body of resignations, one other clear pattern presents itself: the reshuffling of ministerial ranks was greatest at the junior level. On the face of it this may seem less significant than the turnover of Cabinet level officials given the “junior” status of these individuals. But, in fact, the high rate of turnover of junior ministers is at least as significant as those of the senior ministers because of the key role these officials play in the British system of government regarding policy-making. Junior ministers are responsible for driving policies through their ministries and

186

J. ADOLINO

both houses of Parliament. They are also frequently called upon to represent their departments in Parliament, serving as important channels of information and accountability. So, churn at this level clearly diminished the government’s policy-making capacity—with further loss of expertise and experience. A Government in Disarray By early April 2019, as the resignation tide waned, it was clear that the May government’s Brexit policy apparatus in no way resembled a Rolls-Royce operation. Perhaps more importantly, political elites—in the Government, the Parliament and the civil service—were no longer “singing from the same hymn sheet”—as expressed most concretely in unprecedented levels of ministerial turnover. Of course, some might say that the high level of resignations and the political churn they generated—and the sense of political turmoil more generally–is simply business as usual on the European Union front. Since the late 1950s, there is no question that in both the Cabinet and Parliamentary arenas questions surrounding EU membership have routinely been highly contested— concerns about both national and parliamentary sovereignty vis-a-via the EU have been the source of persistent divisions both within and between the two major parties. Time and again, intra-party divisions combine with inter-party adversarial politics to turn European integration into a conflictual issue in domestic politics. Arguably, however, setting the modern record for ministerial resignations over Brexit provides confirming evidence of the deep existing fault lines over the decision to leave the European Union within the Cabinet, the Parliament, and, of course, the British citizenry. By March 29, 2019, May had proven unable to bridge any of these fault lines and thus was forced to ask the EU for an extension to the Article 50 deadline. In July 2019, her premiership came to an end without a Brexit deal in place.

Conclusions and Implications: Brexit, Turnover, and Output Legitimacy Principles of collective responsibility, foundational to the British system of government, hold that government ministers—in particular the cabinet— must publicly agree and abide by government decisions (irrespective of personal views) and that they should not disclose what was said privately

7

MAKING A HASTY BREXIT? TURNOVER AND BREXIT NEGOTIATIONS

187

among ministers in reaching those decisions (Taylor 2019). This principle should have made the UK government’s experience with the Brexit negotiations seamless, unified, and coherent. Instead, the high rate of ministerial turnover laid bare the breakdown of this norm, revealing a deeply divided government, the breakdown of party discipline, and, as a result, the impossibility of strong majoritarian rule under May. Further, successive rounds of personnel turnover demonstrated May’s inability to control her Government or the Conservative Party—both of which were increasingly impaired by indecision, infighting, and instability on Brexit. Indeed, each individual resignation was a public manifestation of the division within her ranks and a direct challenge to her authority, with some leaving the Cabinet over process concerns, others over the form Brexit should take, and others in protest over both. Whatever the reason, cumulatively the resignations served as both a strong indicators and further drivers of government dysfunction. As Schimmelfennig (2018) notes, the process dynamics associated with differentiated disintegration from the outset imparted significant advantages to the EU side. The EU not only recognized but effectively exercised its strong institutional and material bargaining power and took advantage of the EU-favoring asymmetric relationship between the EU-27 and the UK. However, the UK’s position of weakness relative to the EU was clearly exacerbated by the governing turmoil created by ministerial turnover. The steady stream of ministerial resignations further exacerbated the political and institutional chaos that had characterized governance in the UK from 2016 forward, disadvantaging the UK relative to its more stable, composed, and confident EU counterparts. High levels of ministerial turnover subjected May to constant, divisive political pressures that ultimately made it impossible for her government to conduct business in a manner consistent with the UK’s historic EU negotiation record nor able to compensate for structural weaknesses in the British position vis-à-vis the Union that were inherent to this process of differentiated disintegration. More specifically, the unprecedented breakdown in ministerial discipline undermined May’s subsequent ability to control the Brexit narrative, as key members of her team openly rejected her vision, leadership, and authority in pursuit of their own Brexit objectives and political ambitions. In the face of such defiance, which spilled over into the parliament as these ministers returned to their House of Commons seats, it proved impossible for May to set the terms of the debate, to navigate the underlying power

188

J. ADOLINO

dynamics of the negotiations, or to project strength (in either the UK or EU political arena). Turnover, of course, was an articulation of preexisting deep divisions within the Cabinet and governing party, but it also further fractured the UK side, strengthening the hand of those who were most opposed to the Prime Minister’s positions, and further inflaming an already volatile political context, as pre-existing factions within the Conservative Party shaped—and rendered asunder—the UK narrative, crippling May’s ability to convincingly project strength. Not surprisingly, this government turmoil also made it impossible to effectively navigate the Brexit negotiations—at work here were all the perils associated with personnel turnover: amateurism, poor-quality policy, and disunity. The regular exodus of ministers deprived the government of expertise and experience; this loss of expertise, while manifest in the Cabinet itself, was especially critical within the ministries, where junior ministers are relied upon to develop and deliver policies; turnover at this level depleted the knowledge base and made intra- and inter-institutional information-sharing and coordination more difficult. In addition, unable to manage her government at home, May did not arrive in Brussels as a force to be reckoned with, backed by an experienced, unified, focused, or loyal team. Instead, ministerial turnover served as a burning bush for disunity and weakness for all to see, undermining her negotiating position. By early 2019, senior EU diplomats negotiating Brexit declared the UK political arena “a circus that is beyond comprehension” (Barker 2019) while another likened bargaining with the UK to “dealing with a failed state.” All told, the effects of ministerial turnover played a strong role in the government’s failure to conclude a withdrawal agreement and, as such, to deliver the referendum result. The patterns of personnel turnover observed suggest that minimizing turnover among negotiating teams can help to boost output legitimacy, while widespread personnel replacements imperil negotiating success and undermine output legitimacy. More specifically, a lack of personnel turnover on the EU side can be viewed as having enhanced output legitimacy. During this phase of the Brexit negotiations, personnel consistency certainly helped the EU to maintain unity and focus and to effectively leverage its negotiating team’s expertise to achieve its desired outcome. In the end, the EU team appeared strong and in control and, as such, projected power and authority on the world stage. Polls routinely show that European citizens will see the EU as more legitimate if and when EU foreign policy becomes more effective. To the extent that this is the case,

7

MAKING A HASTY BREXIT? TURNOVER AND BREXIT NEGOTIATIONS

189

the EU side’s performance during the May period of the Brexit negotiations suggests that minimizing turnover among foreign policy personnel can help to increase the EU’s output legitimacy. On the other hand, output legitimacy was likely undermined by excessive turnover on the UK side. Beyond the failure to reach an agreement during her tenure in office (thus producing no “output” for citizens to embrace), there is also ample evidence of high levels of dissatisfaction with what the May government had on offer. In the months following May’s January 2017 Lancaster House speech, in which she laid down her “red lines” for the Brexit negotiations, almost half the British population thought the government was handling the negotiations “well.” But, by July 2017, this faith was in steep decline, with a majority of people believing that the Prime Minister would not be able to secure the right deal and that her government was doing a bad job of negotiating Brexit. By February 2019 this opinion was held by an overwhelming majority of 78% (Hobolt 2019). After two years of negotiations, only 13% of people thought the government was doing “well.” By this point, a vast majority–both Remainers and Leavers–demonstrated dissatisfaction with the government’s handling of Brexit. With respect to the Withdrawal Agreement specifically, polls also consistently indicated low levels of support. For example, the polling group YouGov asked people whether they supported or opposed May’s deal no fewer than 11 times between November 2018 and January 2019. The highest level of support for the deal over this period was 27%, while at least 42% were always against it. These levels of opposition were strong among both Leave and Remain supporters (Curtice 2019). Across the board, then, by March 2019, the government’s deal was widely regarded as a failure. Such opinions suggest problems for output legitimacy in the UK: UK citizens did not see the government as acting “for the people.” If output legitimacy is thought to be present when problem-solving quality—the effectiveness and performance—of laws and rules is judged to be high, the failure to produce an agreement that enjoyed a high degree of public support raises legitimacy concerns. Further, the failure to deliver on the referendum result in the form of an agreement with the EU adds to citizens’ feeling that the government was ineffective. In the end, because personnel turnover slowed decisionmaking and weakened government, it no doubt had a detrimental effect on citizens’ attitudes toward their government.

190

J. ADOLINO

Notes 1. Arguably, Brexit is an exceptional negotiating case, with both internal and external dimensions, in particular prior to the approval of the Withdrawal Agreement. During the 2017–2019 talks, the UK was technically still an EU member; Brexit becomes more a matter of foreign policy after the approval of the Withdrawal Agreement in January 2020. Nonetheless, since even early Brexit discussions were about establishing eventual external relations, they arguably had a foreign policy dimension. Further, the requirements of Article 50 forced Teresa May out of European Council meetings, so she effectively dealt with the EU as an external actor facing off against the EU-27. 2. In and of itself, this is not unusual. Prime ministers often look to keep their opponents close within their Cabinets. For example, Tony Blair’s cabinets included Blairites and Brownites. 3. Further complicating Cabinet dynamics, some appointees who originally supported the Remain side subsequently hardened their positions and became strong Brexit advocates (for example, Sajid David, Jeremy Hunt, Gavin Williamson). 4. The ministerial category includes senior ministers (or secretaries of state) who head government departments; they are the most powerful figures in British government. There are three ranks of junior ministers: minister of state, parliamentary under-secretaries of state, and parliamentary private secretaries. The Ministerial Code clearly requires all ministers to support Government in divisions in the House; they cannot retain their positions if they vote against the Government. 5. While high rates of ministerial turnover are not unusual in the British system, ministerial resignations for policy and political disagreements are not commonplace. Between 2017 and 2019, May’s loss of 34 members of her government due to policy and political disagreements exceeded Margaret Thatcher, Tony Blair and Gordon Brown’s combined experience in 22 years in office. In the first three months of 2019 alone, May had already lost more ministers in a single year than any other recent prime minister (with the exception of herself, in 2018) (Lloyd 2019). Her government also saw resignations by at least three ministers in the same 24-hour period three times from 2018 forward. Between 1979 and 2017, this had happened only once, in 1982, with the resignation of Lord Carrington, Humphrey Atkins and Richard Luce in protest over the Falklands mission (Freeguard 2019).

