Representing Europe’s Citizens?
This page intentionally left blank
Representing Europe’s Citizens? Electoral Instit...
17 downloads
437 Views
1MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Representing Europe’s Citizens?
This page intentionally left blank
Representing Europe’s Citizens? Electoral Institutions and the Failure of Parliamentary Representation
David M. Farrell and Roger Scully
1
3
Great Clarendon Street, Oxford ox2 6DP Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York © David M. Farrell and Roger Scully 2007 The moral rights of the authors have been asserted Database right Oxford University Press (maker) First published 2007 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose the same condition on any acquirer British Library Cataloguing in Publication Data Data available Library of Congress Cataloging in Publication Data Data available Typeset by SPI Publisher Services, Pondicherry, India Printed in Great Britain on acid-free paper by Biddles Ltd., King’s Lynn, Norfolk ISBN 978–0–19–928502–0 1 3 5 7 9 10 8 6 4 2
Preface and Acknowledgements
This book is based on a research collaboration that began shortly after we first met in 1998. At the time, our curiosity had been aroused by the British Labour government’s proposal (subsequently enacted) to make a radical change to the electoral system used to elect Britain’s Members of the European Parliament. This set us thinking about the likely consequences of such a reform for how Britain would be represented in the EU’s increasingly important elected institution. We soon found that in some respects—such as how the move to proportional representation from a single-member district system would alter partisan and social ‘representativeness’—the implications of electoral reform were fairly clear and well established. But for other possible implications of reform—notably whether it would prompt change in how elected representatives understood and sought to carry out their role—the academic literature on electoral institutions and political representation offered us far less guidance. This seemed to us a highly unsatisfactory state of affairs, and prompted us to begin work on the research project that, some years later, has now resulted in this book. The scope of the study that we have developed around the themes of electoral institutions and political representation in the European Parliament, along with the main features of our argument, is outlined in the introductory chapter. Rather than rehearse them here, we wish simply to take the opportunity to thank several groups of people and institutions which between them helped us bring this project to fruition.
v
Preface and Acknowledgements
First and foremost we want to thank the many Members of the European Parliament, and their staff, who we interviewed faceto-face and surveyed in the course of our research or who otherwise provided us with assistance. Virtually without exception, and despite the fact that they are among the most surveyed of elite groups, we found them to be courteous and generous with both their time and their thoughts. We are grateful to them all. Second, we are grateful to our respective colleagues at the University of Manchester and the University of Wales, Aberystwyth, for helping provide us with the congenial and stimulating academic surroundings in which this book has been written. Special thanks also to the Center for the Study of Democracy at the University of California, Irvine, which hosted two brief research visits by David Farrell in 2003 and 2005. Third, we wish to thank those colleagues who have read drafts of our work or otherwise commented on the ideas developed in this book. These include Shaun Bowler, Russell Dalton, Bernard Grofman, Simon Hix, Tapio Raunio, Jacques Thomassen, and Martin Wattenberg, as well as participants who commented on draft papers at various conferences (especially those of the American Political Science Association, European Consortium for Political Research, and the Elections Public Opinion and Parties specialist group) and colloquia (notably at Aberystwyth, Essex, and Manchester). We must also thank the following for helping provide us with important information regarding European Parliament electoral systems: Luciano Bardi, Michael Gallagher, Richard S. Katz, Adonis Pegasiou, Philip Stover, and Andreas Wüst. We also wish to record a special thanks to Dermot Scott, Director of the European Parliament UK office, for his considerable endeavours to help ‘open doors’ for us at various stages. Fourth, we offer thanks to our editor, Dominic Byatt, and the rest of the team at Oxford University Press. Their exemplary professionalism at all stages of the production process makes the job of the authors much easier. Fifth, we are very happy to acknowledge the financial support given by the Economic and Social Research Council of the
vi
Preface and Acknowledgements
United Kingdom (grants R000239231 and RES-000-22-1554) and the European Union’s Framework 6 programme (resulting from David Farrell’s involvement in the CIVICACTIVE project directed by Richard Sinnott at University College Dublin), which between them allowed us to conduct the interviews and surveys that this study draws heavily on. We also acknowledge the financial support provided by the Nuffield Foundation, as well as by Claes de Vreese (University of Amsterdam) who supplied some of his budget from a Dutch National Science Foundation grant, and from Susan Banducci (University of Twente). Finally, we record our thanks to our respective families, for their support and love in our academic endeavours. We dedicate this book to our parents who, between them, hold ultimate responsibility for setting both authors on their academic paths. DMF RS
vii
This page intentionally left blank
Contents
List of Figures List of Tables
x xi
1. Introduction: Representing Europe’s Citizens?
1
2. Giving the People What they Want: Public Attitudes to Representation in the EU
17
3. Electoral Institutions and Political Representation
41
4. Representation and Electoral Systems in Europe
63
5. Representation in Europe: The Institution and the Individuals
93
6. Electoral Institutions and Geographical Representation
119
7. Electoral Reform and the British MEP
139
8. Life Under List: Representing a Region
163
9. Conclusion: The Failure of Parliamentary Representation in Europe
197
Appendices Bibliography Index
209 217 229
ix
List of Figures
2.1. Trends in attitudes to the EU over time on three general questions
21
2.2. Perceived importance of the EP (%), 1987–2000
31
3.1. Electoral systems and parliamentary representation
51
3.2. Degrees of ‘openness’ in ballot structure design
53
4.1. Variations in the ballot structures used for EP elections in 2004
78
4.2. Disproportionality trends in EP elections, 1979–2004
80
4.3. The ratio of MEPs to the electorate in the EP, 1979–2004
81
4.4. The effective number of (parliamentary) parties trends in EP elections, 1979–2004
83
4.5. The proportions of women MEPs in the EP, 1979–2004
84
4.6. The consequences of variations in EP electoral system design in 2004
85
7.1. Locating Britain’s electoral reform in the context of the existing EP electoral systems in 1999
149
7.2. Conceptualizing Britain’s EP electoral system change and its consequences for the representative function of MEPs
151
x
List of Tables
1.1. The growth of the EP
6
2.1. Support for EU membership by member state, 1991 and 2003
25
2.2. EU referendums, 1972–2005
27
2.3. Perceived importance of EU institutions, 2005
30
2.4. Public trust in EU institutions, 2005
30
2.5. Levels of voter contact during the 2004 EP election
36
3.1. The five main families of electoral systems
44
4.1. Electoral systems for national and EP elections in 1979
70
4.2. The EP’s electoral systems, 2004
75
4.3. The representativeness of EP electoral systems in 2004
79
4.4. The effects of effective threshold and delegation size on disproportionality: OLS regression
86
5.1. MEPs’ opinions (%) on the powers of the EP
96
5.2. Bivariate correlation coefficients (r ) for EP powers items
97
5.3. OLS regression coefficients (standard errors) for EP powers scale and location of EP plenaries
100
5.4. How important is it to you to represent the following groups of people in the EP? (%)
105
5.5. ‘When thinking about your work as an MEP, how important are the following aspects of your work?’
106
5.6. Correlations of MEP individual role items
108
5.7. Rotated factor loadings for MEPs’ responses to attitudinal items (varimax rotation)
109
xi
List of Tables 5.8. Regression estimates (standard errors) for three representation factors
111
6.1. Time spent on electioneering by MEPs
121
6.2. Types and degrees of campaign activity
122
6.3. Time spent by MEPs in member states
123
6.4. Forms of voter contact
124
6.5. MEP contacts with organized groups and individuals (%)
124
6.6. Scoring the EP’s electoral systems on the intra-party dimension
130
6.7. Regression estimates (standard errors) for eight dependent variables
133
7.1. The 11 British regions for the 1999 and 2004 EP elections
148
7.2. Elections to the EP in Britain, 1994–2004
150
7.3. British MEPs’ view on representation: 1996 and 2000 compared
156
8.1. The representative role of MEPs
166
8.2. Team-related matters
167
8.3. Regional variation in British representation in the 1999–2004 EP
169
8.4. The representative role: regional features (%)
179
8.5. The representative role: party-level features (%)
180
8.6. The representative role: individual-level features (%)
182
8.7. Team-related matters: regional features (%)
190
8.8. Team-related matters: party-level features (%)
190
8.9. Team-related matters: individual-level features (%)
191
9.1. Electoral system design and elements of voter linkage: evidence from the 2004 EP elections
202
Appendices MEPMNP Survey (1996)
210
MEP2000 Survey
211
A.4.1. Disproportionality trends in EP elections, 1979–2004
212
xii
List of Tables A.4.2. Effective number of (Parliamentary) parties trends in EP elections, 1979–2004
213
A.4.3. Proportion of women MEPs elected to the EP, 1979–2004
214
A.4.4. Ratio of MEPs per voting age population in the EP, 1979–2004
215
A.6.1. Country codes for the modified Shugart index
216
xiii
This page intentionally left blank
1 Introduction: Representing Europe’s Citizens?
At 10 p.m. on 29 May 2005, polling stations across France closed their doors. The country had been voting in a referendum on the draft constitution for Europe. Immediately voting ended, exit poll figures on French television showed that the constitution—despite commanding the support of President Chirac, the government, and most of the French political establishment—had been decisively rejected by the people, by 55 to 45 per cent. As innumerable political commentators were quick to point out, it was particularly striking that France, one of the founder members of the European Union (EU) and a traditional leader of the integration project, had said non so strongly. Three days later came another referendum, in the Netherlands. The Netherlands is also an EU founder member, is a country with a strong internationalist tradition, and it has long been a solid stalwart of European integration. Yet the Dutch voted nee even more decisively than had the French—62 to 38 per cent. Though European political leaders vacillated and squabbled in the days and weeks that followed, it seemed clear that the constitution for Europe was now dead. It was not meant to be like this. The development of the EU has been one of the most successful and remarkable political projects in human history. Over the past fifty years, much of Europe has been transformed from a byword for hatreds and armed conflicts into the location of the world’s most advanced and deeply
1
Introduction: Representing Europe’s Citizens?
institutionalized form of international cooperation. The project of European unity made further substantial advances during the 1980s, 1990s, and early 2000s, and only a few months prior to the French and Dutch referendums had accomplished one of its greatest achievements: enlarging EU membership to ten new states, bringing into the fold many former members of the Communist bloc. In the view of its many supporters, the EU has provided much for the people of Europe to be grateful for. And yet, somewhere along the way, the EU had left many of the people behind. To many close observers of the Union, the 2005 referendum results were no shock. They simply confirmed trends that had been evident for many years. Since the early 1990s, public dissatisfaction with the European integration project has generally been on the rise, and the previous ‘permissive consensus’ has eroded. It may once have been true that the public would be willing to accede to whatever amount of integration political elites wished to give them. This is clearly not true now. This study is about one aspect of this problem facing the EU; or, to be more precise, about one putative solution. The European Parliament (EP) is the Union’s only directly elected institution. Many in the EU have believed that declining public support for European integration might be effectively tackled, at least in part, by empowering the democratic ‘voice of the people’ in EU politics. But we argue that while the EP has been a highly successful institution in many respects—in terms of winning greater powers, in establishing working practices that integrate representatives of many countries, parties and languages remarkably efficiently, and even in some ‘descriptive’ aspects of representation—it is a failure as a representative body. The reasons for this go further and deeper than the widely noted problems with EP elections, which have been heavily criticized for their failure to generate public enthusiasm or offer voters the opportunity to choose from alternative visions of the EU (van der Eijk and Franklin 1996). The manner that Europe’s Parliament is elected in many member states is actively obstructive of Europe’s voters being represented
2
Introduction: Representing Europe’s Citizens?
in the way that those voters are most likely to respond positively towards. Voters across Europe have distinct preferences over how they wish to be represented in Europe, and these preferences are similar across member states. But these voter demands are likely to be poorly served by what elected representatives are willing and able to supply. The behaviour of elected representatives is shaped strongly by the electoral institutions under which they operate. Yet in the EP, those institutions push most representatives to behave in ways contrary to what voters desire. The result is a failure of parliamentary representation at a time when links between EU institutions and the citizen have never been more important for the legitimacy of the Union. This opening chapter to the book has four main sections. First, we introduce the study by explaining the intellectual background to it. We discuss the importance of electoral institutions to representative democracy, and briefly address the actual and/or potential links between the two. Thus, the broad theoretical interest of the study is explained. Second, we outline the specific empirical focus of the research, in a brief historical overview of the development of the EP. After this, we elaborate on the research design of the study, explaining how a mixture of cross-national and temporal comparison allows for substantial analytical leverage to be obtained on the questions the study focuses on. Finally, we outline how the rest of the book addresses the key questions we seek to answer.
1.1. Electoral Institutions and Political Representation The defining feature of representative democracy, as a genus of political system, is that the votes of (much of) the populace determine the membership of key political institutions. Such institutions can include those held by single individuals—typically an elected presidency—as well as the multi-member representative parliaments that are a ubiquitous part of representative democracies.
3
Introduction: Representing Europe’s Citizens?
A necessary feature of such democracies is electoral systems, or the means by which votes are translated into seats in the process of electing politicians into office. Electoral systems are more than just mechanical devices that neutrally convert a public input (electoral preferences) into a political output (the holders of office). They can be, and often have been, the subject of furious arguments among politicians and other political activists—an implicit acknowledgement by those most closely involved in political life that they believe that electoral systems have important consequences. A substantial body of scholarly research has validated this belief. Electoral systems based around the principle of proportional representation (PR) have been shown to produce political institutions that are not only more ‘representative’ in partisan terms, but also in terms of members’ social characteristics (Farrell 2001), than are non-proportional systems. Moreover, PR leads not only to parliaments that look more like the societies they represent, but also ones that think more like them: the ‘opinion congruence’ between representatives and voters is notably lower in non-proportional systems than under PR (Huber and Powell 1994). In return, however, non-proportional systems have been argued to have other merits: producing strong and stable oneparty governments (and thus, it is often suggested, leading to better policymaking); and making it more difficult for extreme parties to win political representation. The provenance of some of these latter claims can be disputed. While PR systems do produce somewhat greater government instability on average they appear, if anything, associated with somewhat better policy outcomes (Lijphart 1999; Powell 2000). And some important recent work refutes the widely assumed link between PR and the electoral success of extreme right-wing parties (Carter 2005; although see Norris 2005). But while the precise consequences of electoral systems can be disputed, few question that they do have important consequences. Electoral systems have become a much more mainstream subject of political science research in recent years (certainly more than in the 1980s; see Taagepera and Shugart 1989: ch. 1), and
4
Introduction: Representing Europe’s Citizens?
we now know a great deal about the manner in which electoral arrangements shape who represents the people. However, we know rather less about the manner in which electoral systems affect how people are represented. It has long been accepted that representation is about more than simply a ‘match’ between people and politicians; it is an ongoing, dynamic process (Pitkin 1967). Within this process, how elected parliamentarians view and seek to carry out their role as representatives is of fundamental importance. And any impact that electoral systems have on this is also very important to understand. But previous research has made only quite limited progress in establishing how electoral systems influence the attitudes and behaviour of those who are elected. There is a body of research which suggests that electoral systems that create opportunities and electoral incentives for representatives to try to garner a ‘personal vote’ within a particular geographical constituency will indeed prompt behaviour consistent with those incentives (Ames 1995; Bowler and Farrell 1993; Cox 1990; Shugart 2001). Nonetheless, the evidence for this is limited in scope and fragmentary in nature. The weakness of work in this area is attributable in large part to problems of research design. In most single-country studies, the electoral system is a constant, not a variable. And comparative studies usually have great difficulties in distinguishing electoral system effects from other institutional influences (such as how the organization of particular parliaments shapes representatives’ attitudes and behaviour) and from broader cultural differences in how the representative relationship is defined. Moreover, in the relatively few cases where a major electoral reform has been experienced (such as New Zealand in the 1990s), such changes have accompanied—and often been prompted by—a broader transformation of politics and political culture (Boston et al. 1999). As a result, little established knowledge exists about how electoral systems shape the manner in which elected representatives define and carry out their role. Our study is directly aimed at addressing this deficiency in current understandings of the effects of electoral systems on political representation. We explain below how our research design
5
Introduction: Representing Europe’s Citizens?
attempts to overcome some of the problems that have bedevilled previous work. Before doing this, however, we first consider how, in addition to the general theoretical contribution of this book, it also seeks to make an important addition to understanding the politics of the EU and its elected parliament.
1.2. Electoral Institutions and the EP The 732 Members of the European Parliament (MEPs) elected in the June 2004 elections received mandates to serve in an institution very different from that in which their first predecessors served more than 50 years previously. What is now the EP began life on 10 September 1952 as the Common Assembly of the nascent European Coal and Steel Community (ECSC): a 78-member, parttime and unelected chamber (MEPs were delegates from member states’ national parliaments) wielding few powers. The changes over the intervening period have been about much more than simply a growth in the size of the EP (see Table 1.1 for a description Table 1.1. The growth of the EP Year
No. of MEPs
No. of member states
1952 1958 1973 1979 1981 1986 1994 1995 2004
78 142 198 410 434 518 567 626 732
6 6 9a 9 10b 12c 12d 15e 25 f
a
Status of MEPs Nominated Nominated Nominated Elected Elected Elected Elected Elected Elected
Title of chamber ECSC Common Assembly EC Common Assembly EP EP EP EP EP EP EP
Enlargement to Denmark, Ireland, and UK. Enlargement to Greece. c Enlargement to Portugal and Spain. d German enlargement and seat redistribution. e Enlargement to Austria, Finland, and Sweden. f Enlargement to Cyprus, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Slovakia, and Slovenia. b
6
Introduction: Representing Europe’s Citizens?
of how the EP has enlarged over time). MEPs (Members of the European Parliament) are now directly elected representatives— since June 1979, there have been six rounds of the five-yearly elections to the parliament—and the bulk of them are now fulltime European parliamentarians. A second set of changes has been to the role and status of the chamber. From the early 1970s, treaty amendments, institutional agreements, and other developments have gradually ratcheted up the EP’s powers over the EU budget, European legislation and in terms of oversight of the European Commission. In many respects, the EP can be considered a remarkably successful institution. It has used its status as the voice of the people in the EU very effectively in the battle to win greater prerogatives for itself (Rittberger 2005). As these powers have been won, the parliament has also reformed internally to facilitate using those powers with greater effectiveness (Kreppel 2002). Given its enormous internal diversity in terms of parties (the 2004 elections produced representatives from more than 150 different national political parties), nations, and languages, the EP works remarkably effectively as a functioning political body. One striking measure of this success is the fact that despite their considerable diversity, the multinational party groups display high internal cohesion in parliamentary votes, while divisions within the parliament are overwhelmingly along the left-right lines typical in national legislatures (Hix, Noury and Roland 2006). And there is no denying that, for many people, the existence of the parliament as a multinational, elected democratic institution has a very powerful symbolism.1 Nonetheless, it is far from clear that, in one of its core roles, the EP can be viewed as anything other than a failure. In contemporary Western democracies (a category that includes all member states of the EU), the holding and exercise of political authority is generally regarded as deriving legitimacy in essence from representative democracy. That is, government is legitimate government if, and only if, it is grounded in democratic procedures (Beetham 1991). During the 1980s and 1990s, many thought that this truism could be applied directly to the EU. As the Union developed and
7
Introduction: Representing Europe’s Citizens?
consolidated as a supranational political system, its legitimacy and popularity with the peoples of Europe might, it was believed, be guaranteed by bolstering the democratic credentials of the Union via enhanced status for its directly elected parliament (Herman and Lodge 1978; Williams 1991). The central problem with this strategy is that it does not work. The very substantial increases in the powers of the EP experienced during the 1980s and 1990s have certainly raised the salience of the parliament among political elites—most politicians, government officials, and political lobbyists are very well aware that the EP is now a very significant body in the politics of the EU. But this has had little or no impact on public perceptions. Enhancing the prerogatives of the EP has neither changed the way the public view the parliament (as is discussed in more detail in Chapter 2, public knowledge of the EP has remained low, and turnout in EP elections has fallen even as the powers of the chamber have grown) nor had any discernable impact on more general public attitudes towards the EU and the European integration project. While the EP might be the ostensible voice of the people in the EU, giving that body greater powers does nothing to enhance the public legitimacy of the Union. Why is this so? One possible explanation is that of ignorance. It may be that the public are so disconnected from developments in the EU that they are quite unaware of the growth in the EP’s powers,2 but that many would feel more positively about the EU if they understood more about its elected parliament, and the growing role that parliament has within the Union. Another possible explanation relates to the ubiquitous power of nationalism. Political authority in the contemporary world is legitimated not only via democracy, but also by being based around nation states. Governments are legitimate governments if they represent a particular people. Perhaps a parliament like the EP, which does not represent a single demos, cannot help confer legitimacy in like manner as its national counterparts? The analysis in this study does not attempt to test either of these two potential explanations. It may be that greater public
8
Introduction: Representing Europe’s Citizens?
understanding of the work of the EP would bring more positive attitudes towards it and towards the EU; or it may be the case that the very nature of the EP, as a multinational representative body, inevitably hinders it from legitimizing the wider political system of which it is a part. We remain agnostic on these matters. What we argue is that the EP’s failure as a representative institution follows, at least in part, from what are, on the face of it, rather more mundane reasons. Specifically, we suggest that although the EP can point to some successes in terms of descriptive representation (by which we mean, and shall show in later chapters, that the parliament has come to look rather more like the population of Europe), it is elected in ways that are inimical to people being represented in a manner that they would be most likely to respond positively towards. As well as a perceived ‘legitimacy gap’ (Rittberger 2003), the EU also suffers from a substantial ‘representation deficit’—a mismatch between the growth of the EP’s powers and the representative capabilities of MEPs. This is the result of a fundamental weakness in institutional design: the electoral systems used to elect most MEPs promote the interests of political parties and other organized interests over those of individual voters.
1.3. Research Design, Methods, and Data As has now been made clear, this study has a twin focus. In part, it is about an important aspect of the politics of the EU. But the study is also about the politics of electoral systems, and in addition to advancing an argument about political representation in the EU, we also seek to make a general theoretical contribution concerning political representation tout court. These two foci complement each other well, because the EU’s elected parliament is not only a fascinating experiment in multinational representative democracy. It also provides a laboratory of great interest to scholars of electoral systems. At first glance, this might appear to be a strange statement. After all, in 2002 the EU passed legislation requiring ‘uniform electoral
9
Introduction: Representing Europe’s Citizens?
procedures’ (UEPs) to be applied henceforth in the election of all member states’ representatives to the EP. Taking this at face value, the legislation would seem to make the electoral system a constant, and not a variable, and thereby to limit the scope for the EP to provide any comparative insight into the effects of different electoral arrangements on the attitudes and behaviours of MEPs from different countries. But in fact, as we detail in Chapter 4, the EU continues to allow considerable scope for differentiation along a number of dimensions: ‘uniform procedures’ certainly do not entail identical electoral arrangements. Thus, the EP continues to provide, as it has done for some years, a probably unique opportunity for insight into the role of electoral systems in shaping the attitudes and behaviour of political representatives, containing as it does within a single institution, individuals elected under a considerable variety of different electoral systems. Much of the empirical evidence and analysis presented in this study is based around our ability to compare representatives elected under different types of electoral system. We draw on detailed evidence gathered in several surveys of EP members conducted in recent years (see Appendix for details). These surveys probed into numerous aspects of MEPs’ attitudes and behaviours regarding their role as elected representatives, and the responses received on many of these items can be compared with the information we have on how they were elected to provide powerful cross-sectional evidence of any link between electoral institutions and representative attitudes and behaviours. But in addition, we are also able to draw on evidence from a detailed study of one EU member state, the United Kingdom. In the mid-1990s, the electoral system used for the election of British MEPs was changed.3 Whereas EP elections from 1979 through 1994 had used a version of the ‘single-member plurality’ (SMP) system, from 1999 onwards Britain shifted to a closed party-list system within multi-member regions. This was a rare contemporary example of electoral reform being applied to an existing political institution, and allows for a ‘before-and-after’ comparison of a precision rarely available to students of politics in advanced democracies. To
10
Introduction: Representing Europe’s Citizens?
examine the impact of this change we not only use survey evidence from before and after the change; we also use information gathered from a large number of interviews, and other types of qualitative evidence. Overall, by combining cross-sectional analysis with a more detailed, across-time examination of one particular country, we are able to develop a particularly powerful research design. This combines the strengths of cross-national, large-N analysis with the greater sensitivity and qualitative detail that our focus on the British case permits.
1.4. An Outline of the Study The subject matter outlined above is addressed in eight subsequent chapters. Chapter 2 begins the detailed work of the study by exploring public attitudes to representation in the EU, a topic that has hitherto been sparsely treated in the academic literature (cf. Gabel 2003). In Section 2.1 of the chapter, we review existing evidence about public attitudes to the EU and the integration process. This evidence suggests widespread public ignorance about the EU, but also growing levels of public suspicion and hostility in recent years. In Section 2.2, we then go on to explore public attitudes to representation in the EU. Drawing on Eurobarometer data we argue that, notwithstanding their low levels of knowledge about the EP, voters across the EU seem to have reasonably clear and common preferences as to what they want from their European-level representatives. Voter demands are for representatives who are responsive to the particular problems of their communities and willing to represent these concerns within the EU. Finally, we reflect upon the implications of these findings for the choice of electoral institutions for EP elections. Chapter 3 reviews existing scholarly knowledge about electoral institutions and political representation. We elaborate here on the argument that electoral systems are not some dry, constitutional
11
Introduction: Representing Europe’s Citizens?
detail, but a central and highly consequential political institution. We do so first by examining the diversity of electoral systems that exist, and how they have been classified and understood in previous work. Following on from this, we then explain what have been the major research findings about the political impact of electoral institutions. But in Section 3.3 of the chapter, however, we consider how our understanding of the consequences of two important features of electoral system design—ballot structure (how voters vote) and district magnitude (the place that the representatives represent)—remains underdeveloped. Section 3.4 of the chapter demonstrates that even under previous classifications of electoral systems, knowledge about the relationship between electoral systems and the attitudes and behaviour of political representatives remains scanty. Through reviewing the fragmentary existing research, and detailing the areas that it does not currently address adequately, the contribution of this study will be made more explicit. Chapter 4 begins to link our discussions on electoral systems with the issue of representation in the EU by exploring how the EP is elected. We start by reviewing the history of discussions concerning a ‘uniform’ electoral procedure for the EP, which date from before the first elections to the chamber in 1979, through to the legislation on uniformity enacted in 2003. However, as we then go on to elucidate in Section 4.2, the uniformity provided for in the legislation remains very limited, with considerable diversity persisting in two factors of greatest interest to this study—ballot structure and district magnitude. Finally, we examine the extent to which the EP has been effective in terms of being a ‘microcosm’ of the societies it represents, and the evidence of how this is influenced by the electoral systems used across member states. The conclusion of this section is that in terms of ‘descriptive representation’, the EP has been quite successful; the problems lie in other aspects of representation. Chapter 5 takes our discussion further by beginning to examine the attitudes and behaviour of the representatives themselves. Although the EP has often been accorded prominence in
12
Introduction: Representing Europe’s Citizens?
discussions of ‘democracy’ and ‘representation’ within the Union, very little attention has been paid in previous research to the attitudes of EP members. As a first stage in our empirical analysis of representation at the European level, we look in detail at MEPs’ views in two important areas. First, we explore MEPs’ attitudes to their collective position as members of the EP. What role as a representative institution do they think the Parliament can and should play? Following this, we then examine how MEPs view their role as individual representatives, and their priorities within that role. What or who do they consider it important to represent, and how do they seek to do so? Drawing on evidence from a major recent survey of MEPs, we assess the extent to which their views on these questions are predicted by individual, institutional, and political factors, including electoral system differences. In Chapter 6 we probe further into the relationship between electoral system variations and how politicians interpret their representative role. The focus here is on the phenomenon of ‘geographical representation’: that is, the efforts of elected politicians to develop links with the people of a specific geographic area— usually referred to as ‘constituency representation’ in the British context. We show, first, that there are numerous ‘observable implications’ of greater or lesser degrees of concern with geographical representation among MEPs. Second, building on the discussion in Chapter 3, we specify the plausible links between electoral systems and constituency representation. Finally, we conduct a detailed empirical analysis of survey data relating to a number of different dependent variables. Our analysis shows that while the relationship between electoral institutions and geographical representation is not wholly consistent or simple, there is nonetheless a strong general pattern. More ‘open’, candidate-based electoral systems are associated with greater concern with representing constituents among elected politicians; more ‘closed’, party-based systems with lower levels of constituency representation. In Chapters 7 and 8, the focus of the study shifts from largescale, cross-national analysis to the examination of one particular country. The electoral system used for EP elections in Britain was
13
Introduction: Representing Europe’s Citizens?
changed radically in 1999, with the SMP system being replaced by a PR regional closed-list system. This change represented a very rare case of fundamental electoral reform in an established political system, and the only instance of major electoral reform for EP elections. In Chapter 7 we first explain the extent to which a traditional hallmark of British representative democracy at many levels (from local councillors through to MEPs) has been the constituency-based representative. We then go on to detail the history of British representation in the EP, before explaining the precise nature of the change instituted in 1999, elucidating the differences between the two electoral systems and outlining key features of the new electoral regions instituted under PR. We then examine the major differences produced by the new system in terms of the partisan balance of British representation, as well as other effects on the ‘representativeness’ of the UK contingent (such as gender and racial balance). Finally, we explore survey data regarding British MEPs’ attitudes towards constituency representation before and after electoral reform. This analysis points towards a decline in British MEPs’ constituency focus. Chapter 8 examines more closely how electoral reform has produced changes in the manner in which British people are represented in Europe. Life for British MEPs as representatives is now very different from the way it used to be. Drawing on interviews with MEPs, and material gathered from visits to their regional offices, we explore some of the most salient differences. First, we examine MEPs’ attitudes to the electoral reform and to their representative role. Who do they now feel that they represent? To what extent has the focus of representation changed as a result of electoral reform? We then go on to consider behaviour—how do MEPs seek to carry out their representative duties? Overall, the evidence shows most of them having had to adapt quickly to the new regime, to varying degrees depending on regional and party circumstances, and with varying degrees of willingness depending on personal circumstances. The option of standing still has simply not been available. But as part of the process of adaptation, we suggest, the idea and the practice of representing
14
Introduction: Representing Europe’s Citizens?
the people of a distinct geographic area has declined among British MEPs. Chapter 9 draws together and reflects upon the main findings of the previous chapters. Three key themes are explored. First, we address the principal theoretical lessons offered by asking what the study tells us about the relationship between electoral systems and parliamentary representation. We argue that while the relationship is complex, electoral institutions nonetheless can be shown to have clear and strong effects on how the process of democratic representation is understood and conducted. Second, we consider the practical implications of our findings for the EP and for political representation more broadly. Is there an ‘ideal’ electoral system for the EP, and how might this differ from those most appropriate for other representative institutions? Finally, we reflect on the current failure of the EU’s elected chamber to supply the form of representation that most voters want: what implications does this have for the broader legitimacy of the European integration project in the twenty-first century?
Notes 1. In his acceptance speech for winning the 1998 Nobel Peace Prize, the Northern Irish Politician John Hume observed that: On my first visit to Strasbourg in 1979 as a Member of the European Parliament, I went for a walk across the bridge from Strasbourg to Kehl. Strasbourg is in France. Kehl is in Germany. They are very close. I stopped in the middle of the bridge and I meditated . . . If I had stood on this bridge 30 years before at the end of the Second World War when 25 million people lay dead across our continent and if I had said: ‘Don’t worry. In 30 years time we will all be together in a new Europe, our conflicts and wars will be ended and we will be working together in our common parliament’, I would have been sent to a psychiatrist. But it has happened, and it is now clear that the European Union is the best example in the history of the world of conflict resolution. (Oslo 10 December 1998)
15
Introduction: Representing Europe’s Citizens? 2. It is almost certainly true that most EU citizens are unaware of the EP’s growing powers: according to the Eurobarometer 63 survey, conducted in summer 2005, only 50 per cent of EU citizens are even aware that the chamber is an elected body! 3. Readers unfamiliar with the (many) peculiarities of the United Kingdom of Great Britain and Northern Ireland should note that the electoral procedures for choosing Northern Ireland’s three MEPs have always been different from those for mainland Britain. Northern Ireland has used the single transferable vote (STV) system since 1979, and was unaffected by the 1990s reforms. References to ‘Britain’ throughout this book should be understood as excluding Northern Ireland unless explicitly stated otherwise.
16
2 Giving the People What they Want: Public Attitudes to Representation in the EU
The introductory chapter provided an overview of this study. In doing so, it outlined the central message running through this book—the problematic nature of political representation in the EU. This chapter begins the task of elaborating this central message through the development of more detailed and specific arguments. Later chapters of the study are concerned with exploring how and why electoral institutions are important for shaping practices of political representation, and explore the impact of electoral systems on political representation in the EP. As we show, EU member states have tended to adopt electoral institutions that discourage those elected from supplying the forms of representation that are most demanded by citizens. But to develop this argument of a demand–supply mismatch, we must first attempt to establish what voters actually want from their representatives in the EU. This is the concern of the present chapter, which comprises two main sections. We start by providing a general overview of what we know about public opinion towards the EU and European integration. The extant academic literature, and available evidence, suggests several things about public opinion: growing levels of public hostility towards the EU and integration since the early 1990s; widespread public ignorance
17
Giving the People What they Want
about the EU and a desire for greater information; a greater politicization of ‘Europe’ as a domestic issue; and some suggestions that positive public attitudes, when they exist, are motivated increasingly by instrumentalist reactions to perceived benefits rather than any generalized or principled support for European unity. These findings establish the context within which more specific attitudes to political representation must be considered and understood. In Section 2.2, we then go on to explore in greater detail public attitudes to representation in the EP. In line with the more general literature, we find both substantial levels of ignorance about the EU’s elected chamber as well as declining support for the institution. Research suggests that how people are represented by their MEPs is not a matter of great salience among the public. Nonetheless, when people are prompted to consider these matters, the evidence does indicate that voters across the EU have reasonably clear and broadly common preferences as to what they want from their European-level representatives. These demands are for representatives who are responsive to the particular problems of citizens and their communities, and who represent these concerns within the EU. The chapter concludes with a brief discussion considering the implications of these findings for the chapters to follow.
2.1. Public Attitudes Towards the EU As with much of the academic literature on the EU, what has been devoted to understanding public attitudes to the EU has developed substantially in recent years. This development has encompassed both a growing sophistication in terms of the analytic techniques used (where research has increasingly drawn on concepts and methods developed in the most advanced political science research on public opinion) and also an increasing differentiation of the subject matter. One of us lamented a few years ago that, ‘the study of public opinion on the EU remains focussed almost exclusively on undifferentiated notions of the “EU” or “integration”.
18
Giving the People What they Want
We should surely aim to differentiate more clearly’ (Scully 2001: 521). Yet even as this lament was being issued it had begun to be less true, as scholars began to develop more refined and specialized inquiries. The purpose of this section of the chapter is to review and assess some of the major strands of work on public attitudes towards the EU, and to outline the substantive conclusions that this work points towards. As with research elsewhere, the systematic investigation of public opinion about the EU has depended heavily on representative sample surveys to chart the landscape of popular attitudes. Surveys are not, of course, the only source of valuable evidence. Media commentators and academics alike have long drawn valuable impressions of the public mood from diverse sources in popular culture and contemporary events. Specific political occasions like elections and referendums (and the academic studies that frequently follow them) also provide substantial information. Survey-based research has also been supplemented by alternative methods, such as focus groups and community-based studies. Nonetheless, the survey remains the workhorse of public opinion research. Scholars of the EU are, therefore, extremely fortunate to be able to draw on the Eurobarometer surveys: a series of biannual surveys (supplemented by a large number of one-off studies) of the public in EU member-states dating back to 1973, and containing numerous questions that have been repeated across many of the surveys.1 The Eurobarometer surveys are neither a perfect source of evidence nor the only valuable source, but their longevity does provide invaluable insight into the evolution of public attitudes towards European integration. It is by no means self-evident how one might best measure public attitudes in this area. After all, the EU and European integration are not clearly identified, unambiguously defined objects towards which all or even most people can be expected to possess considered well thought out opinions. The Eurobarometer surveys have attempted to gauge the public mood by asking a number of different questions, which tap into the general phenomenon of the EU in rather different ways. Among the more general questions
19
Giving the People What they Want
that have been repeatedly asked over many years are ones probing respondents’ attitudes to their own country’s membership of the Union (‘Generally speaking, do you think [your country’s] membership of the EU is a “good thing”, a “bad thing”, “neither good nor bad”, “don’t know”?’); whether they perceive their country to have benefited from membership of the EU (‘Taking everything into consideration, would you say that [your country] has benefited from being a member of the EU?’); and their attitude to the hypothetical abolition of the EU (‘If you were told tomorrow that the EU had been scrapped, would you be very sorry about it, indifferent, or very relieved?’). Figure 2.1 charts fluctuations in support for European integration over time by presenting the respective percentages of those that answered positively to these three questions: namely, the percentage of the sample answering that their country’s membership of the EU was a ‘good thing’; the percentage stating that their country had benefited from being a member of the EU; and the percentage saying that they would be ‘very sorry’ to hear that the EU had been scrapped. How should findings like those reported in Figure 2.1 be interpreted? Scholars of public opinion and the conduct of attitude surveys have long made clear that survey responses, in themselves, constitute no more than a set of answers to a series of questions (Weisberg et al. 1996). We cannot simply read off responses as revealing the landscape of public opinion in a wholly unproblematic manner. This is particularly so when—as is the case with opinions regarding the EU (discussed later)—the questions concern matters of, or in which, the majority of the public have only limited knowledge or interest. More subtle interpretations are required, ones that students of public attitudes to the EU have sought to provide. The most obvious thing about the findings reported in Figure 2.1 is the similarity of the trends over time along the three different questions selected. Such trends, indeed, are typical of those obtaining on many of the standard Eurobarometer questions, and point to broad changes in the public mood towards European unity. Much of the early scholarly literature on the subject talked about a
20
Giving the People What they Want
Panel A: If heard EU to be scrapped, would be sorry? 60 50 40 30
Very Sorry
20 10
88 19 90 19 92 19 95
83
85
19
19
77
19
19
19
73
0
Panel B: Has country benefited from EU membership? 70 60 50 40 30
Has benefited
20 10
19
83 19 86 19 88 19 90 19 92 19 96 19 98 20 01
0
Panel C: Is country’s membership of EU a good thing?
Good Thing
19 73 19 78 19 82 19 87 19 92 19 97 20 02
80 70 60 50 40 30 20 10 0
Figure 2.1. Trends in attitudes to the EU over time on three general questions
21
Giving the People What they Want
‘permissive consensus’ in public attitudes towards European integration (Lindberg and Scheingold 1970). Surveys and other evidence indicated that the west European public (the relevant public in the 1960s and early 1970s) was not actively demanding that its political leaders move rapidly ahead with European integration. Main streets and central squares in major cities were not being filled with angry or excited crowds demanding ‘European unity now’! But equally, there was little substantial public opposition to the advancing of integration. Compared with matters such as the economic standard of living, health, education, and crime, Europe was not a major concern for many people. Yet, while voters were not clamouring for greater integration, they were generally favourable to the idea when questioned about it in surveys, and likely to accede if and when their political leaders were able to advance the structures and practices of European cooperation. This sort of landscape of public opinion gave considerable freedom of movement to political leaders—neither being pressed to advance further than the more reluctant might wish, nor being greatly restrained from what the more Euro-enthusiastic might seek. As Inglehart summarized, ‘There was a favourable prevailing attitude towards the subject, but it was of low salience as a political issue— leaving national decision-makers free to take steps favourable to integration if they wished but also leaving them a wide liberty of choice’ (1970: 773). However, the evidence of Figure 2.1 (and other evidence reported in numerous other studies2 ) suggests that there was a darkening of the public mood in the 1970s and into the 1980s. This was the era of what has come to be termed ‘Euro-sclerosis’. The EU states shared in the global economic slowdown experienced at this time, and the political project of integration appeared to most observers to have run out of momentum, with the Community having become bogged down. While some scholarship about this period has challenged the notion that EU decision-making stagnated in these years (Golub 1999), there was certainly a decline in positive responses to questions about the EU and European integration. Inasmuch as these survey responses can be taken to have
22
Giving the People What they Want
reliably tapped into public attitudes, those attitudes had clearly become more negative. From the mid-1980s onwards, however, the public mood became markedly more positive. These were years of economic recovery and growing prosperity in most EU member-states, and also of the renewal of the political project of European integration (often referred to at the time as the ‘relaunch’). While the revival of integration was an essentially elite-led development, it was one not without public resonance: Figure 2.1 shows that there were substantially rising levels of support for European unity during these years. The high-water mark of this period of ‘Euro-phoria’ was the Maastricht Treaty on EU. Signed in early 1992, this treaty promised to create a single European currency and make other important advances in the political unity of the European continent. At the time, such a political project appeared to be responding to, or at the very least to be consistent with, a substantial tide in public attitudes. Yet the signing of the Maastricht Treaty was followed in rapid succession by the emergence of several problems for the EU, and by a steady decline in public support. Denmark required two referendums (in 1992 and 1993) for its political leaders to persuade the Danish people to ratify Maastricht; the French, despite the leading role taken by their political leaders in advancing the agenda of European integration, also almost rejected the treaty in a referendum in September 1992. The Maastricht plans for a smooth transition towards the single currency by the end of the 1990s were severely, if temporarily, disrupted by crises in the European Monetary System that aligned EU national currencies (Dyson 2000; Gros and Thygesen 1998). And these events were accompanied—as Figure 2.1 illustrates—by a substantial decline in public support for European integration, which has never recovered to the same levels experienced in the early 1990s. Some analysts spoke of anti-EU ‘genies’ having been uncorked from bottles by the scope and scale of the Maastricht agenda (Franklin et al. 1994: 470); more generally, the era of the permissive consensus was widely agreed to have come to an end (Franklin et al. 1996; Niedermayer 1995). These
23
Giving the People What they Want
developments have prompted considerable efforts by scholars to chart the extent of, and the reasons for, growing public opposition to the EU and to European integration. One immediate possible response is to question the extent to which much of the change in support for the EU is genuine and not merely an artefact of the enlargement of the Union. After all, the trends reported in Figure 2.1 do not relate to a constant group of states, but rather are for the EU as a whole, and may thus be distorted by the changing membership of the Union during these years. Just as the rise in support shown in the late-1980s was boosted by the accession in 1986 of Spain and Portugal (countries that were, not least for historical reasons to do with putting behind them an anti-democratic past, very enthusiastic about joining the EU), so the figures for the 1990s might be expected to have been depressed by the accession in 1995 of Austria, Finland, and Sweden, three states that, for various reasons, have long been more cautious about integration. But, in fact, enlargement accounts for very little of the trends discussed above. This can be demonstrated by the data reported in Table 2.1, which shows how reported support in individual member states for membership of the EU3 changed between 1991 (or 1995 for Austria, Finland, and Sweden) and 2003. As can readily be seen, support fell in almost all states, and the decline across the EU as a whole was substantial. Much of the academic research attempting to understand the reasons behind the decline in support for the EU has suggested that the salience of EU-related matters has risen among the public. That is, the decline of the permissive consensus is said to be not only about falling levels of support for closer European unity and for the achievements of the EU. An important part of the process, and an important part of the reasoning behind declining public support, is said to be that the public has become more aware of European integration (e.g. Franklin and Wlezien 1997). And an increasingly aware public has proven less willing to go along with an elite-led integration process. These arguments are far from wholly unproblematic. It has long been noted that, at the individual level, greater knowledge of the EU is generally associated with greater
24
Giving the People What they Want Table 2.1. Support for EU membership by member state, 1991 and 2003 Member state
1991a
2003
Change (%)
Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden UK All EU
39 75 62 50 71 70 76 77 79 84 89 79 78 38 57 74
35 56 57 39 44 46 62 73 58 77 62 55 62 40 28 48
−5 −19 −5 −11 −27 −24 −14 −4 −21 −7 −27 −24 −16 +2 −29 −26
a
Figures for Austria, Finland, and Sweden are from spring 1995, the first Eurobarometer survey conducted after their accession to the EU. Source: Adapted from Hix (2005: 153), using Eurobarometer 35 (spring 1991), 43 (spring 1995), and 60 (autumn 2003).
support for integration:4 one might, therefore, have expected any enhanced public awareness to have had a positive impact on support for the Union. But an even more pertinent point is that it is far from clear that the general public did become better informed or more interested about the EU as general attitudes became more critical. The public’s self-reported awareness of central EU institutions like the Commission was no higher by the mid-1990s than it had been pre-Maastricht. (For instance, 51 per cent responded ‘Yes’ in Eurobarometer 44.1, December 1995, to the question ‘Have you recently seen or heard, in the papers, on the radio, or on television, anything about the European Commission in Brussels, that is the Commission of the European Community?’; this was exactly the same percentage as responded ‘Yes’ in Eurobarometer 30 in November 1988). Nor did the public’s own reported interest in EU matters increase: Eurobarometer 30 found that 39 per cent of people responded ‘A Great Deal’ or ‘To Some Extent’ to a question
25
Giving the People What they Want
gauging interest in EU politics;5 that figure actually fell slightly (to 37 per cent) by the final time the question was asked, in Eurobarometer 42, December 1994. The broad picture that emerges from much of the evidence is that the public were not much, if at all, more knowledgeable about, or interested in, the EU by the mid- to late-1990s, when support for integration was low, than they were at the start of the decade when support had been much higher. The increasingly sophisticated analysis of public attitudes to the EU that scholars have developed has coalesced around two broad ‘schools’ of interpretation (Hooghe and Marks 2005). The first of these views attitudes as primarily economic-led. That is, public reactions to the EU are viewed as being shaped primarily by the general economic climate and by the direct impact of EU policies on the economic circumstances of individuals and their communities (Brinegar, Jolly and Kitschelt 2004; Gabel and Whitten 1997; Gabel 1998). The second, increasingly popular school approaches public attitudes to the EU as an ‘identity-led’ phenomenon, with attitudes to European integration seen as heavily conditioned by factors such as: strength of identification with one’s own state (Diez Medrano and Gutierrez 2001); attitudes towards domestic political actors and institutions (Anderson 1998; Rohrschneider 2002); and the sense of ‘cultural threat’ that might emanate from elsewhere in Europe (McLaren 2002). But whichever of these broad explanatory approaches is preferred or given priority by analysts, certain commonalities exist. And one of these is that the decline in popularity of the idea of European unity has generated public reactions that are increasingly critical and instrumentalist in nature (Franklin and Wlezien 1997).6 It is also true that ‘Europe’ has become increasingly politicized as an issue in domestic politics in many EU member states. It used to be the case that, with few exceptions, anti-EU positions were confined to the political margins; they were, to use Paul Taggart’s astute (1998) and evocative phrase, a ‘touchstone of dissent’. This has become much less true: mainstream politicians and parties have become more likely to articulate at least mildly Euro-sceptical positions (Steenbergen and Scott 2004). And where new, populist
26
Giving the People What they Want
political movements articulate such views, the widespread decline of traditional party loyalties in recent decades has made it easier for these movements to gain political traction. ‘Europe’ is more on the domestic political agenda in many member states, and being at least somewhat anti-EU has become a more feasible political position to adopt. Thus, one observes a steady rise in the success of anti-EU or (to varying degrees) Euro-sceptical parties in elections to the EP (reaching about 24 per cent of all MEPs elected in 2004); also the occasional populist movements (such as that of Pim Fortuyn in the Netherlands in 2002) that draw on Euro-sceptical themes as part of their broader appeal; and the growth of specifically EU-related political protest activity (Imig and Tarrow 2001). This is also the context in which one can interpret the trends in EU-related referendums. Table 2.2 reports the results for all statewide referendums held on EU matters (except for membership referendums, where
Table 2.2. EU referendums, 1972–2005 Date April 1972 February 1986 May 1987 June 1989 June 1992 June 1992 September 1992 May 1993 May 1998 May 1998 May 2000 June 2001 October 2002 September 2003 February 2005 May 2005 June 2005 July 2005
Country
Issue
Yes (%)
France Denmark Ireland Italy Denmark Ireland France Denmark Ireland Denmark Denmark Ireland Ireland Sweden Spain France The Netherlands Luxembourg
Enlargement Single European Act Single European Act Negotiating Mandate Maastricht Treaty I Maastricht Treaty Maastricht Treaty Maastricht Treaty II Amsterdam Treaty Amsterdam Treaty EMU Membership Nice Treaty I Nice Treaty II EMU Membership EU Constitution EU Constitution EU Constitution EU Constitution
68.3 56.2 69.6 88.1 49.3 68.7 51.1 56.8 61.7 55.1 46.9 46.1 62.9 42.0 76.7 45.3 38.4 56.5
Note: Negative results are underlined.
27
Giving the People What they Want
populations were being asked to decide on their country’s continuing or future membership of the Union). Although there are too few instances here to draw out statistically significant trends, two things are fairly clear from the table: first, that referendums are increasingly frequent events, but second, that populations are increasingly viewing EU proposals with a sceptical eye. The first referendum that rejected an EU treaty or other major advance in integration was that in Denmark in June 1992. Since then, five further referendums—all of them held since May 2000—have yielded negative results. The general picture that emerges from the evidence, and from the work of scholars to understand it, can be summarized as follows. The EU has declined substantially in popularity among the public since the early 1990s. While this decline might partly be said to be simply a reversion to normality after the substantial rise in popularity during the immediately preceding years—so that it might well be argued that the brief era of Euro-phoria was the truly exceptional period—there is no doubt that attitudes towards the EU have, throughout Europe, come to be tinged with greater scepticism in recent times. This growing scepticism is not a function of growing knowledge: the public has largely remained fairly uninformed about, and mainly uninterested in, the EU. Nevertheless, there has been a growing trend towards a politicization of ‘Europe’ as an issue in many member states. And in a broader context of the decline of traditional party loyalties, the fading of European idealism has rendered voters increasingly responsive to political elites advancing Euro-sceptical positions. It is in this broad context that we approach the more specific task of this chapter: understanding the attitudes of the public towards the EP and towards how they are represented in the EU. A general public mood that appears to balance apathy and scepticism towards the EU would not, prima facie, appear to offer fertile ground on which to build a strong relationship between representatives and the represented. Section 2.2, therefore, moves on to consider what—if anything—voters think about how they are, and how they might be, represented in Europe.
28
Giving the People What they Want
2.2. Representation in the EU: Do Voters Know or Care? As already discussed, it is well established that the majority of the European public do not approach consideration of matters relating to the EU with a substantial amount of enthusiasm, interest, or knowledge. We can certainly expect this to be the case for specific EU institutions, including the EP. Most voters are unlikely to know much about the EP, or to have thought a great deal about it. And the fact that the EP is likely to be of very limited salience to most people necessitates that we exercise some caution in interpreting survey evidence concerning public attitudes to the chamber as a representative body. But none of this requires that the public might not have distinct, if somewhat inchoate and underdeveloped, preferences in this respect; preferences that might be identified from consideration of the available evidence. Standard Eurobarometer surveys have typically asked a number of questions about the EP, and the evidence from these can be complemented with that from various special studies to help build a general picture of public attitudes towards the chamber. Several of the questions used are also asked in equivalent form for other EU institutions, thus giving us some useful comparators with which to evaluate responses about the EU. As a starting point, we can consider questions that examine public perceptions on the perceived importance of not only the EP, but also such other major EU institutions as the Commission, Council of Ministers, and European Court of Justice (‘For each of the following European bodies, do you think it plays an important role or not in the life of the EU?’), and a further set of questions that probe levels of public trust regarding these same institutions (‘And for each [institution], please tell me if you tend to trust it or tend not to trust it’). As the evidence presented in Tables 2.3 and 2.4 suggests, in comparison to other EU institutions, in 2005 the parliament scored reasonably highly with the public. The EP was the most trusted and the highest ranked in terms of perceived importance.
29
Giving the People What they Want Table 2.3. Perceived importance of EU institutions, 2005 Institution
‘Important’ (%)
European Parliament European Commission European Central Bank European Court of Justice Council of Ministers
77 72 67 61 61
Source: Eurobarometer 63 (spring 2005).
Table 2.4. Public trust in EU institutions, 2005 Institution European Parliament European Court of Justice European Central Bank European Commission Council of Ministers
‘Tend to trust’ (%) 52 52 47 46 42
Source: Eurobarometer 63 (spring 2005).
However, a much less positive picture of public attitudes begins to emerge if we look rather more deeply at the evidence specifically concerning the EP. While the parliament may be more trusted than most EU institutions, levels of distrust have been growing over time: the level of those reporting that they ‘Tend not to Trust’ the EP had reached 31 per cent by spring 2005 (Eurobarometer 63), the highest level yet recorded on a question that has been asked consistently since Eurobarometer 39 (April 1993). A similar picture emerges from a question probing respondents’ ‘desired’ level of importance for the EP (‘Would you personally prefer that the EP played a more important or a less important part than it does now?’) that was asked consistently by Eurobarometer surveys from 1983 (Eurobarometer 19, April 1983) through to 2000 (Eurobarometer 53, June 2000). From a high of 61 per cent answering ‘More Important’ in Eurobarometer 35 (spring 1991), ratings fell substantially to a low of 38 per cent in Eurobarometer
30
Giving the People What they Want
48 (autumn 1997), with the four lowest ratings (42 per cent or below) all being registered at the end of the decade. It is only in relation to other EU institutions that the parliament now looks popular. It is also true that the EP has had, and retains, a very low profile among the European public. This is shown by a standard Eurobarometer question which tracks respondents’ self-reported awareness of the parliament. The evidence from this question indicates that even in surveys taken in the immediate aftermath of EP elections, when public awareness of the chamber is close to its peak, nearly half of the public do not report having heard anything about the EP during the relevant period.7 Similarly, Eurobarometer 63, conducted one year after the June 2004 EP elections, found only 50 per cent of respondents aware of the fact that MEPs are elected by the public, and only 29 per cent replying correctly (‘False’) to the statement that ‘The last European elections have taken place in June 2002’.8 Another means of demonstrating essentially the same point is to examine trends in the ‘perceived’ importance of the EP over time. Objectively, the parliament has become a much more important institution within the EU since the passage of the Single European Act in 1987. Yet, as Figure 2.2
70 60 50 40
Imp + V.Imp
30 20 10
M
ay
-9
9
7
5
ay -9 M
-9 ay M
ay
-9
3
1 M
ay -9 M
-8 ay M
M
ay
-8
7
9
0
Figure 2.2. Perceived importance of the EP (%), 1987–2000 Source: Standard Eurobarometer surveys (EB27–EB53).
31
Giving the People What they Want
demonstrates, the substantial increases in powers experienced by the chamber between 1987 and the end of the 1990s made very little impact on public perceptions of the importance of the chamber: the line showing the proportion rating the chamber as either ‘Very Important’ or ‘Important’ ‘in the life of the European Community nowadays’ displays only trend-less fluctuation, and barely increases at all over the entire period. Taking all this evidence together, it is difficult to dissent from the conclusion of one major study, that the chamber suffers from a ‘failure to even begin to penetrate the consciousness of so many of its electors’ (Blondel et al. 1998: 242). An obvious question that arises from this evidence is why and how people answer survey questions about the EP, when it is evident that a great many of them know little or nothing about the institution? As part of the burgeoning literature on public opinion in the EU, analysts have begun to understand more about public attitudes towards particular aspects of the EU. Gabel (2003) provides the farthest reaching exploration yet of reported public attitudes to the EP. He finds that specific answers to questions on the parliament are heavily conditioned by much more general factors, including respondents’ sense (or lack thereof) of European identity, and their instrumental evaluation of the impact of EU membership on their country and ‘citizens’ concerns about the scope of EU authority’ (2003: 305). Although most people are quite capable of answering survey questions regarding the EP, Gabel concludes that, ‘a majority of the public has not formed a clear opinion supporting the EP or a future parliamentary system for the EU’ (Gabel 2003: 306). Given low levels of knowledge about the EP, and the lack of public salience it enjoys, we could not expect the majority of Europeans to possess deeply thought out attitudes regarding how they are represented in or by the institutions. It does not, however, automatically follow from this that people do not care at all about how they are represented in the EU, or that if they were asked to think about the subject that they could not articulate some broad preferences in this regard.
32
Giving the People What they Want
Nonetheless, very little research hitherto has actually sought to explore how people perceive the manner in which they are represented in the EU, and whether they have any clear preferences in this respect. Work on elections to the EP has not dealt much with such matters, having overwhelmingly concentrated on other questions—such as the causes of low (and declining) levels of electoral participation, and the differences between aggregate voting patterns in EP compared to national elections (Blondel et al. 1998; Caramani 2005; van der Eijk and Franklin 1996). However, by drawing from a variety of surveys and other studies, it is possible to identify fragments of evidence that point to some reasonably clear and consistent conclusions. A first point to make is that not only are most people (as demonstrated above) fairly uninformed about the EU, a great many of them recognize that they are uninformed and, when asked about it, express a wish to become better informed. As an illustration, Eurobarometer 63 (spring 2005) found only 2 per cent of respondents willing to claim that they ‘know a great deal’ about the EU, while some 70 per cent acknowledged that they know ‘a bit’ or ‘almost nothing’; the same survey also came across far more people believing that the news media give ‘too little’ information on the EU than believe that they give too much.9 Moreover, there are good reasons to believe that this lack of information is harmful for the EU. Not only does it make it very difficult for citizens to ever develop substantial respect for, or attachment to, EU institutions like the EP, there is also growing evidence that such lack of information generates a sense of public alienation from the Union and major projects associated with it. The Eurobarometer survey conducted shortly before the French and Dutch referendums in 2005 found 28 per cent of opponents of the draft EU constitution responding to an open-ended question regarding their reasons for opposition with concerns about ‘lack of information’ (Eurobarometer 63, spring 2005); the special study conducted after the Netherlands referendum found some 32 per cent offering lack of information as a reason for voting against the constitution—more than any other reason nominated.10
33
Giving the People What they Want
A second important point is that some strong evidence exists that, not least because of this widespread sense of lack of information and knowledge about the EU, many citizens—when questioned about the matter—express a desire for political representatives to help bridge this gap between people and the Union. An important, large-scale qualitative focus group study conducted in 2001 found a strong desire among participants that they should be directly represented in the EU; this was common across different member-states. According to this study the participants felt that these representatives could and should act for them, both as a conduit for information, and as advocates of the EU for their particular interests and concerns.11 Furthermore, and importantly for the purposes of this study, among those that respondents have identified as potential sources of this information, and a direct link between the people and the EU, are MEPs. Eurobarometer 56.3 (May 2002) found that while only 1 per cent of respondents reported that they ever had used an MEP (or a Member of a National Parliament) as a source of information about the EU, some 24 per cent of respondents agreed that MEPs should serve as a source of ‘communication about the EU and its activities’. The same survey also found there to be considerable support for greater information about and from MEPs being available on the internet.12 There appears also to be a significant desire across many states for people to be able to identify with individual MEPs as their representatives.13 Nonetheless, there is also a widespread and common public perception that the EP is not very effective in terms of representing people’s concerns. The post-2004 EP elections Eurobarometer study found that a consistent criticism across all states was the ‘EP’s failure to take account of the concerns of European citizens’.14 Specifically, the survey reported 44 per cent of respondents across the EU rejecting the statement that ‘The European Parliament takes into consideration the concerns of European citizens’, with the figure being slightly higher (47 per cent) and fairly consistent across those EU member states with significant experience of
34
Giving the People What they Want
representation by the EP (namely the pre-2004 accession ‘EU15’ states). Similarly, Eurobarometer 63 (spring 2005) found a clear majority (53 per cent) disagreeing with the statement that, ‘My voice counts in the EU’, with only 38 per cent agreeing; elsewhere, the same survey found 19 per cent volunteering that ‘getting closer to European citizens’ should be one of the three major EU priorities—the same number that suggested ‘fighting terrorism’ and ‘protecting the environment’ as priorities, and a greater number than nominated ‘fighting illegal immigration’ or ‘implementing the single European currency’. Further evidence can be drawn from the post-2004 EP elections Eurobarometer study, which asked a series of questions about the campaign, and particularly about the types of campaign communication received by voters and whether they felt sufficiently informed. Table 2.5 provides some indicative descriptive statistics relating to these questions. These figures reveal very little personal contact with voters (in the form of ‘active’ campaign communications); rather more contact in the form of election leaflets (i.e. ‘passive’ campaign communications); but, nonetheless, a general sense among most voters that they had sufficient information prior to determining their electoral choice. The table also provides the national variations in the responses to the three sets of questions. As we see in later chapters, it is among those countries that score highly on the various items here (as underlined in the table) that we find the cases of electoral systems that tend to promote greater voter contact by MEPs.
2.3. Conclusions The general conclusion to draw from the evidence reported in this chapter is the following. Representation in the EP is not a matter of great salience for most Europeans: few of them know much about the EP or have given substantial thought to how they are represented within it. Yet it does not follow from this that they
35
Giving the People What they Want Table 2.5. Levels of voter contact during the 2004 EP election
Austria Belgium Cyprus Czech Rep. Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Portugal Slovak Rep. Slovenia Spain Sweden UK Average
Candidate called to homea (%)
Voter received election leafletsb (%)
Voter had necessary informationc (%)
3 6 5 3 1 2 0 1 2 4 3 68 14 1 2 2 20 0 1 2 1 0 3 1 14 6
71 86 64 56 49 68 60 77 55 18 81 96 89 55 56 91 94 53 44 40 65 84 83 75 76 68
66 69 62 41 52 62 70 57 60 66 73 73 71 49 64 79 80 50 28 58 42 76 66 44 62 59
a
‘Political parties and candidates campaigned for votes in the European Parliament elections we have just had. For each of the following, please tell me if you have been in this situation or not. . . . Political parties or candidates or their representatives called to your home.’ b ‘. . . .You received leaflets concerning the European Elections in your mailbox.’ c ‘For each of the following propositions, please tell me if it rather corresponds or rather does not correspond to your attitude or your opinion. . . . You had all the necessary information in order to choose for whom you were going to vote in the recent European Elections.’ Note: Those cases with highest proportions answering ‘Yes’ are underlined. Source: Flash Eurobarometer: Post European Elections 2004 Survey, June 2004.
are wholly indifferent to how they are represented in the EU, nor that they would not be more likely to respond positively to some forms of representation than others. There is a widespread public perception, largely common across the EU, that the Union is a
36
Giving the People What they Want
distant organization about which they know little and towards which they can have little connection. When prompted to consider these matters, people do—and again, largely in a common manner across the EU—express a desire to be better informed, and to have direct, identifiable representatives who can act as sources of information and advocates for the interests of themselves and people like them. In short, most people have some, albeit inchoate, desire to have ‘their’ representatives in Europe. The question to be addressed in the remainder of this volume is whether, and to what extent, these desires are likely to be satisfied.
Notes 1. Standard Eurobarometer surveys include 1,000 respondents from most member states (a sample of only 600 is usually obtained in Luxembourg—and, since their accession to the EU, Cyprus, and Malta), although there has long been over-sampling in the UK (to allow for a sample of 300 in Northern Ireland) and in postreunification Germany. As of February 2006, 63 standard Eurobarometer studies had been completed and documented. In addition, 237 ‘special’ studies focusing on particular issues had been reported, along with 176 ‘Flash’ Eurobarometers (surveys conducted rapidly, using telephone polling, in response to urgent issues). The EU also funded special surveys on the accession states between 2002 and their joining the EU in 2004. Further details can be accessed at http://europa. eu.int/comm/public_opinion/index_en.htm. 2. On the evolution of public support for European integration through the 1970s and 1980s, see Anderson (1995), Niedermayer (1995). 3. The question discussed above: ‘Generally speaking, do you think [your country’s] membership of the European Union is a “good thing”, a “bad thing”, “neither good nor bad”, “don’t know”?’. 4. As an example of the relationship between knowledge and support for integration, Eurobarometer 63 (spring 2005) finds that respondents’ support for their country’s EU membership was at 40 per cent among those rated as ‘low’ in knowledge about the Union, 61 per cent among those ‘medium’ in knowledge, but 67 per cent among those scoring a ‘high’ knowledge rating.
37
Giving the People What they Want 5. ‘And as far as European politics are concerned, that is matters related to the European Community, to what extent would you say that you are interested in them?’ 6. In a similar manner, even a recent Eurobarometer report concedes that ‘European citizens are today more critical in their analysis of the European Union, without however calling into question either their membership of the European Union or the European construction itself’ (Eurobarometer 63 Report, September 2005: 5). 7. The standard question is: ‘Have you recently seen or heard in the papers, or in the radio or TV, anything about the European Parliament, that is the parliamentary assembly of the European Community?’. This question was asked consistently in Eurobarometer surveys from 1997 to 2001. In the November 1999 survey (Eurobarometer 52), conducted only a few months after the June 1999 EP elections, 53 per cent of respondents answered ‘Yes’ to this question. It is also true, moreover, that these reported figures for awareness of the EP are almost certainly over-statements of the true figure. Given that the figures are based on self-reports, and that social desirability factors (in this context, the desire among many survey respondents to appear more knowledgeable than they actually are) will lead to inflated estimates, the true figure must be lower—and quite possibly much lower. 8. It is, perhaps, worth noting that the question regarding the date of the previous European elections offered only 3 options to survey respondents; one should therefore expect that purely by random guessing, 33 per cent of respondents should have given the right answer—some 4% more than the number who actually did! 9. Eurobarometer 63 (spring 2005) report, pp. 84–5. Other Eurobarometer studies have also shown substantial public interest, when respondents were prompted to consider the matter, in receiving greater information about the EU (e.g. Eurobarometer 56.3, May 2002). 10. Dutch referendum study, p. 15. 11. Perceptions of the EU: A Qualitative Study of the Public’s Attitudes to and Expectations of the EU in the 15 Member States and in 9 Candidate Countries. 12. Eurobarometer 56.3 (May 2002), Getting Information on Europe, the Enlargement of the EU and Support for European Integration: European Public Opinion Takes the Floor.
38
Giving the People What they Want 13. The Flash Eurobarometer, Post European Elections 2004 Survey found 43 per cent of all respondents assenting to the statement that ‘it is very important for you which particular candidates win seats and become MEPs in the EP elections’. 14. Flash Eurobarometer, Post European Elections 2004 Survey, p. 46.
39
This page intentionally left blank
3 Electoral Institutions and Political Representation
Representation of an individual in a society is a condition which exists when the characteristics and acts of a person in a position of power in the society are in accord with the desires, expressed or unexpressed, of the individual (Gosnell 1969: 105).
Representation is one of the most central, fundamental, and important of political concepts. It is not, however, one of the simplest. Indeed, given that to represent or to ‘re-present’ means to make present (in some sense) that which is (in some other, perhaps more literal sense) not present, one could well say that to speak of representation is to enter the realm of the paradoxical. At the very least, the implication of some of the most outstanding scholarly studies of the concept has been to affirm its complexity (e.g. see Mainsbridge 2003; Pitkin 1967, 1969). Representation can legitimately be conceptualized in a range of different ways and studied from a number of alternative angles; practices of representation can reasonably be evaluated according to varying criteria. This chapter does not attempt to resolve, or even to discuss, many of the issues that animate much of the broader literature on representation. We set our parameters much more narrowly, focusing on one particular question among the many that one can raise about representation: the relationship between electoral
41
Electoral Institutions and Political Representation
system design and the process of representation. In what ways does variation in electoral arrangements appear to shape the links between those who are charged with being representatives, and those thereby said to be represented? In tackling this question, we are fortunate to be able to draw on a body of research on electoral systems that has developed substantially in recent times. Twenty years ago one might justifiably have described research on electoral systems as a marginal and marginalized field of work, populated largely by the academic equivalent of ‘train spotters’ or of quirky electoral reformers. Such a description is emphatically not justified today. The ‘maturation’ of the field of electoral system research has encompassed both considerable developments in researchers’ knowledge of electoral systems (their types and their consequences), and a greatly enhanced appreciation in the broader political science community that electoral systems are a key element of institutional design within a political system (Shugart 2005). Research has shown that the apparently mechanical process by which votes are translated into seats has clear and far-reaching political consequences. Even seemingly technical, mundane differences in electoral systems can matter for the make-up of parliaments, the formation of governments, the nature and range of parties in the political system, levels of governmental stability, and somewhat more indirectly, the public policy process. Our book extends this body of research further. Specifically, we examine, and seek to make an important contribution to our knowledge about, how electoral systems shape the nature and style of parliamentary representation. This chapter begins the task by setting out what is known, and what remains unknown, in this area. We start in Section 3.1 at the most general level, by providing a brief overview of the main strands of electoral system research. As we show, the principal foci of work in this field have traditionally been around how electoral systems can be understood and classified, and exploring their implications for proportionality. In Section 3.2, however, we go on to explore the more recent body of research that has shown how electoral systems can affect
42
Electoral Institutions and Political Representation
political ‘representativeness’ in ways that go beyond the preoccupation with partisan proportionality. As we discuss, there is strong evidence that electoral systems have an impact on ‘descriptive’ representation: that is, on the types of individuals elected to parliament. Section 3.3 moves on from this discussion to then explore the extent to which previous work has identified any consequences for electoral systems in terms of styles of representation: that is, the behaviour of representatives. We first outline the theoretical arguments made as to why electoral arrangements should so influence the behaviour of those elected, and then proceed to consider the empirical evidence adduced to support or contradict these arguments. The empirical evidence here on electoral system effects is, we suggest, rather more limited and fragmentary than in other areas of research. We conclude the chapter, in Section 3.4, by considering how our study can address some of the limitations in previous work.
3.1. The Study of Electoral Systems and their Consequences A major feature of electoral system design is its variability. No two countries have electoral systems that are exactly identical. The variations that exist range from the seemingly trivial (although even that can be open to interpretation) to some that are more obviously fundamental. The ubiquity and range of variability can lead to debates and disagreements over how to classify the different electoral systems that exist, and—relatedly—which characteristics of electoral systems should feature in such classifications (e.g. Blais and Massicotte 1996; Lakeman 1974; Reynolds, Reilly and Ellis 2003). Nonetheless, virtually all analysts (for exceptions, see Lundell 2005; Norris 2004) concur that proportionality is the most basic characteristic by which to classify systems, with extant systems thus falling into two main groups, the proportional and non-proportional—although, inevitably there are some that fall somewhat in-between.
43
Electoral Institutions and Political Representation Table 3.1. The five main families of electoral systems Electoral systems
Proportion of countries using the system
Average disproportionality (GI)
Non-proportional systems 1. Single member plurality 2a. Alternative vote 2b. Run-off
18.6 1.7 3.4
12.3 10.3 21.4
Proportional systems 3. List 4. Single transferable vote
47.5 1.7
3.9 5.4
Hybrid systems 5a. Mixed-member proportional 5b. Mixed-member majoritarian
6.8 20.3
3.8 9.9
Notes: Only including countries with a population greater than two million (1998 estimates), and with a Freedom House score of 4.0 or less (1998–9 rankings). GI = Gallagher (least squares) index of disproportionality. Source: Farrell (2001).
Within this elementary dichotomy, electoral systems can broadly be grouped into five families, as shown in Table 3.1. The proportional group comprise electoral systems that make some effort to match the share of the seats a party wins in parliament with the share of the votes it wins from the electorate. The two long-standing types of system fitting this description are the list electoral systems that predominate across much of Western Europe, and the STV system used in Ireland and Malta. The two main types of non-proportional system are the SMP systems, used notably in the UK, the USA, Canada, and India; and majoritarian systems, which in turn come in two main variants— run-off systems, that are most closely associated with legislative elections in France but that are also commonly used in many presidential elections, and the alternative vote (AV) system used in Australia. The fifth family of electoral systems, mixed-member systems, are a hybrid. One variant (mixed-member proportional) is clearly a member of the proportional camp; the other (mixedmember majoritarian) is just as clearly a non-proportional system. For long this type of system was virtually unique to the post-war Federal Republic of Germany, but from the early 1990s onwards
44
Electoral Institutions and Political Representation
mixed-member systems have become increasingly popular. They are now among the most commonly used type of electoral system, and are certainly favourably regarded by most electoral system experts (Bowler and Farrell 2006; Bowler, Farrell, and Pettitt 2005; Shugart and Wattenberg 2001). Table 3.1 displays information about both the number of countries that use each type of system for their national legislative elections and the level of proportionality associated with each type of system. The classifications of electoral systems into their groups and families are a direct consequence of two main features of electoral system design. The first of these is the electoral formula (EF): literally, the mechanism used to determine when a candidate (or party) has won a seat. The most simple EF is that used under SMP: the formula here is a plurality, whereby a seat is awarded to the candidate who receives more votes than any other candidate or, to use the (somewhat misleading) British vernacular, is ‘first past the post’. As implied by their title, majoritarian systems require a candidate to win a majority of the vote (50% + 1): this can be achieved via a ‘run-off’ second ballot between the leading candidates in the first, or, under AV, by reallocating votes from defeated candidates according to voters’ second (or third, or even lower) preferences. Electoral formulas become more complex with PR systems, and encompass the various divisors used in the different highest average-list systems (including d’Hondt, Sainte-Laguë, and modified Sainte-Laguë), and the quotas used in both largest-remainder list and STV systems (including Hare and Droop). Mixed-member systems, unsurprisingly, use a different formula for the two types of member elected: most typically this is SMP for constituency seats and one of the various list formulas for seats allocated via the list (for more details on EFs, see Farrell 2001). The second central element of electoral system design is district magnitude (M). This is simply the size of the district, or constituency, in which the candidate is elected, measured in terms of the number of seats. Non-proportional systems are frequently characterized by single seat constituencies where M = 1 (indeed, for SMP this is true by definition); nonetheless, there are
45
Electoral Institutions and Political Representation
exceptions, such as the multi-seat Block electoral systems used for some local elections in the UK. In proportional systems, district magnitude is always greater than one. The extreme cases are Israel and the Netherlands, where for national parliamentary elections the country is not divided into districts at all, and M is therefore equal to the size of the parliament.1 Numerous studies have demonstrated that the EF and district magnitude characteristics of an electoral system have important consequences (the classic study is Rae 1967; for recent overviews, see Katz 1997; Lijphart 1994). Until quite recently the particular consequence that the majority of electoral systems research focused on was that of systemic proportionality: that is, the relationship between seats and votes, and its wider implications for the number of parties that secure seats in parliament. This body of research includes the substantial amount of work conducted from the 1950s onwards regarding ‘Duverger’s law’: the contention that the SMP system tends overwhelmingly to produce two-party systems within a country, and the related contention that proportional systems typically generate multiparty electoral politics (for a sample, see Cox 1997; Grofman and Lijphart 1984). Other major works have identified the role played by both EF and M in influencing the level of proportionality that elections produce (e.g. Farrell 2001; Katz 1997; Lijphart 1994). This research strongly indicates that the importance of electoral formulae goes much further than simply defining whether the particular system will be broadly proportional or not.2 Among the different ‘proportional’ formulas there are also significant differences in terms of the levels of systemic proportionality they produce, with SainteLaguë generally viewed as the most proportional of the available formulas and d’Hondt as the least (see Katz 1997; Lijphart 1994a). District magnitude has been shown to have an even greater effect than EF in shaping proportionality, following a general rule (at least for proportional systems) that there is a positive relationship between district magnitude and proportionality: higher levels of M producing more proportional election results (e.g. Farrell 2001; Katz 1997; Lijphart 1994).