7

MAKING A HASTY BREXIT? TURNOVER AND BREXIT NEGOTIATIONS

191

References Atkinson, Michael, and David Docherty. 1992. Moving Right Along: The Roots of Amateurism in the Canadian House of Commons. Canadian Journal of Political Science 25 (2): 295–318. Barker, Alex. 2019. EU Frustration with UK Rises as Brexit Talks Near Final Hours. Financial Times, March 20. Accessed June 2, 2020 at https://www. ft.com/content/80bcdb6a-4b19-11e9-bbc9-6917dce3dc62. Bulmer, Simon, and Martin Burch. 2009. The Europeanization of Whitehall: UK Central Government and the European Union. Manchester, UK: Manchester University Press. Curtice, John. 2019. Brexit: What Voters want MPs to Do Next. UK in a Changing Europe, January 14. Accessed June 2, 2020 at https://ukandeu. ac.uk/brexit-what-voters-want-mps-to-do-next/. European Council. 2017. Guidelines Following the United Kingdom’s Notification under Article 50 TEU. April 29. Accessed June 2, 2020 at https://www. consilium.europa.eu/media/21763/29-euco-art50-guidelinesen.pdf. Francis, Wayne, and John Baker. 1986. Why Do U.S. State Legislators Vacate Their Seats? Legislative Studies Quarterly 11 (1): 119–126. Freeguard, Gavin. 2019. The Scale of Brexit Resignations Is Now so Great That Theresa May Could Run Out of Potential Ministers. Prospect, March 29. Accessed June 2, 2020 at https://www.prospectmagazine.co.uk/politics/the resa-may-government-ministers-resignation-brexit-talent. Henley, Jon, and Dan Roberts. 2018. 11 Brexit Promises the Government Quietly Dropped. The Guardian, March 28. Accessed June 2, 2020 at https://www.theguardian.com/politics/ng-interactive/2018/mar/ 28/11-brexit-promises-leavers-quietly-dropped. Hibbing, John. 1991. Legislative Careers: Why and How We Should Study Them. Legislative Studies Quarterly 20 (7): 149–171. Hobolt, Sara. 2019. Divided Over Brexit. United in Despair Over Politicians’ Handling of Brexit. UK in a Changing Europe, April 12. Accessed June 2, 2020 at https://ukandeu.ac.uk/divided-over-brexit-united-in-despair-overpoliticians-handling-of-brexit/. Kassim, Hussein. 2016. London Calling: Revisiting the National Coordination of EU Policy in the UK. Revue Française d’administration Publique 158 (2): 431–446. Lloyd, Lewis. 2019. The Brexit Effect: How Government Has Changed Since the Referendum. London: Institute for Government, March 29. Accessed June 2, 2020 at https://www.instituteforgovernment.org.uk/publications/brexiteffect. Matland, Richard E., and Donley T. Studlar. 2004. Determinants of Legislative Turnover: A Cross-National Analysis. British Journal of Political Science 34 (1): 87–108.

192

J. ADOLINO

Maurice, Eric, 2018. Brexit: Lessons in Negotiations for the EU. Schuman Foundation, November 26. Accessed June 2, 2020 at https://www.robertschuman.eu/en/european-issues/0494-brexit-negotiation-lessons-for-the-eur opean-union. McConnell, Allen, and Simon Tormey. 2019. Explanations for the Brexit Policy Fiasco: Near-Impossible Challenge, Leadership Failure or Westminster Pathology? Journal of European Politics and Policy 27 (5): 685–702. Niemi, Richard G., and Laura R. Winsky. 1987. Membership Turnover in U.S. State Legislatures: Trends and Effects of Districting. Legislative Studies Quarterly 12 (1): 15–23. Patel, Oliver. 2018. “The EU and the Brexit Negotiations: Institutions, Strategies and Objectives. UCL European Institute, October. Accessed June 2, 2020 at https://www.ucl.ac.uk/european-institute/news/2018/oct/eu-andbrexit-negotiations-institutions-strategies-and-objectives. Perez, Lauren K., and John A. Scherpereel. 2017. Vertical Intra-institutional Effects of Ministerial Turnover in the Council of the European Union. Journal of European Public Policy 24 (8): 1154–1171. Putnam, Robert. 1976. The Comparative Study of Political Elites. Englewood Cliffs, NJ: Prentice Hall. Rosenthal, Alan. 1974. Turnover in State Legislatures. American Journal of Political Science 18 (3): 609–616. Scharpf, Fritz. 2003. Problem-Solving Effectiveness and Democratic Accountability in the EU. Max Planck Institute Working Paper 3(1). Accessed June 2, 2020 at http://hdl.handle.net/10419/41664. Scherpereel, John, and Lauren K. Perez. 2015. Turnover in the Council of the European Union: What It Is and Why It Matters. Journal of Common Market Studies 53 (3): 658–673. Schimmelfennig, Frank. 2018. Brexit: Differentiated Disintegration in the European Union. Journal of European Public Policy 25 (8): 1154–1173. Shin, Kwang, and John Jackson. 1979. Membership Turnover in US State Legislatures, 1931–1976. Legislative Studies Quarterly 4 (1): 95–104. Taylor, Robert Brett. 2019. Brexit and Collective Cabinet Responsibility: Why the Convention is Still Working. LSE Blog, May 20. Accessed June 2, 2020 at https://blogs.lse.ac.uk/brexit/2019/05/20/brexit-and-collectiveresponsibility-why-the-convention-is-still-working/.

CHAPTER 8

Turnover, Conditionality, and Europeanization in the Western Balkans John Hulsey

The European Union (EU) has been a powerful force for political and economic change in Central and Eastern Europe; however, the pace of change in Southeastern Europe has been uneven and halting. While Slovenia raced through the accession process, those countries more directly impacted by Yugoslavia’s violent dissolution have moved much more slowly. Croatia’s accession took a full decade longer than Slovenia’s, and the pace of reform in the remaining Western Balkan countries suggests a long and uncertain path toward accession. Even worse, recent critics of the EU role in the region argue that the EU’s approach increasingly reinforces stagnation through its focus on stability over reform and inconsistent application of accession conditions (Balkans in Europe Policy Advisor Group 2017; Bassuener and Perry, n.d.). The slow pace of reform

J. Hulsey (B) Department of Political Science, James Madison University, Harrisonburg, VA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. A. Scherpereel (eds.), Personnel Turnover and the Legitimacy of the EU, Palgrave Studies in European Union Politics, https://doi.org/10.1007/978-3-030-60052-5_8

193

194

J. HULSEY

and long time horizons raise serious questions about the continued effectiveness of conditionality in bringing about improved democracy and governance in the region. On the one hand, this failure casts doubt on the role of conditionality to handle the kinds of problems faced by the remaining countries of the Western Balkans. On the other hand, long and uncertain time horizons themselves weaken conditionality by decreasing the expected value of accession relative to the immediate costs of change. This chapter examines the declining impact of EU conditionality in the Western Balkans—and, in particular, in Bosnia and Herzegovina—from the perspective of personnel turnover and its resulting threats to the EU’s output legitimacy. Case studies of police reform and constitutional reform efforts in Bosnia and Herzegovina link turnover among the EU personnel responsible for spearheading state-building and accession to a repeated failure to stick to stated conditions for progressing toward EU membership. As a result, Bosnia and other accession countries have progressed toward membership without completing fundamental reforms necessary for membership. The failure of these outputs of the EU’s engagement with the Western Balkans threatens the legitimacy of the accession process (a) among politicians in candidate states, who can claim the rewards of moving forward towards accession without undertaking reforms that would threaten their power, and (b) among the citizens and politicians of the EU, due to the inability to bring about reforms necessary to complete accession (Maier and Rittberger 2008). The results of this lack of legitimacy in the outputs of the accession process are significant: regional politicians are able to flout democratic norms, and public support for accession within the EU decreases (Lavic and Bieber 2020). The length and intensity of international intervention, including the European Union and its member-states, in the Balkans and especially in Bosnia, means that international personalities, who are somewhat obscure on the European stage or even in their home member-states, are important public figures in the Balkans. The most visible example of this is the parallel life of the word “Dayton” in Bosnia as result of the choice of Wright-Patterson Air Force base for the location of the negotiations that brought about the end of the Bosnian war and created the current constitution of Bosnia and Herzegovina. The phenomenon is also manifested by a range of middling member-state politicians with outsized roles including Miroslav Lajˇcák, Paddy Ashdown, Carl Bildt, Christian Schwarz-Schilling, Johannes Hahn, and Lars-Gunnar Wigemark. The accession process is a

8

TURNOVER, CONDITIONALITY, AND EUROPEANIZATION …

195

peripheral issue for EU member-states, but the prospect of EU membership is of core concern to politicians and citizens of Western Balkans countries. This asymmetry gives those member-state and EU officials tasked with promoting democracy and economic development in Bosnia outsized influence and agency. The chapter explores the extent to which turnover among these EU figures may help to explain the variable influence of conditionality. In short, European Union elites who are tasked with promoting democracy and accession have significant influence and freedom for action. Turnover among these EU representatives may also be driving some of the instability in EU policy outcomes related to accession. The chapter proceeds by first examining overall trends in democracy and governance outcomes in the Western Balkans as a way of establishing the lack of progress in the region. It continues with a discussion of the connection between personnel turnover and the existing literature on conditionality and Europeanization. Next, the chapter examines the links between turnover and policy variability in the application of conditionality in Bosnia and Herzegovina. It concludes with reflections on the implications of personnel turnover for the EU’s engagement with Western Balkan countries and for EU output legitimacy more broadly conceived.

Democracy and Governance Outcomes in the Western Balkans Figure 8.1 shows the World Bank’s Worldwide Governance Indicators (http://info.worldbank.org/governance/WGI/) for the countries of former Yugoslavia and Albania. They paint a sobering picture. While poorly performing countries improved across all indicators from 2000 to 2010, the pace of improvement has declined and eventually leveled out in a region-wide pattern. There are many potential explanations for this decline, including that the easiest reforms were accomplished during the early stages of EU partnership and that the underlying standards behind the measures have become tougher, but there is little evidence for the overall effectiveness of the accession process. However, the deceleration also coincides with declines in political competition in Western Balkan countries and a series of crises in the EU that have taken focus away from enlargement. For each country in Fig. 8.1, the vertical lines denote critical junctures in each country’s path toward accession: signing a Stabilization and Association Agreement, Candidacy Status, and membership. Each of these

J. HULSEY

Candidate

SAA Signed

Bosnia and Herzegovina SAA Signed

Albania 1 0 −1

Member

SAA Signed

Kosovo

Candidate

SAA Signed

Croatia 1 0 −1

SAA Signed

Montenegro

Candidate

SAA Signed

N. Macedonia 1 0 −1

1995

2000

2005

2010

Control of Corruption

2015 Rule of Law

1995

Member

Candidate

Candidate

SAA Signed

Slovenia SAA Signed

Serbia 1 0 −1

Candidate

196

2000

Regulatory Quality

2005

2010

2015

Voice and Accountability

Fig. 8.1 Worldwide Governance Indicators for Western Balkan Countries, 1995–2018

steps are focal points for the application of conditionality and represent the fulfillment of formal criteria. Yet, it is notable that a country like Croatia, which moved through the stages of the accession process all the way to full membership, did not experience major improvements in the indicators. These measures comport with scholarly assessments of the political systems of the region (Balkans in Europe Policy Advisor Group 2017; Hulsey 2018).