46
Electoral Institutions and Political Representation
In sum, the cumulative body of research on electoral systems and proportionality is now very well developed and documented; sufficiently so, indeed, that scholars can make fairly strong and reliable predictions about the results that particular configurations of an electoral system will produce. Given any normative position on whether strict proportionality is a ‘good’ or ‘bad’ feature of democracy, it is pretty straightforward to design electoral institutions to suit that position, with appropriate district magnitudes and EFs selected accordingly.
3.2. Electoral Systems and Types of Political Representatives Electoral systems shape representation by influencing the number and proportion of seats won by various parties. But political representation is about far more than just the partisan balance of a legislature. And electoral systems affect the representative process not only via their direct impact on the partisan outcome of elections. Another important aspect of representation—to many people at least—is the type of people who constitute the body of representatives. The essence of this idea is encapsulated in John Adams’s classic reference to parliament as a ‘microcosm of society’: the notion that a representative assembly ought, as far as possible, to represent accurately not only the partisan balance of opinion within a country, but also to be a good match of the society in terms of attitudes and various characteristics. In short (and to borrow a term from the literature on surveys) it should constitute a representative sample of the population. This dimension—what Hannah Pitkin refers to as ‘descriptive representation’ (Pitkin 1967)—provides another means of evaluating electoral systems, if they can be shown to have an influence in this regard. The notion of descriptive representation is not without problems; at least four main criticisms can be outlined.3 First, there is a problem of defining the characteristics according to which
47
Electoral Institutions and Political Representation
representation is to be evaluated. The problem, basically, is where—and on what basis—one draws the line. While greater representation for women and ethnic minorities might command wide support, at what point do efforts to achieve good descriptive representation start to look ridiculous?4 Second, there is the point that ‘being’ does not necessarily equate to ‘doing’: in short, there is no guarantee that a female MP will promote the interests of female voters (e.g. Diamond 1977). Third, there is what Mansbridge (1999: 637) describes as the problem of ‘essentialism’. The notion that ‘only women can understand women’s issues’ that is often deployed in support of the importance of descriptive representation has the unfortunate logical corollary that women cannot adequately represent men, and so on. And a fourth criticism is that, carried to extremes, descriptive representation may actually perpetuate the very problems it seeks to address: that by typecasting representatives (women only representing women, blacks only representing blacks, etc.) one adds to societal divisions or at least erodes ‘the ties of unity across a nation’ (Mansbridge 1999: 639; Phillips 1995: 22 ff.). But despite these criticisms there are still good reasons why many scholars see descriptive representation as important;5 even if, as in the case of Mansbridge (1999), they approve the principle only contingently. First, in a context of distrust it enables ‘enhanced communication’ because in certain circumstances (e.g. in the case of subordinate groups in society) shared societal characteristics between representatives and constituents may make people more prepared to trust representatives (Mansbridge 1999: 641). Second, in cases where interests are not fully formed, it can help to have a representative whose characteristics match those of their constituents. This can result in what Mansbridge terms ‘introspective representation’ (1999: 646–7): where representatives act on the basis of what they believe their core voters would wish.6 Fenno provides a good instance of this in reporting the comments of one black legislator who said: ‘When I vote my conscience as a black man, I necessarily represent the black community. I don’t have any trouble knowing what the black community thinks or
48
Electoral Institutions and Political Representation
wants’ (Fenno 1978: 115). For Mansbridge, there is a crucial link between uncrystallized interests and introspective representation: On the many issues relating to gender, for example, where views are changing and policies developing in a relatively ad hoc way to meet a rapidly evolving situation, descriptive representatives are, other things equal, more likely than non-descriptive representatives to act as their descriptive constituents would like them to act (1999: 646; also Phillips 1995).7
There are, in short, some good reasons to value a high degree of descriptive representation, even if the notion is by no means simple or unproblematic. How, then, is descriptive representation facilitated? A number of factors can shape the extent to which a legislature looks like the society it exists to represent. Legal provisions can be one important factor. Laws that make it difficult or even illegal for certain groups in society to stand as candidates or to vote (such as was the case for non-whites in Apartheidera South Africa, and for African-Americans in the pre-civil rights era South) render descriptive representation of those groups next to impossible. Alternatively, legal provisions may make descriptive representation of some groups mandatory—one clear example being the seats in the New Zealand parliament that are reserved for representatives of the Maori communities. In the absence of legal provisions, social norms or societal pressure may achieve a rather similar impact. This can readily be seen in the case of female representation in parliaments. In some societies, female representation may be encouraged as merely one part of broader societal attempts to promote women’s rights and prominence in society; at the other end of the spectrum, women seeking to stand as political candidates may be subject to extreme intimidation in a manner continuous with other efforts in those societies to marginalize women.8 And descriptive representation may also vary considerably within a society across different political parties, with some promoting the candidacy from some groups by positive action or quotas, and other parties actively resisting pressures to move in this direction.9
49
Electoral Institutions and Political Representation
However, in addition to all these factors, there is also a further important one to consider, namely the electoral system. Previous research has shown that electoral systems not only affect which parties are represented, and in what proportions, but also whom is elected on behalf of these parties.10 The basic rule is that more proportional voting systems generate parliaments that are more descriptively representative of a society: as well as being more closely aligned to the partisan preferences of the people, these parliaments are more similar in terms of objective social characteristics, and also in relation to measurable political attitudes (e.g. Huber and Powell 1994; Lijphart 1999; Powell 2000). The relationship is not an inevitable one: it is certainly possible to have nonproportional electoral systems that produce a legislature scoring highly on descriptive representation, while the use of an electoral system that guarantees a high degree of partisan proportionality does not necessitate that women and minorities will be present in the parliament. Nonetheless, the relationship is an empirically strong and consistent one. In short, then, the basic argument is that the nature of the parliamentary representative (who she or he is) shapes the nature of parliamentary representation, and both in turn (as summarized in Figure 3.1(a)), are affected by the proportionality of the electoral system. Electoral systems, it should by now be abundantly clear, have numerous important political consequences. They shape directly the nature and outcomes of electoral contests, and they also have more distal effects. More proportional systems tend to promote a more consensual, accommodating style to politics (Lijphart 1999), and also to be associated with a style of policy process that seeks to meet the interests of wider groupings of voters (Powell 2000). As we have seen, more proportional systems are likely to facilitate higher levels of descriptive representation—parliaments more like the societies that they represent. In the remainder of the chapter we explore whether it is likely that electoral system effects go even further, to an influence on the behaviour of individual representatives; and, if so, how and why such effects might be manifested.
50
Electoral Institutions and Political Representation (a) Proportionality of electoral system District magnitude …in combination effect…
Types of representatives
Electoral formula
(b) Degrees of personalization Openness of ballot structure …in combination effect…
Styles of representation
Nature of district
Figure 3.1. Electoral systems and parliamentary representation
3.3. Electoral Systems and Representative Behaviour Whereas study of the proportional consequences of electoral systems tends to lead to a concentration on the district magnitude and EF characteristics of electoral systems, a shift of focus to the behavioural consequences of electoral systems requires us to pay attention to other features of electoral system design. The factors that previous scholars have drawn attention to in this regard are ballot structure and, to a somewhat lesser extent, a particular aspect of district magnitude that is distinct from the aspects of M relevant to assessing proportionality. Taken together, these factors have been hypothesized to influence the behaviour (the ‘style’) of elected representatives (Figure 3.1(b)). We deal with each of these characteristics in turn.
51
Electoral Institutions and Political Representation
From Douglas Rae (1967) onwards, scholars studying electoral systems have distinguished between categorical and ordinal ballot structures. The former allows for only a single, undifferentiated vote for either a party or candidate; the latter allows voters—to varying degrees—to influence which actual candidates get elected. This distinction is a logical way of separating different types of ballot structures, and—despite Rae’s own misgivings—has been shown to be a useful variable for explaining variation in the systemic proportionality of voting systems (Lijphart 1994a). However, the categorical–ordinal distinction is insufficiently nuanced to delineate the full range of variation in ballot structures. For instance, it lumps SMP systems together with closed-list systems: yet these two systems differ not only with regard to district magnitude, but also in terms of the apparent incentives they impose on representative behaviour (Katz 1980). Research suggests that representatives working under SMP systems are much more likely than those under closed-list systems to feel the need to nurse a personal vote (e.g. Bowler and Farrell 1993); equally, voters under SMP systems are more likely to pay attention to the work of individual MPs. The categorical–ordinal distinction also treats those systems where voters vote for closed party lists as equivalent to systems where voters can vote only for one candidate from a party list. Both of these systems involve a categorical choice, but in the latter case there is a motivation for candidates to chase personal votes in the hope of leapfrogging over those candidates higher in the list and thus secure a seat (as, for instance, an adult movie star did so colourfully in the early 1980s in Italy). In addition to the categorical–ordinal distinction, another means of differentiating among types of ballot structure is whether the voting act is candidate based or party based (Bowler and Farrell 1993). This relates to a key feature of electoral system variation, the extent to which the fate of individual candidates is determined by personal votes, or as Carey and Shugart put it (1995: 419), ‘the degree to which electoral systems reward politicians’ personal reputations’. Adding this additional dimension, one can produce
52
Electoral Institutions and Political Representation
Candidate-based
Ordinal
Categorical
…to ‘open’…
…from ‘closed’… Party-based Figure 3.2. Degrees of ‘openness’ in ballot structure design
a two-by-two typology of electoral systems based on electoral structure characteristics, which is summarized in Figure 3.2. The heavy grey arrow indicates the flow from ‘closed’ to ‘open’ electoral systems relating to the nature of the ballot structure design. As the system becomes more open, greater emphasis is placed on individual politicians, who in turn, it can be hypothesized, place greater emphasis on the representation of individual constituents and on personal vote chasing. In addition to developing this more refined understanding of ballot structure, an adequate understanding of the possible influence of electoral system design on representative behaviour must also take into account the size of the region that is being represented. District magnitude, understood in this sense, has if anything featured even more prominently in previous research
53
Electoral Institutions and Political Representation
than has ballot structure (e.g. Cox 1997; Wessels 1999), and close analysis of representative roles has found evidence of significant differences based on size of region characteristics (Bowler and Farrell 1993; Cowley and Lochore 2000; Lancaster and Patterson 1990; Lundberg 2002). The important point to stress is that the district characteristic is conceptualized here in quite a different manner to how it was referred to above in our discussion of systemic proportionality. Quite apart from the simple count of how many representatives there are in a district (the essence of ‘district magnitude’), electoral districts can vary in a range of other ways that have important implications for parliamentary representation. We would therefore argue that allowance also has to be made for a district’s geographical characteristics, such as its physical size, and population density. There is also a peculiarity of the electoral system used for electing Italian MEPs that is relevant here. While we examine this in more detail later (see Chapter 4), it is important at this stage to set out the main point of relevance. This relates to a disjuncture between the size of the district in terms of calculating the distribution of seats to parties (i.e. the district magnitude, or M) and the size of the district that the Italian MEPs are actually elected to (we refer to this as the ‘locus of representation’, or LoR). As was seen in Section 3.2, the standard measure of M is based on the level at which seats are allocated to the parties. In elections to Ireland’s national parliament, for example, seats are allocated to the parties in a given multi-member constituency based on their votes in that constituency, and therefore the average M for Ireland as a whole is based on the range across the various constituencies. By contrast, in Israel’s national elections, seat allocations are based on the parties’ national votes, and so there is just one M (= 120) for the entire country. In Italy’s EP elections, however, this picture is somewhat complicated by the fact that while seats are allocated nationally (M = 87), the national vote proportions are based on aggregate party votes from each of Italy’s five regions (producing an average LoR of 17.4); and the seats are allocated to the parties’ candidates
54
Electoral Institutions and Political Representation
based on their rankings on the regional list and (crucially for our discussion) the personal votes received by the candidates in the region. So, while it is correct to talk of M = 87 for the purposes of analysing aggregate proportionality in Italian EP elections, in this specific case it would not be correct to use M as the basis for examining the effects of electoral system design on styles of parliamentary representation. To the best of our knowledge there is no other electoral system anywhere in the world that shares this feature.11 This is because, first, there are very few cases of national-list (as opposed to regional-list) electoral systems, and, second, of those other cases none shares the additional feature of an open ballot structure. The two cases that come closest to Italy’s EP system are the Dutch national electoral system, where the parties’ lists are drawn up regionally, and the German EP electoral system, where the parties can opt to vary their lists regionally. However, in both cases closed(Germany) or ordered-(the Netherlands) list systems cancel out a potential candidate-oriented campaign, or concomitant representation styles, because personal votes are either unavailable or do not matter for prospects of electoral success.12 To date, only limited systematic research has examined the impact of electoral systems on representative behaviour. Our established knowledge of constituency representation comes mainly from research conducted in the US and British contexts, where electoral systems have been a constant, not a variable, factor. Much derives from the work of Richard Fenno (1978) on representatives’ ‘home styles’. Constituency representation, for Fenno, was not only something that representatives put a great deal of time and effort into; the manner and forms of that representation could tell scholars much about how representatives perceived their constituency and understood the representative relationship: [W]e cannot know all we need to know about House members in Washington unless we move out beyond the capitol city into the country and into its congressional districts. Washington and home are different
55
Electoral Institutions and Political Representation milieus . . . But they are not unconnected worlds. The theory and practice of a representative form of government links them one to the other. Though a congressman be immersed in one, he remains mindful of the other. (Fenno 1978: 214)
Although clearly aware of possible links between home styles and representatives’ behaviour in the chamber, Fenno placed greater emphasis on the electoral implications of constituency representation (see also Kuklinksi 1979). The ‘electoral connection’ angle on constituency representation was further developed by Cain, Ferejohn, and Fiorina (1987). An important part of the stimulus for constituency service behaviour, they suggested, was the electoral benefits accruing to representatives in terms of a ‘personal vote’. British MPs and US Congressmen, operating under a SMP system, appeared to be motivated in their constituency service activities to a high degree by strategic-electoral considerations of vote-maximization. By making oneself known in a district, and particularly known as someone who worked hard for the interests of the district, both US representatives and, it was suggested, (although to a rather lesser degree) British MPs, could enhance their electoral prospects. The implication was that under different electoral rules, different behaviour would follow from representatives, with constituency representation likely to be downgraded or to at least take different forms: ‘a polity’s electoral process, its policy processes, and the finer details of its institutional structure are bound together. If one changes the others adjust accordingly’ (Cain, Ferejohn, and Fiorina 1987: 9). Consistent with this perspective, Bowler and Farrell’s study (1993) of MEPs argued that party-list systems impose a need for those seeking re-election to orient their activities around the needs of a party leadership, and less of an incentive for non-partisan constituency service activities: ‘it is relatively easier for legislators to shirk in satisfying voter demands under some electoral systems than others’ (1993: 53). However, a cross-national collaborative research project in the 1980s reached the rather contrary conclusion that ‘electoral systems are not fundamental in determining parliamentarian/constituency
56
Electoral Institutions and Political Representation
relationships . . . electoral systems are, perhaps, rather more passive elements . . . than either supporters or opponents of electoral reform tend to believe’ (Bogdanor 1985: 299). In addition to a considerable body of literature exploring constituency representation further in the US context,13 other work conducted in the UK has reinforced the finding that constituency representation is increasingly central to the role of British MPs (e.g. Norton and Wood 1993). Some doubt has, however, been cast on the degree to which this behaviour is motivated primarily by vote-winning considerations (Searing 1994), leading some scholars to term the growing constituency-related activity of most UK MPs ‘the puzzle of constituency service’ (Norris 1997). Indeed, the puzzle grows given the limited awareness of most MPs by their constituents (Crewe 1985; Weir and Beetham 1999: 70), while the work of Gaines (1998) has produced little support for the idea that growing levels of constituency service activity by MPs has led to an increased ‘incumbency advantage’. Other work conducted in Ireland, however, has found high levels of constituency activity by representatives, and a possible link to the candidate-centred electoral system operating there (Farrell 1985; Wood and Young 1997; though see Gallagher 1987),14 while research in Australia, Canada and Germany has also found some suggestive evidence of the personal vote phenomenon (Bean 1990; Ferejohn and Gaines 1991; Lancaster and Patterson 1990).15 Finally, a very limited body of work has examined constituency representation in the context of the EP. Bowler and Farrell (1993) used evidence from a mail survey of MEPs that indicated a strong empirical relationship between the electoral systems used in EU member states and aspects of MEPs’ constituency service behaviour. Where voting is linked to individual candidates rather than parties, such as in Britain with SMP (at the time of that survey) and Ireland with STV, representatives placed greater emphasis on developing contacts with individual voters via constituency casework, personal appearances, or maintaining a constituency office. Those elected under PR list concentrated more on building links with their party machinery and with large-scale organized interests.
57
Electoral Institutions and Political Representation
However, this work was limited in that only a small range of behaviours was examined, not much attention was paid to individuallevel variance, and (as acknowledged by the authors) they had little ability to control for factors like political culture that might offer alternative explanations of apparent electoral systems’ effects. Indeed, more recent evidence from the 1994 European Elections Study has led Katz to support the view that cultural factors are rather more important than strategic-electoral considerations promoted directly by an electoral system. Katz therefore concludes that the stronger constituency emphasis among British MEPs compared to MEPs from other member states has little to do with electoral systems effects, but rather ‘is suggestive of a cultural effect’ (1997a: 218; also Katz 1999). He therefore warns against predicting that electoral reform will generate a mechanistic response from representatives: ‘ “political culture” plays an important part in determining the political consequences of electoral systems; the same institutions may be associated with quite different outcomes if the actors pursue a different mix of objectives’ (1999a: 16). To summarize, we now have considerable knowledge about constituency service activity by representatives. However, scholars remain some way from achieving consensus on the factors that shape legislators’ understanding of their representative role and that promote greater levels and different forms of constituency service effort from elected representatives. At least three alternative sources of variance are clearly identifiable in extant work in this area:
r Individual-level: individual differences in interpretation of the role of representative (e.g. Searing 1994), or in actual or perceived electoral vulnerability (Fenno 1978),16 are one likely source of variation in the quantity and form of constituency representation.
r Electoral System: Broad differences following the strategic-
electoral incentives created by an electoral system are a second
58
Electoral Institutions and Political Representation
likely source of variance identified (Bowler and Farrell 1993; Cain, Ferejohn, and Fiorina 1987).
r Cultural: General cultural differences across country or region in the expectations and demands placed on elected representatives are a third source of variance identified (Katz 1997, 1999).
Unfortunately, while previous work has been able to identify these different factors as plausible explanations of variation in constituency service activity, there has been little progress made in understanding when, and to what extent, each is important. This failure on the part of previous work has largely not been a function of sloppy or incomplete analysis of available data. Rather, the problem has usually been the more fundamental (and intractable) one of limitations in research design. For instance, while research in a single political system (e.g. Fenno 1978; Norris 1997; Searing 1994) and even some comparative analysis (notably Cain, Ferejohn, and Fiorina 1987) has been able to examine individual sources of variance in behaviour—such as electoral vulnerability, role choices, and so on—such work, however, is generally unable to explore strategic-electoral or cultural influences, because such factors are largely, if not wholly, constant. Needless to say, crossnational work does open up the possibility of being able to explore cultural and electoral systems variables. However, several factors have often been at play in limiting the conclusions drawn from cross-national comparisons: (a) cross-national work that compared members of different political institutions might be ignoring a further important source of variance—behavioural imperatives in some way imposed on representatives by the institution itself;17 (b) much cross-national work (Bowler and Farrell 1993; Katz 1997a), in concentrating on the broader comparative perspective, has ended up paying only limited attention to individual differences, and (c) a fundamental problem has often existed in attempting to disentangle electoral system effects from cultural ones. Often, the two may appear largely coterminous in particular cases, particularly as in the long run an electoral system plausibly shapes the broader political culture of a country.
59
Electoral Institutions and Political Representation
Thus, despite the accumulation of work in this area, some key questions remain essentially unanswered. Is the constituency service phenomenon inherently bound up in a broader political culture that defines the representative relationship in particular ways, or is such behaviour by politicians primarily driven by individual strategic considerations of maximizing their own electoral prospects by seeking a personal vote? How, then, might the behaviour of representatives vary across different electoral systems and political cultures? And how does their broader interpretation of the representative role differ in response to such factors?
3.4. Conclusions and Implications It should be clear from the above discussion that scholars have developed substantial knowledge about electoral systems and their effects. It should also be clear from the discussion in the final section of the chapter that we now have considerable knowledge about constituency service activity by representatives. However, we remain some way from achieving consensus on the factors that shape legislators’ understanding of their representative role and their behaviour, and the extent to which the electoral system is one of the most important of those factors. Many of the remaining chapters in this study are devoted to exploring this question. Chapter 4 begins this task, in the context of representation in the EP, by documenting the electoral systems used for EP elections, and some of the immediate consequences that these systems would appear to have.
Notes 1. We should mention, at least in passing, a third characteristic of electoral systems that influences systemic proportionality, namely assembly size (AS), or the number of MPs in the parliament. As first pointed out by Lijphart (1994a), larger assemblies are associated
60
Electoral Institutions and Political Representation
2.
3. 4.
5. 6. 7.
8.
with greater proportionality. Despite its clear and obvious ramifications for the wider effects of electoral systems (e.g. there is not much that a ‘large’ district magnitude can achieve in raising overall proportionality in a parliament of few members), strictly speaking this is less a feature of electoral system design per se than of the size of a nation’s population—smaller countries tend to have smaller parliaments. Indeed, according to Katz (1997), when looking at proportional systems on their own, EF tends to have a greater effect on systemic proportionality than does M. This discussion draws heavily on Mansbridge’s masterful overview (1999). Charles Pennock puts this question rather more colourfully in asking whether morons should be represented by morons (1979: 314; also Grofman 1982). For a notable recent example, see Phillips (1995). This is also encapsulated by Goodin’s notion (2003) of ‘democratic deliberation within’. Mansbridge (1999: 628) refers to two other merits of descriptive representation, that, while not adding significantly to the substance of representation, are nevertheless of some significance: (a) creating a sense among groups that have been long disadvantaged that they have an ‘ability to rule’, and (b) increasing a polity’s legitimacy in a context in which there had previously been discrimination. In subsequent work, Mansbridge further develops the notion of introspective representation by reference to what she calls ‘gyroscopic representation’, in which the representatives rotate ‘on their own axes, maintaining a certain direction, pursing certain built-in (although not fully immutable) goals’ (2003: 520). These representatives are not accountable in the sense (as we explored in Section 3.3) of seeking to maintain close and explicit relations with constituents, rather they operate for their own ‘internal reasons’. Mansbridge suggests that this form of representation is likely to be common to party-list systems, in which representatives, she argues, pay more attention to party interests than to the interests of individual constituents (also Esaiasson 1999). An obvious example of women’s candidature being subject to a hostility that reflects broader social problems would be the cases of those
61
Electoral Institutions and Political Representation
9.
10.
11.
12.
13. 14. 15.
16.
17.
62
women seeking to stand as candidates, or otherwise be involved in political debates, in the elections held in Afghanistan in September 2004: for a discussion, see Human Rights Watch (2004, 2005). A good illustration of this is the use of women quotas by the British Labour Party in the 1990s (see Norris 1994). More generally, see Lijphart (1986). It is worth noting in passing that the issue of whom is elected can also involve questions over whether the ‘correct’ candidate (the Condorcet winner) is actually the one elected (e.g. Saari 2001). It is interesting to see that this Italian variant (originally introduced for the 1979 direct elections to placate those calling for an element of regional representation; for details, see Cotta 1980), attracted considerable interest among other member state MEPs. A subcommittee of the EP’s Political Committee, established in the wake of the first direct election with the remit of producing proposals for UEPs, offered this system as one means of marrying the twin desires of proportionality and local representation. In addition, in the Dutch case, the parties tend not to vary their lists across the nineteen different electoral districts; by default, therefore, they produce national lists (see Andeweg 2005: 493–7). For a useful overview, see Smith (2003). See also Chan (1988) on Korea and Hazan (1999) on Israel. Bean’s evidence in favour of a personal vote in Australia is somewhat contradicted by the findings of Studlar and McAllister (1996) who find that local constituency service activity is actually negatively correlated with voting support for members of the House of Representatives. Fenno observes of members of Congress that ‘their perception of a reelection constituency is fraught with uncertainty . . . House members see electoral uncertainty where outsiders would fail to unearth a single objective indicator of it’ (1978: 10–11. Emphasis added). As Bowler and Farrell observe, ‘while legislators face electorallyimposed incentives, it is also the case that they face incentives imposed by the chambers in which they work . . . it is all too easy— especially when comparing across different nations—to forget factors which affect the behaviour of parliamentarians that are more related to the legislature in which they work’ (1993: 48–9).
4 Representation and Electoral Systems in Europe
In June 2002, the EP approved a Council of Ministers amendment to the 1976 legislation governing EP elections. Agreement on this amendment was a historic step. After a quarter of a century of failed initiatives, the EU had finally achieved legislative agreement on UEPs for its elected institution. The timing of the legislation was particularly propitious. Given the imminence of EU enlargement, new member states would now have no choice but to adopt electoral systems for choosing their MEPs that met the new regulations. In this chapter, we do three things. First, we review the historical debates surrounding the question of electing the EP, debates that started barely before the ink was dry on the Rome Treaties, and that continued down a slow and, at times, tortuous path before their culmination in the 2002 legislation. As we see, one thing that was apparent from the outset was the intention that EP elections should be conducted via PR. Second, we provide a comprehensive account of the rules under which the EP was elected in 2004 after the passage of the UEP legislation. We explain the main features of the legislation, and then explore the electoral systems used by all twenty-five member states in the most recent elections. We show that meeting the ‘uniformity’ requirements was still consistent with a substantial degree of variety, in several respects, in electoral
63
Representation and Electoral Systems in Europe
arrangements adopted in both pre- and post-enlargement member states. Third, we then go on to address an important question raised by the latter discussion: if the electoral systems now used for EP elections are some way from being identical in form, do they also differ significantly in their effects? To answer this question we examine evidence concerning the proportionality and ‘representativeness’ of the electoral outcomes produced by the different systems in 2004. Finally, in the conclusion we consider the implications of the EP’s current electoral arrangements for political representation in the EU.
4.1. The Road to Uniformity From the outset, in the Treaty of Paris (1951), it was always envisaged that the EP (the ‘Assembly’ as it was originally called) should be directly elected.1 It is in the Treaty of Rome (1957) that we see the first actual reference to UEPs. It is worth quoting the relevant article in full: The Assembly shall draw up proposals for elections by direct universal suffrage in accordance with a uniform procedure in all Member States. The Council shall, acting unanimously, lay down the appropriate provisions, which it shall recommend to Member States for adoption in accordance with their respective constitutional requirements. (Treaty of Rome, 1957, a.138(3))
There are two significant points here. The first is that the responsibility was given to the Assembly to draw up the proposals for direct elections according to ‘a uniform procedure’; the second that the Council would have to agree these proposals unanimously. The former was always going to be easier to achieve than the latter. Despite repeated efforts by the EP to develop viable proposals, it would take the best part of half a century before member states would sign up to them. The development of events can be grouped into two
64
Representation and Electoral Systems in Europe
chronological stages: pre- and post- the first direct elections to the EP in 1979.
4.1.1. 1958–79 The EP2 took its responsibilities under a.138 very seriously. Once its members had been appointed (as they then were), one of the first acts of the new chamber in 1958 was to establish a working party chaired by Ferdinand Dehousse to consider the question of direct elections. Dehousse’s working party consulted the six member state governments and produced a report that was adopted by the Political Affairs Committee in the form of a draft convention in 1960. This dealt with a range of matters relating to the EP and its election, including the numbers of MEPs to be elected, that elections should occur on the same day in all members states, and that the elections might be phased in gradually (with some MEPs continuing to be appointed in the interim). Significantly, the report did not make specific recommendations regarding UEPs. Dehousse was a pragmatist: he was on record in expressing the view that ‘uniformité ne veut pas dire identité’ (quoted in van den Berghe 1981: 11). Rather than attempting to force an agenda for the adoption of UEPs, his working party suggested instead that, in the first instance, member states should determine their own electoral systems. In other words, Dehousse accorded priority to sorting out of direct elections ahead of dealing with the thornier issue of UEPs.3 Despite the EP’s best efforts (and following what was to become a regular pattern), member state governments showed little appetite for addressing this issue. This was principally because of French President de Gaulle’s well-known objections to any moves that might further encroach on national sovereignty. Over the following years the EP, and some of the other member state governments, attempted in vain to shift the General from his perch of intransigence; and before very long, the entire issue was swallowed up in the mid-decade ‘empty chair’ crisis.
65
Representation and Electoral Systems in Europe
As the decade drew to a close, in 1969 an exasperated EP threatened to take the Council of Ministers to the Court of Justice for non-compliance with its obligations under a.138. This prompted the member states to take action, and over the subsequent three years representatives of the Parliament and Council held a series of meetings that attempted to come to some accommodation on the issue. However, this was to no avail. As 1972 drew to a close, and after the Paris Summit of that year had yet again failed to resolve the problem, the EP expressed its regret ‘that no definite dates have been laid down for the general and direct election of MEPs and no instructions given to solve the remaining difficulties’.4 Ironically, soon after the EP made this statement, the omens started to look better. The change of presidency in France, first to Pompidou and then to the more communitaire Giscard d’Estaing, produced a fresh impetus for institutional and policy reform in the Community, including the granting of increased budgetary powers to the EP; this in turn prompted among many the feeling that the Parliament needed greater democratic legitimacy in order to deploy these powers. Furthermore, the first enlargement saw the arrival of two new member states—Denmark and the UK—with more forthright views about the need to address the democratic deficit in the Community. The EP decided to start with a clean sheet and so the Political Affairs Committee gave the task to Schelto Patijn to draw up fresh proposals. Patjin’s 1974 report, in large part, mirrored that of Dehousse a decade earlier, the principal point of distinction being that he now set out a formal two-stage process in which direct elections (Patijn suggested in May 1978) would first occur under electoral systems to be designed by the member states; only then would a newly elected EP draft proposals for UEPs be brought forward (Patjin proposed that this be done by 1980). After some prevarication—caused principally by differences of opinion over how many MEPs each member state should have—the Council ultimately endorsed these proposals in 1976, thus paving the way for the first direct elections to the EP. After a last minute hitch due
66
Representation and Electoral Systems in Europe
to delays in passing enabling legislation in some member states, the first direct elections were held in June 1979. The lack of agreement about UEPs for EP direct elections threw the process of electoral system design back to the member states, and throughout the middle part of the 1970s a flurry of debates ensued in national capitals. Three of these are especially noteworthy: the German and Italian debates because individually they (especially the latter) resulted in important new features in electoral system design which had some bearing on subsequent debates over UEPs; and the British debate, which showed a surprising willingness on the part of senior British politicians at that time to come to some accommodation over the thorny issue of proportionality for EP elections, an initiative which, had it succeeded, would undoubtedly have accelerated subsequent efforts to move to UEPs. Germany and Italy are interesting because they provide an example of how electoral system design can square the supposed circle of maximizing proportionality (which generally requires a large M) while at the same time also maximizing a strong constituency orientation on the part of the elected politicians (which requires a small M). The Italian EP electoral system was born out of a compromise between the needs of smaller parties, which required maximum proportionality—and hence the biggest districts possible—if they were to win any seats, and the desires of prominent elements in other parties [notably within the Christian Democrats (DC)] to preserve the Italian tradition of preferential voting, which of itself necessitates smaller districts. This was the era of the compromesso storico, when a grand coalition was formed, led by the DC, that relied on the support of a series of smaller parties and, critically, the Communists. A spirit of compromise was in the air, and nowhere was this more apparent than in the legislative shenanigans surrounding the design of this electoral system (a process that was so long-winded that, at one point, there were doubts over whether Italy would be ready in time for the 1979 direct elections; see Cotta 1980). The deal that was finally struck was that the allocation of seats among the parties would be determined nationally
67
Representation and Electoral Systems in Europe
(i.e. treating the entire country as one electoral district), while the allocation of seats within the parties (between the competing candidates) would be determined in five separate regions, this latter feature enabling preferential voting. Germany, similarly, faced a debate over electoral system design between those parties that wanted a single national list, and those favouring regional lists. But in this case, the cause of the disagreement was not over preferential voting; rather it related to the national ambitions of a prominent regional party leader—the CSU’s Franz Josef Strauss. Relations between the CDU and CSU at that time were quite strained, and the CDU was especially exercised over the possibility that its Bavarian junior coalition partner might ‘go federal’. Adopting a national-list system for German EP elections was fraught with danger for the CDU, in that it might impel a first CSU move in this direction. Running in the contrary direction, however, were the views of the SPD, which wanted an electoral system that would not ‘accentuate regional and land issues in a supranational campaign’ (Lodge 1980: 88), and those of the FDP, which needed a national list system if it was to have a realistic chance to win any EP seats. The elegant compromise that was struck was for parties themselves to decide whether to run national or linked-regional lists; the practice ever since has been for the CDU/CSU to follow the latter course. An often forgotten feature of the debates over EP electoral system design is how close the UK actually came to adopting a PR electoral system at the outset. The 1970s was a tumultuous period in British electoral politics. Two general elections in quick succession, combined with (indeed, caused by) growing electoral support for the (then) Liberal Party and the rise of nationalist parties in Scotland and Wales (in turn sparking a heated devolution debate), placed electoral reform for the House of Commons high on the political agenda (Bogdanor 1981).5 In such a context, there was bound to be greater willingness to consider alternatives to SMP for the EP as well. Furthermore, the Labour government that emerged as a result of the second 1974 election had only a bare parliamentary majority—one which, with the inevitable process of by-election
68
Representation and Electoral Systems in Europe
attrition, soon disappeared. Within a few years Britain was forced down the road of its own compromesso storico in the form of the ‘Lib-Lab pact’, formalized in 1977. At the heart of the pact was an agreement by the Labour government to allow the Liberals input into the deliberations over electoral system design for the UK’s EP elections. The result was a proposal for a regional-list PR system for EP elections. But this proposal ran into a wall of internal divisions within the Labour Party and steadfast opposition from the Conservatives. The Labour Party at that time was riven with deep factional disputes over European integration (just as heated as the divisions within the Conservative Party some twenty years later), and to prevent embarrassing defeats on the matter, the party leadership allowed a free vote in the House of Commons on EP electoral system design. Meanwhile, the Conservatives, having disposed of the communitaire Edward Heath in favour of Margaret Thatcher, were far less supportive of any steps that might be construed as deepening integration, and were also implacably opposed to any moves away from the SMP electoral system. In consequence, the proposed regional list system—described in derisory terms by one senior Conservative as having being ‘discovered by a Liberal professor in the Forests of Finland’ (quoted in Hagger 1980: 226)—was defeated in the House of Commons by 321 votes to 225. It was decided, instead, that British MEPs would be elected by SMP, while those from Northern Ireland would be elected by STV. The Commons vote left Britain out of step with her Community partners, all of whom had selected some form of PR system for electing their MEPs. While it clearly helped that in virtually all the other cases (the exception was France), PR systems were already de rigueur, adaptations were nevertheless made to existing national electoral practices by many states, and none more so than Germany and France (see Table 4.1). By sticking with a nonproportional SMP system, Britain was the clear outlier in the 1970s, thus blocking efforts to adopt UEPs for electing the EP. As we see in Section 4.1.2, the irony is that a quarter of a century later (and solely for domestic reasons) it was Britain that would ultimately
69
Representation and Electoral Systems in Europe Table 4.1. Electoral systems for national and European Parliament elections in 1979 Member state
National electoral system
EP electoral system
Degree of difference
Belgium Denmark France Germany Ireland Italy Luxembourg Netherlands UK
Regional ordered list Two-tier open list Run-off Mixed-member proportional Single transferable vote Two-tier open list Regional open list Two-tier ordered list Single-member plurality
Regional ordered list National open list National closed list Nationala closed list Single transferable vote Nationalb open list National open list National ordered list Single-member pluralityc
None Minor Major Major None Minor Minor Minor None
Notes: For explanations of electoral systems, see Farrell (2001). Parties can opt to have linked regional lists (see text for details). b Regionally anchored national lists (see text for details). c STV for electing Northern Ireland’s MEPs.
a
break the impasse by unilaterally reforming its EP electoral system, thereby paving the way for UEPs.
4.1.2. Post-1979 With the first direct elections behind it, the EP wasted no time in flexing its legislative and political muscles, developing an agenda based around increasing its power and influence within the Community framework (Rittberger 2005). It sought also to keep UEPs on the political agenda, and to that end, within three months of the elections, the Political Affairs Committee established a subcommittee tasked with bringing forward proposals. Jean Seitlinger was appointed as rapporteur, and after some deliberation his subcommittee produced two alternative options for UEPs: a mixedmember proportional system (MMP, based on the German national electoral system), and a variant of the electoral system used for electing Italian MEPs (as described above). What both options shared in common was the feature of combining national lists (i.e. maximum proportionality) with constituency/regional representation (i.e. scope for voter linkage). After some discussion,
70
Representation and Electoral Systems in Europe
the MMP system was ruled inappropriate for EP elections due to complications surrounding the fact that it can produce an anomalous scenario in which more politicians are elected than there are seats to fill (what in Germany is known as the Überhangmandat),6 something which cannot be catered for in the EU, where the national allocation of EP seats is the result of highlevel intergovernmental bargaining. Based on the recommendations of Seitlinger’s subcommittee in early 1982, the EP proposed the adoption of UEPs based on a variant of the Italian EP electoral system.7 The by-now-familiar cycle then followed in which, after detailed discussion and negotiation by a Council working party, no agreement was forthcoming, and in May 1983 the Council formally ‘parked’ the issue until after the 1984 EP elections. Following these elections, the Political Affairs Committee tasked R. Bocklet with bringing forward fresh proposals. Bearing the hallmarks of Dehousse’s pragmatic approach, Bocklet’s 1985 report to the Political Affairs Committee reintroduced the idea of a stagedapproach to UEPs. But the report never made it to the floor of the Parliament. In a clear difference of opinion, the Legal Affairs Committee took the position that electoral system uniformity should mean just that: ‘as a matter of law, the requirement for a “uniform electoral procedure” means not merely that principles, objectives or results of the electoral systems shall be uniform, but that the actual procedures by which those principles, objectives, and results are put into effect shall be uniform’ (quoted in Grunert 1990: 497). Given the differing views of these two committees a joint working party was established to find a consensus position designed to maximize support in the EP. It reported in mid-1986, proposing once again a variant of the Italian EP electoral system (i.e. mixing national seat allocation with regional representation), with much stress on the proposal that the lists would be ‘open’— this latter feature designed to bring British MEPs on board on the grounds that it would facilitate a constituency orientation among MEPs. But despite their best efforts, the working party’s proposal attracted derisory support among the wider EP membership and
71
Representation and Electoral Systems in Europe
the EP was unable to come to any agreement on the matter in advance of the 1989 elections. In the wake of the 1989 elections, yet another attempt was made—this time under the stewardship of the Belgian Liberal, Karel de Gucht. On this occasion, the report was drawn up under the auspices of the Institutional Affairs Committee. The cyclical pattern of moving between Dehouuse-like pragmatic generalities on the one hand, and specific proposals for a uniform system on the other continued, with on this occasion de Gucht, at least initially, preferring the former—his 1991 report (following the then prominent principle of ‘subsidiarity’) proposing that member states should continue to have considerable leeway in designing the procedures for electing their MEPs. However, in the light of pressure from British Liberals—who had unsuccessfully challenged the constitutionality of Britain’s SMP system for EP election—de Gucht’s working party produced a subsequent report in 1993 calling for the adoption of a PR system for EP elections. The member states would be free to opt for national or regional lists, and— as a sop to the British—they would also be free to opt for list or mixed systems, on the proviso that, in the latter case, at least onethird of the seats would be list seats. The proposal ran into the usual blockage of dithering in the Council and, in consequence, the 1994 elections were fought once again using existing national systems. The subsequent 1994–9 period was influenced by two main events: a significant Treaty change, and the election of a new government in the UK committed (in its early years at least) to an ambitious package of institutional and constitutional reforms. The first event was an important Treaty change, under Amsterdam, which qualified the issue of electoral system uniformity as stated in a.138(3). Now, all that was required was that the EP electoral system should be ‘in accordance with principles common to all member states’. This lower common denominator avoided the need for UEP legislation to be overly prescriptive in pushing for complete uniformity. This change, combined with the 1997 election of a British Labour government committed to
72
Representation and Electoral Systems in Europe
large scale institutional reform—a commitment which resulted in the shift from SMP to a regional-list system for electing Britain’s MEPs8 —opened the way for yet another attempt by the EP to get UEP proposals through the Council. The Institutional Affairs Committee established another working party, led by Giorgios Anastassopoulos, and his 1998 report proposed the adoption of a list system for electing MEPs.9 For the most part, the precise nature of the list system would be left to individual member states to determine; although, somewhat controversially, there was a specific proposal for regional representation in member states with more than 20 million inhabitants. One other feature was new to this proposal, namely that there should be a 10 per cent Euro-wide top-up. In other words, the Anastassopoulos report was proposing a mixed-member electoral system. In the light of spirited objections, that specific proposal was withdrawn—with an agreement to reconsider it on a later occasion.10 The Anastassopoulos report was considered by the Council, which passed a number of (non-controversial) amendments and finally reached agreement (after the 1999 elections) in 2002, sending it back to the EP for it to give its assent.11 The Constitutional Affairs Committee gave the task to José Maria Gil-Robles Gil-Delgado to consider the Council’s amended proposals. He recommended that the EP give its formal assent, thereby passing the legislation into law.
4.2. The 2002 Legislation and the Notion of Electoral System Uniformity The main provisions of the 2002 legislation can be summarized quite clearly and simply:12
r That EP elections shall be held under a PR electoral system, using either a list system or the STV;
r That there is scope for member states to adopt some form of preferential voting, but this is not a requirement;
73
Representation and Electoral Systems in Europe
r That member states shall be free to establish regional lists, providing that these do not affect the proportional nature of the voting system;
r That a legal minimum threshold for representation of parties may be set, though it should not exceed 5 per cent;
r That subject to the provisions of the Act, there is scope for national legislation to take account of the specific situation of a member state, but this cannot affect the proportional nature of the electoral system. The legislation on UEPs, coming after many years of failed attempts, was undoubtedly significant in embedding a common set of principles by which all representatives would be elected to the EU’s democratic chamber. Nonetheless, the fact remains that the legislation encompasses considerable scope for member states to operate electoral systems that are far from identical. This scope was to a substantial extent utilized in June 2004 (Table 4.2). Examining the electoral arrangements deployed in the 25 member states (the table comprises 26 cases, because of the internal differentiation in the UK between mainland Britain and Northern Ireland), we can detail differences across the systems according to three main features of electoral systems: the EF, district magnitude, and ballot structure. In terms of EF, two levels of difference can be observed. There is the obvious distinction between STV and list systems. But within the list systems we see a mixture of highest average (d’Hondt, Sainte-Laguë, modified Sainte-Laguë), and largest remainder (Hare, Droop) formulas. It is well established in the voluminous electoral systems literature (see Summary in Farrell 2001: 156) that the choice of EF can have important implications for the overall proportionality of the electoral system; in this regard, it is interesting to note how a plurality of member states (twelve in all) have opted for the least proportional formula available, namely d’Hondt. As far as district magnitude is concerned, variation is less than in the past: since Britain’s shift to list PR in 1999, no state operates a single-member constituency-based system. The most pertinent
74
Table 4.2. The EP’s electoral systems, 2004 Country
Austria Belgium Britain Cyprus Czech Rep. Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands N. Ireland Poland Portugal Slovak Rep. Slovenia Spain Sweden
No. of Electoral formula MEPs 18 24 75 6 24 14 6 14 78 99 24 24 13 78 9 13 6 5 27 3 54 24 14 7 54 19
d’Hondt d’Hondt d’Hondt Hare d’Hondt d’Hondt d’Hondt d’Hondt Hare/d’Hondt Hare-Niemeyer Largest remainder-Droop d’Hondt STV-Droop Hare St. Laguë Hare d’Hondt STV-Droop Hare/d’Hondt STV-Droop d’Hondth d’Hondt Largest remainder-Droop d’Hondt d’Hondt Modified St. Laguë
Ballot structure
Orderedb ; single vote Ordered; multi-votec Closed; single vote Ordered; multi-vote Ordered; multi-vote Open; single vote Open; single vote Open; single vote Closed; single votec Closed; single vote Closed; single vote Closed; single vote Open; multi-vote Open; multi-votec Ordered; multi-vote Open; multi-voteg Open; multi vote Open; multi vote Ordered; single vote Open; multi vote Closed; single vote Closed; single vote Ordered; single vote Ordered; single vote Closed; single vote Ordered; single vote
No. of Mean districts M 1 3 11 1 1 1 1 1 8 1e 1 1 4 1f 1 1 1 1 1 1 1h 1 1 1 1 1
18 8.0 6.8 6 24 14 6 14 9.8 99 24 24 3.3 78 9 13 6 5 27 3 54 24 14 7 54 19
Eff thres (Teff )a 4.0 (3.9) 5.0 (8.3∗ ) 9.6∗ 1.8 (10.7) 5.0 (3.0) 5.0 10.7 5.0 5.0∗d (6.9) 5.0 (0.7) 3.0 (3.0) 5.0 (3.0) 17.4∗ 0.9 5.0 (7.5) 5.0 (5.3) 10.7 12.5 2.6 18.7 5.0 (1.3) 3.0 5.0 (5.0) 4.0 (9.3) 1.3 4.0 (3.7)
Notes: a Effective Threshold: Teff = 75%/(M + 1). Effectively a mirror image of M, the lower the Teff , the more proportional the system. Because it is seen as an approximation, Teff is recorded to just one decimal point (* indicates those cases where it is based on a national average). In those cases where there is a legal threshold—which is usually (the exceptions are underlined) greater than Teff —the legal threshold is reported (and Teff is provided, for the sake of record, in brackets). b In ordered list systems, the rules vary regarding the proportion of ‘personal votes’ a candidate requires in order to win a seat regardless of where she or he is ranked. Information is patchy, but the rules we are aware of are as follows: Austria, 7%; Belgium, d’Hondt quota of the party vote; Czech Republic 5%; the Netherlands, 10%; Sweden 5%. c Multi-vote implies that voters can express a vote/preference for more than one candidate. Gender equality law applies requiring the parties to balance their lists. d 5% in the region. e Parties can balance lists to ensure a fair regional representation. This option tends to be used by CDU and CSU. f In this table, Italy is treated as having one national constituency (mean M = 78) reflecting the fact that the seat allocation is based on national votes. However, for the purpose of examining MEPs’ representative roles, it should be noted that Italy is divided into five regions (mean M = 15.6). g Parties can opt to have closed lists, an option used by the Labour Party. All other parties operated open lists. h Available sources are unclear, but it seems that the seats are allocated nationally using d’Hondt and are then filled within each of the thirteen regions (using Hare-Niemeyer). Source: Various. In particular: www.elections2004.eu.int; www.electionsineurope.org; www.MEPs.org.uk; http://www.sussex.ac.uk/sei/; Lodge (2005); Maier and Tenscher (2006); information supplied by Allan Sikk, Philip Stöver, Ingrida Unikaite, and Andreas Wüst.
75
Representation and Electoral Systems in Europe
distinction now in types of electoral system is that between regional and national systems. The 2002 UEP legislation refers to the possibility that member states might consider regionalizing their representation in the EP (i.e. through the use of regional list systems or STV).13 As we saw, this was stated more prescriptively in earlier drafts of the legislation—requiring larger states (those with more than 20 million citizens) to move in this direction, a provision that would have necessitated changes on the part of larger member states. But these proposals were watered down, in anticipation of likely national resistance; in 2004 France was the sole pre-accession member state to voluntarily ‘regionalize’ its system. As with EFs, current district magnitudes vary substantially— something that can be seen with reference to both the direct measure of magnitude (M) and the measure of effective threshold (Teff ; essentially the mirror image of M). The latter measure is included as it is intuitively easier to interpret, and it also facilitates easy comparison with legal thresholds wherever these are applied. The variations in Teff follow from three factors:
r In part, they result from the small number of MEPs elected in some places [notably Luxembourg and Northern Ireland whose Teff s of 10.7 and 18.7 respectively are very much on the high (i.e. disproportional) side];
r They also result from some member states electing MEPs in geographical regions (resulting in higher than average Teff s in Britain, and Ireland);14
r Finally, they are also affected by the use of legal thresholds in fourteen cases (Austria, Belgium, Cyprus, Czech Republic, France, Germany, Greece, Hungary, Latvia, Lithuania, Poland, Slovak Republic, Slovenia, and Sweden), although bizarrely in six of these cases the legal threshold is actually set lower than the Teff (Belgium, France, Latvia, Lithuania, Slovenia, and most dramatically of all, Cyprus).
76
Representation and Electoral Systems in Europe
As we see in Section 4.3, variation in district magnitude has significant implications for the proportionality of the electoral systems. Apart from the mechanical effects of electoral systems on who is elected to parliament, there are also potentially important consequences of electoral systems for how representatives operate once elected. The third of our major electoral system features— ballot structure—could be expected to have an impact in both respects. Ballot structure is referred to in the new legislation, which states that member states ‘may authorize’ preferential voting. In earlier legislative drafts, the case was put more strongly in favour of requiring member states to move in this direction, but this was soon watered down once it was clear that most member states were not sympathetic. Following the ballot structure typology set out in Chapter 3 (Figure 3.2), Figure 4.1 plots the main areas of variation in ballot structure design in the 2004 EP elections, showing three main forms of variation: open, ‘ordered’, and ‘closed’. The open systems—in which the candidates’ electoral fates are affected by their personal vote-chasing activities—are used in nine cases, three of these recent EU entrants (Estonia, Malta, and Lithuania15 ). At the other extreme, closed systems—in which candidates’ electoral fates are determined by their party list placement—are used in eight member states. Finally, there are ordered list systems, in which there is some, limited scope for candidates to improve their list placement through personal votes. These are used in nine member states, five of which joined in the most recent accession (for more discussion on ballot structure design, see, inter alia, Shugart 2005). As the discussion in this section has demonstrated, the provisions of the 2002 UEP legislation left significant scope for variation in the electoral systems used by EU member states for EP elections. Furthermore, across all three dimensions of electoral system discussed—EF, district magnitude, and ballot structure— substantial variation was indeed experienced. In the remainder of this chapter, we review evidence as to the impact of these electoral system differences.
77
Representation and Electoral Systems in Europe Candidate-based
STV with candidate ranking: Ireland, N. Ireland, Malta
Ordered list; single candidate vote matters: Denmark, Estonia
Ordered list; multi-candidate vote matters: Italy, Lithuania, Luxembourg
Ordered list; single candidate vote option does not matter: ‘Ordered’ Austria, The Netherlands, Slovenia, Slovak Rep., Sweden
Ordered list; multi-candidate vote option does not matter: Belgium, Cyprus, Czech Rep., Latvia
Categorical
‘Closed’
Ordinal
Open list; single vote: Finland
‘Open’
Closed list: Britain, France Germany, Greece, Hungary, Poland, Portugal, Spain
Party-based
Figure 4.1. Variations in the ballot structures used for EP elections in 2004 Note: Regional list cases are underlined, including Italy and Germany whose MEPs are regionally anchored. Source: As for Table 4.2.
4.3. Uniform Procedures and Diverse Electoral Outcomes A large body of academic literature on the effects of electoral systems on political representation (for recent examples, see Farrell 2001; Katz 1997) leads us to expect that the electoral system differences set out in Table 4.2 should be associated with substantive differences among the body of MEPs elected in 2004. In this section, we review the main strands of evidence on whether such expectations were, in fact, realized.16 Table 4.3 reports the standard constellation of aggregate electoral system consequences based on the 2004 EP elections. The evidence
78
Representation and Electoral Systems in Europe Table 4.3. The representativeness of EP electoral systems in 2004 Country
Austria Belgium Britain Cyprus Czech Rep. Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta The Netherlands N. Ireland Poland Portugal Slovak Rep. Slovenia Spain Sweden
Disproportionality (GI)a
Effective no. parliamentary parties (Ns )b
Proportion of women (%)
4.72 4.52 8.24 10.32 6.97 4.48 13.12 6.55 10.21 4.00 3.15 3.75 9.77 1.50 13.21 9.15 11.59 10.54 3.88 17.78 3.65 2.83 8.64 12.43 2.56 3.34
3.45 8.02 4.12 3.60 4.18 4.91 3.00 4.46 4.14 3.96 2.94 2.50 3.76 5.78 3.53 4.32 3.00 1.92 5.58 3.00 6.12 2.50 4.91 3.76 2.41 5.91
38.89 29.17 25.33 0 20.83 35.71 33.33 35.71 43.59 31.31 29.17 33.33 38.46 19.23 22.22 38.46 50.00 0.00 44.44 33.33 12.96 25.00 35.72 42.86 33.33 57.90
Notes: a b
√ [ (v−s)2 ]/2. Ns = 1/ si2 . GI =
Source: http://www.elections2004.eu.int/ep-election/sites/en/; http://www.parliament.uk/commons/lib/research/rp2004/rp04-050.pdf; http://www.interieur.gouv.fr/rubriques/b/b3_elections/b32_resultats/france.pdf; http://www.bundeswahlleiter.de/wahlen/europawahl2004/ergebnisse/bundesergebnisse/be_ tabelle_99.html; http://ekloges.ypes.gr/en/index.htm; http://cedweb.mininterno.it:8886/europee/euro040612/Z000000.htm; http://www.tageblatt.lu/dossier/article.asp?DossierArticleId=253; http://www.doi.gov.mt/EN/elections/2004/06EUparl/eu_parl1.asp; http://www.elecciones.mir.es/elecjun2004/peuropeo/DPE999999.htm; http://www.minbzk.nl/wwwkiesraadnl/verkiezingen/verkiezingen/uitslag_verkiezing; http://electionsireland.org/; http://www.sussex.ac.uk/sei/documents/epernep2004finland.pdf; information supplied by Michael Gallagher and Adonis Pegasiou.
79
Representation and Electoral Systems in Europe (a) National trends in 1979 and 2004 (among the original ‘10’) 25 1979
2004
20 15 10 5 0 ly
Ita
er
G
th
Ne
a
Fr
l
Be
n
De
Irl
x
Lu
N.
Irl
B
G
(b) Mean averages over time GI
10 9 8 7 6
Mean (‘10’) Mean (All)
5 4 1979
1984
1989
1994
1999
2004
Figure 4.2. Disproportionality trends in EP elections, 1979–2004 Source: Times Guides to the European Parliament (various); www.europarl.eu.int.
in this table is supplemented by two sets of figures: (a) Figures 4.2–4.5, which summarize the main changes between 1979 and 2004 for those countries that have been EU member states since the first direct elections in 1979, and which also track the average trends across all elections; and (b) Figure 4.6, which contrasts the trends across the 26 electoral systems in the 2004 elections. We start with the temporal comparisons. In line with the aspirations of the early proponents of UEP for EP elections (see the
80
Representation and Electoral Systems in Europe (a) National trends in 1979 and 2004 (among the original ‘10’) 700,000 1979
2004
600,000 500,000 400,000 300,000 200,000 100,000 0 x
Lu
Irl
n
De
l
Be
N.
Irl
th
Ne
a
Fr
B
G
ly
Ita
er
G
MEPs/VAP
(b) Mean averages over time 420,000 400,000 380,000 360,000 340,000 Mean (All)
320,000
Mean (‘10’) 300,000 1979
1984
1989
1994
1999
2004
Figure 4.3. The ratio of MEPs to the electorate in the EP, 1979–2004 Source: Times Guides to the European Parliament (various); www.europarl.eu.int
discussion in Bardi 1990), the EU’s elected institution has clearly, if gradually, become more ‘representative’ of the general population. Electoral systems research has typically found a link between partisan proportionality and the extent to which a parliament is a ‘microcosm’ of the people it exists to represent (Farrell 2001), and
81
Representation and Electoral Systems in Europe
such is indeed the case for the EP despite the fact that, due to the most recent enlargement, there are now more electors per MEP. The relevant trends are summarized in Figure 4.2(a). Across the nine longest established member states, even though in virtually all cases there has been a decline in the number of MEPs per member state (Figure 4.3(a)), aggregate disproportionality (measured via the Gallagher index; see Gallagher 1991) declined in most cases, with a fall in the average from 8.94 in 1979 to 7.60 in 2004. The trends are tracked over time in Figure 4.2(b), indicating three interesting patterns. First, the decline in disproportionality in most cases was associated with the accession of new member states operating more proportional electoral systems than many of the existing member states. Second, the exception to this general pattern was in 2004 when many of the newer member states opted for electoral systems that tended to be less proportional than the average. Third, and not surprisingly, the largest drop in disproportionality was in 2004 in large part due to Britain’s switch to a regional list system.17 Levels of disproportionality along party lines in Britain plummeted (the Gallagher index of disproportionality recorded 8.2 in 2004—little more than a third of the 1994 figure of 23.7).18 Consistent with these disproportionality trends, the number of parties (as measured by the Laakso and Taagepera [1979] effective number of parties index) rose across the period from an average of 3.68 in 1979 to 4.63 in 2004. With the exception of Denmark, across the nine member states (and Northern Ireland, if we continue to count this as a separate electoral system) that were involved in the 1979 direct elections, the numbers of parties was greater in 2004 (Figure 4.4(a)). And the underlying trends (Figure 4.4(b)) reveal a pretty steady growth throughout the first twenty years, with much of this growth occurring in the more established member states. The rise in the index is rudely interrupted in 2004 when, across the board, there was a drop in the numbers of parties (see also Appendix Table A.4.2). In similar fashion to the trends in proportionality and numbers of parties, the proportion of women MEPs nearly doubled from 16.1 per cent in 1979 to nearly a third of the parliamentary
82
Representation and Electoral Systems in Europe (a) National trends in 1979 and 2004 (among the original ‘10’) 9
1979
2004
8 7 6 5 4 3 2 1 0 n
De
l
Be
ly
Ita
a
Fr
Irl
th
Ne
N.
Irl
er
G
x
B
Lu
G
(b) Mean averages over time
Ns
5.5 5 4.5 4 3.5 3 Mean (All) Mean (‘10’)
2.5 2 1979
1984
1989
1994
1999
2004
Figure 4.4. The effective number of (parliamentary) parties trends in EP elections, 1979–2004 Source: Times Guides to the European Parliament (various); www.europarl.eu.int.
membership (30.5 per cent) in 2004.19 Comparing 1979 with 2004, we see a rise in all ten cases (Figure 4.5(a)); and the underlying averages reveal a steady growth pattern across all members states over time (Figure 4.5(b)).20 Today, with the exceptions of Cyprus and Malta which have no women MEPs (clearly in part a product of electing so few MEPs overall), in all the other member states between one-fifth and one-third of the delegations are women, and
83
Representation and Electoral Systems in Europe (a) National trends in 1979 and 2004 (among the original ‘10’) 50 45
1979
2004
40 35 30 25 20 15 10 5 0
N.
Irl
l
Be
ly
Ita
Irl
B
G
er
G
x
Lu
th
Ne
a
Fr
n
De
% Women
(b) Mean averages over time 35 30 25 20 15 10 Mean (all) 5
Mean (‘10’)
0 1979
1984
1989
1994
1999
2004
Figure 4.5. The proportions of women MEPs in the EP, 1979–2004 Source: Times Guides to the European Parliament (various); www.europarl.eu.int
in two cases (Luxembourg and Sweden) women are now at parity with men (see also Appendix Table A.4.3). While the evidence in Table 4.3 and Figures 4.2–4.5 does point towards interesting differences over time, a second story also emerges. This latter story is cross-sectional: the large degree of variation between our 26 cases. There were significant
84
Representation and Electoral Systems in Europe (a) Relationship between effective threshold and disproportionality 20 GI 18 16 14 12 10 8 6 4 2 0
(b) Relationship between delegation size and disproportionality GI 20 15 10 5 0 0
5
0
15
10
20
40
60
100
80
120
20 No. MEPs
(c ) Relationship between disproportionality and the effective number of parties
(d ) Relationship between disproportionality and the proportion of women MEPs
9 8 7 6 5 4 3 2 1 0
% Women
Eff. no. parties
Effective threshold
0
5
10
15
20
GI
70 60 50 40 30 20 10 0 0
5
10
15
20
GI
Figure 4.6. The consequences of variations in EP electoral system design in 2004 Source: As for Tables 4.2 and 4.3.
variations in the types of representatives elected to the EP in 2004. Levels of partisan disproportionality range from the rather substantial in Northern Ireland (17.78), Estonia (13.12), and Slovenia (12.43), down to some very low levels (Italy, with a GI score of only 1.50). There is also considerable difference in the number of parliamentary parties represented (ranging from Malta’s effective number of parties score of 1.92 to Belgium’s 8.02); and, as we have already seen, in terms of the representation of women. That electoral systems are at least partially responsible for the differences in the types of representatives elected to the EP in 2004 is given support by the evidence presented in Figure 4.6. The relationships revealed here are very much consistent with the predictions that would be derived from the electoral systems literature. The steepness of the regression line in Figure 4.6(a)
85
Representation and Electoral Systems in Europe Table 4.4. The effects of effective threshold and delegation size on disproportionality: OLS regression
Constant Effective threshold Delegation size Adjusted R2 ∗
B
Std. error
5.24 0.55∗ −0.04
1.38 0.14 0.02
0.48
Significant at 0.01 level.
demonstrates the strong relationship between Teff and disproportionality. The (Pearson) correlation between these two variables is 0.73 (significant at 0.01 level, 2-tailed). Another electoral system variable that can have an important influence over aggregate proportionality is AS (see Lijphart 1994); the equivalent in this context is delegation size.21 As Figure 4.6(b) shows, there is, indeed, a significant relationship between delegation size and disproportionality (Pearson correlation = −0.46; significant at 0.05 level, 2-tailed). However, once we examine the combined influence of Teff and delegation size on disproportionality, the latter becomes non-significant (see Table 4.4).22 The trends are more mixed with regard to other expected electoral system consequences. Consistent with expectations, there is a relationship (as indicated by the regression line) between disproportionality and the effective number of parties (Figure 4.6(c)); however, in this case the correlation is not significant. The weakness of this relationship could well, in part, reflect the fact that party systems are still largely determined nationally (e.g. Lipset and Rokkan 1967), and therefore there is limited scope for influence here by EP electoral system design. And national influences are clearly an even more powerful factor in explaining the complete lack of relationship between disproportionality and the proportion of women MEPs (as shown by the flat regression line in Figure 4.6(d)). The higher than predicted proportion of French women MEPs is, to some degree, a likely result of recent gender
86
Representation and Electoral Systems in Europe
equality legislation, whereas the lower than predicted proportion of Maltese women MEPs is consistent with a national political culture that does not favour female representation (Lane 1995).
4.4. Conclusion The EU now has a legislative framework for UEPs governing elections to the EP; however, the systems used for those elections remain somewhat less than entirely uniform. Does this mean that the 2002 UEP legislation was worthless? Not necessarily: at the very least, the legislation guards against future change in the direction of non-proportionality. This matters because, as we have seen, the trend of member states towards using more proportional electoral systems has produced a set of parliamentary representatives that is increasingly politically and socially more representative and by implication also attitudinally more congruent—something closer to the ideal classically expounded by John Adams of a parliament that is ‘an exact portrait, in miniature, of the people at large’ (quoted in McLean 1991: 173; see also Pitkin 1967). There currently appears little realistic prospect of further reform of the UEP legislation, apart from the promise to revisit the question of a 10 per cent Euro-constituency top-up before 2009.23 This does not, of course, mean that potential reforms cannot be evaluated in terms of their likely consequences and thus their overall desirability. In the context of ongoing debates about how citizens can be most effectively represented in the EU, and particularly given the competitive relationship between the EP and the Council, notably over the design of EU legislation, and further taking into account the continuing weakness of the Committee of the Regions, there is a strong case that the EP’s representative role should emphasize a regional perspective, to counteract the national orientation of the Council. However, while such a regional emphasis is consistent with current legislation, making it a more-or-less universal requirement on member states would require further legislation.
87
Representation and Electoral Systems in Europe
Notes 1. Article 21 of the Paris Treaty (1951). 2. For ease of exposition throughout this discussion we shall refer to the Assembly as the EP. Formally it did not assume this title until 1962. 3. It is interesting to note that such a two-stage process of electoral system design is consistent with the practice of newly developing federations. For instance, in 1902, the members of the newly formed Australian federal parliament were elected by the electoral systems then used in their states. The intention of the constitution framers was that it should be the newly elected parliament that should then design UEPs for subsequent elections (see Farrell and McAllister 2006). 4. Doc 194/1972. 5. At the same time, the breakdown of devolved government in Northern Ireland culminated in the British government’s decision to introduce PR (in the form of STV) for local and regional elections in the Province. 6. For details, see Farrell (2001: 105). 7. Official Journal C 87, April 4, 1982, p. 61. 8. It is worth noting that while this reform provided a convenient means for the New Labour government to signal to its partners that it wished to engage more constructively in the EU than its Conservative predecessors had done, this was by no means the sole motivation. Mr Blair’s government was, for a while at least, far more ready than any recent predecessor to engage with the agenda of electoral reform. Thus, the new devolved institutions in Scotland and Wales were given proportional MMP systems; there were (failed) efforts to introduce (partial) elections for the House of Lords; and there was a highlevel, government-sponsored investigation of alternative options for electing the House of Commons (the ‘Jenkins process’). All this was in addition to the shift from SMP to PR-list for the EP (for details, see Farrell 2001a). In short, the new electoral system for electing British MEPs was introduced in large part for domestic political reasons. 9. DOC 354/354956; PE224.331/fin.
88
Representation and Electoral Systems in Europe 10. At the time of writing, the current proposal is that this 10 per cent top-up proposal should be considered in time for the 2009 elections. 11. The Maastricht Treary had since included UEPs in the list of items covered by the Assent Procedure. 12. The following summary is based on a note from the General Secretariat of the Council of the EU on February 22, 2002 (6151/02 PE 14 INST 21). This memo anticipated the subsequent Assent of the EP in June 2002. 13. See the note from the General Secretariat of the Council of the EU on February 22, 2002 (6151/02 PE 14 INST 21). 14. Note that France, which changed from national to regional list in 2004, adopted a 5 per cent legal threshold (at regional level) at the same time, which ironically was lower than the Teff (of 6.9) that would have been required for parties to win seats. Had the proposal in the original draft of the EP legislation requiring larger member states to adopt regional lists been implemented, this would have required changes by all the larger member states, including Germany (which permits a regional balance on the party lists, through the use of combined Land lists, an option that tends to be used by the CDU and CSU). The interesting case here is Italy, whose electoral system incorporates a national list calculation for determining seat allocation with a five-region system for locating the MEPs. In terms of measuring overall proportionality, therefore, the Teff is a very low 0.9; however, this masks the fact that the MEPs are regionally anchored. 15. Lithuanian parties are free to opt for either open or closed list design. With the exception of the Labour Party, all the other parties opted for open lists (information supplied by Ingrida Unikaite). 16. This section focuses on the ‘mechanical’ effects of electoral systems as measured by aggregate proportionality and most closely associated with the district magnitude and EF variations shown in Table 4.2. The ‘strategic’ effects of electoral systems are the primary focus of attention in the remainder of this volume. 17. Note, however, that Britain was not the only cause of this shift in 2004. That election also witnessed more proportional election results
89
Representation and Electoral Systems in Europe
18.
19.
20.
21.
22.
90
in Belgium, Denmark, France, and Germany (for individual county details, see Appendix Table A.4.1) Consistent with this, the effective number of parliamentary parties in the British case more than doubled (from 1.7 in 1994 to 3.1 in 1999, with a further rise to 4.1 in 2004); and there was also a notable increase in the proportion of women MEPs (up from 19 per cent in 1994 to 25 per cent in both 1999 and 2004). The ‘mechanical’ effects of the 1999 change to PR, then, were entirely consistent with what would have been predicted from the electoral systems literature (see also Farrell and Scully 2003). We note that the trend towards greater female representation is unlikely to be solely due to the impact of electoral system effects; it can also be accounted for in part by the general process of modernization of representative politics (see Cotta and Best 2000). The two lines in Figure 4.5b are based on the mean average percentage of female MEPs from, respectively, the ‘original ten’ and across the entire EP membership at the time of each set of elections. This figure is thus ‘weighted’ to allow for the varying size of memberships from the different member states, and therefore calculated on a different basis from Figures 4.2b and 4.3b. In the latter cases, given our interest in the impact of electoral systems on proportionality, the lines represent the un-weighted mean averages of the GI and Ns figures. Unlike at national level where the national political elite can essentially pick the AS they wish, in the case of the EP delegation size is the product of intense negotiations between member states, each anxious to grab the maximum number of EP seats possible. In passing, it is interesting to note how, as a result of the most recent enlargement, it is no longer the case that Germany, the largest member state, has a proportionate disadvantage in the ratio of MEPs to voters. As Figure 4.3a reveals, Italy now holds that position, with France in second place. Figure 4.3b indicates how the established member states have tended to jealously guard their delegation size in periods of enlargement: while average ratios drop after 1989, among the ‘original’ 10 the ratios actually continue to rise. The multivariate findings reported in Table 4.4 might simply reflect the substantial statistical and substantive association between
Representation and Electoral Systems in Europe Effective Threshold and Delegation Size. Delegation Size is an important factor shaping Teff in many member states; thus, it is unsurprising that our measures of the two variables are moderately well correlated (r = −0.35). The exact interpretation of the coefficients in Table 4.4 remains open for dispute; what we can be clear about is that no distinct relationship between delegation size and proportionality remains once Teff has been accounted for. 23. See the note from the General Secretariat of the Council of the EU on February 22, 2002 (6151/02 PE 14 INST 21).
91
This page intentionally left blank
5 Representation in Europe: The Institution and the Individuals
Chapter 4 documented the electoral systems that are used for EP elections. The discussion in that chapter demonstrated that— notwithstanding the long battle fought by many to implement UEPs for EP elections, and the 2002 legislation to this effect— there remained, even in the June 2004 European elections, substantial variation in the electoral arrangements deployed across the twenty-five member states for choosing their members of the EU’s elected chamber. Chapter 4 also showed that some of the expected ‘mechanical’ effects of different electoral systems were manifest in EP election results. While performing increasingly well in terms of ‘micro-cosmic’ representation, in the sense that the EP has come to look more like the population of the EU, there still remain crossnational differences in the ‘representativeness’ of EP delegations; these differences, moreover, are very much in line with what the electoral systems literature would lead us to expect. As we observed in Chapter 3, however, there is a great deal more to political representation than just the micro-cosmic dimension. Representation is a complex and much-contested concept, encompassing important questions about who should be represented, by whom, and how. Disagreements about representation, moreover, are not argued out merely at the level of normative political theory, or even of general debates about the political structures, such as the electoral institutions, that should exist in a society. Choices
93
Representation in Europe: The Institution and the Individuals
between alternative visions and practices of representation are a daily, practical reality for elected politicians. Facing potentially infinite calls on a finite quantity of time and energy, elected politicians have to prioritize their activities. And in practice, as much previous research has shown, parliamentarians differ considerably in the priorities they select and the models of representation that they follow (Wahlke et al. 1962; Muller and Saalfeld 1997). In this chapter, we start the process of examining the extent to which electoral institutions may shape political representation. In this chapter and the next, we draw in particular on a recent (2006) survey conducted with MEPs from all 25 EU member states, and use the information gathered from this survey to examine parliamentarians’ attitudes towards their role as representatives. Chapter 6 explores in close detail the possible links between variation in electoral institutions, on the one hand, and the attitudes and behaviour of elected representatives on the other. The current chapter serves as something of a prelude to Chapter 6, by considering MEPs’ attitudes more broadly. Little attention has been paid in previous research to their attitudes to practices of representation in the EU. As a first stage in our analysis of political representation in the EU, we explore MEPs’ views in two central areas. First, we consider their collective position as members of a representative institution, the EP. What role as a representative body do they think this chamber can and should play? Second, we then move on to address how MEPs view their individual roles as representatives. We consider how they understand this role, what are their priorities within it, and finally begin to explore the factors that might explain differences between them. The conclusion to the chapter then summarizes the implications of our findings for the following chapter.