Europeanization, Domestic Politics, and Personnel Turnover The political and economic transformation of Eastern Europe is a clear example of the power of external actors to drive democratization. The European Union has been the lead actor in that transformation both for the countries that have achieved member-status and those whose membership is called into question or thrown into doubt altogether. However, democratic backsliding in Central Europe and the slow pace or even reversal of change in the Balkans suggest that there are powerful limits to the impact of international intervention (Vachudova 2015). The most influential work on external influence by the European Union focuses on the ways in which the rewards and requirements of the

8

TURNOVER, CONDITIONALITY, AND EUROPEANIZATION …

197

accession process create incentives for political elites in target countries to adopt laws and practices in order to better align with the EU acquis communautaire as well as to the demands of EU bodies and memberstates. The more that material benefits to the target state exceed the costs of adopting EU rules, the greater the incentive to adopt those rules (Schimmelfennig and Sedelmeier 2004; Schimmelfennig et al. 2003). This core finding has been supplemented by work demonstrating the ways in which states’ capacity for bringing about reform varied significantly across states and across reform areas (Zhelyazkova et al. 2018: 16–17). In particular, the states of the Western Balkans are poorer than the Central European states. They face issues with the rule of law and contention over the fundamental structure of the state that impede the effectiveness of conditionality independent of the determinacy and credibility of the accession process (28–29). External influence also plays out in the context of internal party politics whereby the success of conditionality can powerfully shape party competition in target countries just as national politics can powerfully condition the impact of conditionality. On the one hand, both the material draw of the EU and socialization of elites worked to push parties to adopt a pro-EU approach. On the other hand, the impact of these incentives on party competition played out in complex ways. In some contexts, it seems also to have created political opportunities for extreme parties (Grzymala-Busse and Innes 2003). In others, vibrant political competition was a key contributing factor that pushed parties to adopt pro-EU policies (Vachudova 2005). Given the demonstrated power of the EU to bring about political and economic change, why has this influence stalled in the Western Balkans? Existing explanations fall broadly into two categories—those that focus on differences between the Western Balkans and earlier enlargement countries, and those that focus on changes in the implementation from the side of the EU. Most importantly, the EU accession process has struggled to come to grips with fundamental challenges to the rule of law and democratization, which threatens to undermine the de facto impact of de jure EU rule adoption (Kmezic 2018: 93). While these concerns are rooted in the Copenhagen Criteria, over time the EU has become much more systematic in tying rule of law questions to other areas of the acquis that are built upon the assumption of the rule of law (Zhelyazkova et al. 2018: 24–25). However, that questions of the rule of law pose less of a problem of technical alignment and more a fundamental political problem means

198

J. HULSEY

that the tools that have worked in aligning economic rules have been less effective with regard to the rule of law (Kmezic 2018). In fact, questions of the rule of law and suppression of corruption cut to the heart of national leaders’ bases of power and livelihood, which strongly divides the interests of leaders in power from the interests of the people as a whole (Vachudova 2018: 66, 73). To the extent that payoffs from conditionality are broadly shared while the costs of implementing the rule of law are born by government elites, the fundamental mechanism of conditionality breaks down. A second major challenge for the Balkans is the way in which required changes, as well as the fundamental freedoms of the Single Market, come into conflict with questions of national sovereignty. While such concerns were present for earlier entrants to the EU, the recentness and violence of the dissolution of Yugoslavia means that fundamental questions of statehood are still contested. As a result, those countries with the least state contestation have had easier paths toward Europeanization (Dzankic and Keil 2018). Additionally, these two challenges reinforce one another, as corrupt political leaders use national questions both as a means of electoral mobilization and as a shield against too-aggressive EU intervention. As a result, efforts to promote state stability, both through power-sharing political settlements and efforts at state capacity-building serve to strengthen established leaders and their clientelist networks. By working through clientelist leaders, the EU may be strengthening the greatest threat to the rule of law in the Balkans (Balkans in Europe Policy Advisor Group 2017: 93; Richter and Wunsch 2019).

Personnel Turnover and EU Leverage This chapter adds to the vibrant literature on EU influence in the Balkans by linking to an emerging literature that focuses on the effects of personnel turnover on EU politics. Recent work on the Council of the European Union has shown that turnover is high among ministers in the Council of the European Union and that configurations with higher levels of turnover suffer in negotiations with the other institutions of the European Union (Scherpereel and Perez 2015). Furthermore, high turnover is associated with lower levels of ministerial control in the Council (Perez and Scherpereel 2017). EU rules, including EP elections, the formation of the Commission, and the rotating presidency of the Council of the EU further introduce periodicity to the work of the EU that may reduce the

8

TURNOVER, CONDITIONALITY, AND EUROPEANIZATION …

199

EU’s effectiveness in formulating and enforcing conditions as part of the accession process. High levels of turnover in EU institutions may affect the performance of the EU in tackling the complex work of fostering democracy, the rule of law and EU governance in the Balkans in several ways. First, turnover may be partially to blame for the inconsistency in the enforcement of conditions that has undermined conditionality (Vachudova 2018: 66). The roots of this inconsistency could stem from either differences of opinion over time in the proper conditions to be put in place or in attempts to complete tasks within the time in office of a particular officeholder or country. Such inconsistency over time undermines conditionality by calling into question the credibility of EU commitments. Second, high turnover among EU officials contrasts with low levels of turnover among national-level officials in the Balkans. The weakening of democratic opposition and the stability of elite clientelistic networks in Serbia and Bosnia raises the possibility that the passage of time will weaken the hand of EU negotiators relative to experienced and entrenched national elites. In this way, the stability approach chosen by the EU threatens to do long-term damage. In order for turnover in EU officials to play a meaningful role, those officials and their organizations need to have autonomy both from member-states and other international actors. Such autonomy has been a goal of the development of the Common Foreign and Security Policy of the European Union, as shown by the establishment of the High Representative for Foreign and Security Policy and the establishment of the European External Action Service (Morgenstern-Pomorski 2018). As the case studies that follow show, the autonomy and profile of EU officials has increased over the last two decades, which has coincided with the disengagement of the United States from the Balkans after its lead role in ending the conflicts there. Today, the European Union is the dominant external actor in the Balkans. The analysis that follows implies variability in the autonomy of EU actors like the EU Special Representative for Bosnia and Herzegovina as well as instances of distinct disagreement between EU officials’ decisions coming from Brussels’ institutions. Further, this variability may be tied to turnover, as newly arrived officials often come from Brussels or non-Balkan postings. The legitimacy of the policy outputs of the EU accession process, in the form of accession conditions, are undermined when the conditions are changed or not enforced. EU actors are caught between two distinct

200

J. HULSEY

audiences for their policy output: the European Union and the publics and politicians of potential member-states in the Western Balkans. In order for conditionality to be effective, the potential member-states must believe that the conditions offered by EU officials will be enforced and the rewards delivered, despite the fact that the final decision on those matters is not controlled by the negotiators but is instead subject to a vote of the member-states. Turnover—either of negotiators or of decision-makers in Brussels—empowers new actors. Additionally, different understandings of the accession process and/or the nature and/or urgency of policy change often accompany the turnover process. While turnover on the side of the EU influences the consistency of EU policy and the leverage of conditionality, the counterpart to EU turnover is turnover in the leadership of candidate countries, which varies dramatically across the region depending on the quality of democratic competition. While government turnover in North Macedonia was associated with a breakthrough on its name dispute with Greece, leadership turnover in Serbia, Montenegro, and Bosnia has been rare and looks unlikely in the future. The following section shows the ways in which elite turnover on the side of the EU has been a source of inconsistency in the application of accession criteria as part of the EU’s enlargement policy, which has empowered local elites to resist those aspects of the process, particularly improvements in the rule of law, that most threaten their political power.

Turnover and Leverage in Bosnia and Herzegovina EU involvement in Bosnia and Herzegovina takes place in the context of a broader multilateral effort. In important ways, the enlargement process expands on and replaces previous attempts to bring about democracy and effective governance in Bosnia. As a result, the impact of the enlargement process in Bosnia must be understood in the context of the overall effort, and in the changing role of the EU over the period of Bosnia’s postconflict history. The core problem of international intervention in Bosnia has been the failure to create lasting coalitions for reform among the elected politicians of the country and therefore the degree to which it is appropriate to reopen constitutional questions inherent in the Dayton Peace Agreement (DPA). The first phase of international intervention after the signing of the DPA in 1995 was characterized by the military and security implementation of the agreement. The DPA is very specific with regard to

8

TURNOVER, CONDITIONALITY, AND EUROPEANIZATION …

201

the military implementation of the agreement. The Inter-Entity Boundary Line between Republika Srpska and the Federation of Bosnia and Herzegovina became a buffer zone delineated and patrolled by international forces. At the same time elections were held according to the structures laid out by the DPA, including for Canton, Entity and BiH-level institutions (the central government for the whole country of Bosnia and Herzegovina). The DPA and resulting constitution contain elements of both territorial division and unification. The process of negotiation and postwar implementation heavily favored territorial division not only between the two entities, but also between Croat and Bosniak areas within the Federation. The initial focus was on enforcing the territorial boundaries created by the agreement. The DPA, however, also included elements—such as the creation of the BiH level government and the prospect of refugee return—that held out the possibility of recreating a Bosnian state. During the first years after the war, wartime ethnic political leaders succeeded in translating their wartime power and control over pseudo-states into postwar dominance of sub-territories. To an extent, this was reinforced by the international community, which gave preference to the stability offered by territorial division over the right of refugee return (Toal and Dahlman 2011). The failure of the newly-created BiH government to agree to even basic legislation, such as an appropriate flag or an election law, led to a more aggressive international intervention. In 1997, the Peace Implementation Council of countries participating in the international intervention in Bosnia agreed that the Office of the High Representative (OHR) had the power to enforce the civilian implementation of the agreement and could remove officials and enact legislation if necessary. These became known as the “Bonn powers” after the location of the meeting. The Office of the High Representative was created as an afterthought to the Dayton Agreement, partly to carve out a role for European countries who had been effectively excluded from the final phase of the US-dominated peace process. Eventually, however, this afterthought became the driver of a more aggressive external intervention. The High Representative, always a European diplomat, enacted basic legislation and removed politicians, particularly Canton and municipal-level politicians who disregarded either BiHlevel laws or the Dayton constitution, particularly with regard to refugee return. In this way, the central state institutions in Bosnia were built.