5.1. The EP as a Representative Institution We begin our examination of the attitudes of MEPs towards political representation by considering what one might term issues of
94
Representation in Europe: The Institution and the Individuals
‘macro-representation’. That is, we explore what the attitudes of MEPs are towards the institution of the EP, and to its position as a representative body within the EU. These matters are important not only for what they tell us about how MEPs view the EU and representative structures within the Union; it also matters because how they view the role of their institution may have a significant impact on how MEPs understand and interpret their individual roles. While there have been sporadic investigations of MEPs’ general attitudes towards integration (for a review, see Scully 2005: ch. 3), the extant academic literature in the area we examine here remains sparse in the extreme. We thus have little established knowledge to build on, and much of what we seek to do here in investigating the landscape of attitudes among MEPs and the reasons for differences among them must be descriptive and exploratory. To conduct this investigation, we draw on a major academic survey of all MEPs that was conducted in 2006. (For a detailed description of this survey, see the Appendix.) In particular, in this section of the chapter, we utilize responses to a number of questions that were included in the survey regarding the powers and position of the EP. To begin with, in Table 5.1 we present the aggregate results from our sample of MEPs on seven items. As can readily be observed, opinion among MEPs across these different items is strongly skewed towards support for an enhanced EP role in most respects. However, there is also a persistent body of opposition. Moreover, when presented with a particularly radical (and arguably unrealistic) option—empowering the EP to shape Eurozone interest rates, something that contravenes established international models of monetary policymaking—most MEPs oppose this proposition. In other words, while the majority of MEPs support the EP becoming a more powerful representative body within the EU, the evidence of our survey suggests that their support is not indiscriminate or uncritical.1 What individual differences, and what general patterns, lie behind these overall findings? To explore MEPs’ attitudes beyond
95
Representation in Europe: The Institution and the Individuals Table 5.1. MEPs’ opinions (%) on the powers of the EP Item
The EP should have more power to influence interest rates in the Eurozone The EP should have the right to initiate legislation The EP should have equal power with the Council in all areas of law-making The EP should have equal power with the Council to amend all areas of expenditure in the budget The Commission President should be nominated by the EP, rather than the European Council The EP should be able to remove individual Commissioners from office The EP should be allowed to hold all its plenary sessions in Brussels
Strongly Agree Neither Disagree Strongly No. of agree agree nor disagree respondents disagree 6.1
16.7
19.4
30.6
27.2
180
37.8
37.2
9.6
11.2
4.3
188
40.2
32.3
10.1
14.3
3.2
189
46.0
33.3
8.5
10.6
1.6
189
32.3
26.5
19.6
17.5
4.2
189
38.7
33.5
11.0
13.6
3.1
191
54.7
17.4
11.6
8.4
7.9
190
Note: For all items in this table, the survey question was worded as ‘To what extent do you agree or disagree with the following statements about the powers of the European Parliament?’, except for the first item, where the relevant question was ‘To what extent do you agree or disagree with the following statements about Economic and Monetary Union and EU monetary policies?’. Source: 2006 MEP Survey.
the broad aggregate results, we first construct a correlation matrix of the various items to test their patterns of association. The bivariate correlation coefficients for responses to these seven questions are reported in Table 5.2. The results show that MEPs’ answers on all the items are positively and at least moderately strongly interrelated, with one exception—the question concerning EP plenary
96
Representation in Europe: The Institution and the Individuals Table 5.2. Bivariate correlation coefficients (r) for EP powers items Initiate legislation
Eurozone Initiate legislation Equal power law-making Equal power budget Commission President nomination Individual Commissioners
.39∗
Equal Equal Commission Individual Brussels power power president Commissioners plenaries law-making budget nomination .27∗ .42∗
.27∗ .45∗
.32∗ .47∗
.24∗ .31∗
−.12 −.01
.68∗
.53∗
.27∗
.00
.55∗
.29∗
.07
.33∗
−.21∗
.11
sessions. While supporting greater powers for the EP in legislation is positively associated with also supporting greater powers over the EU budget, over the Commission and in managing the Eurozone (and the converse is also true: an MEP inclined to oppose one of these options is also likely to oppose the others), this pattern does not hold true for the question of the parliament’s location. While the latter might intuitively appear to be, no less than the other six questions, a matter concerning the status and autonomy of the EP, positive answers on the other items are not significantly associated with supporting the EP’s right to locate itself where it chooses. This finding is confirmed by a reliability analysis, which indicates that without the EP plenary item included, the remaining six items generate a highly reliable Likert additive scale of MEPs’ opinions on the position of the EP within the institutional structures of the Union.2 To probe further into the reasons for differing levels of support for empowering the parliament among MEPs, we next conduct a multivariate analysis. We specify two dependent variables. One is an additive scale of responses to the six highly correlated survey items on EP powers identified above, with responses coded so that the most ‘pro-EP’ positions (‘strongly agree’) are scored a ‘5’ for
97
Representation in Europe: The Institution and the Individuals
all items, and the least pro-EP position (‘strongly disagree’) scores a ‘1’. The additive scale thus runs from 6 to 30.3 The second dependent variable comprises responses to our other question on the location of EP plenary sessions, with responses coded from ‘5’ for ‘strongly agree’ (that ‘The EP should be allowed to hold all its plenary sessions in Brussels’) down to ‘1’ for ‘strongly disagree’. We then specify an explanatory model that seeks to tap into systematic sources of possible variation in MEPs’ attitudes on these two measures. This model includes potential explanatory variables grouped under four main headings. First, we specify several variables concerned with potential individual sources of variation. To control for generational differences among representatives, we include a measure of the age (in years) of each MEP. We also include a dummy variable for gender (coded ‘1’ for female MEPs, ‘0’ for men). We specify the number of years (not necessarily continuous) an individual has served as an MEP, to test for any impact of internal ‘institutional socialization’ on MEPs’ views (cf. Scully 2005), and we include two dummy variables measuring whether an individual has been a national parliamentarian or a national government minister to test for any national-level socialization effects. Second, we specify several variables that seek to identify general political sources of variation in attitudes. We include a measure of MEPs’ self-placement on the left–right ideological continuum, specified in two distinct forms: both directly, and in a form ‘folded’ around the mid-point of the scale, to tap into differences between centrists and extremists (with extremists scoring highest on this scale).4 In addition, to allow for a possible relationship between attitudes towards European integration and political representation at the European level, we include a variable based on MEPs’ self-placement on integration.5 Third, we explore the possibility that differences in the interpretation and understanding of the representation in the EU may be promoted by institutional features of the political environment— specifically, the electoral system differences that we have discussed earlier in the book. Following the discussion in Chapter 4, and
98
Representation in Europe: The Institution and the Individuals
referring to the two features of electoral system design that are most likely to affect MEP attitudes towards representation (ballot structure and district magnitude; see Figure 3.1(b)), we specify electoral system differences in the following way. First, in recognition of STV’s location at the top right quadrant (most ‘open’) of the ballot structure typology presented in Figure 4.1), we provide a basic dichotomous ballot structure dummy variable that distinguishes between STV (coded ‘1’) and the various List systems (coded ‘0’)— in the tables that follow this is referred to as ‘STV/List’. Second, we include a more nuanced ballot structure variable that distinguishes between the three main types of ballot structure that MEPs were elected under—open (coded ‘2’), ordered (coded ‘1’), and closed (coded ‘0’).6 Third, we include a variable for district magnitude (the mean M for each member state). Finally, as a residual category to allow for broad national differences in attitudes, we include a set of dummy variables specifying the nationality of an MEP.7 All four sets of variables are included in OLS regression analyses, the results for which are reported in Table 5.3. As can be seen from the table, we specify a series of alternative models for both of our dependent variables; all the models include our individual control variables; we then separately include our political, institutional, and national variables in further models, as well as in a combined model.8 The results reported in Table 5.3 indicate—as one would have expected given the weak inter-relationship between our two dependent variables—that the factors that best explain MEPs’ attitudes towards a general empowerment of their institution are quite different from those that condition their views on the location of the chamber. As the first panel in the table indicates, broadly political factors have the greatest explanatory leverage in accounting for variation on the EP Powers Scale. With the exception of a modest gender effect—female MEPs being more favourably disposed towards empowering the parliament—individual background factors, including political experience within and outside the chamber, do not appear of great importance.9 Nor does nationality per se appear to play much of a role in explaining differences: when
99
Representation in Europe: The Institution and the Individuals Table 5.3. OLS regression coefficients (standard errors) for EP powers scale and location of EP plenaries Variables
Model 1
Model 2
Model 3
Model 4
Combined Model
(a) EP Powers Scale Age .03 (.04) .01 (.04) .03 (.04) −.01 (.04) .01 (.04) Gender 1.64 (.74)∗∗ 1.12 (.68) 1.72 (.74)∗∗ 1.93 (.75)∗∗ 1.20 (.68)∗ Years as MEP −.04 (.07) −.07 (.06) −.03 (.07) −.03 (.07) −.06 (.06) Former nat. MP −1.26 (.79) −1.22 (.73)∗ −1.28 (.80) −.87 (.86) −1.01 (.75) Former nat. minister −.90 (1.09) −.92 (1.04) −.76 (1.09) .03 (1.21) −.98 (1.05) Left-right stance −.40 (.16)∗∗ −.45 (.17)∗∗∗ Left-right stance (folded) .10 (.31) .10 (.31) Integration self-placement .70 (.14)∗∗∗ .64 (.15)∗∗∗ STV/list −2.65 (2.47) −6.04 (3.15)∗ Ballot structure .64 (.49) .35 (.46) District magnitude .02 (.01)∗ .02 (.01) Belgian 1.64 (2.91) British −3.30 (2.52) Cypriot .02 (3.96) Czech −2.69 (2.81) Danish −4.11 (3.09) Estonian .75 (3.53) Finnish −1.80 (3.05) French 2.08 (2.61) German .50 (2.51) Greek 2.41 (5.09) Hungarian −3.71 (3.22) Irish −3.51 (3.19) Italian 1.47 (2.60) Latvian −5.97 (3.31)∗ Lithuanian 2.00 (3.04) Luxembourg −1.79 (4.12) Maltese (dropped) Dutch −2.53 (3.01) Polish −1.32 (2.58) Portuguese −1.14 (2.90) Slovak 1.26 (3.89) Slovene −.73 (3.93) Spanish 2.48 (2.94) Swedish −3.39 (2.77) (Constant) Adjusted R2 N (b) Location of EP plenaries Age Gender Years as MEP Former nat. MP Former nat. minister
100
20.81 (1.98) 19.47 (2.60) 19.58 (2.08) 23.32 (3.06) 19.39 (2.63) .03 .21 .04 .13 .22 189 186 189 189 186 −.01 (.01) −.16 (.20) −.03 (.02) −.14 (.22) −.14 (.30)
−.01 (.01) −.19 (.21) −.03 (.02) −.15 (.23) −.12 (.33)
−.01 (.01) −.23 (.20) −.02 (.02) −.23 (.22) −.21 (.30)
.01 (.01) .05 (.17) −.03 (.02) −.38 (.20)∗ −.11 (.28)
.01 (.01) −.02 (.17) −.03 (.02)∗ −.40 (.20)∗∗ −.01 (.29)
Representation in Europe: The Institution and the Individuals Table 5.3. (Cont.) Variables
Model 1
−.04 (.05) −.02 (.10) .01 (.04)
Left-right stance Left-right stance (folded) Integration self-placement STV/list Ballot structure District magnitude Belgian British Cypriot Czech Danish Estonian Finnish French German Greek Hungarian Irish Italian Latvian Lithuanian Luxembourg Maltese Dutch Polish Portuguese Slovak Slovene Spanish Swedish (Constant) Adjusted R2 N ∗
Model 2
4.63 (.55) −.00 190
4.84 (.81) −.02 186
Model 3
−.79 (.68) .23 (.14)∗ −.01 (.00)
4.68 (.57) .02 190
Model 4
Combined Model −.06 (.04) .05 (.08) .10 (.04)∗∗
−.46 (.67) −.72 (.68) .21 (.58) .16 (.58) .31 (.91) .12 (.90) −1.11 (.65)∗ −1.35 (.64)∗∗ .61 (.71) .41 (.70) −.39 (.81) −.78 (.82) .00 (.70) −.40 (.71) −2.71 (.60)∗∗∗−3.11 (.61)∗∗∗ −1.29 (.58)∗∗ −1.56 (.58)∗∗∗ .27 (1.17) −.53 (1.17) .04 (.74) .23 (.73) −.79 (.74) −.88 (.88) −.38 (.60) −.78 (.60) −1.04 (.76) −1.09 (.75) .37 (.70) .20 (.69) −1.62 (.95)∗ −1.90 (.94)∗∗ (dropped) (dropped) .23 (.69) .02 (.69) −.11(.60) −.28 (.59) −.20 (.69) −.68 (.67) .64 (.90) .45 (.88) −.30 (.91) −.63 (.90) .03 (.68) −.16 (.67) .46 (.64) .21 (.64) 4.29 (.70) .38 190
.42 186
< .10; ∗∗ < .05; ∗∗∗ < .01.
these are added to personal factors, there is only one, limited and modest, significant coefficient for the national dummy variables (with Latvian MEPs tending to oppose greater EP powers). There is more indication of a relationship between attitudes to the powers of the EP and electoral-institutional factors. In particular, the district magnitude measure (M) shows a positive relationship with the dependent variable. However, the impact even of this variable is weak in comparison to the major political
101
Representation in Europe: The Institution and the Individuals
variables specified—left–right position and an MEP’s general stance on European integration. The latter two variables are strongly associated with attitudes to the institutional position of the parliament. Specifically, opposition to empowering the EP comes substantially from MEPs on the political hard left; while, unsurprisingly, those representatives who are generally supportive of integration are also very likely to support specific enhancements to the role of the EU’s elected chamber. These relationships are confirmed in a combined model where the personal, institutional, and political variables are all included: while electoral-institutional variables, as well as gender, do attain statistically significant values, general political attitudes have by far the greatest predictive power. Things are very different as far as attitudes to the location of the EP’s plenary sessions are concerned. Here, individual factors offer no empirical leverage by themselves, while general political attitudes are also of marginal impact. And while there is a modest statistical association between one electoral-institutional variable—ballot structure—and the dependent variable, this would seem to be a spurious association, caused solely by the fact that certain nationalities are among the group of countries with ‘closed’ ballot structures. This is revealed by the model that includes national dummy variables, which has a greatly increased model fit. When the national dummies are included in a combined model with the personal and political variables, we see some significant coefficients emerging among the latter—longer serving MEPs, and those with former national parliamentary experience, tend not to support any change to the location of EP plenary sessions, while those generally supportive of further integration do. But the main driver of MEPs’ views on where their institution should be located is nationality: French MEPs, joined to some extent by their colleagues from Germany, Luxembourg, and Denmark, are the main opponents of a change to the location of EP plenaries. Given the geographical location of Strasbourg, the current base for most EP plenary sessions, such a finding is not difficult to interpret or understand.
102
Representation in Europe: The Institution and the Individuals
To conclude, this section of the chapter has shown several things. It has shown that most MEPs have clear views on the role of their institution. And, while the balance of opinion within the EP is very much in favour of granting the chamber further powers, such views do not appear to reflect mindless self-aggrandisement. MEPs’ attitudes on a variety of questions in this area are systematically related, and most MEPs reject unrealistic suggestions regarding how their institution might be empowered. Finally, while nationality is the dominant factor in explaining views on the symbolic question of the location of the chamber, on more substantive questions about its role, attitudes are shaped far more by broad political ideas and orientations, as well as to a limited extent by the electoral institutions from which MEPs emerge.
5.2. Individual Representation In this section of the chapter, we now turn from examining MEPs’ attitudes to the role of their institution to begin exploring their attitudes to their own roles. In other words, rather than examining the representative role of MEPs collectively, we are now concerned with trying to understand them as individuals. How do MEPs understand their position as representatives, and what are their individual priorities as representatives? We could reasonably expect that there will be substantial differences among MEPs in how they understand and seek to carry out their role as elected representatives. In addition to individual and ideologically-based differences, such as one could expect to exist among the membership of any legislative assembly, MEPs are also likely to differ because of the influence of two further factors. While national differences were of some importance in accounting for MEPs’ attitudes towards the status and location of the EP, we might expect them to matter even more so here. Coming from 25 different member states with widely varying political histories and domestic political institutions, we can plausibly expect that there should be substantial differences in the political cultures from
103
Representation in Europe: The Institution and the Individuals
which MEPs emerge that would have implications for political representation. It seems unlikely that prevailing expectations and beliefs concerning the role of the elected representative are wholly invariant throughout the EU. Second, while electoral-institutional differences appear to matter little for explaining attitudes towards the position of the EP, there are much stronger reasons, relating to the electoral-political incentives under which they operate, which we have already considered in previous chapters of this book, to expect such factors to shape individuals’ attitudes towards how they understand their own role as elected representatives. Current knowledge of how MEPs understand and approach their role as individual representatives is distinctly limited. Few previous explorations of this topic have been carried out, and none could really be said to constitute major studies. Two pieces by Richard Katz have examined the ‘role orientations’ of MEPs: in his first paper, Katz (1997) compares MEPs with other candidates for the EP; in a subsequent paper, he contrasts MEPs with national parliamentarians (Katz 1999). While some attempt is made to explain differences among representatives, and to consider the relationship of MEPs’ role orientations to their broader preferences on integration, in neither study are MEPs’ attitudes explored in any great depth.10 Similarly, a study by Bernhard Wessels (1999) that examines the ‘Focus of Representation’ is based around a comparison of MEPs with national parliamentarians, and makes only very limited use of multivariate analysis in exploring parliamentarians’ attitudes. There is, in short, much more we can, and should, know about how MEPs see their position as individual, elected representatives. Among the matters addressed in the 2006 MEP survey were MEPs’ attitudes to representation—in particular, whom did most MEPs regard themselves as representing, and what aspects of their work as elected representatives did MEPs view as the most important? Results from a question probing the former (‘How important is it to you to represent the following groups of people in the EP?’), where several different response categories were offered, are presented in Table 5.4.
104
Representation in Europe: The Institution and the Individuals Table 5.4. How important is it to you to represent the following groups of people in the EP? (%) Response category
Of little importance 1
All people in Europe All people in my member state All people in my constituency/region All people who voted for my party My national party My EP party group
Of great importance 2
3
4
5
Mean importance (out of 5)
No. of respondents
10.6 1.0
8.7 18.3 31.3 6.7 13.5 21.2
31.3 57.7
3.64 4.28
208 208
2.0
4.9 12.3 17.2
63.5
4.16
203
1.4
6.8 14.5 29.0
48.3
4.35
207
2.0 6.8
6.4 17.7 34.5 3.9 26.3 26.8
39.4 36.1
4.03 3.81
203 205
Source: MEP 2006 survey.
At the aggregate level (as indicated by the mean averages for each category) the data show some significant, although far from vast, differences in the importance accorded to the representation of different groups. The representation of ‘all people in my member state’, and ‘all people in my constituency/region’ are accorded the highest levels of importance, while representing the EP party group and ‘all people in Europe’ attain the lowest ratings. The differences here, however, are fairly modest (a mean difference of .71 on a 5-point scale between the highest and lowest rated items), and even the lowest rated groups score well above the mid-point on the scale. This suggests that most MEPs have little problem in acknowledging that their representative role is one with numerous dimensions—or to put it in the language of much contemporary political science literature, these ‘agents’ accept that they have important responsibilities to represent multiple ‘principals’. An additional question in the 2006 survey probed further into MEPs’ priorities as representatives, by asking them directly about the importance they accord to different activities and areas of work that representatives typically engage in (‘When thinking about your work as an MEP, how important are the following aspects
105
Representation in Europe: The Institution and the Individuals Table 5.5. ‘When thinking about your work as an MEP, how important are the following aspects of your work?’ Response category
Of little importance
Working on legislation Parliamentary oversight Articulation of important societal needs and interests Developing common strategies for EU policies Mediation between different interests in society Representation of individual interests of individual citizens
Of great importance 4
5
Mean importance (out of 5)
No. of respondents
1
2
3
1.9
2.4
8.1 23.2
64.5
4.46
211
5.3
10.2 20.9 26.2
37.4
3.80
206
5.7
3.3 13.3 33.8
43.8
4.07
210
4.3
9.5 15.2 34.8
36.2
3.89
210
9.1
16.7 29.2 28.2
16.7
3.27
209
13.9
18.8 26.0 20.7
20.7
3.15
208
Source: MEP 2006 survey.
of your work?’). Aggregate results for the six response categories offered in the survey are shown in Table 5.5. These items produce somewhat clearer distinctions than our previous set of questions: while the mean average for all the responses categories is above the mid-point of the 5-point scale, there is a substantial and significant gap between the average importance accorded to the highest rated item (‘legislating’, which has a mean rating of nearly 4.5 of 5) and that given to the lowest rated matters (‘representation of individual interests of individual citizens’, and ‘mediation between different interests in society’, which attain average ratings of 3.15 and 3.27 respectively). While almost 90 per cent of MEPs indicate that they rate legislative work above the mid-point on the 5-point scale, onethird of them score the representation of individual citizens below the mid-point. Their duties in parliament, and the representation of broad social interests, appear to most MEPs as being clearly of
106
Representation in Europe: The Institution and the Individuals
greater importance than representing the smaller-scale concerns of individual citizens. Beneath these aggregate responses, however, how do MEPs differ in their individual attitudes? Are there any observable patterns in terms of different types of MEPs responding differently? To explore such matters, we examined the bivariate correlations between the individual survey items; the full correlation matrix of all the items from Tables 5.4 and 5.5 is presented in Table 5.6. The results here show a large number of significant correlations between items. Such results are not unexpected; it is intuitively unsurprising that MEPs’ attitudes to various aspects of their role should be empirically associated. Taken together, the results are suggestive of three broad factors along which MEPs’ attitudes seem to vary. First, we can observe in the table several fairly strong positive correlations (a Pearson’s r coefficient greater than 0.3) among items that seem to relate to variation among questions that tap into attitudes concerning core parliamentary activities of legislating, parliamentary oversight, and policy development. Second, we observe the strongest individual correlations among items that appear to reflect varying degrees of concern with Europe-wide matters: EU policy development and the representation of both the EP party group and all people across Europe. Third, we can see strong correlations also for attitudes related to local and national representation: a positive relationship between attitudes towards the representation of the constituency/region, voters in the member state, and an MEP’s national party. To test the validity of this interpretation of the correlations, the 12 items in Tables 5.1 and 5.2 were subjected to a principal components factor analysis.11 Results from this analysis are reported in Table 5.7. These results confirm that three main factors account for the majority of variance in MEPs’ responses to these survey items, and that no other factors approach the same level of importance;12 the results also strongly indicate that the three factors can be interpreted in the manner already suggested. The rotated factor loadings reported in the table are labelled to summarize the variables that load most heavily on them. The first factor
107
Parliamentary Articulation Developing Mediation Representation All All All people All people My My EP oversight of important common between of individual people people in my who national party societal strategies different interests of in in my constituency/ voted party group needs and for EU interests individual Europe member region for my interests policies in society citizens state party Working on legislation Parliamentary oversight Articulation of important societal needs and interests Developing common strategies for EU policies Mediation between different interests in society Representation of individual interests of individual citizens All people in Europe All people in my member state All people in my constituency/region All people who voted for my party My national party
.35∗
.35∗ .33∗
.38∗ .28∗
.30∗ .26∗
.10 .20∗
.25∗ .23∗
.10 .04
.15 .20∗
−.00 .23∗
.02 .08
.16 .14
.41∗
.28∗
.10
.21∗
.14
.07
.03
.15
.17
.53∗
.03
.63∗
.17
−.07
−.07
−.05
.33∗
.25∗
.37∗
.13
−.01
.02
.04
.37∗
.34∗
.23∗
.17
.10
−.10 .21∗
−.05 .21∗
−.05 .26∗
.37∗ .19∗
.40∗
.35∗
.22∗
.57∗
.25∗
−.09
.06
.38∗
.27∗
Representation in Europe: The Institution and the Individuals
108
Table 5.6. Correlations of MEP individual role items
Representation in Europe: The Institution and the Individuals Table 5.7. Rotated factor loadings for MEPs’ responses to attitudinal items (varimax rotation) Item
Role Legislating Oversight Societal needs/interests Policy strategies Social mediation Individual citizens
Parliamentarism
Local/Nat. representation
European orientation
.68 .74 .61 .54 .58 .47
−.02 .15 .09 −.21 −.05 .41
.11 .09 .16 .63 .44 −.28
Representing All in Europe All in member state National party voters People in const/region National party EP party group
.22 −.04 .02 .17 −.03 .18
−.21 .36 .81 .72 .78 .33
.81 .62 .10 −.09 .14 .58
% of variance explained
19.54
19.20
17.69
is thus labelled ‘Parliamentarism’: scores on this factor, which accounts for slightly under one-fifth of explained variance, are very strongly associated with attitudes towards traditional parliamentary activities—legislating, representing individual interests, and parliamentary oversight, as well as with policy development and interest mediation. The second factor, which we label ‘Local/National Representation’, is most strongly associated with three survey items (the representation of national party voters, voters in the constituency/region, and representation of the national party itself), and also moderately strongly with according importance to the representation of individual citizens as a feature of an MEP’s work. This factor, like the first identified, also accounts for almost 20 per cent of variance in MEPs’ responses to these surveys items. The third factor, responsible for another 18 per cent of variance, is labelled ‘European Orientation’: this represents differences among MEPs with regard to the importance (or otherwise) of representing ‘all people in Europe’ and the European party group, and
109
Representation in Europe: The Institution and the Individuals
developing EU-wide policies, although the item for representing ‘all people in my member state’ also loads quite strongly on this factor. Specifying the main dimensions of variation in MEPs’ attitudes towards their role as representatives does not, however, do much to explain the reasons for such differences. To explore the reasons behind variation in MEPs’ attitudes, we develop a multivariate analysis, with the dependent variables being each representative’s factor scores on the three factors extracted in our previous factors analysis. (High scores on each of the factors equate to high scores on the items that load most heavily on that factor. Thus, a high score on the Parliamentarism factor would mean that a respondent probably offered ‘Great Importance’ responses for the survey items on ‘legislating’ and ‘parliamentary oversight’, a high score on the European Orientation factor that a respondent offered such responses for items like the ‘All People in Europe’ one.) We specify a very similar set of independent variables to those deployed in the analysis conducted in Section 5.1 of the chapter. An identical set of individual variables are included. The political variables are also identical, except that we additionally include our scale of attitudes to the empowerment of the EP, in order to test whether views on the representative role of the institution influence MEPs’ views on their own representative role. Finally, we also include identical electoral-institution, and national variables to those deployed previously, to see whether the incentives promoted by electoral systems, or broad national differences in political culture, might be important in explaining differing interpretations of the representative role among MEPs. The results of OLS regression analysis on our three dependent variables are presented in the three panels of Table 5.8. The results generally show a fairly modest degree of fit for the models specified, indicating that it is not very easy to explain individual, subjective attitudes to the role of elected representative with the broad, objective criteria employed here. Nonetheless, several important and interesting findings emerge.
110
Representation in Europe: The Institution and the Individuals Table 5.8. Regression estimates (standard errors) for three representation factors Variables
Model 1
Model 2
Model 3
(a) Parliamentarism Factor Age −.01 (.01) −.01 (.01) −.01 (.01) Gender .23 (.16) .16 (.16) .29 (.16)∗ Years as MEP .03 (.01∗∗ .04 (.01)∗∗ .03 (.01)∗∗ Former nat. MP −.03 (.17) .04 (.17) .03 (.17) Former nat. minister .08 (.23) .04 (.25) .08 (.23) Left–right stance .05 (.04) Left–right stance (folded) −.00 (.07) Integration self-placement .03 (.04) EP powers scale .04 (.02)∗∗ STV/list 1.04 (.53)∗ Ballot structure −.21 (.10)∗∗ District magnitude .00 (.00) Belgian British Cypriot Czech Danish Estonian Finnish French German Greek Hungarian Irish Italian Latvian Lithuanian Luxembourg Maltese Dutch Polish Portuguese Slovak Slovene Spanish Swedish (Constant) Adjusted R2 N
.20 (.42) −1.20 (.68) .02 .06 182 160
(b) Local/National representation factor Age −.02 (.01)∗ −.01 (.01) Gender −.16 (.16) −.15 (.16) Years as MEP −.00 (.01) .00 (.01)
.25 (.43) .04 182 −.02 (.01)∗ −.18 (.16) .00 (.01)
Model 4
Combined Model
−.01 (.01) −.01 (.01) .30 (.17)∗ .19 (.16) .04 (.02)∗∗ .03 (.01)∗∗ −.07 (.20) .11 (.17) .00 (.27) .08 (.24) .07 (.04)∗ .02 (.07) .05 (.04) .05 (.02)∗∗∗ 1.31 (.97) −.28 (.10)∗∗∗ −.00 (.00) −.53 (.58) −.47 (.50) .00 (.74) −.57 (.61) −.66 (.65) .07 (.75) −.55 (.63) −.29 (.52) −.13 (.51) .57 (.85) −.78 (.85) .17 (.75) −.78 (.53) −.34 (.70) −.48 (.73) .16 (.88) 1.57 (1.11) −.02 (.62) −.07 (.53) .19 (.64) .59 (.84) .15 (.64) .50 (.59) −.61 (.59) .49 (.65) −1.44 (.68) .02 .10 182 160 −.01 (.01) −.20 (.17) −.02 (.02)
−.01 (.01) −.20 (.17) .00 (.02) (Continued)
111
Representation in Europe: The Institution and the Individuals Table 5.8. (Cont.) Variables
Model 1
Model 2
Model 3
Model 4
Former nat. MP Former nat. minister Left–right stance Left–right stance (folded) Integration self-placement EP powers scale STV/list Ballot structure District magnitude Belgian British Cypriot Czech Danish Estonian Finnish French German Greek Hungarian Irish Italian Latvian Lithuanian Luxembourg Maltese Dutch Polish Portuguese Slovak Slovene Spanish Swedish
.00 (.17) .44 (.23)∗
−.02 (.17) .23 (.25) −.00 (.04) −.07 (.07)
−.05 (.17) .39 (.23)
−.17 (.20) .42 (.27)
(Constant) Adjusted R2 N
.77 (.43) .02 182
−.14 (.04)∗∗∗ −.02 (02)
2.01 (.70) .12 160
−.09 (.18) .16 (.25) −.01 (.04) −.10 (.07) −.15 (.04)∗∗∗
.24 (.53) .09 (.10) −.00 (.00)
.79 (.44) .01 182
−.70 (.58) −.07 (.50) −.94 (.74) −.57 (.61) −.38 (.64) .46 (.75) −.68 (.63) −1.00 (.52)∗ −.57 (.50) .42 (.85) −.25 (.84) .24 (.74) −.33 (.53) −.66 (.69) −.13 (.72) .05 (.88) −.54 (1.10) −.70 (.62) −.90 (.53)∗ −.07 (.64) .22 (.83) −.30 (.64) .04 (.58) .10 (.59) 1.22 (.64) .03 182
(c) European orienatation factor Age .02 (.01)∗∗∗ .02 (.01)∗∗ .02 (.01)∗∗∗ .02 (.01)∗∗∗ Gender .14 (.15) .09 (.16) .13 (.15) .07 (.14) Years as MEP −.03 (.01)∗∗ −.05 (.01)∗∗∗ −.03 (.01)∗∗ −.01 (.01) Former nat. MP .11 (.16) .18 (.17) .05 (.17) −.19 (.16) Former nat. minister .35 (.22) .27 (.25) .35 (.22) .41 (.22)∗ Left–right stance .00 (.04) Left–right stance −.08 (.07) (folded)
112
Combined Model
−.02 (.02) −.92 (1.01) .19 (.10)∗ −.00 (.00)
2.20 (.71) .13 160 .02 (.01)∗∗ .09 (.16) −.02 (.01) −.09 (.17) .37 (.26) −.04 (.04) −.08 (.07)
Representation in Europe: The Institution and the Individuals Table 5.8. (Cont.) Variables
Model 1
Integration self-placement EP powers scale STV/list Ballot structure District magnitude Belgian British Cypriot Czech Danish Estonian Left–right stance Left–right stance (folded) Integration self-placement EP powers scale STV/list Ballot structure District magnitude Belgian British Cypriot Czech Danish Estonian Finnish French German Greek Hungarian Irish Italian Latvian Lithuanian Luxembourg Maltese Dutch Polish Portuguese Slovak Slovene Spanish Swedish (Constant) Adjusted R2 N ∗
Model 2
.15 (.04)∗∗∗ .01 (.02)
.00 (.04) −.08 (.07) .15 (.04)∗∗∗ .01 (.02)
Model 3
Model 4
Combined Model .07 (.04)∗∗ −.02 (.02)
−.10 (.50) .25 (.10)∗∗∗ .01 (.00)∗∗
−.70 (.47) −.78 (.56) −2.01 (.41)∗∗∗−1.93 (.48)∗∗∗ −.09 (.60) .24 (.72) −.12 (.49) −.14 (.54) .09 (.52) .05 (.57) −.28 (.60) −.34 (.66) −.04 (.04) −.08 (.07) .07 (.04)∗∗ −.02 (.02)
−.10 (.50) .25 (.10)∗∗∗ .01 (.00)∗∗
−1.09 (.41) −1.90 (.69) −1.37 (.41) .07 .17 .12 182 160 182
−.70 (.47) −.78 (.56) −2.01 (.41)∗∗∗−1.93 (.48)∗∗∗ −.09 (.60) .24 (.72) −.12 (.49) −.14 (.54) .09 (.52) .05 (.57) −.28 (.60) −.34 (.66) −.36 (.51) −.43 (.57) −.43 (.42) −.30 (.49) −.68 (.41)∗ −.65 (.47) −.06 (.69) −.43 (.94) .19 (.68) .04 (.73) −.55 (.60) −1.29 (.94) −.50 (.43) −.45 (.49) −.83 (.56) −.80 (.61) .39 (.58) .47 (.62) −.69 (.71) −.82 (.76) −.59 (.89) (dropped) −.04 (.50) .06 (.58) −.74 (.43)∗ −.62 (.48) .22 (.52) .07 (.58) .08 (.67) .11 (.70) −.49 (.52) −1.23 (.72)∗ −.09 (.47) .05 (.54) −1.34 (.47)∗∗∗−1.36 (.53)∗∗ .44 (.52) .37 182
.12 (.86) .39 160
< .10; ∗∗ < .05; ∗∗∗ < .01.
113
Representation in Europe: The Institution and the Individuals
As far as our parliamentarism factor is concerned, we observe some impact for the individual variables specified. A minor impact is observed for gender; but a much stronger and consistent association can be seen for the variable specifying experience in the EP. Previous work (Scully 2005) and our earlier findings have shown that length of service in the EP does not have any consistent relationship with MEPs’ support for greater EP powers, or for closer integration in general. However, it does correlate positively with according importance to traditional parliamentary activities. Our political variables have no notable impact except for the variable measuring MEPs’ positions on empowering the chamber further: supporting a powerful EP is positively associated with according importance to the traditional parliamentary activities (activities which would themselves assume even greater importance in a more powerful EP). And while the national dummy variables fail to identify any meaningful national differences among MEPs, there is a definite impact for our electoral-institutional variables, and particularly ballot structure. The coefficient for this variable indicates a strong tendency among those MEPs from less ‘open’ systems to accord greater levels of importance to traditional parliamentary activities. For the Local/National Representation factor, we observe no major impact for our individual variables. Nor do our political variables wield much empirical leverage, with the exception of general attitudes to European integration—where the sign of the coefficient indicates that those according a high degree of importance to representing their locality and nation tend to be those scoring lower on enthusiasm for closer integration. Our national dummy variables also do little to explain variance in MEPs’ views, with only two moderately significant coefficients (for France and Poland, suggesting that MEPs from these countries accord less importance to local and national representation). And there is also only a moderate impact for our electoral-institutional variables, with only one significant coefficient here—although one in a theoretically plausible direction, with MEPs from more ‘open’ systems according greater importance to representing their locality and nation.
114
Representation in Europe: The Institution and the Individuals
For our European Orientation factor, we observe a positive association with age: MEPs from older generations tending to accord greater importance to European-level representation and policy development. Among our political variables, the only important association is with general attitudes to integration; here, the nature of the relationship is the intuitively expected one, with those more favourable to integration scoring higher on European Orientation. We also observe an impact for two of our electoral-institutional variables. A more open ballot structure and a greater district magnitude are positively associated with European Orientation. However, there is a large increase in model fit when we include national dummy variables, with particular associations that persist even in a model that combines these variables with both the personal and political ones. In particular, we see British and Swedish representatives being disposed not to adopt a European Orientation. To summarize, the findings of this section of the chapter show several things. European parliamentarians have strong views on their role as representatives. Although aware of responsibilities to represent multiple ‘principals’, they prioritize between different representative activities. The differences between MEPs in the priorities they set as representatives are to some degree explicable by systematic, measurable variables. And among those variables that appear to feature strongly are the electoral systems under which MEPs are elected.
5.3. Conclusions This chapter has investigated MEPs attitudes to the representative role of their institution, the EP, and also to their own individual representative role. In both areas our previous knowledge was sparse and we cannot claim to have offered the final word. Nonetheless, our exploratory analysis has produced some interesting findings, both in regard to the landscape of MEPs’ attitudes, and concerning the factors that help explain differences in those attitudes. Most MEPs want to work in a powerful EP, and most of
115
Representation in Europe: The Institution and the Individuals
them have fairly clear attitudes regarding the different potential priorities within their own role. Among the factors that are prominent in explaining these attitudes, particularly with regard to their own role, are the institutional arrangements under which MEPs have to compete for election. Having established in this chapter that electoral institutions have a role to play in this regard, the objective of Chapter 6 is to assess how significant a role that is. Therefore, Chapter 6 will conduct a more detailed assessment of the exact nature of how electoral system differences can, and do, shape how MEPs understand and seek to carry out their role as elected representatives.
Notes 1. Previous surveys of MEPs offer a very similar picture to that presented here, indicating that MEPs are generally supportive of further empowerment of their institution, but not uncritically so. Moreover, the differences between their attitudes and those of national MPs on such questions are generally very small (Scully 2005: ch. 5). 2. The Cronbach’s alpha coefficient of reliability for a scale of all the seven items is .70. While this indicates that the seven items form an acceptable scale—the general rule is that items are regarded as forming a reasonable additive scale if the alpha coefficient is at or exceeds .7 (Pennings et al. 1999: 96–98)—the scale value rises substantially, to .79, if the item on the location of EP plenaries is excluded. 3. The other responses are coded as follows: ‘Agree’ scores 4, ‘Neither agree nor disagree’ scores 3, and ‘Disagree’ scores 2. Where respondents had completed three or more of the scale items, missing cells were filled with the mean value on an item; where respondents had failed to complete at least three items, they were excluded from the analysis. 4. The question asked: ‘Where would you place yourself on the left-right spectrum?’ Respondents were invited to tick one box on a 10-point scale (1 = ‘left’; 10 = ‘right’). 5. The question asked: ‘Where would you place yourself on the question of European integration?’ Respondents were invited to tick one box
116
Representation in Europe: The Institution and the Individuals
6. 7. 8.
9.
10.
11. 12.
on a 10-point scale (1 = ‘European integration has gone much too far’; 10 = ‘The EU should become a federal state immediately’). The national values for all the codings on the electoral system variables are as indicated in Chapter 4. The comparison category—for which no national dummy variable is specified in any model—is Austria. Because of collinearity between our electoral institutional variables and the national dummy variables, our combined models include either the institutional or the national variables, depending on which the earlier analyses have indicated have the greater explanatory leverage. That length of service in the EP is not a significant predictor of attitudes towards the powers of the EP fits broadly with a range of empirical findings consistently produced from other evidence in previous research. (For a detailed discussion, see Scully 2005). Interestingly Katz (1997) does find three factors among responses to a battery of eight questions investigating MEPs’ and candidates’ role priorities: these distinguish between representatives’ attitudes to the importance of national or European priorities, representing constituency interests (an ‘agent’ role), and the use of representatives’ own judgement (a ‘trustee’ orientation). On parliamentary norms in the EP, see also Bowler and Farrell (1999). The principal components analysis employed varimax rotation, with listwise deletion of missing cases. The three factors for which loadings are reported in Table 5.7 were the only ones attaining an eigen value above 1.0 and the only ones accounting for more than 10 per cent of variance in MEPs’ reported attitudes.
117
This page intentionally left blank
6 Electoral Institutions and Geographical Representation
Chapters 4 and 5 in this study have, respectively, examined in detail the electoral systems used for elections to the EP and considered how members of that institution understand and interpret their representative role and that of the parliament. The analysis in Chapter 5 concluded with some indications that electoral systems are among the principal factors that influence MEPs’ orientations towards key aspects of the representative process. This chapter draws together the threads of these previous discussions. Specifically, this chapter is devoted to examining whether there is an observable relationship between, on the one hand, the institutions under which MEPs are elected, and on the other, the manner in which they seek to represent those to whom they owe their election. It was noted in Chapter 2 that, to the extent that they can be deduced, public preferences appear to be strongly in favour of a style of representation at the European level that is visible and that helps to connect citizens to the EU. In this chapter, we assess the degree to which such representation is actually being provided by MEPs—and whether electoral arrangements help or hinder in the provision of such practices. The chapter proceeds in three main sections. The first examines some broad descriptive data, drawn mainly from our 2006 survey of MEPs, about the extent to which MEPs actually engage in various forms of what we term ‘geographical representation’
119
Electoral Institutions and Geographical Representation
(i.e. representation on the ground; referred to in the UK as ‘constituency representation’), and their attitudes towards such matters. This is then followed, in Section 6.2, by a discussion of the electoral systems used for EP elections in relation to their expected consequences for MEPs’ attitudes and behaviour. This section builds on discussion in Chapter 3 to establish some specific observable implications of electoral system differences that are then tested in Section 6.3 of the chapter. Section 6.4 deploys multivariate statistical analysis of the survey data to consider whether electoral institutions are indeed linked, in some systematic manner, to the attitudes and behaviour of representatives. The conclusion then summarizes the main findings of the analysis.
6.1. Geographical Representation in the EP As we have previously explained, examining the representative role of MEPs entails a focus on their attitudes and behaviours with regard to the geographic area that they represent, and the individuals within it. In Chapter 3 we referred to this feature of representation as ‘constituency representation’. Given the fact that, apart from those cases such as Ireland that have adopted small regional ‘constituencies’ most member states have opted for national lists, it would be misleading to apply this term here. Instead, we prefer to refer to this as geographical representation (or representation of individual voters on the ground), as opposed to thematic or functional representation (which tends, on the whole, to privilege organized interests). What would constitute evidence of a strong degree of geographical or constituency focus among MEPs? Drawing on the discussion in earlier chapters, we suggest that several features of MEPs’ attitudes and behaviour (two of which were outlined in Chapter 5) can be regarded as relevant. Election Campaigning. The first area that we can look at is evidence relating to the Election Campaign. The campaign prior to winning election is an obvious starting point for a putative
120
Electoral Institutions and Geographical Representation
representative to begin building a favourable profile with their constituents. Given that electoral incentives are generally understood to be a central part of the motivation for geographical representation (referred to by Carey and Shugart (1995) as ‘personal vote chasing’), it is reasonable to regard evidence of a serious commitment of substantial time and other resources to the election campaign as broadly indicative of a greater geographical focus in representation. But we should also pay attention to the particular activities that representatives engage in during the campaign. Do they involve themselves in candidatefocused election activities that might develop an individual profile with those they represent, or is their main concentration elsewhere, such as on relations with their party organization and leadership? The 2006 MEP survey gathered substantial evidence on MEPs’ campaign activities.1 As is shown in Table 6.1, which reports the amount of their own time per week that MEPs state that they spent on campaigning activities in the period immediately before the 2004 EP election, most of them report having spent a substantial amount of time on campaigning. Our 2006 survey of MEPs also enquired about the amount of effort that those elected had devoted to a number of different Table 6.1. Time spent on electioneering by MEPs Response
(%)
Less than 2 hours 2–10 hours 11–20 hours 21–30 hours More than 30 hours
1.4 5.9 5.0 9.1 78.6
N = 220 Note: The question asked was: ‘In the final weeks of the election campaign how much time did you spend per week on the campaign?’ Source: 2006 MEP survey.
121
Electoral Institutions and Geographical Representation Table 6.2. Types and degrees of campaign activity
Telephone canvassing Door-to-door canvassing Direct mail Party meetings Public meetings Press conferences Media relations Campaign website Direct email
A Lot
Some
Very little
None
N
11.9 25.0 29.1 63.6 76.0 39.5 61.6 40.5 19.1
22.9 24.1 40.8 24.8 17.2 42.3 31.1 30.2 26.5
24.8 21.7 16.0 10.3 6.3 15.3 5.9 13.5 24.5
40.5 29.2 14.1 1.4 0.5 2.8 1.4 15.8 29.9
210 212 213 214 221 215 219 215 204
Note: The question asked was: ‘How much effort did you and your team put into the following activities in the 2004 election campaign?’ Source: 2006 MEP survey.
campaign activities. The aggregate results from this question are reported in Table 6.2. The figures show (perhaps surprisingly) a substantial degree of variation across many of the items. While most MEPs report having spent considerable time on speaking to party and public meetings, and engaging in some form of relations with the news media, a significant number are also willing to admit to not having spent a great deal of effort on many typical campaign activities. Of particular interest with regard to the provision of any form of individual representation to local communities is the large number of MEPs who admit to not having engaged much in those activities most likely to develop a personal connection between the representative and their individual constituents (telephone and door-to-door canvassing). In the multivariate analysis developed in Section 6.3 of this chapter, it will be important to examine whether electoral system variables are able to predict systematic differences in both the quantity and the type of electoral campaigning that MEPs engage in. Attitudes to Representation. As we discussed in Chapter 5, there are numerous different activities that MEPs can engage in and prioritize between; there are also many different groups or entities that they can view themselves as representing. Evidence of a high degree of constituency focus among an MEP would be provided by
122
Electoral Institutions and Geographical Representation
that representative according a high priority to the representation of the particular geographic area for which they are responsible, and similarly to according importance to representing the concerns of individual constituents. As was seen in Chapter 5, many MEPs—though not all—do give priority to such activities; the task for later on in this chapter will be to consider whether electoral system differences help explain the likelihood of a representative adopting such attitudes towards representation. Representational Activities. As we also discussed in Chapter 5, once they have been elected MEPs can decide to give time to any of an almost infinite number of activities. But an MEP with a high degree of constituency focus should be expected, ceteris paribus, to spend significant amount of time on political work in their domestic base, rather than concentrating solely or mainly on work inside the EP. We should also expect such MEPs to maintain an active political base (such as a well-resourced office) in their constituency/region; and to be in frequent contact with ordinary, individual citizens (in addition to the many organised groups and political actors that all MEPs will frequently interact with). Tables 6.3 to 6.5, respectively, display aggregate information from the 2006 MEP survey on the extent to which MEPs do, in fact, report spending substantial time on domestic political work, maintain a full-time office of their own within their constituency and other forms of voter contact, and are frequently in contact with ordinary citizens (with additional Table 6.3. Time spent by MEPs in member states Response
(%)
Most of my time each week Some of my time each week Limited time, mostly at weekend Little or no time Varies too much to say N = 214
9.3 48.1 33.2 2.8 6.5
Note: The question asked was: ‘How much time do you spend on political work in your home country rather than work at the European Parliament?’ Source: 2006 MEP survey.
123
Electoral Institutions and Geographical Representation Table 6.4. Forms of voter contact Response
(%)
Access via a permanently staffed office of my own Access via a permanently staffed office shared with other MEPs Personal consultation sessions for individual voters
86.0 8.4 60.3
N = 214 Note: The question asked was: ‘Which of the following forms of contact with individual voters do you have?’ Source: 2006 MEP survey.
information, for the purposes of comparison, on the frequency of MEPs’ contact with other individuals and groups). The task for later on in the chapter will be to consider whether electoralinstitutional factors help explain variance among MEPs on these measures. In this section of the chapter we have shown that there are a large number of potentially observable implications of a high degree of constituency focus among MEPs. We have also demonstrated that there is a substantial degree of variation on measures relating to these observable implications. The task of Section 6.3 of this chapter will be to explore whether electoral-institutional Table 6.5. MEP contacts with organized groups and individuals (%)
Ordinary citizens Organized groups Lobbyists Journalists National party members National party executive National MPs National ministers
At least once a week
At least once a month
At least every three months
At least once a year
Less often/no contact
N
73.1 52.1 40.3 55.5 57.0 36.1 23.8 9.5
21.7 42.2 32.3 38.8 36.2 37.0 48.1 31.4
3.8 12.8 16.4 4.8 4.3 18.3 20.0 30.5
0.9 1.9 5.0 0.5 0.5 4.3 3.3 14.8
0.5 0.9 6.0 0.5 2.0 4.3 4.8 13.8
212 211 201 209 207 208 210 210
Note: The question asked was: ‘How frequently are you in contact with the following groups, people or institutions?’ Source: 2006 MEP survey.
124
Electoral Institutions and Geographical Representation
variables might explain some of this variation. Before we begin that analysis, however, we must first consider exactly how the electoral-institutional variables must be specified for our analysis— revisiting some of the discussion in Chapters 3 and 4. This is the task for Section 6.2 of the chapter.
6.2. The Potential Impact of Electoral Systems The discussion and classification in Chapter 4 of the electoral systems used for EP elections, along with the extant literature on electoral system effects on representation, allow us to develop several possible hypotheses as to the potential link between electoral system differences and MEPs’ attitudes and behaviour. This section of the chapter specifies how we classify the features of electoral systems that we will consider in our empirical analysis. Our earlier discussions (in Chapters 3 and 4) indicated that, for the purpose of examining their effect on styles of representation, there are two major features of electoral system that feature prominently (especially Figure 3.1(b)). The first of these is Ballot Structure, which, as in Chapter 5, can be categorized in two different ways. First, we take account of the basic distinction between the most ‘open’ ballot structure design (STV—used in Ireland, Northern Ireland and Malta) and all list systems (and refer to this in the tables that follow as ‘STV/list’). Our hypothesis with regard to this distinction (see Figure 4.2) would be that the greater scope for personal voting and intra-party competition created by STV ought to generate a greater focus on constituency representation by MEPs elected under these systems. In Chapter 5 (and Figure 4.2) we proposed a more nuanced measure of Ballot Structure, where the systems are categorized into three groupings: open, ordered, and closed. The ‘open’ category includes those states using STV, as well as some list systems that heavily promote candidate-based voting—here the exemplar is Finland in which voters select the candidate of their choice from among the alphabetical lists of candidates on offer and enter the candidate’s
125
Electoral Institutions and Geographical Representation
code on the ballot paper (i.e. there is no prior ranking of candidates by the parties). The polar opposites are the closed-list systems, where all that the voter can do is choose one party list. Short of joining a party and somehow getting involved in the candidate selection process, the voter is powerless to influence the ranking of candidates and thereby the determination of which candidates get elected and in what order. Lying between closed- and open-list systems are ordered-list systems, in which the parties draw up a rank-order of candidates, with the intention that those located towards the top of the list will be elected—the details are provided in Figure 4.1. Unlike closed systems, however, voters can declare a personal vote for individual candidates, giving scope for them to influence the rank-order if a sufficient proportion of them declare a personal vote for one candidate to move that candidate up the rankings. In most cases, however, the ordered-list system takes account of party votes as well as personal votes, and therefore it is rare for a low-placed candidate to accumulate sufficient personal votes to move high enough up the rankings to get into a winnable position.2 Denmark stands out in this regard, because under its electoral law, the parties are able to decide whether to use party votes to top up the personal votes of the candidates at the top of the list (as happens, for instance, in Belgium and the Netherlands), or to allocate seats to candidates according to the number of personal votes each receives (in essence, what happens in Finland).3 In the past, the Danish Socialist People’s Party tended to take the first option, but in recent elections it has fallen in line with all the other parties in opting for what is in essence the open-list version (albeit one where the very act of parties putting up ordered lists still connotes an advantage for those candidates ranked higher on the ballot paper). Given its similarity with the Finnish system, we have located Denmark (and Estonia) in the cluster of ‘open systems’. Similarly, since candidates’ personal votes matter greatly in Irish, Maltese, Italian, Luxembourg, and Lithuanian elections, we also include them on the open cluster (see also Bardi 1987). Our working hypothesis is that we should expect to find significant
126
Electoral Institutions and Geographical Representation
differences in ‘geographical representation’ by MEPs depending on which of the three electoral system clusters they were elected under, with the relationship being broadly linear—from higher levels of such representation under open systems, through more modest levels under ordered systems, and lower still in closed systems. The second important dimension for analysing potential electoral systems effects on representative roles is District Magnitude— the size of the area they are representing. Any individual voterorientation promoted by a more open ballot structure might be limited in the case of national-list systems (which inevitably require a large district magnitude). For this reason, earlier studies have tended to place great stress on district magnitude; indeed, if anything this has featured even more prominently in previous research than has ballot structure (e.g. Cox 1997; Wessels 1999), and close analysis of representative roles has found evidence of significant differences based on district magnitude (M) or size of region characteristics (Bowler and Farrell 1993; Cowley and Lochore 2000; Lancaster and Patterson 1990; Lundberg 2002). At this juncture it is important to remember how we deploy the concept of district magnitude. The standard measure of M is based on the level at which seats are allocated to the parties; in Ireland, for example, the seats are allocated to the parties in a given multi-member constituency based on their votes in that constituency, and therefore Ireland’s M varies across each of its four constituencies (M averages 3.8 across the country). By contrast, in Poland seats are allocated based on parties’ national votes, and so there is just one M (= 54) for the whole country. In Italy, however, this picture is somewhat complicated by the fact that while seats are allocated nationally, the national vote proportions are based on aggregate party votes from each of Italy’s five regions (producing a national average M of 17.4), and the seats are allocated to the parties’ candidates based on their rankings on the regional list and (crucially for our discussion) the personal votes received in the region. For this reason, it is more appropriate
127
Electoral Institutions and Geographical Representation
in our analysis to treat Italian MEPs as if they are elected regionally. In their theoretical modelling of electoral system effects on incentives for politicians to cultivate a personal vote, Carey and Shugart propose a modification of the relationship between M and MPs’ personal vote-chasing activities, based on the degree of openness of the ballot structure: their central idea is that a non-linear relationship is likely to operate, with a representative’s ‘personal reputation’ being worth less and less as M rises in closed systems, but more and more in open systems (1995: 431). This suggests a lower emphasis on ‘constituency service’ activities in the closed/national-list cases, but also implies the need for an interactive term in multivariate analysis that combines M with electoral system type. In the analysis that follows we therefore include variables for ballot structure (closed, ordered, and open) and district magnitude (variations in M) separately and in combined interaction terms. But in order to provide an additional test of how these variables, operating in conjunction with each other, may interact with politicians’ representative roles, we also take account of recent work by (Carey and Shugart 1995; Shugart 2001) to develop an index of ‘intra-party efficiency’. This index takes account of ballot structure and district magnitude variations to categorize the EP’s electoral systems in terms of a single measure. The index is based on three main characteristics that Shugart terms Ballot, Vote, and District, in which higher scores across these components are indicative of a candidate-centred ordinal system and lower scores of a partycentred categorical system (see also Farrell and McAllister 2006: ch. 7). The Ballot component measures the degree of party versus voter control over the ballot placement of candidates. This characteristic incorporates features of electoral system design as well as parties’ candidate selection rules. The rationale here is that the lower the extent of party control, the greater the potential incentive for candidates to develop personal reputation. The coding is as follows:
128
Electoral Institutions and Geographical Representation
1. Ballot access dominated by parties, and voters may not disturb order of list; 2. Ballot access dominated by parties, but voters may disturb list; 3. Ballot access nearly unrestricted. The Vote component distinguishes between systems that require voters to vote for party lists or candidates (a nominal vote). Following Farrell and McAllister (2006), our coding of this component differs from Shugart’s with regard to where to locate STV and open list systems. Accordingly, our adapted Vote coding is as follows: 1. Vote for list only; 2. Vote is list or nominal, but list votes predominate; 3. Vote is nominal or list, but nominal votes predominate and pool to other candidates; 4. Vote is nominal only, but vote may pool or transfer to other candidates. Finally, the District component takes account of the potential effect of district magnitude, and the likelihood that this can vary depending on the nature of the ballot structure. In systems where voters cast party-based votes, they find that the personal reputation of the candidate declines in significance as district magnitude rises, whereas in systems characterized by candidatebased (nominal) votes, as district magnitude rises and candidates face more inter- and intra-party competitors, the incentives for personal vote chasing increase. This characteristic is coded as follows: 1. District magnitude is greater than one, with Vote < 3; 2. District magnitude is greater than one, with Vote > 2, provided that Ballot > 1. Using this coding scheme, we categorize the 26 electoral systems for EP elections as shown in Table 6.6 (see Appendix Table A.6.1 for details on how the codings for individual cases were derived.). We
129
Electoral Institutions and Geographical Representation Table 6.6. Scoring the EP’s electoral systems on the intra-party dimension Index Component scores
Description of system
9 8 7
3, 4, 2 2, 4, 2 2, 3, 2
5
2, 2, 1
STV Quasi-list Open list, panachage Ordered list
3
1, 1, 1
Closed list
Cases
Ireland, N. Ireland, and Malta Finland Denmark, Estonia, Italy, Lithuania, and Luxembourg Austria, Belgium, Cyprus, the Czech Rep., Latvia, the Netherlands, Slovakia, Slovenia, and Sweden Britain, France, Germany, Greece, Hungary, Poland, Portugal, and Spain
Number of cases in 2006 MEP survey 7 5 56 74
201
include this modified Shugart index, along with the other electoral system variables indicated above, in the multivariate analysis that follows.
6.3. Analysing Electoral System Effects As indicated in Section 6.1 of the chapter, there are multiple observable implications of geographical representation among MEPs for which we have available survey evidence. We do not propose to try to isolate one or two of these observable implications as the most important and examine only them; rather, we have chosen to run a parallel series of explanatory multivariate models on numerous dependent variables, and to then examine the overall patterns emerging from these multiple indicators. The electoral system-related variables included in the analysis are specified as follows. First (and following the practice in Chapter 5), we include two sets of variables for ballot structure differences: (a) our dichotomous ‘STV/list’ measure (coded ‘1’ for an MEP elected under STV, ‘0’ otherwise); (b) our more nuanced
130
Electoral Institutions and Geographical Representation
ballot structure variable (based on Figure 4.2), which we have coded as two dummies: ‘1’ for those elected under open and ordered systems, ‘0’ otherwise, with MEPs elected under closed systems serving as the comparator category. Second, we include a variable for the average district magnitude for each of our cases.4 Third, to allow for the possibility that there is an important interrelationship between ballot structure and district magnitude, we include two interaction terms, where our measure of M is combined with our main (nuanced) ballot structure variable. Fourth, we also specify a variable for our modified Shugart index, coded in the manner indicated previously. And finally, we include a dummy variable for MEPs from Britain—to capture the possibility that there might be a persisting culture of constituency representation here that is not present in many of the other member states operating closed electoral systems for EP elections. For each dependent variable, we specify three versions of the multivariate model. The first includes only our basic measures of STV/list, ballot structure, and district magnitude, plus the dummy variable for British MEPs. The second model includes all of these variables, plus our modified Shugart index. The final model omits the Shugart index but includes the interaction terms combining ballot structure with district magnitude. These models are specified for eight different dependent variables, which draw on the indicators presented in Section 6.1 of the chapter. Three of the dependent variables are concerned with campaigning activities: the amount of time an MEP reports having put into campaigning, and the amount of effort they placed on telephone and door-to-door canvassing. Two further dependent variables are concerned with MEPs’ attitudes to the representation of individual constituents. The final three dependent variables concern their reported behaviours once in office: the amount of time that they spend on political work at home, and whether their have an individual office of their own, and whether they conduct personal consultation sessions for individual voters. OLS regression estimates for the explanatory
131
Electoral Institutions and Geographical Representation
models applied to these several dependent variables are reported in the multiple panels of Table 6.7. The general expectations concerning relationships between our independent variables and the various dependent variables specified should by now be clear. If more open electoral systems do promote greater geographical representation and a more active effort by MEPs to project an individual presence, we should observe such patterns in the results presented in Table 6.7. Thus, we should expect MEPs from more open systems to campaign more vigorously in elections, and particularly to engage in those types of campaigning that connect them to individual voters. We should also expect them to accord greater importance to representing individual constituents, and to spend more time and effort on domestic political work and in upholding their regional presence. If such patterns do generally prevail, then their manifestation in our respective independent variables may be somewhat complex, due to both the close interrelationships between some of our predictor variables (notably between that for our two main ballot structure variables) and because the specification of interaction terms can complicate the interpretation of the original ‘main effects’ variables (Black 1999: 488–513). But the broad understanding of the hypothesized relationship is apparent. The various models for which results are presented show a generally modest ‘fit’: electoral system effects do not come anywhere close to wholly accounting for MEPs’ attitudes and behaviour— nor would we have expected them to. Nonetheless, there are interesting and important results for some of our electoral system variables. Panel A shows that the only variable having a consistent impact on the amount of election campaigning conducted is the national dummy for British MEPs—the sign of the coefficient indicating them to have campaigned harder than most of their counterparts from other member states—although the coefficients for ordered systems (both the main effects coefficient and the interaction) are also modestly significant in our third model. But we see, as suggested above, much stronger effects when we look at the type
132
Electoral Institutions and Geographical Representation Table 6.7. Regression estimates (standard errors) for eight dependent variables Variables
Model 1
Model 2
Model 3
Panel A: Time spent on campaigning in final weeks of 2004 EP election campaign STV/list Open system Ordered system District magnitude British Modified Shugart index Open∗ M interaction Ordered∗ M interaction (Constant) Adjusted R2 N
0.47 (0.37) 0.06 (0.20) −0.24 (0.19) 0.00 (0.00) 0.51 (0.23)∗∗
0.81 (0.90) 0.80 (10.81) 0.13 (0.89) 0.00 (0.00) 0.51 (0.23)∗∗ −0.18 (0.44)
0.91 (0.56) −0.60 (0.60) −0.85 (0.34)∗∗ 0.00 (0.00) 0.456 (0.23)∗∗ 0.04 (0.04) 0.04 (0.02)∗∗
4.47 (0.18) 0.04 220
5.02 (10.32) 0.04 220
4.53 (0.18) 0.06 220
Panel B: Effort made on telephone canvassing in 2004 EP election campaign STV/list Open system Ordered system District magnitude British Modified Shugart index Open∗ M interaction Ordered∗ M interaction (Constant) Adjusted R2 N
0.356 (0.42) 0.13 (0.22) −0.36 (0.20)∗ −0.01 (0.00)∗∗ 0.88 (0.25)∗∗∗
2.21 (0.19) 0.19 210
−0.01 (0.95) −0.65 (1.90) −0.74 (0.93) −0.01 (0.00)∗∗ 0.88 (0.25)∗∗∗ 0.19 (0.46)
1.45 (0.61)∗∗ −1.33 (0.64)∗∗ −0.48 (0.37) −0.01 (0.00)∗∗∗ 0.85 (0.24)∗∗∗
1.64 (1.38) 0.19 210
2.25 (0.19) 0.21 210
0.10 (0.04)∗∗ 0.01 (0.02)
Panel C: Effort made on door-to-door canvassing in 2004 EP election campaign STV/list Open system Ordered system District magnitude British Modified Shugart index Open∗ M interaction Ordered∗ M interaction (Constant) Adjusted R2 N
1.58 (0.42)∗∗∗ −0.54 (0.23)∗∗ −0.98 (0.21)∗∗∗ −0.01 (0.00)∗∗∗ 0.36 (0.26)
3.00 (0.20) 0.23 212
4.40 (0.98)∗∗∗ 5.67 (1.98)∗∗∗ 2.03 (0.98)∗∗ −0.01 (0.00)∗∗∗ 0.36 (0.26)∗∗ −1.51 (0.48)∗∗∗
2.54 (0.63)∗∗∗ −1.86 (0.67)∗∗∗ −1.19 (0.40)∗∗∗ −0.01 (0.00)∗∗∗ 0.33 (0.26)
7.51 (1.45) 0.26 212
3.04 (0.21) 0.24 212
0.09 (0.04)∗∗ 0.01 (0.02)
Panel D: Importance accorded to representing all people in constituency/region STV/list Open system
0.57 (0.44) −0.01 (0.23)
0.74 (1.00) 0.36 (2.00)
0.11 (0.65) 0.62 (0.68) (Continued)
133
Electoral Institutions and Geographical Representation Table 6.7. (Cont.) Variables Ordered system District Magnitude British Modified Shugart index Open∗ M interaction Ordered∗ M interaction (Constant) Adjusted R2 N
Model 1 −0.17 (0.22) 0.00 (0.00) 0.49 (0.26)∗
4.26 (0.20) 0.02 203
Model 2 0.01 (0.99) 0.00 (0.00) 0.49 (0.26)∗ −0.09 (0.48)
4.53 (1.46) 0.02 203
Model 3 −0.16 (0.38) 0.00 (0.00) 0.50 (0.26)∗ −0.04 (0.04) 0.00 (0.02) 4.25 (0.20) 0.02 203
Panel E: Importance of representing citizens’ individual interests STV/list Open system Ordered system District magnitude British Modified Shugart index Open∗ M interaction Ordered∗ M interaction
2.26 (0.60)∗∗∗ 0.14 (0.29) 0.21 (0.26) 0.01 (0.00)∗∗ 1.33 (0.32)∗∗∗
(Constant) Adjusted R2 N
2.57 (0.25) 0.12 208
1.46 (1.27) −1.64 (2.50) −0.64 (1.23) 0.01 (0.00)∗∗ 1.33 (0.32)∗∗∗ 0.43 (0.60)
1.29 (1.82) 0.12 208
1.20 (0.83) 1.58 (0.85)∗ 0.22 (0.47) 0.01 (0.00)∗∗ 1.36 (0.32)∗∗∗ −0.10 (0.06)∗ 0.00 (0.02) 2.55 (0.25) 0.13 208
Panel F: Time spent on political work in home country STV/list Open system Ordered system District magnitude British Modified Shugart index Open∗ M interaction Ordered∗ M interaction
0.60 (0.30)∗∗ 0.23 (0.16) 0.29 (0.15)∗ 0.00 (0.00) 0.30 (0.18)∗
(Constant) Adjusted R2 N
2.50 (.14) 0.04 200
0.57 (0.75) 0.17 (1.50) 0.26 (0.74) 0.00 (0.00) 0.30 (0.18)∗ 0.02 (0.37)
1.00 (0.47)∗∗ −0.31 (0.50) 0.28 (0.26) 0.00 (0.00) 0.29 (0.18) 0.04 (0.00) 0.00 (0.01)
2.45 (1.09) 0.03 200
2.51 (0.14) 0.04 200
Panel G: Does MEP maintain full-time office of own? STV/list Open system Ordered system District magnitude British Modified Shugart index Open∗ M interaction
134
0.36 (0.20)∗ 0.08 (0.09) 0.15 (0.08)∗ 0.00 (0.00) 0.28 (0.10)∗∗∗
−0.25 (0.48) −1.23 (0.94) −0.49 (0.47) 0.00 (0.00) 0.28 (0.10)∗∗∗ 0.32 (0.23)
0.13 (0.29) 0.38 (0.29) 0.26 (0.15)∗ 0.00 (0.00) 0.29 (0.10)∗∗∗ −0.02 (0.02)
Electoral Institutions and Geographical Representation Table 6.7. (Cont.) Variables
Model 1
Model 2
0.42 (0.07) 0.02 220
−0.54 (0.70) 0.02 220
Ordered∗ M interaction (Constant) Adjusted R2 N
Model 3 −0.01 (0.01) 0.41 (0.07) 0.02 220
Panel H: Does MEP conduct consultations with individual voters? STV/list Open system Ordered system District magnitude British Modified Shugart index Open∗ M interaction Ordered∗ M interaction
0.24 (.19) 0.06 (0.08) 0.16 (0.08)∗ 0.00 (0.00) 0.11 (0.10)
(Constant) Adjusted R2 N
0.27 (0.07) 0.00 220
∗
−1.13 (0.47)∗∗ −2.87 (0.92)∗∗∗ −1.28 (0.45)∗∗∗ 0.00 (0.00) 0.11 (0.09) 0.72 (0.22)∗∗∗
−1.88 (0.67) 0.03 220
−0.14 (0.28) 0.58 (0.28)∗∗ 0.34 (0.15)∗∗ 0.02 (0.01)∗ 0.13 (0.10) −0.03 (0.02)∗ −0.01 (0.01) 0.25 (0.07) 0.01 220
< 0.10; ∗∗ < 0.05; ∗∗∗ < 0.01
of election campaigning conducted. Efforts to connect with individual voters, via telephone and doorstep canvassing, are strongly predicted not only by the dummy variable for British MEPs but also by the STV/list variable (with those elected under STV much more likely to engage in such activities), and by the more nuanced ballot structure variable and district magnitude once these latter two factors are specified together in an interaction term. More open electoral systems are associated with greater personal campaigning, in particular for MEPs representing larger regions. Our two dependent variables on MEPs’ attitudes towards representing their constituents produce rather inconsistent findings. The model for ‘representing all people in the constituency/region’ has a very poor model fit with only the British dummy variable emerging with a (weakly) significant coefficient. However, when we examine attitudes on the ‘importance of representing citizens’ individual interests’ we find not only a strong positive coefficient for British MEPs, but also a similarly strong effect for the STV/list
135
Electoral Institutions and Geographical Representation
variable in our simple model, and a somewhat weaker effect for our general measure of ‘open’ systems in our fuller model, with those elected under STV/Open systems again according greater importance to this aspect of representation. When we examine our final three dependent variables, which are all concerned with MEPs’ representative behaviour, we find only modest model fits. Nonetheless, on each occasion electoral system variables still emerge as significant predictors in all three cases. Our models for ‘time spent on political work in an MEP’s home country’ produce positive and significant coefficients not only for the British dummy variable, but also in two of the three models for STV/list (again those elected under STV giving more time to domestic political work) and in our first, simple model for open and ordered electoral systems. Our model for whether an MEP maintains a full-time office again produces a positive coefficient for British MEPs, but also, in our simplest model, for STV/list and in our final model (once the interaction terms are included) for open and ordered systems: suggesting that MEPs from such systems are more likely to maintain offices, particularly in smaller district magnitudes. Our model for the ‘conduct of consultations with individual voters’ also produces a somewhat complex pattern of coefficients, but one indicating that such behaviour is engaged in most by MEPs from open (and ordered) systems, and particularly when they operate in smaller district magnitudes. Overall, our findings do not suggest that electoral system-related factors are all-important for shaping the attitudes and behaviours of MEPs with regard to constituency representation. Electoral system factors do not, for instance, explain the persistence of the ‘British effect’ that we have observed in nearly all of our multivariate models. But there are electoral system effects, and they tend to be very much in the direction hypothesized. MEPs elected under more open electoral systems, and particularly STV, are more likely to engage in personal election campaigning, to accord importance to representing individual constituents, and to
136
Electoral Institutions and Geographical Representation
engage in post-election activities that maintain and build their presence among those voters who may shape their re-election prospects.
6.4. Conclusion In this chapter we have explored evidence as to whether electoral system-related factors shape geographical representation in the EP. We began by considering the numerous possible observable implications of greater or lesser concern with this dimension of representation. We also spent some time assessing how electoral system differences might be specified for our empirical analysis. Finally, we conducted multivariate analyses which showed that, while the relationship is not a particularly simple one, there is a clear empirical link between the systems under which MEPs are elected and their orientation towards constituency representation. However, there remain some important factors that our analysis has not adequately explained. One of these is the persistent distinctiveness of British MEPs—elected under a closed-list system, yet tending to accord a high priority to geographical representation. In the following two chapters, we explore why this might be the case.
Notes 1. In the literature, the tendency has been to explore campaign activities at national level (e.g. see Lodge 2005; Maier and Tenscher 2007). This is one of the few instances of analysis of individual-level (MEP) campaign activities. 2. The proportions of personal votes that candidates need to leapfrog up the list and secure a seat vary from one case to the next. This is usually determined by straightforward percentile thresholds, such as: Australia, 7 per cent; the Czech Republic, 5 per cent; the Netherlands, 10 per cent;
137
Electoral Institutions and Geographical Representation and Sweden, 5 per cent. In Belgium, however, the threshold amounts to a d’Hondt quota of the party vote. 3. Lithuania also allows its parties to opt for open or closed lists, and all bar the Labour Party have opted for open lists. 4. For all the analyses reported, we did also try running the models with the measure of district magnitude specified as the natural log of M, rather than M itself. However, this made no substantive difference to any of the important findings, and tended to lower the model fit. We have therefore reported the analyses conducted using M, rather than the logged form.
138
7 Electoral Reform and the British MEP
Chapters 5 and 6 of this book have examined attitudes to political representation among MEPs. In Chapter 5, we explored in general terms the understandings of representation held by MEPs. Then, in Chapter 6, we brought together our previous exploration of the electoral institutions used across EU member states with a specific focus on geographical representation. The empirical findings of Chapter 6 indicate that there is an important relationship between electoral systems on the one hand and the attitudes and behaviour of politicians on the other. The relationship is not a wholly simple or linear one. Nor do electoral systems explain everything about how politicians relate to those they represent. Nonetheless, our findings did strongly indicate that the sort of constituency-focused political representation that we discussed in Chapter 3 tends to be promoted by electoral systems that are more open, and thus provide greater incentives for vote-seeking politicians to develop a visible constituency presence. There is, nonetheless, an obvious shortcoming with the findings presented in Chapter 6. This is that it is difficult to disentangle causes and effects in the connection between electoral systems and styles of political representation. It may be, as discussed in Chapter 3, that particular electoral institutions are chosen in particular countries at least in part because of assumptions about the nature of political representation that are deeply embedded in those political cultures. The practical implication of this is that while
139
Electoral Reform and the British MEP
the attitudes (and associated behaviours) of representatives may be correlated with particular electoral systems, this could occur without the electoral systems actually exerting any direct causal influence on what elected representatives think and do. Thus, any apparent effects of electoral systems would be spurious. A related shortcoming is focused specifically on one case, Britain, whose MEPs constituted a collective outlier. Although they were elected from a closed system, they nonetheless tended to score highly on ‘constituency focus’. At the very least, this anomaly in our findings warrants further investigation. The purpose of this chapter and the next is to address these deficiencies in the previous analysis. We do so by shifting our focus, and examining some very different evidence to that considered in Chapters 5 and 6. Having conducted a large-scale analysis of cross-national data concerning MEPs from 25 countries, we now examine in greater detail evidence regarding one EU member state, Britain (excluding Northern Ireland, which uses STV). There are two principal reasons for focusing so closely on Britain. The first is simply that it was an outlier in Chapter 6, and possible reasons why this was so ought to be explored. (Moreover, it is widely recognized that exceptional cases can often be informative because of their peculiarity (e.g. Rogowski 2004).) The second and more important reason, however, is that Britain provides a very valuable opportunity for investigating the consequences of electoral systems for political representation. The electoral system used in Britain for EP elections was changed radically in 1999, the SMP system being replaced by a PR, regional closed-list system. This change represented a very rare case of fundamental electoral reform in an established democratic political system, and by far the most substantial electoral reform ever introduced for EP elections. Britain, therefore, provides a near-ideal opportunity with which to examine the direct impact of electoral reform on the attitudes and behaviour of elected representatives. This chapter and the next thus explore the repercussions of this electoral reform—both the immediate implications in terms of who was elected, and also the longer-term and (at least for the
140
Electoral Reform and the British MEP
purposes of this study) more important consequences of reform for how those elected understood and sought to carry out their role as representatives. We draw on a mixture of different types and sources of evidence. This includes quantitative data on EP election results before and after electoral reform, and further quantitative data drawn from surveys of MEPs in 1996 and 2000—that is, preand post-reform. But we also explore in detail a large amount of qualitative evidence, which is drawn from face-to-face interviews conducted by the authors with the majority of British MEPs elected in 1999, from interviews with members of their staff, and also from visits to and observations of a number of MEPs’ constituency offices. We preface the examination of this evidence in the next section of this chapter, with a discussion of the notion of constituency representation as it has developed historically within Britain. We consider here the extent to which the constituency-based representative has been a traditional hallmark of British representative democracy at many levels (from local councillors through to MEPs), and how prevailing understandings of the appropriate role of this representative may have developed over time. This discussion is essential to establish a sense of the broad political–cultural milieu within which electoral reform was experienced. As part of this discussion, we also review the limited evidence on the style of representation adopted by British MEPs prior to electoral reform. Section 7.2 of the chapter then outlines the precise nature of the change instituted in 1999, elucidating the differences between the two electoral systems, outlining how key features of the new system can be expected to have an impact on the nature of the representative process, and tracing the immediate consequences of reform for the social and political representativeness of the British cohort of MEPs. In Section 7.3 we then examine some survey evidence on the differences in attitudes to representation among MEPs that were produced by the new electoral system. Following all of this, in Chapter 8 we move on to explore in detail our qualitative evidence regarding electoral reform and its impact on the attitudes and behaviour of British MEPs.