202

J. HULSEY

In 2000, this effort was buttressed by the Constituent Peoples’ Decision of the BiH Constitutional Court, which established that Bosniaks, Croats and Serbs constituted not only BiH as a whole, but both Entities and all sub-units, some of which had previously been explicitly defined as ethnic enclaves. The OHR was most aggressive under Wolfgang Petrisch (1999–2002) and Paddy Ashdown (2002–2006), who used their Bonn powers widely, especially to finally return property to so-called “minority” returnees, who sought to reclaim property in an area where they were no longer the dominant ethnicity, and from which they had been ethnically cleansed. This was made possible not only by the Bonn powers, but by the expansion of monitoring at the local level (Toal and Dahlman 2011). While Bonn powers can still technically be used by the OHR, they haven’t been used since the end of Paddy Ashdown’s tenure as OHR. The European Union accession process, instead, should be the means of changing the incentives of still recalcitrant Bosnian politicians.

Police Reform and a Stabilization and Association Agreement The OHR model had come under criticism for its colonial nature amid concerns that heavy-handed external influence might be both running roughshod over the rule of law and allowing Bosnian politicians to sidestep difficult decisions by passing them on to the OHR (Knaus and Martin 2003). However, early attempts at the shift from neo-colonial imposition to EU conditionality, which started already during the last years of Ashdown’s time in Sarajevo, did not go smoothly. Ashdown had already been double-hatted as the EU Special Representative for Bosnia and Herzegovina. As a result, he represented both the Office of the High Representative, with a mandate from the Peace Implementation Council, which was made up of all countries contributing to the peacebuilding mission in Bosnia and the European Union. In this role, he stated that a restructuring and significant centralization of Bosnia’s police forces would now be a pre-condition for moving forward with the accession process. Up to this point, most police were under the control of Canton and Entity governments and were often a source of harassment to “minorities” who returned or traveled to ethnically cleansed areas. Ashdown’s declaration resulted in a political crisis in Bosnia, during which the President of Republika Srpska, Milorad Dodik, threatened to march on Sarajevo if Ashdown used his Bonn powers to force a centralization of the police.

8

TURNOVER, CONDITIONALITY, AND EUROPEANIZATION …

203

The inclusion of police centralization as a criterion for accession was weakened by the fact that several EU member-states had very decentralized police forces. Additionally, a genuine reform had the potential to cut to the very heart of the local power bases of political parties and leaders. The police forces are not only an important source of power and influence because of their essential function but because of the large number of jobs in that sector. The local opposition to Ashdown’s declaration and the attendant crisis dragged on beyond Ashdown’s term as High Representative. The next High Representative, Christian Schwarz-Schilling, rejected the use of the coercive OHR power and instead opted for negotiation among the parties in Bosnia. This approach failed to produce movement on a number of fronts, including police reform, and Schwarz-Shilling’s mandate was not renewed after his two year term (Peter 2015: 140–141). His replacement as OHR in July 2007, the Slovak diplomat Miroslav Lajˇcák, brought significant EU foreign policy connections, having served as Javier Solana’s special envoy for Montenegro’s independence referendum in 2006. While offering EU accession as the justification for the reforms, the new OHR reached back to the more aggressive approach used by Ashdown by threatening to use imposition to bring about police reform. This pressure produced a stalemate, whereby Lajˇcák’s threat to use the OHR’s powers was called and shown to be empty in the face of determined opposition from Serb parties (Bennett 2016: 188–189). In the end, a face-saving agreement was negotiated after nearly four years and three High Representatives/EU Special Representatives. The result was a watered-down reform that shifted little power to the center and, as a result, did little to undermine the power bases of local ethnonational politicians, who had been the target of the reform in the first place. In exchange, the EU agreed to initial the SAA with Bosnia, having achieved little from the EU’s leverage (Bennett 2016: 190). There are a host of factors at play in the failure to hold to the conditions initially set out by Ashdown. The nature of the reforms both struck at the heart of ethnonationalist power and fit awkwardly in existing EU practice, giving local opponents both the incentive and the means to oppose the reforms. The latter stages of the negotiation process also took place during a broader context of Kosovo’s independence declaration, which raised awkward questions about Republika Srpska’s own claims for independence, as well as the perceived need to provide political cover for more moderate parties in Serbia. However, EU leadership and policy turned over while the constellation of local leaders and approaches did not.

204

J. HULSEY

´ Constitutional Reform and the Sejdic-Finci Case The pattern shown in police reform has been repeated since Ashdown’s departure from OHR in 2006. The most prominent example includes attempts at constitutional reform. Like police reform, attempts at constitutional revision straddle the OHR approach to peacebuilding focused on imposition and the EU approach to peacebuilding focused on conditionality. When faced with significant obstacles, the OHR approach could resort to the direct removal of politicians and the enactment of legislation. In other words, it was possible to bypass or override democratic political institutions. In contrast, the EU approach seeks to exert leverage on politicians and thereby shift outcomes. Conditionality has been successful in bringing about regulatory reforms and many improvements to life in Bosnia. However, EU leverage has been unsuccessful when the accession process comes up against reforms that have implications for the balance of power between the central BiH institutions and the power of the Cantons and Entities. These constitutional questions take on an ethnonational character, so the incentives offered by the EU do not outweigh the potential losses entailed in giving up political power and regional autonomy. Power that is shifted from the BiH level to the entities gives Serb-dominated Republika Srpska greater independence, while Croat ethnic parties push for more devolution to Cantonal governments within the Federation. The attempt at police reform that began in 2006 coincided with the beginning of the first of two major efforts at externally brokered constitutional reform, aimed at making the state more centralized and efficient by modestly curtailing some of the power-sharing and territorial autonomy requirements. In both efforts, the focal point that put constitutional change on the agenda resulted from cases before international courts. The first attempt at reforms responded to the Venice Commission’s identification of violations of the European Convention for the Protection of Human Rights and the European Charter of Local Self-Government (Perry 2015: 492–494). In a report instigated by a challenge to the powers of the OHR, the Venice Commission pointed out the obvious undemocratic nature of the OHR, but emphasized that the state structure in the constitution also included a number of provisions that were in violation of international charters, particularly veto rights and patterns of representation based on ethnic group and not the individual (Venice Commission 2005).

8

TURNOVER, CONDITIONALITY, AND EUROPEANIZATION …

205

At first, the constitutional reform effort was spearheaded by the United States. It was put forward not only on democratic grounds, though, but also as necessary step for the creation of a central government capable of meeting the demands of the acquis and, therefore, of EU accession. The result of months of intense engagement between American diplomats and parties in government resulted in a set of constitutional changes called the “April Package,” which would have removed some of the ethnic power-sharing requirements of the constitution and streamlined legislation (Perry 2015: 493). The American attempt at streamlining BiH legislative processes failed after Bosniak parties that were initially sympathetic to the effort abandoned it at the last minute in the hope of electoral gain. While there were follow-on attempts, the failure of the constitutional reform process marked the end of American leadership of international intervention in Bosnia and a decisive move toward EU leadership and focus on accession conditionality as the primary means of leverage. This change represented not only a change in leadership but in the tools to be used, as the failure of the “April Package” did not result in the imposition of constitutional changes. The efforts failed to pass the democratic institutions set out by the existing constitution and therefore were abandoned. The shift was marked by the initialing of Bosnia’s Stabilization and Association Agreement with the European Union in 2007, despite the failure of police reform and the “April Package.” However, the implementation of the SAA was held up by a new round of constitutional problems arising from the Sejdi´c-Finci decision of the European Court of Human Rights in 2009, by which time Lajˇcák had become the foreign minister of Slovakia and been replaced in Sarajevo by Austrian diplomat, Valentin Inzko. As members of the Roma and Jewish communities in Bosnia, respectively, the plaintiffs secured a decision establishing that ethnic representation in the (BiH) House of Peoples and Presidency violated the human rights of BiH citizens who did not identify as one of the three constituent peoples defined in the DPA (Bosniaks, Croats, Serbs). In response, the European Commission delayed the implementation of the not-yet-ratified SAA until the concerns of the court were addressed, thereby creating a renewed focal point for leverage to bring about constitutional change around the SAA (Bennett 2016: 212–213). Even though the Sejdi´c-Fenci ruling focuses only on the rights of the “Others,” the debate in Bosnia quickly shifted to a fight over the fundamental structure of the country, the balance of power between the three

206

J. HULSEY

constituent peoples, and particularly the role of the Croat constituent people, whereby Croat political leaders sought a third Croat-dominated Entity. Despite threats that the accession process would be held up and even that elections held under the old constitution would not be recognized, no constitutional change has been put through to rectify the situation (Perry 2015: 498). In 2011, the OHR and EU Special Representative jobs were split. Valentin Inzko continued as the head of a much-weakened and sidelined OHR, while Peter Sørenson was named EU Special Representative. The SAA had been ratified by the EU member-states, but could not come into force because of the failure to implement the ECHR ruling. The European Council decided that the SAA could come into force if “credible effort” were made toward implementing the ruling (Bennett 2016: 216). However, even this low bar could not be reached by the time Sørenson left office in 2014. Following Sørenson’s departure, the EU approach to Bosnia underwent both a change in leadership—with the arrival of Lars-Gunnar Wigemark as the new EU Special Representative in March 2015—and a change in policy. Beginning in December 2014, the German and British governments put forward a joint plan to abandon any attempts at constitutional reform and instead revive Bosnia’s EU accession process by focusing on less-sensitive structural socio-economic issues. This approach was formalized in the Reform Agenda for BiH 2015–2018, whereby Bosnian party leaders agreed on a set of economic reforms in exchange for implementation of the SAA (Belloni 2019: 221). A resolution to Sejdi´cFinci was dropped from the list of required conditions. Furthermore, the requirements for structural changes that replaced the demand for constitutional change were weakened between the kick-off of the Reform Agenda and the signing of the long-delayed SAA by allowing for “meaningful progress” as opposed to “full implementation” (Weber 2018: 30). Once again, international intervention was caught between a focus on stability and the demands of accession. The EU’s shift in approach in 2015 opened the door to unfreezing the SAA. However, constitutional questions again reared their head with the issue of the coordinating mechanism for accession negotiations. Who would have authority to carry out the detailed negotiations necessary to move forward toward membership? In the absence of a strengthened supremacy clause in the Bosnian constitution that would have permitted agreements reached by the BiH-level government to be enforceable on the entities and Cantons, Bosnia was required to create a “coordinating

8

TURNOVER, CONDITIONALITY, AND EUROPEANIZATION …

207

mechanism” that reflected Bosnia’s political structure in its negotiations with the EU (Perry 2015: 498, 502–503). Of course, such representation is the purpose of the Bosnian constitution as a whole, so the creation of a streamlined process implies a major constitutional change and, potentially, a major centralization of power. As a result, the coordination mechanism approved by Bosnia’s Council of Ministers in Spring of 2016 was publicly rejected by Milorad Dodik, President of Republika Srpska and the most powerful Serb politician. Further negotiations between the Bosniak member of the BiH Presidency and Dodik brought about an agreement that did finally clear the way for Bosnia’s application for EU membership to be accepted. Currently, Bosnia is moving toward official candidate status.

Conclusions Bosnia’s experience with police reform and constitutional reform show clearly the inconsistency in the application of EU conditionality. More fine-grained research is necessary to firmly establish that elite turnover plays a decisive role in this policy inconsistency, but the evidence from Bosnia does suggest that turnover effects should be considered as the accession process drags on. Part of this consideration must be a discussion of the degree to which policy change is desirable. Criticisms of EU conditionality in Bosnia come from two directions. The first focuses on the repeated pattern of softening or removing conditions in order to bring about some progress. In this explanation, personnel turnover, either by shifting the individuals in place or through officials’ desire to show progress during their mandate, weakens conditionality by allowing progress even though the conditions are not substantially met. These questions of mandates extend beyond officials posted to Sarajevo to Council presidencies, including the accusation that Bosnia’s progress toward the implementation of its SAA was greatly facilitated by the desire of the Greek Prime Minister Samaras to show progress on that priority of the Greek Presidency of the Council of the EU during the first half of 2014. The opposing axis of criticism argues that those areas where conditionality has been softened were misguided to begin with in that they represent constitutional issues that are core to Bosnia’s democracy and should not be reopened as part of the accession process.

208

J. HULSEY

Stagnation in the accession process is surely an overdetermined phenomenon. The lack of variation that serves as the evidence for the problem also impedes decisive analysis. Long-term structural factors place high obstacles in the way of reform. However, patterns of rotation and replacement in those responsible for EU policy in the Balkans further weakens the most important tools for leverage on Balkan political systems by undermining the enforcement of conditionality. Local leaders can count on being able to outlast their counterparts and await a new policy, which raises interesting implications for the populist and feminist critiques of turnover in two directions. With regard to EU bodies, critics’ shared push for greater turnover may be weakening the EU’s accession-related efforts to support democratization in candidate countries. However, applying the same critiques to the candidate countries themselves highlights the decreasing turnover and alternation in power in the Western Balkans, particularly by parties who share the populists’ preference for national-level politics.

References Balkans in Europe Policy Advisor Group. 2017. The Crisis of Democracy in the Western Balkans: An Anatomy of Stabilitocracy and the Limits of EU Democracy Promotion. Edited by Marko Kmezic and Florian Bieber. Bassuener, Kurt, and Valery Perry (n.d.). Erratic Ambiguity: The Impact of Trump’s Unpredictable Foreign Policy in the Western Balkans. Democratization Policy Council. Available at http://www.democratizationpolicy.org/sum mary/erratic-ambiguity-kb-vp/. Accessed July 13, 2020. Belloni, Roberto. 2019. The Rise and Fall of Peacebuilding in the Balkans. London: Palgrave Macmillan. Bennett, Christopher. 2016. Bosnia’s Paralyzed Peace. Oxford: Oxford University Press. Dzankic, Jelena, and Soeren Keil. 2018. The Europeanisation of Contested States: Comparing Bosnia and Herzegovina, Macedonia and Montenegro. In The Europeanisation of the Western Balkans: A Failure of EU Conditionality?, ed. Jelena Dzankic, Soeren Keil, and Marko Kmezic 181–206. Cham, Switzerland: Palgrave. Grzymala-Busse, Anna, and A. Innes. 2003. Great Expectations: The EU and Domestic Political Competition in East Central Europe. East European Politics and Societies 17 (1): 64–73. Hulsey, John. 2018. Institutions and the Reversal of State Capture: Bosnia and Herzegovina in Comparative Perspective. Southeastern Europe 42 (1): 15–32.

8

TURNOVER, CONDITIONALITY, AND EUROPEANIZATION …

209

Kmezic, Marko. 2018. EU Rule of Law Conditionality: Democracy or ‘Stabilitocracy’ Promotion in the Western Balkans?. In The Europeanisation of the Western Balkans: A Failure of EU Conditionality?, eds. Jelena Dzankic, Soeren Keil, and Marko Kmezic, 87–110. Cham, Switzerland: Palgrave. Knaus, Gerald, and Felix Martin. 2003. Travails of the European Raj. Journal of Democracy 14 (3): 60–74. Lavic, Milan, and Florian Bieber. 2020. Shifts in Support for Authoritarianism and Democracy in the Western Balkans. Problems of Post-Communism. Online Pre-Print. Maier, Jurgen, and Berthold Rittberger. 2008. Shifting Europe’s Boundaries: Mass Media, Public Opinion and the Enlargement of the EU. European Union Politics 9 (2): 243–267. Morgenstern-Pomorski, and Jost-Henrik. 2018. The Contested Diplomacy of the European External Action Service: Inception, Establishment and Consolidation. London: Routledge. Perez, Lauren K., and John A. Scherpereel. 2017. Vertical Intra-Institutional Effects of Ministerial Turnover in the Council of the European Union. Journal of European Public Policy 24 (8): 1154–1171. Perry, Valery. 2015. Constitutional Reform in Bosnia and Herzegovina: Does the Road to Confederation Go Through the EU? International Peacekeeping 22 (5): 490–510. Peter, Mateja. 2015. No Exit: The Decline of the International Administration in Bosnia and Herzegovina. In State-Building and Democratization in Bosnia and Herzegovina, ed. Soeren Keil and Valery Perry, 131–149. Farnham: Ashgate. Richter, Solveig, and Natasha Wunsch. 2019. Money, Power, Glory: The Linkages Between EU Conditionality and State Capture in the Western Balkans. Journal of European Public Policy 27 (1): 41–62. https://doi.org/10.1080/ 13501763.2019.1578815. Scherpereel, John A., and Lauren K. Perez. 2015. Turnover in the Council of the European Union: What It Is and Why It Matters. JCMS: Journal of Common Market Studies 5 (3): 658–673. Schimmelfennig, Frank, and Ulrich Sedelmeier. 2004. Governance by Conditionality: EU Rule Transfer to the Candidate Countries of Central and Eastern Europe. Journal of European Public Policy 11 (4): 661–679. Schimmelfennig, Frank, Stefan Engert, and Heiko Knobel. 2003. Costs, Commitment and Compliance: The Impact of EU Democratic Conditionality on Latvia, Slovakia and Turkey. JCMS: Journal of Common Market Studies 41 (3): 495–518. Toal, Gerard, and Carl T. Dahlman. 2011. Bosnia Remade: Ethnic Cleansing and Its Reversal. Oxford: Oxford University Press.

210

J. HULSEY

Vachudova, Milada Anna. 2005. Europe Undivided: Democracy, Leverage, and Integration After Communism. Oxford: Oxford University Press. Vachudova, Milada Anna. 2015. External Actors and Regime Change: How PostCommunism Transformed Comparative Politics. East European Politics and Societies 29 (2): 519–530. Vachudova, Milada Anna. 2018. EU Enlargement and State Capture in the Western Balkans. In The Europeanisation of the Western Balkans: A Failure of EU Conditionality?, ed. Jelena Dzankic, Soeren Keil, and Marko Kmezic, 63–86. Cham, Switzerland: Palgrave. Venice Commission. 2005. Opinion on the Constitutional Situation in Bosnia and Herzegovina and the Powers of the High Representative. Council of Europe. Available at http://www.venice.coe.int/webforms/documents/default.aspx? pdffile=CDL-AD(2005)004-e. Accessed July 13, 2020. Weber, Bodo. 2018. The EU’s Failing Policy Initiative for Bosnia and Herzegovina: A Reform Agenda and Questionnaire Monitoring Report. Democratization Policy Council. Zhelyazkova, Asya, Ivan Damjanovski, Zoran Nechev, and Frank Schimmelfennig. 2018. European Union Conditionality in the Western Balkans: External Incentives and Europeanisation. In The Europeanisation of the Western Balkans: A Failure of EU Conditionality?, ed. Jelena Dzankic, Soeren Keil, and Marko Kmezic, 15–38. Cham, Switzerland: Palgrave.

PART V

Conclusion

CHAPTER 9

Turnover, Legitimacy, and the EU Timescape John A. Scherpereel

How does personnel turnover affect the legitimacy of the EU? Chapters 2 through 8 of this volume offer a range of insights into this question. The current chapter summarizes these insights. Collectively, the book’s analyses suggest that personnel turnover’s effects vary significantly across the three facets of EU legitimacy. Contributors find that turnover has limited effects on input legitimacy. They also find—in a way that corroborates and pushes forward existing academic studies of turnover—that turnover has significant effects on throughput and output legitimacy. In addition to developing and discussing the repercussions of these findings, the chapter reflects on the volume’s implications for populist and

Electronic supplementary material The online version of this chapter (https://doi.org/10.1007/978-3-030-60052-5_9) contains supplementary material, which is available to authorized users. J. A. Scherpereel (B) Department of Political Science, James Madison University, Harrisonburg, VA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. A. Scherpereel (eds.), Personnel Turnover and the Legitimacy of the EU, Palgrave Studies in European Union Politics, https://doi.org/10.1007/978-3-030-60052-5_9

213

214

J. A. SCHERPEREEL

feminist critiques of EU politics: collectively, the contributors find limited empirical support for populist propositions and important, if still preliminary, support for feminist propositions. The chapter concludes by placing studies of personnel turnover in Europe in a broader theoretical perspective, suggesting that turnover is an important component of what Goetz and Meyer-Sahling (2009, 2012) call the EU timescape. The chapter concludes by identifying a number of potentially productive avenues for future research.