141
Electoral Reform and the British MEP
7.1. The ‘British Tradition’ of Constituency Representation An important part of British thinking about political representation over a long period has been the concept of constituency representation: a broad notion encompassing the idea that elected representatives are linked to specific geographic areas, and further that those representatives can be expected to pursue, in their political activity, the interests and concerns of those within those areas. Tracing the historical roots, the changing strength and the substantive content of such a tradition is, however, complicated by both the long and complex evolution of the contemporary British polity from out of the expanded English state, and the inherently problematic nature of investigating a tradition, like constituency representation, in a polity like Britain where the very idea of tradition has been subject to substantial, self-conscious veneration. Nevertheless, at least some valid generalizations about the historical evolution of this tradition can be made. As Judge’s authoritative (1999) and perceptive overview makes clear, contemporary understandings of political representation in Britain have both distant and more recent antecedents. There are identifiable elements of continuity with the centuries-old tradition of the English parliament as a representative chamber for various elements of society, an institution that had an important role in the articulation of local grievances and concerns, as well as to legitimize the exercise of public authority, pass laws and raise taxes. A great deal of the development of the British political system, indeed, can be understood as the process whereby more recently accepted notions of democracy have been grafted onto the long-standing notion of the representation of societal interests to those in authority. As Judge observes, ‘When “democracy” came in Britain, therefore, it came late. And it came as an appendage to an established system of representative government’ (1999: 17). Understandings and practices of representation within Britain have therefore become deeply intermingled with the various paradoxes and contradictions of the British polity. But constituency
142
Electoral Reform and the British MEP
representation has been one of the most persistent elements within this evolving and complex tradition: that representatives should represent a particular area was an idea that won acceptance long before the idea that representation should accord with modern concepts of democracy. As we observed in Chapter 3, electoral laws tend to be heavily influenced by (although they may also, in turn, play an important role in further influencing) prevailing ideas in a society about political representation: about who should represent whom, and how. It is in this context that we can understand the historical development of electoral institutions for the (gradually democratizing) House of Commons, as well as for other elected bodies such as local government. While it may be unfair to claim about parliamentary representation, as did Norton, that ‘in both British political thought and practice, the argument for a socially typical House has never taken root’ (1981: 55), it is true that arguments for PR, although a long-standing strain within British political thought, have been until very recent years largely unsuccessful. Representation in parliament has continued to be, at least predominately, something that obtains from particular localities. Indeed, the thinking on representation of Edmund Burke—and in particular Burke’s advocacy of the representative as a quasiautonomous ‘Trustee’—was, as Judge (1999: ch. 3) notes, developed in the context of, and was to a large degree predicated on, the idea of the representative as representing a territory. The recognition of the locally-based nature of elective representation has long been institutionalized into the very practices of the House of Commons: in a manner quite distinct from that pertaining in the House of Lords, MPs refer to each other in the chamber with reference to the constituency they represent (e.g. ‘the Honourable Gentleman the Member for Bogsworth’). The electoral system that has encapsulated this tradition is single-member plurality. SMP was far from universally applied for House of Commons elections until well into the twentieth century, with University seats (elected by University graduates alone) and some constituencies electing two MPs (under plurality rules) being features that remained in
143
Electoral Reform and the British MEP
place in Britain until the end of the 1940s. Moreover, while local government elections have also long been conducted by plurality, in many parts of Britain these occur in districts (‘wards’ as they are generally called) returning two or three councillors.1 But SMP has provided the dominant institutional context within which contemporary thinking and practice about parliamentary representation has developed. While parliamentary representation in Britain has long been based on the constituency, this does not mean, however, that most elected representatives have actually been active on behalf of their constituents. Until about the 1960s, there appears to have been little expectation that MPs would actually maintain much of a presence in their constituency, or do very much in terms of constituency service work. The very opposite appears to have been the case: Norton and Wood (1993: 39–41) discuss the tradition that persisted across much of Britain until well after the Second World War of the MP making an annual visit to their constituency! Representing one’s constituents—to the extent that MPs worried about it at all—appears to have been primarily about articulating their concerns and interests, or at least what the MP believed were their interests, in parliament. Even many of those MPs in the 1970s who were classified by Searing (1994: ch. 4) as ‘Constituency Members’ do not seem to have maintained a huge workload on behalf of the localities to which they evinced a substantial attachment. Evidence from the most detailed behavioural studies (Cain et al. 1987; Norton and Wood 1993) suggests that things began to change from the 1960s onwards. Demands on, and expectations of, MPs rose: a more educated and less deferential public became more demanding of their politicians, while the problems associated both with the consequences of growing economic instability and with citizens’ dealings with the welfare state generated a greater need for the services of people, like MPs, who could intercede with public bodies on behalf of members of the public. At roughly the same time, an increasingly well-educated and professional body of parliamentarians emerged (King 1981): politicians who were both more able and more interested in supplying an active form of local
144
Electoral Reform and the British MEP
representation. Thus, not only did MPs find that their volume of mail and other public contact was increasing substantially; greater numbers of them began to maintain a home in their constituency, to hold regular ‘surgeries’ for constituents with problems, to raise parliamentary questions on behalf of constituency interests, and to target local newspapers more professionally in order to raise their profile. Whatever may have been the dominant prevailing motivations for these changes in MP behaviour, the changes themselves have been substantial.2 Various estimates suggested that by the late-1990s, constituency-related work constituted around onethird of MPs’ workload, in addition to it consuming a substantial share of their administrative resources—resources that, while still being small in scale compared those enjoyed by their German or American counterparts, were greatly increased from those of even 20 years previously (Judge 1999: ch. 7). To summarize, the British polity sustains a long-standing tradition of constituency-based political representation, a tradition that has been given new life in the past 30–40 years with the growth in constituency activity of MPs. In this context, it is rather unsurprising that when MEPs were first elected in 1979, many of those from Britain should have sought to replicate some aspects of such a tradition. There were obstacles in the way: the need for MEPs to spend considerable time in Strasbourg and Brussels, taking them even further away from their constituencies than Westminster does for most MPs; the very low profile and public awareness of MEPs and their institution within Britain; and the size of MEPs’ constituencies, equivalent to 7–8 Westminster constituencies on average. Nonetheless, as we review in more detail in Chapter 8, the constituency role was viewed as important by many of those elected. But even so, the returns for their activity in terms of public recognition, still less personal votes, were probably meagre. Shephard (1997) found that only about 1 in 240 British electors contacted MEPs every year, compared to 1 in 25 who approached Westminster MPs. And levels of public awareness of individual MEPs in their constituencies remained low: a survey by MORI in October 1992 found only 7 per cent of respondents
145
Electoral Reform and the British MEP
able to correctly name their constituency MEP, compared to 51 per cent who could so name their representative at Westminster.3 (While the figure for MPs was perhaps inflated by the fact that the survey was conducted only six months after a general election, the disparity was still very large.) Furthermore, while it is deeply rooted in British political culture, the tradition of constituency-based representation via SMP has come under renewed challenge in the last decade or more. After many years in which the subject had been politically marginalized and was in every sense academic, by the mid-1990s electoral reform had come much more to the forefront of the UK political agenda. Not only was the partisan disproportionality produced by SMP subject to continued public challenge by reformist groups such as Charter 88 and the Electoral Reform Society. Plans being developed during the mid-1990s for devolved institutions in Scotland and Wales came to assume that these bodies would be elected under systems that guaranteed at least an element of proportionality. The May 1997 election brought in a new Labour government that was committed to serious consideration of electoral reform— with some senior figures in the government, including then Foreign Secretary Robin Cook, known to be convinced supporters of the merits of electoral reform. SMP did not simply fade away. The system, and the principle of constituency-based representation associated with it, retained powerful support. Thus, although the Labour Party’s 1997 election manifesto had made a commitment to hold a UK-wide referendum on reform for Westminster elections, and the Blair government subsequently established a Commission under Lord (Roy) Jenkins of Hillhead to devise the alternative system that would be put to the vote, that referendum never materialized. While proportionality was brought in for the MMP electoral systems introduced for the devolved Scottish Parliament and National Assembly for Wales, the majority of members of these new bodies were still to be elected under SMP rules. And subsequent proposals to change the electoral system for these bodies to STV have been emphatically rejected by the Labour government.4
146
Electoral Reform and the British MEP
Thus far, there has only been one instance of an existing nationwide political institution to which British voters elect representatives actually experiencing electoral reform. This is the EP. On four occasions between 1979 and 1994, British members of the chamber (MEPs) were elected under SMP. This all changed for the June 1999 elections when, for the first time, they were conducted in 11 multimember regions, with seats allocated proportionally to party lists. As was pointed out in Chapter 4, the cause of this change had far less to do with British desires to converge on common electoral rules with those of its EU partners than it had to do with the Labour government wishing to make concessions to proponents of electoral reform in the UK (Farrell 2001a). But whatever the reasons for the change, it was all-but-certain to make a substantial difference in terms of who would be elected from Britain to the EP. What we investigate in the remainder of this chapter, and in the next one, is whether the change from SMP to regional-list PR also had a substantial impact on how those elected sought to carry out their role as representatives.
7.2. The New Electoral System for British EP Elections and Its Consequences for Representation The 1999 EP election was the first time that Britain had used PR on a nationwide basis. Although uniformity in EFs across member states had long been a stated goal of the EU, in 1994, British MEPs were still being elected under SMP, from 84 constituencies (each averaging in size between 7 and 8 Westminster parliamentary constituencies).5 But in 1999, the country was divided into eleven regions, which would elect between 4 and 11 MEPs each; with the subsequent accession of ten new member states, regional sizes reduced in 2004 (see Table 7.1 for details). Under the new system, seats were allocated between party lists on a proportional basis, via the d’Hondt formula. The MEPs chosen under this new system were elected from closed lists, meaning that candidates were ranked solely by the parties’ selectorates; thus, and unlike in many
147
Electoral Reform and the British MEP Table 7.1. The 11 British regions for the 1999 and 2004 EP elections
Eastern East Midlands London North-East North-West Scotland South-East South-West Wales West Midlands Yorkshire & Humberside
1999 seats
2004 seats
8 6 10 4 10 8 11 7 5 8 7
7 6 9 3 9 7 10 7 4 7 6
countries, the list could not be altered by individual votes (see also Table 4.2). Figure 7.1 locates the new British electoral system in comparison both to its predecessor and to those used by other countries for EP elections in 1999.6 The adoption of PR list for electing British MEPs provides a golden opportunity to investigate some of the consequences of electoral reform. The electoral system change in Britain allows for a ‘before-and-after’ comparison of a precision rarely available to students of politics in advanced democracies. Holding the wider political context and culture relatively constant, amidst this substantial reform of the electoral system and the potential incentives for behaviour promoted by it, allows for an unusually clear insight into not only the immediate, mechanical consequences of electoral system change, but the longer-term consequences as well. In this section, we examine the former, before turning to a consideration of the latter. The shift from a non-proportional to a proportional electoral system had a clear and substantial impact on who was elected (see Table 7.2). In 1994 only four parties won seats, with Labour (on a minority of the vote) winning 73 per cent of the total. In 1999 the number of parties represented jumped to seven, with no party even coming close to winning 50 per cent of total representation.
148
Electoral Reform and the British MEP
Ballot structure
Constituencybased Regional National
Locus of representation
Closed
Ordered
Open
Britain1a
Britain2
Belgium
Ireland Italyb
France Germany Greece Portugal Spain
Austria The Netherlands Sweden
Denmark Finland Northern Ireland Luxembourg
Figure 7.1. Locating Britain’s electoral reform in the context of the existing EP electoral systems in 1999 a
Britain1 (SMP) is categorized as ordered rather than closed, because while the ‘list’ of 1 was essentially a closed list, voters nevertheless were voting for individual candidates, and therefore there is somewhat more scope for candidate-orientation in representative roles than under closed list systems. For this reason, we include SMP in the same category as list system like Belgium’s. b Seat allocation in Italy is based on national votes, but the MEPs are elected on regional lists (in five regions).
This more even and widespread distribution of seats meant that the British contingent in the EP was now, in terms of partisan balance, much more similar to those from the other member states. It also meant that, for the first time in its history, Britain had a parliamentary representation (albeit one at the EU level) that looked like the multiparty pattern that has long obtained across most of Western Europe. In the subsequent 2004 EP elections these trends were carried even further. Once again, seven parties won seats, and the distribution between the parties was even more balanced than in 1999.
149
Electoral Reform and the British MEP Table 7.2. Elections to the EP in Britain, 1994–2004 1994
1999
2004
Electoral reform Electoral formula Number of constituencies Average district magnitude (M) Effective threshold (Teff )a
plurality 84 1 37.5
d’Hondt 11 7.60 8.7
Parties’ share of seats (%) Conservatives Labour Liberal Democrats Greens UK Independence Scottish National Party Plaid Cymru
21.4 73.8 2.4 0 0 2.4 0
42.9 34.5 11.9 2.4 3.6 2.4 2.4
36.0 25.3 16.0 2.7 16.0 2.7 1.3
Aggregate consequences Disproportionality (GI)b Effective number of parliamentary parties (Ns )c Women MEPs (%) Ethnic minority MEPs (No.)
23.73 1.69 19.05 1
8.30 3.13 24.14 4
8.24 4.12 25.33 5
d’Hondt 11 6.82 9.5
a
Teff = 75%/(M + 1). Effectively a mirror image of M, the lower the Teff , the more proportional the system. Because it is seen as an approximation, Teff is recorded to just one decimal point. √ b GI (Gallagher index) = {[ (v − s)2 ]/2}, where v = votes and s = seats. 2 c Ns = 1/ si .
As Table 7.2 shows, the shift to PR in 1999 thus produced a model set of aggregate electoral consequences. Levels of disproportionality along party lines (gauged using the standard measure, the Gallagher index) plummeted (the index score falling from 23.73 to 8.30); the effective number of parliamentary parties doubled (up from 1.69 to 3.13); and there was also a notable increase in the proportion of women MEPs (up from 19 to 25 per cent), while the number of MEPs from ethnic minority backgrounds rose from one to four. The ‘mechanical’ effects of the 1999 change to PR, then, were entirely consistent with what would have been predicted in the electoral systems literature; and they continued through to the subsequent 2004 election.7 Of course, as we saw in Chapter 4, even though the ‘representativeness’ of Britain’s EP delegation improved between 1994 and 1999, Britain still lags behind other EU member
150
Electoral Reform and the British MEP Features of electoral system change
• Dramatic increase in ‘constituency size’ • Loss of sole ‘ownership’ of constituency • Switch from candidate to party emphasis
Consequences for....
• Representative roles Who is being represented How much representation • Team-related matters The importance of ‘teamwork’ Intra-party relations Cross-party relations
Figure 7.2. Conceptualizing Britain’s EP electoral system change and its consequences for the representative function of MEPs
states in many respects. Britain’s disproportionality score of 8.30 in 1999 (8.20 in 2004) compared with an average of 4.76 for the other 14 member states (only Luxembourg, with a mere 6 MEPs, had higher disproportionality, of 10.95). Britain was also significantly lower than average with regard to the effective numbers of parliamentary parties and proportion of women MEPs (averages in the other 14 member states were 4.96 and 33.73 per cent respectively). What about the broader consequences for the representation process that might follow from the electoral reform introduced in 1999? Several aspects of the new system can be identified that might be expected to induce change in representatives’ attitudes and behaviour. These are summarized in Figure 7.2 and discussed briefly here. The first and most obvious change is with regard to the size of the electoral constituency or region. Although not particularly large by some international standards, the geographic spread of many of the new regions is greater than British politicians have traditionally been used to. Some MEPs’ previous single-member constituencies were themselves large in land area (notably the Highlands and Islands constituency in Scotland, and the Mid and West Wales one), but the new regional system has substantially exacerbated these problems. Size can also be viewed in population terms: British parliamentary constituencies have traditionally been measured in the tens of thousands (for Westminster), or low hundreds of thousands (for the EP previously), but MEPs now have to represent regions of several million people. A third issue is the diversity of many regions. While both Scotland and Wales form
151
Electoral Reform and the British MEP
‘natural’ political units, they also share highly uneven population distributions, diverse economies, and poor transport links with much of their respective peripheries. Among the English regions, only London and the North-East region could seriously be argued to have any genuine sense of identity and an obvious ‘centre’; among many of the others, most notably the South-West and South-East regions, identity and centre are conspicuously absent.8 A related change concerns the shift from single- to multimember constituencies. As we have already discussed, much has often been made in the UK of the parliamentarian being the single representative of a particular geographical area, and of the ‘sacred bond’ between a constituency and its member. While there is room for considerable scepticism about the more florid celebrations of this feature of British representative life, one could nonetheless expect a move from single-member to multiple-member constituencies to alter the representative relationship. Among representatives there may be less pressure to represent everyone in the particular area on all issues, and the opportunity to be, in various respects, more selective in their activities, albeit for a larger area and greater number of people. There may also be a sense of competition induced by the fact that the MEP no longer enjoys ‘sole ownership’ of the constituency, but must work with (and, indeed, probably against) others. Finally, there has been a shift in the voting act from one where electors cast votes for candidates to one where they vote for parties. Under SMP, participating voters cast their ballots for individual candidates (albeit ones with a clear party label). The adoption of a closed-party list system removed any semblance of voting for an individual rather than a party; the fact that the list is closed rather than open means that voters cannot favour particular candidates on the list.9 Thus, representatives lose any possibility of garnering ‘personal votes’, unless their mere presence on the party list is enough to persuade voters to support the list in toto. To win (re-)election for the EP now, by far the most important thing for a candidate is to achieve a prominent position on their party’s list for a region and/or to maximize the party vote.
152
Electoral Reform and the British MEP
What collective impact on the interpretation of their role as representative, and their behaviour, could we therefore expect the electoral system change to have? We can expect change in two broad areas. In the first instance, there are likely to be consequences for the representative roles of MEPs, in a number of respects. For instance, the electoral system change may reduce the incentives for MEPs to seek to build a personal vote. This might mean a downgrading of constituency work at the expense of other potential priorities—such as the pursuit of ‘office’ and/or ‘policy’ objectives within the EP. The electoral system change might also promote differences in terms of whom MEPs seek to represent. Under SMP, the consensus view (one infrequently questioned) has been that representatives can and should seek to represent all voters. But one way of making sense of the new electoral system for representatives—in reaction to the problems of seeking to represent large and diverse regions—could be a narrowing of the focus of their efforts. Thus, MEPs might choose to target their efforts at a particular geographical area within their wider region, possibly one containing much of their potential support; alternatively, they might identify particular social groups as including many of ‘their’ voters, or focus quite explicitly on supporters of their party. A further consequence of the changing electoral system is that the incentives for (re-)election imposed on MEPs are now party-based to an even greater extent than previously. With a reduction in any potential for accruing a personal vote, and with candidate (re)nomination now to be in the hands not of constituency parties but of the party leadership or party members on a much larger scale,10 MEPs have an incentive to work to secure support from within the party, certainly to a greater degree than from the broader public. This shift in incentives might be expected to be reflected in changed attitudes and behaviours. The second area of change follows on directly from the shift from single-member constituencies to multi-member regions. The existence of multiple members within each region opens up the possibility of forms of ‘teamwork’ within parties (with representatives in some way dividing responsibilities between them), and/or rivalry between representatives of different parties, competing to
153
Electoral Reform and the British MEP
serve their constituencies better and achieve higher levels of public visibility. In most instances (at least in the case of the two larger parties) there is more than one MEP from the same party in a region. They need to find an appropriate modus operandi, and one that can survive the inevitable strains of the re-selection process when incumbents will be scrambling for high placements on the party lists—a competition which of itself places good regional party relations high on the list of MEP concerns. But the stresses and strains are not only within parties. In all cases, MEPs have to find a way of operating in the region with fellow MEPs from other parties.
7.3. ‘Before-and-After’: Attitudes to Representation In Chapter 8, we assess our two sets of expectations in relation to a large body of qualitative evidence. Before that, however, in this section we examine some relevant quantitative survey data. While the qualitative evidence in Chapter 8 has considerable strengths in terms of subtlety and detail, our survey evidence permits some precise comparisons on standardized questions. Furthermore, drawing on these different sources allows us to adhere to Putnam’s counsel that ‘[t]he prudent social scientist, like the wise investor, must rely on diversification to magnify the strengths, and to offset the weaknesses, of any single instrument’ (1993: 12). Differently to Chapters 5 and 6, where we used evidence from a survey of MEPs conducted in 2006—which allowed us to compare MEPs from 25 member states, and 26 electoral systems—here we use data drawn from earlier surveys. Specifically, we use data taken from a survey of MEPs conducted in 1996, and from another such survey that was implemented in 2000. (For further details on these two surveys, see the Appendix.) The importance of these two dates for our current purposes is that they come, respectively, before and after the electoral reform introduced in Britain for the 1999 EP election. Those British MEPs who responded to the 1996 survey had been elected to the EP under SMP; the equivalent respondents
154
Electoral Reform and the British MEP
in 2000 had been elected (or at least, in their most recent election) under regional-list PR. Our analysis here therefore focuses on the British respondents to these two surveys of European parliamentarians, and it compares their responses in relation to a number of questions. While these surveys differ in many respects, two batteries of questions that were included in the 1996 survey—one concerned with aspects of the MEP’s role, the second with perceived ‘targets’ for representation— were then replicated in the subsequent survey. Our comparisons must be made with caution: the response categories were not identical across the two surveys; furthermore, the fairly low number of British MEPs who responded make any advanced statistical analysis difficult if not impossible. That said, these data still allow for some useful insight into potential changes among British MEPs across the two parliaments. Table 7.3 reports simple aggregate findings for British MEP respondents on our two batteries of questions. Although we must be cautious about inferring much from these figures, given the fairly low numbers of respondents, they do seem to indicate considerable continuity between the two surveys. On most items, the pattern of aggregate opinion appears broadly similar. Such continuity prevails despite the substantial individual-level turnover among British MEPs between the two surveys, as well as the significant changes in partisan composition of this cohort. Thus, in both 1996 and 2000, British MEPs rate ‘All people in my constituency’ most highly in terms of those whom they regard it as important to represent, with ‘My national party’ rated second out of the available response categories in both years, and ‘All people in Europe’ scoring lowest on both occasions. Similarly, for the question that probes MEPs’ priorities among different aspects of their role, ‘Legislating’ and ‘Parliamentary Oversight’ score highly in both 1996 and 2000, and are rated well above ‘Mediation between different interests’ in importance. But notwithstanding the overall pattern of continuity seen in Table 7.3, one interesting difference emerges—and one broadly in line with our existing hypotheses. This is the fall in the importance
155
Of little importance (%)
1996 How important to represent the following groups of people? All people in Europe All people in my member state All people in my constituency All people who voted for my party My national party My EP party group How Important as part of MEP’s work? Legislating Parliamentary oversight Representing individual citizens Social group representation Mediation between different interests
Of great importance (%)
Mean (out of 7)
Normalized scorea
N
1
2
3
4
5
6
7
7.4 3.7 3.6 3.8 7.1 3.7
7.4 3.7 0 0 0 11.1
14.8 3.7 0 3.8 3.6 11.1
22.2 18.5 7.1 15.4 7.1 11.1
18.5 14.8 10.7 23.1 21.4 29.6
18.5 25.9 14.3 38.5 35.7 18.5
11.1 29.6 64.3 15.4 25.0 14.8
4.37 5.33 6.21 5.31 5.43 4.67
0.56 0.72 0.87 0.72 0.74 0.61
27 27 28 26 28 27
0 0 0 0 3.7
7.1 0 3.7 3.7 7.4
3.6 7.1 3.7 3.7 14.8
7.1 7.1 7.4 11.1 33.3
10.7 21.4 18.5 25.9 18.5
28.6 25.0 29.6 18.5 14.8
42.9 39.3 37.0 37.0 7.4
5.79 5.82 5.78 5.63 4.30
0.80 0.80 0.80 0.77 0.55
28 27 27 27 27
Electoral Reform and the British MEP
156 Table 7.3. British MEPs’ view on representation: 1996 and 2000 compared
Of little importance (%)
a
Mean (out of 5)
Normalized scorea
N
1
2
3
4
5
24.2 0 0 3.0 0 3.0
18.2 6.1 0 3.0 0 6.1
36.4 27.3 0 18.2 30.3 48.5
15.2 51.5 30.3 54.5 42.4 27.3
6.1 15.2 69.7 21.2 27.3 15.2
2.61 3.76 4.70 3.88 3.97 3.45
0.40 0.69 0.93 0.72 0.74 0.61
33 33 33 33 33 33
0 0 3.0 3.0 9.7
0 3.0 3.0 12.1 19.4
9.1 3.0 33.3 27.3 38.7
21.2 36.4 24.2 33.3 29.0
69.7 57.6 36.4 24.2 3.2
4.61 4.48 3.88 3.64 2.97
0.90 0.87 0.60 0.66 0.49
33 33 33 33 31
Normalization allows any score to be mapped to the relevant point between 0 and 1, reflecting the location of that score between the minimum and maximum possible scores available. It is calculated as follows: (the score in question—the minimum possible score)/(the maximum possible score—the minimum possible score). Sources: Members of the European Parliament Survey, 1996; The MEP Survey 2000.
157
Electoral Reform and the British MEP
2000 How important to represent the following groups of people? All people in Europe All people in my member state All people in my constituency All people who voted for my party My national party My EP party group How Important as part of MEP’s work? Legislating Parliamentary oversight Representing individual citizens Social group representation Mediation between different interests
Of great importance (%)
Electoral Reform and the British MEP
assigned to ‘Representing individual citizens’ as an aspect of MEPs’ work between the 1996 and 2000 surveys. A comparison of the normalized scores on this item across both samples (dropping from 0.80 in 1996 to 0.60 in 2000) shows the largest change between the two surveys on any of the items considered here. In 1996, four items—legislating, participating in parliamentary oversight, the representation of individuals, and the representation of social groups—were rated roughly equal in importance by British MEPs. By 2000, legislation and oversight activities are now ranked as being substantially more important than the others, with the representation of individuals and social groups being regarded as relatively less important. Thus, we have at least some tentative evidence of the representation of ‘extra-parliamentary’ interests apparently declining in importance among British MEPs— as we hypothesized might be promoted by the change in electoral system that occurred between the two surveys. As this difference in survey responses cannot be attributed to broader changes affecting all MEPs,11 it does suggest that, even within a few months of their introduction, the new electoral institutions were having some impact on British MEPs’ attitudes to representation.
7.4. Conclusion The electoral reform introduced in Britain in 1999 for elections to the EP opened up considerable possibilities for investigating some of the broader impacts of electoral reform on the process of political representation. Our evidence here indicates that the change to PR in 1999 had substantial, and entirely predictable, effects in terms of who was elected. The British presence in the EP became much more ‘representative’ in party terms than it had been previously; the representation of women and the ethnic minorities among the British cohort was also improved. But our major interest is in what occurred after the election, specifically with regard to the anticipated changes set out earlier.
158
Electoral Reform and the British MEP
It is plausible that we might find changes in the nature and quantity of an MEP’s ‘constituency presence’, both with regard to who is being represented (the targets of representation) and also relating to the amount of constituency work engaged in. We can also expect to find changes concerning relations between MEPs in the same and other parties (teamwork vs. rivalry), which in turn might be affected detrimentally by re-selection pressures. Our initial exploration of survey data has offered evidence that is broadly in support of these expectations, and particularly the former. The comparison of survey responses among British MEPs in 1996 and 2000 has shown a clear shift in their priorities, with a greater emphasis being placed on legislation and oversight activities, and a reduced emphasis on constituency work. However, we emphasize the limitations to these survey data. A fuller and more adequate exploration of our expectations requires that they be examined in relation to a more detailed and subtle body of evidence. This examination is the task of the following chapter.
Notes 1. Of course, local elections in Northern Ireland have used STV since the 1970s. In 2007, Scottish local elections will also use STV for the first time. 2. Indeed, there is scope for debate regarding the extent to which MPs’ increased willingness to engage in constituency work has been primarily driven by electoral motives (Cain et al. 1987) or is genuine ‘role preference’ largely divorced from purely electoral considerations (Searing 1994). Furthermore, the evidence on the electoral benefits of constituency work is somewhat ambiguous (Gaines 1998; Norris 1997; Norton and Wood 1993). 3. Source: MORI poll for The Sun, October 1992. We are grateful to Roger Mortimore of MORI for supplying this information. 4. The Richard Commission on the Powers and Electoral Arrangements of the National Assembly for Wales, which reported in 2004, came
159
Electoral Reform and the British MEP
5.
6. 7.
8.
9.
10.
11.
160
out strongly in favour of STV for devolved Welsh elections (Richard Commission 2004). But this proposal was not carried forward by the government. As we saw in Chapter 4, Britain was the only place where SMP was used for EP elections. All other countries used some form of PR— either STV in Ireland (North and South) or a list system—within multi-member regions or across the entire country. In the subsequent 2004 EP elections, France joined Britain as the only other case of closed regional list PR. Needless to say, we are well aware of the fact that other cultural and party-related factors also play a role in determining the types of individuals elected to public office. For discussion, see Farrell (2001: ch. 7). See Anthony Heath, Catherine Rothon, and Lindsey Jarvis, ‘Centre and Periphery: Regional Variations within England’, Paper presented to the annual conference of the Elections, Public Opinion and Parties specialist group of the Political Studies Association, Salford, 2002. This feature of the electoral system prompted criticism when it was introduced for denying voters the opportunity to cast their ballots for particular individuals—although one could well argue that under SMP voters also face a closed list (or what we refer to it in Figure 7.1 as an ordered list): they must either support the candidate chosen by a party or cast their vote for another party. The Labour Party’s candidates for the 1999 EP elections were chosen by the national/regional party leadership; those for the Conservative Party by meetings of party members within each region; and those for the Liberal Democrats by ballots of party members within the region. In the selection of candidates for the 2004 election, Labour shifted to a system of one-member-one-vote postal ballots (with inbuilt protection for incumbents and gender balance rules). The other parties stuck with the same selection systems as before. It does not appear that these attitudinal changes among British MEPs simply reflect broader changes common to all European parliamentarians. One possibility would be that exogenous events—particularly the increases in legislative and oversight powers acquired by the EP
Electoral Reform and the British MEP between 1996 and 2000 in the Amsterdam Treaty—might explain the relative importance given to the different activities in the two surveys. However, when we look at responses from other countries, we find that the decline in the importance of individual representation is specific to British MEPs—across the EP as a whole, this activity only accorded a mean importance rating of 3.97 (out of 7) in 1996.
161
This page intentionally left blank
8 Life Under List: Representing a Region
[W]e have had constituencies that have been represented by an individual ever since Simon de Montfort summoned his first parliament in the 13th century. This isn’t just a kind of abstract concept . . . I think the move away from that has done some damage. (Conservative MEP)1 I know that my colleagues who were under the old system get all dewy-eyed and like to make out to young pups like me that under the old system they were held aloft on the shoulders of voters every time they stopped to buy a pint at noon. I think that’s Bollocks! (MEP from one of the smaller parties)
In Chapter 7 we discussed, and began to assess the consequences of, the electoral reform that was introduced for EP elections in Britain in 1999. As we showed, the move from SMP to regional closed-list PR constituted a substantial change in the electoral institutions deployed for these contests. And as we further demonstrated, this change in electoral institutions had some fairly predictable consequences in terms of who was elected to the EP from Britain in 1999 (and again in 2004). The outcomes of electoral reform with regard to partisan proportionality, the number of parties represented, and the representation of women and minorities were all very much in line with what a wide range of previous
163
Life Under List: Representing a Region
research on the consequences of electoral systems had led us to expect. But, as we have emphasized throughout this book, it is rather less clear what we should expect the implications of electoral reform to have been for the attitudes and behaviour of those elected. There are some indications from the existing literature that we could anticipate the different incentives produced by the new electoral system introduced in Britain in 1999 to influence the focus and activities of political representatives. In line with these suggestions, we developed two sets of expectations about representatives’ focus and activities in Chapter 7. These were that the move to regionallist PR might have prompted change in: (a) the nature and quantity of MEPs’ constituency presence, and the main ‘targets’ of their behaviour—that is, on who they represent, and (b) the emphasis on ‘teamwork’, on relations with MEPs from other parties, and particularly internally with those of their own parties—in sum, how they represent. Some exploratory analysis in Chapter 7 of available survey data did indeed indicate that British MEPs had a lower level of constituency focus after the move to PR than they had done beforehand. However, a serious assessment of our expectations requires a much richer and more complete body of evidence than the limited survey data that were available. To this end, the current chapter draws on a range of qualitative data that we have accumulated. Between 2001 and 2003, we interviewed 61 British MEPs—some 73 per cent of the total British contingent in that parliament. These interviews were conducted at a time that allowed sufficient time for the 1999 reforms to have become embedded, and for representatives to have acquired a reasonable degree of experience in working under the new electoral system. The interviews conducted with MEPs were open-ended, but followed a common structure; they were recorded, and the transcripts have been coded to permit some analyses of broad trends. But in addition to speaking to a large number of MEPs, we also visited twenty MEPs’ offices, and spoke to members of their staff. Pooling all this information gives us a substantial body of data with which to assess how
164
Life Under List: Representing a Region
electoral reform may have altered representatives’ attitudes and behaviour. The chapter proceeds as follows. Referring back to the expectations about behavioural consequences of the electoral reform that were outlined in the Chapter 9, in Section 8.1 we set out our main hypotheses and review the factors (regional, party, and individual) that may interplay with them. We then move, in the following two sections, to examine in detail the evidence concerning each of our expectations: (a) how much, and what type, of constituency presence MEPs have continued to maintain, and who are the targets of their representational efforts, and (b) whether there has been any shift in their representational styles with regard to team-related activities. Section 8.4 of the chapter summarizes the discussion, and lays out the major implications of the empirical findings presented.
8.1. Anticipated Behavioural Consequences of the 1999 Electoral Reform The change of electoral system for EP elections in Britain undoubtedly had some important consequences, and we have already reviewed several of them. Towards the end of Chapter 7 we set out two sets of expectations relating to how the shift from SMP to a closed regional-list electoral system might affect the behaviour of British MEPs. We can express these more formally, as two main hypotheses: H1: The targets of representation: The new electoral system results in a shift in the MEPs’ representative role, with less attention given to individual voters and, in particular, a greater focus inwardly on party interests. H2: The methods of representation: The new electoral system requires MEPs to place greater emphasis on ‘teamwork’, particularly (where relevant) with fellow MEPs from the same party, but also to a degree with their counterparts from other parties.
165
Life Under List: Representing a Region
H2 is not entirely straightforward. Given the closed nature of the electoral system, in the sense that the fate of individual candidates is determined by their ranking on the ballot paper, the success and survival of intra-party teamwork is likely to be affected detrimentally during reselection periods, resulting in the following expectation: H2a: Teamwork suffers in the lead up to candidate (re)selection. A central aim of our interviews was to elicit MEPs’ attitudes to the regional-list system: whether they liked it; how they thought it had affected their representative role; whether they had experienced an increase in their constituency workload; and how they worked with other MEPs in the region. In line with our two main hypotheses, the coded summaries of our interview transcripts are grouped into two Tables 8.1 and 8.2, setting out consequences Table 8.1. The representative role of MEPs (%) MEP’s role (multiple options) Voter representation Policy formation Party representation Ideological representation
75 52 16 13
Who do MEPs represent (multiple options) Europe Britain Party Region Social group/ideology
3 20 48 74 31
Believe their role is different from MPs (% yes) Believe their role is different from other MEPs (% yes) MEP’s representative role has changed (% yes) New electoral system has affected MEP’s role (% yes) Does not like PR list Logistical problems representing region (% yes) ‘Constituency work’ has increased (% yes) Relationships with organized interests (% yes) Regret the lost of ‘sole ownership’ of constituency (% yes)
66 84 79 87 59 75 8 54 33
Source: Interviews of British MEPs by authors (N = 61).