The Effects of Personnel Turnover on the Legitimacy of the EU While there is still much work to be done to understand the ways that turnover affects the dimensions (attitudinal, behavioral) and facets (input, throughput, output) of legitimacy, Chapters 2 through 8 provide five important initial insights. (1) Turnover has few effects on the EU’s input legitimacy Attitudinally, input legitimacy pertains when citizens appreciate the value of political institutions; behaviorally, it pertains when citizens’ actions reflect an appreciation of the relevance of those institutions. Daniel and Metzger (Chapter 2) and Clark and Scherpereel (Chapter 3) prioritize, respectively, attitudinal and behavioral dimensions. Together, the chapters suggest that personnel dynamics have relatively little effect on input legitimacy. Daniel and Metzger develop the idea that memberstates whose MEPs treat the EP as a second-order institution will have citizens who see the EP the same way. The results of their pooled model provide support for this idea, but their more time-sensitive analyses show the association disappearing in recent years. Actors who would like to increase citizens’ appreciation for and attachment to the EP, the chapter suggests, need not be particularly concerned with the fact that significant numbers of MEPs leave the institution during the course of a five-year session. Clark and Scherpereel spotlight mass political behavior. While they find that gendered personnel dynamics are associated with voter turnout, their null findings with respect to both within-session and across-session

9

TURNOVER, LEGITIMACY, AND THE EU TIMESCAPE

215

turnover are also notable. The EP has matured in recent decades, and citizens continue to recognize the EP as the quintessential EU institution. But turnover among MEPs does not seem to affect the likelihood that citizens will cast ballots in EP elections. Of course, there is more to input legitimacy than attitudes about whether an institution should play a stronger role or voter turnout. A fuller account of input legitimacy might also address citizens’ satisfaction with parties’ role in linking them to EU decision processes. It might also investigate whether turnover affects the quantity or quality of public discourse about EU affairs. Overall, though, the volume’s empirical analyses suggest that the link between turnover and input legitimacy is not particularly strong. (2) Turnover has significant effects on the EU’s throughput legitimacy The volume’s middle Chapters (4, 5, and 6) suggest that turnover has more significant effects on throughput legitimacy. This facet of legitimacy deals most directly with goings on in the Brussels bubble. Throughput legitimacy pertains when EU governing processes work—when relevant concerns are aired, when institutions effectively tackle issues of concern, when decision processes play out in time to address public problems and promote social flourishing. Perez and Scherpereel focus on the efficiency of the legislative process. They find, consistent with expectations, that the ordinary legislative procedure becomes less efficient as levels of turnover in the Commission and the Council rise. But their finding with regard to the Parliament, where higher levels of turnover are associated with faster legislative processes, is counterintuitive. While Perez and Scherpereel focus on one aspect of turnover— quantity—both Aldrich and Perez (Chapter 5) and Salvati and Vercesi (Chapter 6) spotlight both quantity and quality. For Aldrich and Perez, the primary quality of interest is gender. Focusing on the Council, they find that the legislative process is more efficient when there are more women in the Council and that retaining more experienced women on the Council pushes the legislative process forward. When the Council has more women, and when women stay in the Council longer, the EU’s legislative process tends to work better. For Salvati and Vercesi, the primary quality of interest is professional experience. They find that while experienced MEPs are more likely than newcomers to be appointed

216

J. A. SCHERPEREEL

as committee chairs, turnover does not necessarily produce the peril of amateurism. There is, rather, an important distinction between incoming MEPs who have gained experience at the national and sub-national level, on one hand, and MEPs who are entering the EP as political newcomers, on the other. While MEPs with pre-EP political experience tend to be quite capable in key roles, those that lack experience struggle to overcome learning curves. Together, then, the volume’s middle chapters suggest that turnover in the Commission and the Council can undercut throughput legitimacy. At least in the case of the Council, though, turnover’s gendered aspect makes a difference: configurations that are gender-imbalanced and/or lose experienced women tend to face more operational hurdles than configurations that are gender-balanced and/or retain experienced women. For the EP, turnover does not necessarily imperil efficiency. In fact, the legislative process may accelerate as turnover increases and would be particularly likely to do so when experienced politicians—regardless of where that experience has been gained—are appointed as policy point people. Three points about throughput legitimacy bear emphasis before addressing output legitimacy. The first is that future researchers would do well to determine the extent to which the findings of Chapters 5 and 6 travel beyond the institutional cases from which they have been generated. Are gender imbalance and the loss of women incumbents as impactful in the European Council, the Commission, and the EP as they are in the Council of Ministers? Is sub-European political experience the same kind of asset in the Commission and the Council as it is in the EP? The second point is that while the volume’s middle chapters focus largely on the legislative process, throughput legitimacy is not solely about legislation. EU institutions do not simply legislate: the European Council deliberates and decides on hugely consequential strategic priorities, for example. The Foreign Affairs Council reaches decisions that do not result in EU laws. EP delegations establish relationships with counterpart delegations of national parliamentarians. And so on. Attention to the ways that the broad range of non-legislative processes play out are important in assessing the extent of the EU’s throughput legitimacy. The third point, stressed already by Salvati and Vercesi, is that even in the restricted realm of purely legislative activity, the policy process is a broader and more temporally expansive phenomenon than the legislative process. To assess the efficiency and inclusiveness of the decision process as a whole (as opposed to the limited legislative sub-phase of that process),

9

TURNOVER, LEGITIMACY, AND THE EU TIMESCAPE

217

one would be well-served to recall textbook admonitions: the policy process is a cycle. It can be difficult, in the absence of clear temporalbureaucratic markers (e.g., recordings of the date that the Commission forwarded a legislative text to the EP and the Council), to assess the point at which a particular decision process has begun. It is clear, though, for effectively all legislation, that key actors in multiple institutions (and generally in interest organizations and advocacy groups) will have a sense of a legislative text well in advance of the Commission’s formal initiation of the legislative process. Insight into throughput legitimacy also requires attention to exigencies that unfold after the signature of a legal act. EU directives, for example, must be transposed into national legal orders, and all “completed” laws require monitoring and evaluation in one form/frequency or another. While these facts present methodological challenges, future studies would do well to investigate the extent to which turnover affects them. (3) Turnover has significant effects on the EU’s output legitimacy Chapters 7 and 8 focus on output legitimacy—that facet of legitimacy that Scharpf (1970) originally emphasized in the Community context. Output legitimacy grows when institutions solve public problems and address public concerns. In their approach to output legitimacy, both chapter authors focus on relations with potentially liminal countries—Adolino’s UK as a country on the way out, Hulsey’s Bosnia as, potentially, a country on the way in. These contributions (which examine non-legislative policy outputs) are timely, not only because EU citizens routinely voice a desire for the EU to do more to project power globally, but also (and relatedly) because the current (2019–2024) Commission president has pledged to lead a “geopolitical Commission.” Together, Chapters 7 and 8 show, at least with regard to the EU’s interface with liminal countries, that turnover undermines output legitimacy. In the case of Brexit, a number of recent contributions (e.g., De Vries 2017) have suggested that EU-27 citizens are looking to the UK for clues about what leaving the EU might mean for their countries. They are “benchmarking Brexit,” using the UK as a test case to size up the costs and benefits of an “Italeave,” a “Polout,” a “Frexit,” etc. Adolino, however, implies that EU citizens are also using Brexit to evaluate their (EU) processes and structures vis-à-vis those of the UK. During

218

J. A. SCHERPEREEL

the period covered by the May governments, the EU side of the negotiations was better-organized, more consistent, and more coherent than the UK side. Prime Minister May decided, early in her first term, to include prominent Leavers and Remainers in her cabinet. As time moved forward, ministers came and went, but fundamental disagreements among cabinet members persisted. This was an eloquent example of the peril of disunity. The disunity and its negative effects were clear to the Brexit process’s principals and to citizens on Europe’s streets—Leavers and Remainers in the UK, citizens throughout the EU-27. The fact that the EU-27 proceeded with constancy and professionalism, and that the UK side was plagued by turnover, may help to legitimize the EU in the eyes of its citizens and to deepen the suspicion that UK citizens feel toward the institutions of the UK state. The EU side produced a consistent policy with which EU-27 citizens could identify. The churn on the UK side alienated UK citizens and encouraged EU-27 citizens to see Brexit, at least in the short term, as a cautionary tale. In Hulsey’s Bosnia (Chapter 8), the “one side stable, one side dynamic” picture is flipped on its head: here, long-lived principals on the liminal side have outlasted ephemeral and occasionally inconsistent actors on the EU side. While the EU’s steady Brexit negotiation process may help to legitimize the EU in the eyes of EU-27 citizens, its longerterm, less stable negotiation processes in the Western Balkans are unlikely to win it plaudits at home. In the Western Balkans, the kinds of reform sequences that the EU saw in previous rounds of candidates have not (yet) materialized. In Bosnia, the list of factors militating against governance progress is lengthy. But the EU’s long time horizons, the waning levels of enlargement buy-in among EU-27 elites and masses, and the rotating cast of diplomatic chiefs have all made it difficult for the Union to promote progress. Stabilitocracy does little to burnish the EU’s credentials with audiences in the Western Balkans, audiences in the EU-27, or audiences further afield. Overall, then, the chapters suggest that personnel turnover has limited effects on input legitimacy and significant effects on throughput and output legitimacy. Additional insights into these relationships would emerge from deeper scrutiny of the attitudinal dimension of legitimacy’s throughput and output dimensions. How, for example, do politicians that retain seats and non-institutional actors (e.g., interest groups) evaluate the effects of personnel turnover?

9

TURNOVER, LEGITIMACY, AND THE EU TIMESCAPE

219

Turnover, Populism, and Feminism European populism and feminism are both big tents, and each tent’s shade covers actors who disagree on many issues. Most actors in each tent would likely agree on two things. First, they would prefer not to be associated with actors in the other tent—populists would denounce feminists; feminists would denounce populists. Second, and more significantly, they would both call for major reforms in the ways that the EU institutions are peopled, the manner by which the institutions work, and the voices that the institutions amplify. The volume’s empirical chapters support two additional conclusions related to the two contemporary critiques. (4) The populist critique of EU personnel dynamics receives little empirical support European populists have vacillated, over time, between outright dismissal of the European project and belief in the project’s fundamental salvageability. In recent years, this pendulum has swung in the direction of salvageability (see, for example, Van Kessel et al. 2020). The vignettes from the 2019 EP campaign trail that were discussed in Chapter 1, the ironic fact that right-wing populist parties have consistently done better in EP elections than they have in national elections, the general recognition that the process of negotiating an EU exit thrust an exiting state into deep constitutional, economic, and geopolitical uncertainty, and the fact that the EU’s response to the Covid-19 pandemic will open up new sources and mechanisms of funding have all encouraged populists to reconcile themselves with some vision of European integration. Acceptance is a far cry from satisfaction, though, and reconciliation is a far cry from redemption. Indeed, it is the chasm between reconciliation and redemption that the populist critique seeks to close. “Redemption through replacement” suggests that a more democratic and legitimate EU will only be achieved when a fundamental power shift from “the elite” to “the people” takes place. Practically, such a shift requires the ejection of entrenched incumbents and the entrance of officeholders who transform “the people’s values” into policy. The chapters in this volume that have focused on the quantity of turnover suggest that such redemptive sequences are unlikely. According to Daniel and Metzger, for example, a link between within-session turnover and attitudinal support for the EP could be observed in the years immediately

220

J. A. SCHERPEREEL

following 1979. That link has not existed in recent years. Clark and Scherpereel, similarly, find no evidence that “kicking the bums out” or “getting the bums to leave” catalyzes an uptick in voter turnout (at least when the analysis examines all MEPs). Yes, Perez and Scherpereel find that turnover among MEPs is associated with a more efficient legislative process, but that association is reversed in the Commission and the Council. Salvati and Vercesi’s deeper look at MEPs’ professional backgrounds, for its part, suggests that efficiency gains are more likely when empowered MEPs are not “full outsiders,” e.g., when they come to Brussels/Strasbourg with significant political experience. Thus, while the weight of the volume’s evidence casts doubt on the empirical accuracy of redemption through replacement, additional research—particularly via studies that focus on the quality of turnover— would help to throw more light on the relationship between populism and EU governance. Studies that compare the pre-EP professional trajectories of MEPs in populist party groups, for example, with the professional trajectories of their mainstream colleagues could generate additional insights into the impacts that different kinds of MEPs have. In the Council/European Council contexts, deep scrutiny of the behaviors of populist and non-populist leaders would lead to better understanding of the extent to which populist politicians propel meaningful changes in the ways the EU system works. And the effects of populist politicians on EU policy outputs and output legitimacy are ripe for future study. Do MEPs from populist parties, for example, behave differently than MEPs from traditional parties when they serve as (shadow) rapporteurs? Do the outputs on which they work look substantially different from those overseen by more centrist and/or experienced politicians? (5) The feminist critique of EU personnel dynamics receives early empirical support Despite the multivocality of feminist critiques of the EU, many feminist critics have derided the institutions for perpetuating gender imbalances. In the same years that European populists have been making electoral gains, European feminists have been scoring some victories: EU attention to gender inequities has risen, and institutional actors have devoted more attention to gender balance. The achievement of gender balance across the “top jobs” allocated in 2019 was a small victory for advocates

9

TURNOVER, LEGITIMACY, AND THE EU TIMESCAPE

221

of “redemption through gender equality.” Feminist critics have defended the equal descriptive representation of the genders for various reasons, e.g., as a matter of justice, as a matter of promoting high-quality decision processes, as a matter of assuring that institutions look like society, as a matter of promoting desirable policy ends like peace, good governance, public health, and domestic and international security. The top jobs breakthrough of 2019 notwithstanding, it is important to recall that equality in the institutions is still a distant goal: even the EP, which has led the pack in terms of descriptive representation, still has internal institutional features that affect genders differently. And, across the institutions, significant gender breakthroughs are often followed by setbacks. Ursula von der Leyen, who formally introduced herself to MEPs by pledging to “ensure full gender equality” in her college of commissioners, ultimately came up just short of that goal in 2019. Is there empirical evidence of “redemption through gender equality”? Findings from parts II and III of the book (on legitimacy’s input and throughput facets) corroborate aspects of the feminist critique. While Clark and Scherpereel find that (ungendered) within-session and acrosssession turnover are not associated with voter turnout, for example, they also find that that citizens in member-states where women gain more seats vote in higher numbers. Future research might focus more on the distinction between within-session turnover and across-session turnover in hopes of generating additional insight into whether, when, and how these two specific mechanisms of gender empowerment affect citizen behaviors. When it comes to throughput legitimacy, Aldrich and Perez find that integrating and retaining women in the Council make the EU system work better. Together, these findings help to underscore a point stressed by many other analysts of gendered political dynamics: representational empowerment is not simply important as a matter of justice. It is important because it can promote other critical ends—efficiency, effectiveness, consensus, cohesion, and others—that liberal democratic communities strive to achieve. While these insights are significant, there is ample room for further research into the gendered implications of turnover in EU institutions and the accuracy of propositions about “redemption through gender equality.” Does empowering women affect different segments of (voting) populations differently? How do EP and Commission career paths differ by gender, and how might such differences affect policy processes and inter-institutional relationships? Is recruiting and retaining women heads

222

J. A. SCHERPEREEL

of government, commissioners, central bankers, MEPs, and judges as substantively consequential as recruiting and retaining women ministers? How, most broadly, are EU institutions and the relationships between institutions and societies transformed when they become more (or less) gender-balanced?

Institutions, Turnover, and the EU Timescape Personnel turnover is an inherently temporal phenomenon: it can only be grasped by comparing an institution’s personnel composition at one point in time with the institution’s composition at one or more earlier points in time. The volume has shown that this temporal phenomenon has important political effects: it influences the ways that EU institutions interact with each other, the kinds of policies they produce, and the legitimacy of the EU. The notion that a temporal phenomenon has critical political effects recalls other work on what Goetz and Meyer-Sahling (2009, 212) have called the EU timescape. Like all robust political systems, today’s EU bears its birthmarks and has evolved over time. It maintains many of the features of the system that the founding six member-states established. But as the EU polity has matured, actors have used new treaties to merge institutions, to establish and strengthen new institutions and to otherwise recalibrate institutional relationships. They have also established a range of formal rules beyond the treaties—rules of procedure, interinstitutional agreements, Court of Justice rulings, and others—that have further (re-)structured the EU political system and (ultimately) affected the three facets of legitimacy. Goetz and Meyer-Sahling (2009, 2012), however, are not primarily interested in the gradual evolution of institutional relationships or policy regimes over a span of linear time. Rather, they are interested in the steps that EU actors take to institutionalize time and the role that time considerations (e.g., electoral calendars, temporally structured budgetary and monitoring frameworks) play in structuring institutional balances (and, indirectly, in affecting EU legitimacy). Borrowing from sociologist Barbara Adam (1998, 2004), they define a timescape as a “cluster of temporal features, each implicated in all the others, but not necessarily of equal importance in each instance” (Adam 2004: 143; quoted in Meyer-Sahling and Goetz 2009: 326). To grasp the role that time considerations play in affecting institutional balances in the EU polity, consider four (from among many potential) examples.

9

TURNOVER, LEGITIMACY, AND THE EU TIMESCAPE

223

First, as noted in Chapter 1, EU treaties and secondary rules lay down a particular system of term lengths : EP elections take place on a quinquennial calendar, for example. The presidency of the Council moves from one member-state to another on January 1 and July 1 of each year. Court of Justice judges serve six-year terms that are renewable once. Second, formal rules require the staggering of terms (Willumsen and Goetz 2015). The aforementioned ECJ terms, for example, are arranged such that half of the seats are up for renewal/replacement every three years. Third, there are issues related to interinstitutional synchronization. In recent decades, for example, the quinquennial “nines and fours” calendar (1979, 1984, 1989 …) has become a more general organizing principle in EU politics. In the wake of the 2019 EP elections, for example, the European Council met to consider the distribution of top jobs across three institutions—the European Council, the European Commission, and the European Central Bank. In the case of the Commission, the term of the politician the heads of state and government nominated (Ursula von der Leyen) will wrap up in the same year as the EP session ends (2024). The European Council president that the national leaders appointed (Charles Michel) will be subject to a fourth temporal characteristic—rhythmicity— that mirrors the EP’s “drumbeat.” Heads of state and government will use a 2.5-year temporal marking point to decide whether to reappoint Michel. MEPs will use the same temporal marking point to carry out their own mid-session leadership changes. If Michel, like the first two permanent European Council presidents, is renewed, the remainder of his term will have a similar (2.5-year) time structure as the remainder of the EP session. According to Goetz (2009), term lengths, staggering, synchronization, rhythmicity, and other time considerations together constitute a “temporal constitution of the EU.” This temporal constitution is an integral part of the EU’s dynamic institutional order. It affects the strategic calculations and broader orientations of actors with a stake in EU politics. It helps to determine actors’ sense of whom they represent and whether, why, when, where, and how certain goals should be pursued. It generates opportunities and imposes constraints. It grants advantages to some actors and handicaps others. So, how does personnel turnover fit into the EU timescape? In conceptualizing the timescape, Meyer-Sahling and Goetz (2009: 326–327) stress issues of layering and embeddedness: “the timescape concept invites a focus on the linkages and interdependencies between different dimensions

224

J. A. SCHERPEREEL

of the temporal constitution of a political system.” The volume’s chapters have examined the ways that turnover affects relationships, e.g., relationships between the institutions and EU citizens, relationships among the institutions, and relationships linking institutions, policies, and broader publics. Terms lengths, staggering, synchronization, and rhythmicity are all elements of the institutionalization of time. The comings and goings that comprise turnover interact with other features of the institutional system—including the timescape. An example helps to illustrate the ways that turnover plays into the broader EU timescape. Consider a career-oriented politician who has previously served as a national member of parliament or a regional minister. The politician earns a place on a national party’s EP electoral list and begins a five-year term as a member of the European Parliament (MEP). The MEP sees her new position as a stepping-stone to a more prestigious national office. That vision, in turn, affects her approach to her EP duties. At some point during the EP term, national general elections take place, and an attractive higher-profile position in the national government becomes available. The incoming head of government invites the MEP to join the government. The MEP resigns from the Parliament and is eventually replaced with a substitute. In this hypothetical sequence, multiple temporal orders—the EP’s calendar, the MEP’s career development plan, the calendar of general elections in the MEP’s home country—intersect. Similar temporal intersections occur throughout the EP’s term and, indeed, throughout the terms of every EU institution. Collectively, these intersections can affect the internal compositions of the institutions (e.g., is a departing member replaced with a woman or a man? With someone with consequentially different professional experiences?), the organization of the institution (e.g., which MEPs are appointed rapporteurships now that particular MEPs have exited?), the efficiency of decision processes (e.g., how long does it take the actors empowered by the MEP’s departure to master new briefs?), the relationships among the institutions, and potentially, the ways that European citizens view the EU. To summarize: the EU’s institutional system is especially dynamic. Relationships among the institutions, between the institutions and the member-states, and between the institutions and citizens change over time in response to various formal adjustments and informal processes. In addition, EU institutions are populated by actors who have distinctive professional goals and face a range of life exigencies—new professional

9

TURNOVER, LEGITIMACY, AND THE EU TIMESCAPE

225

opportunities, family obligations, terminal illnesses, and beyond. The comings and goings of the latter actors affect the composition of EU institutions over time. They constitute a component of the EU’s timescape that is predictable in general (incumbents will always be cycling in and out) but difficult to predict in particular (it is hard to predict who will leave, when they will leave, and/or who will replace them). Incumbents’ comings and goings can affect the ways that citizens evaluate and respond to the institutions, the ways that the institutions operate and relate to each other, and the kinds of policies and actions that institutions ultimately produce.

Future Research on Turnover, Legitimacy, and the EU Timescape The volume’s chapter authors have investigated a range of connections between turnover and legitimacy. Their methodological approaches have varied: some have employed quantitative methods, and others have engaged in qualitative process tracing. Quantitative and qualitative approaches can both help to identify the ways that turnover affects EU legitimacy and interacts with other aspects of the EU timescape. Here, particular methodological approaches have been associated with particular facets of legitimacy: authors studying input and throughput facets have prioritized quantitative analysis, and authors studying output legitimacy have used qualitative methods. There is no necessary connection, however, between particular research methods and particular facets of legitimacy. For example, researchers might engage in non-participant observation or convene focus groups to gain insights into the relationship between turnover and input legitimacy. Such methods would help to expose differences, for example, between the interactions that long-serving and freshmen MEPs (and/or women and men MEPs) have with citizens. They would help to show whether and how those interactions affect citizens’ satisfaction with and approaches to EU institutions. Semi-structured interviews and/or qualitative analysis of memoirs, to take another example, might push forward understandings of the connection between turnover and throughput legitimacy. They could throw light on the ways that a particular government leader’s behaviors— including her level of comfort, confidence, and influence—develop over the course of her service in the European Council (Tallberg 2008) or the ways that individual Court of Justice actors (judges, advocates general) or

226

J. A. SCHERPEREEL

collective chambers develop over the course of their tenures. Quantitative analyses can also illuminate the relationship between turnover and output legitimacy. Recent survey-based research on EU regional policy (Capello and Perucca 2019a, b; Dellmuth and Chalmers 2018), for example, examines whether citizens evaluate the EU positively when policy outputs are directed toward addressing specific problems. How might turnover and the timing of policy cycles play into such dynamics? Are seasoned local, regional, national, and/or European actors more or less likely to target interventions in ways that improve citizens’ attitudes about the EU’s added value? Regardless of the methodological approach ultimately employed, inquiry into turnover’s relationship to legitimacy and role in the EU timescape requires information about incumbents’ “comings and goings.” To that end, data on incumbents and their dates of entry/exit are available in this book’s online appendix. These data cover all seven institutions established by Article 13 TEU. Descriptive statistics on turnover in the seven institutions have been presented in the volume’s first chapter. Beyond that, however, the volume’s chapters have focused quite tightly on the institutions of everyday decision-making—the Commission, the Council, and the Parliament. While these analyses provide insight into the ways that turnover affects EU legitimacy, the number of institutions left relatively unexamined (four) still eclipses the number of institutions spotlighted (three). Of the four relatively neglected institutions, the European Council stands out as particularly prominent. It has become especially active, visible, and influential in recent years (Bickerton et al. 2015; Puetter 2014), playing a key role in shaping EU responses to, inter alia, the financial, migration, and Covid-19 crises. As of 2020, three European Council principals—Angela Merkel, Viktor Orbán, and Mark Rutte—have served for over ten years. What imprint have they left on this dynamic institution? What does their eventual departure—and the departures of other heads of state and government—mean for the European Council’s institutional position, influence, spot in the hearts or minds of EU citizens, and orientation to time? While they may occupy less headline space, the Court of Justice (ECJ), the European Central Bank (ECB), and the Court of Auditors (ECA) figure prominently in the EU’s political system and timescape. Are the comings and goings of specific judges associated with discernible

9

TURNOVER, LEGITIMACY, AND THE EU TIMESCAPE

227

shifts in the Court’s jurisprudential outputs? To what extent has Christine Lagarde’s ascension to the presidency of the ECB—the EU’s most gender imbalanced institution—affected the governing council’s operating style and approach? To what extent have populist critics integrated technical institutions like the ECA into their more familiar critiques of the Commission, and what might the political effects of such targeting be? It is worth recalling, along similar lines, that all institutions are internally articulated and complemented by a range of agencies and advisory bodies. The Council system, for example, includes the General Secretariat, the committee of permanent representatives, and a range of working groups, some of which are composed predominantly of members based outside of Brussels. Do such features affect power balances and/or the ways that time is structured? In sum, personnel turnover is an important component of the EU timescape that affects the legitimacy of the EU—particularly its throughput and output facets. There is considerable room for continued investigation into the ways that turnover interacts with other constitutional and temporal components of the ever-evolving EU system.

References Adam, Barbara. 2004. Time. Cambridge: Polity Press. Adam, Barbara. 1998. Timescapes of Modernity. London: Routledge. Bickerton, Christopher J., Dermot Hodson, and Uwe Puetter, eds. 2015. The New Intergovernmentalism: States and Supranational Actors in the Post-Maastricht Era. Oxford: Oxford University Press. Capello, Roberta, and Giovanni Perucca. 2019a. Citizens’ Perception of Cohesion Policy: From Theory to Empirical Evidence. Regional Studies 53(11): 1520–1530. Capello, Roberta, and Giovanni Perucca. 2019b. Special Issue, From Cohesion Policy Implementation to European Identity. Regional Science Policy and Practice 11(4). De Vries, Catherine E. 2017. Benchmarking Brexit: How the British Decision to Leave Shapes EU Public Opinion. JCMS: Journal of Common Market Studies 55: 38–53. Goetz, Klaus H. 2009. How Does the EU Tick? Five Propositions on Political Time. Journal of European Public Policy 16 (2): 202–220. Goetz, Klaus H., and Jan-Hinrik Meyer-Sahling. 2009. Political Time in the EU: Dimensions, Perspectives, Theories. Journal of European Public Policy 16 (2): 180–201.

228

J. A. SCHERPEREEL

Goetz, Klaus H., and Jan-Hinrik Meyer-Sahling, eds. 2012. The EU Timescape. London: Routledge. Meyer-Sahling, Jan-Hinrik and Klaus H. Goetz. 2009. The EU Timescape: From Notion to Research Agenda. Journal of European Public Policy 16 (2): 325– 336. Puetter, Uwe. 2014. The European Council and the Council: New Intergovernmentalism and Institutional Change. Oxford: Oxford University Press. Scharpf, Fritz W. 1970. Demokratietheorie zwischen Utopie und Anpassung. Konstanz: Universitätsverlag. Tallberg, Jonas. 2008. Bargaining Power in the European Council. JCMS: Journal of Common Market Studies 46 (3): 685–708. Kessel, Van, Nicola Chelotti Stijn, Helen Drake, Juan Roch, and Patricia Rodi. 2020. Eager to Leave? Populist Radical Right Parties’ Responses to the UK’s Brexit Vote. The British Journal of Politics and International Relations 22 (1): 65–84. Willumsen, David M., and Klaus H. Goetz. 2015. Staggered Political Institutions: Design and Effects. Journal of European Public Policy 22 (7): 1040–1051.

Index

A Ashdown, Paddy, 194, 202–204 Athens Declaration, 13

B Bannon, Steve, 10, 27 Barclay, Stephen, 185 Barnier, Michel, 176–178 Bebb, Gutto, 183 Bildt, Carl, 194 Bonn powers, 201, 202 Bradley, Ben, 184 Brexit, 169–190 Bridges, Lord George, 182

C Cameron, David, 179 Codecision. See Ordinary legislative procedure Collective efficacy (group efficacy), 65, 75

Conditionality, 194–200, 202, 204, 205, 207, 208 Conservative Party (UK), 178, 180–182, 184, 187, 188 Conservatives for a People’s Vote Group (PVG), 182 European Research Group (ERG), 180, 184 Copenhagen Criteria, 197 Costa, Alberto, 184 Council of the European Union, 23 configurations of, 114, 129 Court of Justice of the European Union, 17

D Davis, David, 179, 181, 182, 185 Dayton Peace Agreement (DPA), 200, 201, 205 Democratic deficit, 138 Department for Exiting the EU (DEXEU), 182, 185

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. A. Scherpereel (eds.), Personnel Turnover and the Legitimacy of the EU, Palgrave Studies in European Union Politics, https://doi.org/10.1007/978-3-030-60052-5

229

230

INDEX

Differentiated disintegration, 170, 172, 187

E Economic and Social Committee, 93, 99 Elite circulation, 18 EU enlargement, 50 European Central Bank (ECB), 14, 17, 19, 21, 24, 25 European Commission, 13, 14, 17–19, 23, 24 Taskforce 50, 176 European Council, 141, 156, 157 European Court of Auditors (ECA), 21, 24–26 European Parliament committee chairs in, 139–141, 144, 145, 150, 151, 158 committees in, 158 early exit from, 36, 37, 40, 42 rapporteurs in, 145 European Parliament (EP), 10, 17–19 Euroskepticism, 39

F Farage, Nigel, 11 Federation of Bosnia and Herzegovina, 201 Feminist critique of EU institutions, 3, 8, 12–16, 28 Fidesz (Hungary), 10 Five Star Movement (Italy), 10 Fox, Liam, 179 Freedom and Direct Democracy (Czechia), 10

G Gove, Michael, 178 Gyimah, Sam, 183

H Hahn, Johannes, 194 House of Commons (United Kingdom), 181–183, 187 I Independence and Democracy group, 11 Institutionalization of time, 224 rhythmicity, 223, 224 synchronization, 223, 224 term lengths, 223 J Johnson, Boris, 179, 181 Johnson, Jo, 183 Juncker, Jean-Claude, 11, 19, 25, 28 L Lajˇcák, Miroslav, 194, 203 Law and Justice (Poland), 10 Lee, Phillip, 183 Legislative duration, 87, 94, 96, 97, 99, 100 Legislative efficiency, 86, 101 Legitimacy attitudinal, 5 behavioral, 4–7 defined, 4 dimensions of, 4, 5 facets of, 8, 12, 15, 18 input, 6, 8, 11, 15, 26 output, 6–8, 11, 15, 16, 27 throughput, 6–8, 11, 15–17, 27 Le Pen, Marine, 11 M Macron, Emmanuel, 11, 24 Majoritarian institutions (MIs), 20 May, Theresa, 169, 172

INDEX

Merkel, Angela, 11, 24 Meuthen, Jörg, 11 Michel, Charles, 223

N Newton, Sarah, 184 Non-Majoritarian institutions (NMIs), 8, 10, 20, 21, 26

O Office of the High Representative (OHR), 201–204, 206 Orbán, Viktor, 226 Ordinary legislative procedure, 91, 101

P Pareto, Vilfredo, 18 Personnel turnover, 4, 6, 8, 11, 17, 18, 26, 27 across-session turnover, 19 aspects of, 4, 18 defined, 6 EU timescape and, 214, 222–225, 227 mechanisms of, 19 peril of amateurism and, 16 peril of disunity and, 16 peril of poor-quality policy and, 16 quality of, 6, 16, 27 quantity of, 19–21, 27 within-session turnover, 19, 21 Petrisch, Wolfgang, 202 Political careers, 141, 148 Political leadership, 137 Populist critique of EU institutions, 9

Q Qualified majority voting (QMV), 87

231

R Raab, Dominic, 181, 182, 185 Rees-Mogg, Jacob, 181, 183 Republika Srpska, 201–204, 207

S Salvini, Matteo, 11, 12 Schwarz-Schilling, Christian, 194, 203 Second-order elections (SOEs), 38 Seeuws, Didier, 177, 178 Sejdi´c-Finci case, 204 Soros, Georg, 11 Spitzenkandidat process, 14, 28 Stabilization and Association Agreement, 195, 202, 205

T Timescape, 222–227 EU timescape, 214, 222–227 Treaty Establishing the European Economic Community (EEC), 12 Treaty on European Union, 5, 17 Article 13, 17 Article 2, 5 Article 50, 172, 175, 186 Treaty on the Functioning of the European Union Article 255, 26 Trilogues, 112 Trump, Donald, 10 Tusk, Donald, 176

V Van Rompuy, Herman, 11 Vara, Shaleish, 184, 185 Velvet triangles, 13 Venice Commission, 204 Verhofstadt, Guy, 177 von der Leyen, Ursula, 14, 19, 28 Voter turnout, 60

232

INDEX

W Wallström, Margot, 14 Weber, Max, 4, 6

Western Balkans, 194, 195, 197, 200, 208 Wigemark, Lars-Gunnar, 194, 206 Wilders, Geert, 11