166
Life Under List: Representing a Region Table 8.2. Team-related matters (%) Good teamwork with MEPs from same party (% yes) Nature of teamwork—Geographical Functional Mix Sharing of office/staff (% fully/partially) Concern over reselection (% yes) Impact of reselection on MEP relations (% yes) Good relations with MEP’s party in region (% yes) Good relations with MEPs from other parties (% yes)
62 26 16 26 27 37 33 59 44
Source: Interviews of British MEPs by authors (N = 61).
respectively for representative roles and team-related matters. The figures presented in the tables reveal just how significant an effect the MEPs themselves believed that the new electoral system had had on how they operate. We examine the evidence in more detail below, but here it is worthwhile outlining some of the main trends. As we see in Table 8.1, more than three-quarters of those interviewed (79 per cent) felt that their representative role had changed, while an even greater proportion (87 per cent) credited the electoral system with causing change (among the other stated reasons were the increased legislative powers of the EP and the moves towards regional devolution in parts of the UK). As one longstanding MEP put it: I have almost given up trying to be a [name of constituency] MEP in a sense. I now see my work as Brussels-focused because—maybe I just have a constituency withdrawal on that and haven’t adjusted as well as others—but I just don’t see how I can represent from XXX down to XXX and really have a meaningful relationship with that size of a constituency.
Overall, the majority of British MEPs interviewed (59 per cent) did not like the new electoral system, with the most commonly voiced regret—one particularly noticeable among the pre-1999 generation—being the loss of sole ownership of the constituency
167
Life Under List: Representing a Region
(something mentioned by 33 per cent of all the MEPs we spoke to). This point was perhaps best put by one Conservative MEP: I would give my back teeth to have a single member constituency where people can turn round and say ‘He is our MEP; he is the chap that I turn to’. I can represent all of those people in that constituency; I would have a finite area; it would be mine. As it is, we are all over the place at the moment.
Even if MEPs objected to the change, the signs are that many were adapting their modus operandi to suit the new system. This is shown, for instance, by the 62 per cent who felt that they had good teamwork with fellow MEPs (see Table 8.2), and the 42 per cent who claimed to have good relations with MEPs from other parties. But all was not entirely rosy. At the time of our interviews, candidate reselection was on the horizon, which clearly was putting relations under some strain— as shown by the 37 per cent who expressed concerns over reselection. For the 84 British MEPs elected under the regional-list PR system in June 1999, there were numerous factors which had shaped their experiences of the election and would be likely to influence their experiences of, and attitudes towards, regionally based representation after their election. One factor was prior expectations. The certain expectation of winning a seat that applied to candidates at the top of a regional list for a major party stood in stark contrast to the experiences of others—notably some from minor parties, and a few Conservatives who were elected after the Tories’ surprisingly strong showing in the EP elections—whose shock at election victory was often still visible some weeks after the event. But among the eighty-four were also representatives of greatly differing regions; politicians coming from very different political parties; and individuals with widely varying backgrounds and differences in attitudes, abilities, and characters. All these factors taken together mean that how MEPs approached and experienced life under the regional-list system could be expected to vary very
168
Life Under List: Representing a Region Table 8.3. Regional variation in British representation in the 1999–2004 (EP)
Eastern East Midlands London North-East North-West Scotland South-East South-West Wales West Midlands Yorkshire & Humberside
1997 electorate
Land areaa
Seats allotted
Electorate/ seatsb
Land area/ seatsc
4,005,677 3,159,678 4,940,271 1,981,509 5,204,719 3,984,406 5,951,137 3,736,230 2,222,533 4,023,038 3,804,298
19,126.64 15,626.62 1,573.44 8,591.74 14,163.13 78,822.01 19,085.21 23,851.50 20,778.32 13,003.80 15,411.41
8 6 10 4 10 8 11 7 5 8 7
500,710 (−) 526,613 (+) 494,027 (−) 495,377 (−) 520,472 (+) 498,051 (−) 541,013 (+) 533,747 (+) 444,507 (−) 502,880 (−) 543,471 (+)
2,390.83 (+) 2,604.44 (+) 157.34 (−) 2,147.94 (−) 1,416.31 (−) 9,852.75 (+) 1,735.02 (−) 3,407.36 (+) 4,155.66 (+) 1,625.48 (−) 2,201.63 (−)
a
Land area measured in km2 . Symbols in brackets indicate whether below (−) or above (+) average for proportion of electors per seat, based on mean of 512,065. c Symbols in brackets indicate whether below (−) or above (+) average for proportion of km2 covered per seat, based on median (Yorkshire & Humberside) of 15,411.41. b
considerably. In the following paragraphs we try to identify some of the main points of contrast between MEPs, as well as identifying some commonalities. One respect in which the experiences of the MEPs certainly differed concerned the regions that they represented. As we discussed briefly in Chapter 7 (see also Chapter 4, pp. 74–7), there was considerable variation due simply to the politics of scale—namely, the very sizes of the regions in question. As Table 8.3 (which summarizes the state of play for the 1999–2004 EP) suggests, there were substantial differences between the British regions, ranging from the South-East region which in 1999 had almost six million voters down to the North-East with less than two million voters. There was also regional variation in terms of land area with Scotland (78,822.01 km2 ) at one extreme and London (1,573.44 km2 ) at the other. As Matthew Shugart and his colleagues have shown (Carey and Shugart 1995; Shugart 2001; Shugart, Ellis, and Suominen 2003), regional size, or ‘district magnitude’, can have significant and complex effects on the role of elected representatives. One can
169
Life Under List: Representing a Region
readily understand, for instance, how the sorts of distances MEPs must travel compare in urban London versus (largely) rural Scotland. We can expect regional variation to have significant impacts on the representative role of MEPs, most notably with regard to such issues as the degree and nature of intra-party teamwork, or relations with other parts of the party in the region. Before we can assess this, however, we need to give some thought to what is meant by ‘regional size’ in this context. Variations in electorate size traditionally form the basis for determining the allocation of seats across the different regions. In 1999 the seat allocation for the United Kingdom was done in two stages: first, the Northern Irish, Scottish, and Welsh seats were allocated separately, giving voters in these nations a better ratio of seats per voter than the rest of the UK; second, seats for the English regions were allocated using the Sainte-Laguë formula.2 But, when we reexamine the seat allocations according to variations in the regional size based on land mass, Wales and especially Scotland do stand out as good cases for ‘over-representation’. As the final column in Table 8.3 indicates, when we calculate regional land areas as a proportion of seats in the region, Scotland and Wales were less well represented on average than the English regions—Scotland, at 9,852.75 km2 per seat, markedly so. Given that seats were allocated on the basis of variations in electorate size it is not surprising to see a far lower dispersion across the different regions on this measure (the smallest ratio of electorate to seats—Wales—represented 81.5 per cent of the largest—Yorkshire & Humberside) than on the ratio of land area to seats (the smallest ratio of land area to seats— London—represented just 1.6 per cent of the largest—Scotland). If we take both ratios together, the regions with fewer representatives than the average were East Midlands and the South-West, whereas the regions with more representatives than the average included London, the North-East and the West Midlands. A second major difference between MEPs elected in 1999 that had a direct bearing on how they could and did seek to carry out their role as representatives was their party affiliation. This was important in some distinct ways. First, MEPs from different
170
Life Under List: Representing a Region
parties tended to come to the EP with rather different political attitudes, at least some of which had an obvious relevance to how many MEPs interpreted their job. While virtually all Labour and Liberal Democrat MEPs, as well as many Conservatives, had a broadly positive attitude to the institution to which they had been elected, some other Tories, and the UKIP representatives, viewed an important part of their role as being to oppose the dominant ethos of the chamber and to bring its failings to the attention of the British public. The words of one Euro-sceptical Conservative MEP could not have been more plain: ‘My most important role as an MEP is to work towards the independence and self-determination of my country, and to unscrambling this mess which we call the EU.’ But partisan affiliation also had a direct impact on MEPs’ experiences of regionally-based representation in other respects. Representatives of minor parties—UKIP and the Greens, and to a lesser extent the Scottish and Welsh Nationalists—tended to have a high status and broad responsibilities within their parties, given the paucity of prominent elected figures within those parties. Perhaps most importantly of all, the nature of regional representation for an MEP could differ greatly depending on the success of their party in a particular region. Many individuals (fourteen, to be precise) were the sole MEP for their party in a region, but others had party colleagues elected for the same region. In the case of the Conservatives in the South-East England region, each Tory had four party colleagues who shared the region with them. Thus, the regional-list system created opportunities both for rivalry between elected representatives who could each claim a mandate within a particular geographical area, but also in some cases for teamwork in sharing the responsibility for representing a large region. A third factor that very obviously varied greatly among British MEPs was their level of previous political experience, most particularly whether they had previously been MEPs and thus elected under the old SMP system. Of the 84 representatives elected in June 1999, 41 had been outgoing members of the previous parliament. Having previously been representatives for a single-member
171
Life Under List: Representing a Region
district, returning MEPs now had to adjust to a very different model of representation—a model under which they would be merely one of between four and eleven representatives in an area, alongside colleagues from other parties and perhaps also one or more from their own party. As some of the returning MEPs that we interviewed freely volunteered, the adjustment to multi-member regional representation was not always easy. One Labour MEP observed: [I]n some cases . . . people write to all [number of MEPs] in our region and then try to do a sort of Dutch auction in the replies they get, and they are quite open about doing that, which of course vastly magnifies the correspondence and is irritating because they really want to contact the Tory. Others, of course, are put off in one way or another from notifying you at all of something in which you would want to participate. The result is that I do far fewer school and college meetings than I used to and that is because people are nervous that if they invite one they must invite all.3
For newer MEPs, there was less of an adjustment in that they had not been socialized into the previous system. But several, as former Westminster MPs, had had considerable experience of working under SMP. And for all the new members, the normal process of adjusting to the considerable complexities of life as an MEP (Scully 2005) was being compounded by the fact that they were learning one important aspect of their role in an atmosphere of general confusion, as everyone was learning to live with the new arrangements. To summarize, the effects of the reform of the electoral system on MEP behaviour are likely to be conditioned by three main factors: the size of the electoral regions, the size and status of their party in the region, and the political and parliamentary experience of individual MEPs. The analysis that follows on the consequences of the electoral reform for the representative behaviour of British MEPs therefore needs to take account of these factors. In Section 8.2, we start with the first of our two hypotheses, relating to the ‘who’ of representation.
172
Life Under List: Representing a Region
8.2. Changes in the Representative Roles of British MEPs As we discussed in Chapter 4, the change of electoral system from SMP to regional-list PR produced some rather different electoral incentives for British MEPs. One plausible implication of this change is a very simple one—namely, that MEPs would respond by reducing the time and effort they put into developing and maintaining a visible constituency presence. The reasoning for developing this expectation about the impact of the 1999 electoral reform is clear. While it was, as we have seen, difficult for British MEPs to attain a significant public profile and aspire to any substantial personal vote under the SMP system, they may nonetheless have believed that efforts in this direction could have some positive pay offs for them—perhaps less through the MEP being able to attract large numbers of personal votes directly, and more through the impact of a visible presence with constituency parties and workers producing greater efforts on their part to mobilize the vote come election time. The regional system had not only eliminated any sense that the public could vote directly for individual candidates; by broadening representation to larger regions it also weakened MEPs’ links to local communities and local party organizations. Thus, it is very possible that some MEPs may have responded simply by reducing their emphasis on keeping up any ‘constituency’ presence, and decided to concentrate their limited energies and resources more on work within the EP and/or on more general political efforts for their parties. In line with what they report to quantitative surveys, British MEPs tend to place great stress on their link with voters. Even if the notion of a constituency is now more complicated, most clearly emphasize this side of their work as being of some significance: ‘Well, firstly I have got three full time people up in XXX, up there every week . . . I reply to every letter and the website, email and surgeries, I write articles every two weeks for a series of local newspapers’ (Liberal Democrat MEP). But to what extent do the MEPs’ expressed views about their ‘constituency presence’ actually tally with the facts? The degree
173
Life Under List: Representing a Region
to which MEPs focus attention on their constituency presence is determined by (a) what resources they can bring to bear, and also (b) on how much attention is paid to them by their electorates. The available resources are quite considerable. All MEPs receive allowances to cover the cost of running offices in their locale and to pay the salaries of their staff (with just about enough salary resources to employ the equivalent of about three full-time staff members). Apart from ensuring that the budgets are used for the correct purpose (e.g. the office budget cannot be used to pay salaries), the EP traditionally has had rather a laissez-faire attitude with regard to how the monies are spent; indeed, it is only relatively recently that serious attention started to be given to the need to audit MEPs’ usage of their accounts. The vast bulk of British MEPs has a constituency office of some sort. Of the 84 MEPs elected in 1999 only 7 (3 Conservatives, 3 Liberal Democrats, and 1 UK Independent) had no publicly advertised office address in their region. In the 1994–9 parliament there were only 6 British MEPs without an office.4 While this represented an increase in the numbers without offices, which would be consistent with the notion that constituency presence would be downgraded by the electoral reform, the change was very slight. Needless to say, of course, the definition of what constitutes a constituency office tends to be quite loose, as we witnessed on our visits to MEPs’ offices during the course of this research. At one extreme, some MEPs’ ‘offices’ consist of nothing more than a single desk in the corner of their party’s regional office, with one part-time assistant there to take calls and ‘file’ correspondence (in some cases, there may be no assistant, and the calls forward automatically to the MEPs’ home; in some cases the desk is located in the MEPs’ home; in some cases there is nothing more than a postal address, and all correspondence is sent automatically to the Brussels office). At the other extreme are MEPs who take a suite of offices, perhaps shared with other MEPs from the same party in the region, or other party politicians (MPs, members of the Scottish parliament, local councillors, etc.), with a team of assistants and banks of PCs.
174
Life Under List: Representing a Region
There were certainly some suggestions in our interviews that the visibility of MEPs with the public has diminished: Under the PR system how has it changed? You have far less contact with the public, one of the reasons is that, not that one is less accessible, that simply they are not aware of the new system . . . So in other words you have virtually got to do more work to encourage people to make contact with you . . . . I think there is a danger of being slightly over romantic about how hard working members are for their constituents. When they say their constituents what they mean is a certain number, a very small number of their constituents (Conservative MEP).
Certainly, the move to larger multi-member regions has not resulted in larger postbags. Consistent with evidence from interviews with British MEPs’ staff members, the vast bulk of MEPs is quite prepared to admit that their constituency workload has not increased (a mere 8 per cent claimed that it had; see Table 8.1); if anything, the tendency appears to be that, consistent with the reduced constituency orientation of the new electoral system, the constituency caseload has also declined. As one long-standing Conservative MEP readily admitted, ‘I find that I am getting less than a quarter of the letters that I once got’. Similar views emerged in our interviews with MEPs’ assistants: the degree of constituency correspondence appears to be relatively small, and in many instances probably declining: I don’t think we get as much correspondence as MPs . . . Constituents don’t have a clear idea, unfortunately, of what MEPs do. (Conservative assistant) We tend not to get huge volumes of correspondence by any means, but they do get masses of stuff, especially from big companies who have got vested interests and things. They are continually being lobbied either by letter or by email or people holding receptions and so on (Minor party assistant).
To some extent this reflects a greater use of emails, but more often than not the basic explanation is that, simply, there is less contact from voters:
175
Life Under List: Representing a Region I think that [the MEP who employed me prior to 1999] was much more identified; people personally identified with him and therefore he did a lot more. He had a lot more people contacting him . . . He dealt with a lot more casework (Labour assistant).
Some assistants were able to give us more accurate figures on the actual amount of correspondence (letters, not emails) received on an average month. In one case this amounted to just 65 letters, in another 76; the largest amount was one case where there were 95 letters in a month (i.e. just over three per day!). In two instances, we had sufficient information to be able to calculate a rudimentary breakdown of the types of correspondence received and this provides firm evidence in support of the above quotes. In the case of the MEP’s office that received 95 letters in a sample month, 34 per cent of these were circulars and newsletters; 13 per cent were invitations; 12 per cent were office-related administrative matters. At most, 44 per cent (40 letters) could be construed as related to the representative role of the MEP, and even then about half of these were mass-produced lobbying letters (of which few would have been from actual constituents). In the case of the MEP’s office that received 65 letters in a sample month, the proportions were very similar: 25 per cent were circulars and newsletters, 26 per cent were invitations, 6 per cent were office-related; the remaining 43 per cent consisted of constituency-related and lobbying matters. There are also some indications, from at least a few MEPs, and some assistants, that the representative function is increasingly taking a back seat to the legislative function: Constituency activities are not the top priority. It is the actual activity in the parliament that occupies the MEP; it is not the constituency . . . I believe, I can’t say this absolutely definitely, but I believe that xxxx [the predecessor MEP from the same party], for example, when [s/he] was in parliament probably spent a good deal more time on constituency things, but at that time you had a constituency (Assistant).
We are concerned, however, not only with the quantity of MEPs’ representational activities, but also with their direction. Traditional
176
Life Under List: Representing a Region
understandings of constituency representation in Britain have placed great emphasis on the representation of individual citizens—with representatives being expected to spend considerable time dealing with the concerns and problems of individual constituents. And empirical investigation of constituency representation in Britain has also identified the importance placed on other small-scale activities with local authorities and communitybased groups (Cain, Ferejohn and Fiorina 1987; Norton and Wood 1993). Our expectation here is that the changing scale of representation implied by the move to regional representation may have prompted a change in the scale of their activities, with a reduced focus on the individual and the local group, and MEPs concentrating their efforts on activities and organizations more likely to have a reach extending throughout the region. To explore these matters, we started each of our interviews with a deliberately simplistic and direct question (whose seeming vagueness in a few instances elicited incredulous reactions), asking MEPs how they envisaged their role (i.e. an unprompted question). Secondly, and later in the interviews, we followed the lead of a series of MEP surveys by asking MEPs to rank order their representation priorities. Finally, we sought information on how MEPs felt they compared with British MPs and also with MEPs from other countries. The summary trends can be seen by referring back to Table 8.1. Both the unprompted and prompted questions about MEPs’ roles reveal a very high emphasis on the representation of voters in the region, consistent with the evidence from surveys of MEPs. There was also a very high proportion of British MEPs (52 per cent) in our interviews who, unprompted, place great emphasis on the role of policy formation. The majority of British MEPs (66 per cent) believe that their role is substantially different from that of a national MP, as was perhaps most crisply encapsulated in the words of one Conservative MEP: ‘The Westminster Parliament is a complete waste of space really: our MPs from Westminster have very little to justify their existence’. The most commonly stated differences5 were that MPs have
177
Life Under List: Representing a Region
a far less influential committee role (mentioned by 20 interviewees) and that MPs have more constituency casework (mentioned by 19; although there were two MEPs who argued that MEPs have more constituency casework than MPs). Even though most British MEPs believe that they have less contact with voters than their national counterparts, the general consensus is that, if anything, the degree of voter contact they have is still far greater than for their counterparts from other EU member states. More than four in five British MEPs (84 per cent) believe that their role differs from that of other MEPs, and of those who were more specific on what the difference is, 34 referred to the greater degree of constituency contact of British MEPs (albeit with some suggesting that, post1999, this was now changing). The sense of many British MEPs is that voter contact is a much lower priority for MEPs from other member states. As one Conservative MEP commented: ‘I think a lot of the continental MEPs would more or less happily spend their entire lives here and get thoroughly engrossed in every aspect of their parliamentary duties’. Similarly, if an MEP represents a smaller party that has few if any national-level MPs in the region (or a larger party that is traditionally electorally weak in that region), then she or he is likely to face greater demands for constituency activity from supporters of the party, and a more positive reception generally—due to the fact that there are fewer toes to tread on. As one Labour MEP in the south of England commented: ‘Where I do have a good relationship is off in the sticks, often in areas where we have one local councillor out of 60. I mean there I am the first parliamentary XXX representative 70 per cent of my region has ever had. So, I mean, there I am like “Wow, I can do no wrong!”.’ In an effort to examine the interview evidence more systematically, Tables 8.4–8.6 disaggregate the figures from Table 8.1, to allow us to assess differences according to, respectively, the type of region represented, the MEP’s party, and individual-level differences. Table 8.4 considers regional variation using both of our measures of regional size. In the first part of the table we can observe that
178
Life Under List: Representing a Region Table 8.4. The representative role: regional features (%) Regional size (land area)a
Regional size (electorate)b
Small
Medium
Large
Small
Medium
Large
MEP’s role (multiple options) Voter representation Policy formation Party representation Ideological representation
50 80 20 0
79 47 14 16
75 50 25 0
75 75 25 0
85 44 15 18
58 58 16 11
Who do MEPs represent (multiple options) Europe Britain Party Region Social group/ideology
0 0 60 90 40
5 28 49 72 33
0 0 38 63 13
0 0 25 100 13
6 26 44 65 29
0 16 63 79 42
90
56
88
88
65
58
90
81
88
88
85
79
70
79
88
100
71
84
80
86
100
100
82
100
20 60
37 81
13 63
13 75
24 71
53 84
0
9
13
0
9
11
50
56
50
100
56
32
30
37
13
13
24
58
Believe their role is different from MPs (yes) Believe their role is different from other MEPs (yes) MEP’s representative role has changed (yes) New electoral system has affected MEP’s role (yes) Does not like PR list Logistical problems representing region (yes) ‘Constituency work’ has increased (yes) Relationships with organized interests (yes) Regret the lost of ‘sole ownership’ of constituency (yes) a b
Small = London and North-East; Large = Scotland. Small = Wales and North-East; Large = North-West, London, and South-East.
voter representation assumed greater importance in larger regions based on land area (medium = 79 per cent; large = 75 per cent), but also in smaller regions based on electorate size (medium = 85 per cent; small = 75 per cent). There is a similar division with regard to the importance attached to party representation: it features more strongly in smaller regions based on land area (60 per cent)
179
Party
No. of MEPs of same party in region
No. of MPs of same party in region
Lab.
Con.
Oth.
1
2
3+
0
1–10
11+
MEP’s role (multiple options) Voter representation Policy formation Party representation Ideological representation
71 67 10 10
72 40 8 12
87 53 40 20
71 50 21 21
100 64 9 0
69 50 17 14
60 40 20 40
76 52 14 14
74 54 17 9
Who do MEPs represent (multiple options) Europe Britain Party Region Social group/ideology
10 14 52 71 24
0 20 36 52 28
0 27 60 80 47
7 36 71 79 43
0 0 18 81 9
3 14 31 47 14
0 20 40 20 60
0 24 38 76 33
6 17 49 74 26
71 86 95 95 30 80
60 80 76 84 46 71
67 87 60 80 12 77
71 86 57 79 0 71
73 82 100 100 9 82
61 83 81 86 50 75
60 80 40 60 0 100
76 86 81 95 19 71
60 83 83 86 43 74
19 65 38
4 38 48
0 65 0
0 79 0
9 73 9
8 39 53
0 80 0
5 48 19
11 54 46
Believe their role is different from MPs (% yes) Believe their role is different from other MEPs (% yes) MEP’s representative role has changed (% yes) New electoral system has affected MEP’s role (% yes) Does not like PR list Logistical problems representing region (% yes) ‘Constituency work’ has increased (% yes) Relationship with organized interests (% yes) Regret the lost of ‘sole ownership’ of constituency (% yes)
Life Under List: Representing a Region
180
Table 8.5. The representative role: party-level features (%)
Life Under List: Representing a Region
but larger regions based on electorate size (63 per cent). These apparently contradictory patterns are largely driven by London MEPs, representing as they do a region that is both one of the largest in terms of electorate but the smallest in land area. It is not surprising that voter and party representation should loom larger in smaller geographic areas where MEPs have greater scope for meeting with individual constituents than do their counterparts in far-flung regions like Scotland. Among the other items worth singling out from this table are the greater emphasis on policy formation in smaller regions (by both measures), and the greater priority being attached generally to regional representation in smaller regions (again by both measures). Finally, in relation to the overall impact of the new electoral system, MEPs from larger regions (as measured by electorate size) are more likely to regret the loss of ‘sole ownership’ of the constituency (58 per cent), to encounter logistical problems representing the region (84 per cent); and to not like the new electoral system (53 per cent). Once again the two measures of regional size produce countervailing trends, with a greater tendency to see the MEP’s role as having changed in larger regions based on land area (88 per cent, with 100 per cent crediting the new electoral system as the prime reason) and small regions based on electorate size (100 per cent and 100 per cent, respectively). The influence of party-level features on the representative role of an MEP is dealt with in Table 8.5. Three interesting points stand out regarding MEPs from different parties: (a) Labour MEPs placed higher stress on policy formation (67 per cent) than do MEPs from other parties (perhaps indicating a greater emphasis on legislative activities among old hands, given that the bulk of incumbents in our sample were Labour MEPs); (b) Conservative MEPs placed the lowest emphasis on party representation (36 per cent) and also on representing the region (52 per cent) (in addition, fewer Conservative MEPs thought that their role was different from that of an MP (60 per cent) than is the case for MEPs from other parties); and (c) MEPs from smaller parties placed the greatest emphasis on party representation (60 per cent)—something to be expected
181
Length of service in EP
Former MP
New
Incum
Yes
No
MEP’s role (multiple options) Voter representation Policy formation Party representation Ideological representation
78 56 14 19
72 48 20 4
78 44 11 22
75 54 17 12
Who do MEPs represent (multiple options) Europe Britain Party Region Social group/ideology
0 22 47 83 42
8 16 48 60 8
0 22 33 56 33
4 19 50 75 31
Believe their role is different from MPs (% yes) Believe their role is different from other MEPs (yes) MEP’s representative role has changed (% yes) New electoral system affected MEP’s role (% yes) Does not like PR list Logistical problems representing region (% yes) ‘Constituency work’ has increased (% yes) Relationships with organized interests (% yes) Regret the lost of ‘sole ownership’ of constituency (% yes)
69 89 64 78 14 64 3 53 17
60 80 100 100 56 92 16 56 56
78 78 78 89 33 78 0 44 33
63 87 79 87 31 75 10 56 33
Life Under List: Representing a Region
182 Table 8.6. The representative role: individual-level features (%)
Life Under List: Representing a Region
given that in most cases they were the sole party representative in their region—as well as on representing a social group or ideology (47 per cent), which might be expected from UKIP, Green, or nationalist MEPs. The latter trends among smaller parties—of placing greater stress on party and ideological representation—were also evident when we considered the variations in party size (as measured by number of MEPs in the region, or number of MPs from same party in the region). Party size produced one other interesting trend: where there were three or more MEPs in a region, there appeared to be a lower emphasis on representing voters (voter representation = 69 per cent; representing the region = 47 per cent) although this did not coincide with any greater emphasis on party representation. Finally, regarding the impact of the new electoral system, Labour MEPs (reflecting that they had the largest grouping of incumbents) were more inclined to stress that the MEP’s role had changed (95 per cent) and that the new electoral system was the main reason for this change (95 per cent); there were also a significant number of Labour MEPs who claimed that their constituency work had actually increased (19 per cent). Consistent with the party’s line, Conservative MEPs were most hostile to the new system (46 per cent) and most likely to voice regret over the loss of sole ownership of the constituency (48 per cent). Where the MEP was the sole party representative in the region, they were less likely to feel that the role of an MEP had changed (57 per cent); similarly consistent with expectations was the trend to dislike the new system (50 per cent) and the loss of sole constituency ownership (53 per cent) where an MEP had three or more colleagues in the region (again these trends are evident when there were 11 or more Westminster MPs from that party in the region). The third factor to consider is individual-level effects (see Table 8.6). We might expect that those elected for the first time in 1999, having had no experience of life under the SMP system, would find it easier to adapt to the new system. Similarly, MEPs who had not previously been national MPs might place less stress on the constituency orientation.6
183
Life Under List: Representing a Region
Consistent with the expectation that a new electoral system can attract a new sort of politician, newer MEPs were more likely than incumbents to place stress on ideological (19 per cent) and social group representation (42 per cent). Of course, some of this may simply reflect a higher degree of enthusiasm generally for the job among newer recruits, because the new MEPs were also more likely to place stress on representing the region (83 per cent). Former MPs were more likely than MEPs who were not previously MPs to see the role of an MEP as different from that of an MP (78 per cent); while, MEPs who were not previously MPs were more likely to place stress on representing the region (75 per cent) and the party (50 per cent). Finally, concerning attitudes to the new electoral system, the distinction between new and incumbent MEPs was entirely consistent with expectations: incumbents felt the role has changed (100 per cent), that they faced greater logistical problems (92 per cent), and that the new electoral system was responsible for the change (100 per cent); they regretted the loss of sole ownership of the constituency (56 per cent); encountered greater logistical problems representing the region (92 per cent); and did not like the new system (56 per cent). The profile of answers for new MEPs on these matters was strikingly different.
8.3. The Voters or the Party? Our second hypothesis follows from the fact that in addition to changes in the quantity and direction of British MEPs’ representational activities, there are also consequences for how they go about their representative role. Post-electoral reform, MEPs no longer have a sole mandate over particular places—they are no longer ‘the MEP’ for an identifiable geographic area; rather, they share the region with several others, some from other parties. It is plausible that MEPs might come to think less in terms of representing all of ‘their constituents’, and more to be concerned with ‘their voters’: that is those voters in the region that had supported their party list and had thereby contributed to electing them.
184
Life Under List: Representing a Region Now I represent 51 constituencies along with six other people all of whom are not of my party. In that sense I don’t have a very meaningful relationship with my constituents. Instead I have to build a relationship with my party members and I have 17,000 party members spread over an area the size and population of Denmark (South-West MEP).
The second hypothesis, therefore, is centred on the issue of what we refer to as ‘team-related’ matters. The shift from SMP to regional list could be expected to have an effect on the interrelations between MEPs within and across parties and also in their relations with their parties in the wider region. Two main factors are likely to drive this. The shift from single-member to multi-member regions requires MEPs at the very least to have some awareness of the activities of the other MEPs in the area; indeed, in certain circumstances—such as a major crisis in the region—it is likely to produce joint activities. Furthermore, given the increase (in land mass and population) in the size of the representative domain, MEPs by default are likely to adopt more of a regional ambassadorial role (in recognition of the fact that they represent far more than one parliamentary constituency and local authority area): this is likely to have an impact on their relations with other levels of the party in the region. We therefore devoted part of our interviews to teasing out how the teamwork dimension was panning out in reality, with regard to (a) intra-party, (b) cross-party, and (c) intraparty/intra-regional aspects. While, as we saw above, many MEPs bemoan the loss of their sole ownership of a constituency, the more positive side of the coin is the potential that List PR offers in terms of teamwork, providing economies of scale in the representative process, and giving voters greater choice as to which representative they might prefer to approach. This should be most evident with regard to how MEPs from the same party interrelate in the region, with efforts to carve up the constituency geographically, or even to share out the representative load in terms of the thematic or functional expertise of the individual members (McCarthy 1998). While the majority of MEPs claim to have good relations with the other MEPs on the
185
Life Under List: Representing a Region
party team (62 per cent), and in 27 per cent of the cases share office or staff facilities (see Table 8.2), the picture is not entirely rosy. There are undoubtedly cases of where the teamwork does seem to be working well such as one team which claims to operate ‘like an old married couple . . . . We even do joint Christmas cards’ (one of two MEPs from the same party in a region); however, the more common description is along the lines suggested by this Labour MEP: ‘[I]t is not working too badly, but it is not working brilliantly’. Where cooperation was possible, it certainly appeared to help in representing a large region: You try at least to work together as a team . . . among members of your own party because it makes sense not to duplicate each other. For instance on dealing with local authorities which we all deal with, it would be silly for us all to go to the same meeting with one local council and then later in the same day another local council and so on around our region. We divide that up—you deal with this one, I’ll deal with that one, then there is one person they can turn to which is an advantage for them . . . Reporting back to CLPs . . . all of us in my region going round 56 of them it wouldn’t be very easy, so we take so many each—it is so much easier (Labour MEP).
We found substantial evidence that, where this option was open to them, parties with more than one MEP in a region did seek to achieve a degree of teamwork within regions. In a few instances this included joint offices within the region and even staff jointly paid out of two or more MEPs’ office expenses. More commonly, it appeared to involve regular consultations and attempts at coordinating and dividing up work. It was not always clear exactly what form the intra-party teamwork tended to take, but where it is possible to be more specific, we can see that the most common division of labour (26 per cent) was based on geography, or some mix of geographical and functional teamwork (26 per cent), rather than a purely thematic–functional division (see Table 8.2). On the whole, therefore, cooperation tended to follow different patterns—joint planning, division of
186
Life Under List: Representing a Region
labour by functional or geographical criteria, or on a more ad hoc basis: XXX and I get on very well but we just sort of seem to do our own thing. We haven’t divided the region geographically—we cover the ground as we see it, but we do some joint things together, we put out some glossy leaflets together from time to time. Other people I know from other parties in other regions are much more structured. We didn’t feel that would really work (Liberal Democrat MEP).
This is not to say that cooperation always proceeded smoothly; in one region, for instance, relations between the representatives of a major party were at one point so bad that a professional intermediary was required to help resolve differences. One commonly stated explanation for intra-party rivalry was the onset of reselection for the 2004 election—which was starting for all the parties around the time of our interviews. As one Conservative MEP commented: ‘[A]s we get nearer to reselection that process is breaking down, of course, as it is in every region. Our individuals are starting now to compete against each other’. Similar points were made by a Labour MEP: ‘the team system in theory works well but in practice it would be interesting to watch it disintegrate around re-selection’. Another Conservative MEP put things rather more colourfully: ‘[W]e are all running around like blue-arsed flies round our region to get coverage, to get publicity’. The potential for inter-party rivalry is clearly an ever-present factor, so it is quite striking to find in Table 8.2 that 44 per cent of our interviewees claim to have good relations with MEPs from other parties in the region. The responses to this question varied much more starkly than for the question on intra-party relations. On one extreme there was the suggestion that cross-party contacts were few and far between: ‘[W]e see each other like passing ships in the Channel, a distant light going up and down in the distance and that is about it’ (Conservative MEP). On the other extreme was the view of another Conservative MEP who tells his constituents: ‘you had one MEP representing you before and now you have got eight; you are eight times better off’. One newly elected Labour
187
Life Under List: Representing a Region
MEP concurs: ‘So, in a way, cross party working is probably quite exciting and I quite enjoy the work that I have been doing because I think we have actually made a difference’. In some cases, the view was that while there may, indeed, be potential for cross-party collaboration, this is often frustrated by the unwillingness of MEPs from other parties to collaborate, such as the following comment of an MEP from one of the smaller parties: There is an incredible amount of duplication and confusion, frankly. . . . There is absolutely no interest from most of the other political groups—not all, but most—in having any kind of cooperation at all. So we have letters coming in from constituents who start their letters by saying ‘I don’t know which local MEP to write to so I have written to all of you’, and then you can imagine that if not 11 officers that at the very least, you have four officers from different political groups all writing back.
Though most MEPs recognized that the electoral system created the potential for cross-party cooperation to occur, few seemed eager to grasp those opportunities: There is not a great link to be fair and I don’t think that is a problem. We do have a very substantially different political beliefs, certainly when in the past certain organizations such as the CBI would have all of us there or representatives of all of us there . . . it was ghastly, because whilst particularly we were there to listen, usually what would happen is one of us, it could be a Tory it could be Labour, would make some comment, immediately the MEPs from one of the other parties would say ‘Bollocks’ and it all ended up being very negative (Conservative MEP).
Cooperation appeared to be episodic, rather than something that was planned or which, once having tried, MEPs seemed particularly keen to continue. The major exception to this generalization was Scotland: there, several MEPs from different parties reported that relations across party lines were good, and that cooperation occurred across a reasonably broad front. One reason given for this was the recognition of a ‘national’ dimension for Scotland:
188
Life Under List: Representing a Region I do not know a lot about other MEPs that are in the position of representing a whole country . . . and representing a whole country like Scotland is a different ball game altogether. I mean it is not only vast population but it has so many diverse interests and now of course has its own parliament. So I find that we are in a kind of unique position (Scottish MEP).
However, such considerations of national unity and prestige appeared to have markedly little impact on MEPs from Wales. Here cooperation was uniformly reported to be very limited, with antagonism between the Labour Party and Plaid Cymru the main inhibiting factor: ‘Getting Labour and the Nationalists to agree sometimes is very, very difficult on these things. The factors that cause people to be distrustful are going to increase. It seems to me, they are going to develop’ (Welsh MEP). Finally, concerning the notion of MEPs as ambassadors of their region, the interviews provided some support for this. The comment of one Conservative MEP was not untypical: ‘My view is that I am operating like a senator, only not quite the same as a senator, but I am certainly working on the basis of an overview of the region’. One area where this feature is most prominent is with regard to relations MEPs have with other parts of their party in the region. While, overall, 59 per cent of our interviewees claimed to have good relations with their regional parties, there were clearly significant variations across different parties and different regions—so that, while an MEP from one of the smaller parties could claim that ‘I am considered to be the senior party person in the XXX of England’, by contrast, a Labour MEP in another region felt that, if anything, ‘we have disappeared off the map, off everybody’s radar’. As in Section 8.2, in Tables 8.7–8.9 we have disaggregated the trends to take account of regional, party, and individual-level variations. Table 8.7 reports on the regional trends, revealing that team-related matters are for the most part consistent with expectations. MEPs in larger regions (by both measures) tend to experience better relations with fellow-party MEPs, other levels of the party in the region, and also with MEPs from other parties.
189
Life Under List: Representing a Region Table 8.7. Team-related matters: regional features (%) Regional size (land area)a Regional size (electorate)b Small Medium Good teamwork with MEPs from same party (yes) Nature of teamwork—Geographical Functional Mix Sharing of office/staff (fully/partially) Concern over reselection (yes) Impact of reselection on MEP relations (yes) Good relations with MEP’s party in region (yes) Good relations with MEPs from other parties (yes) a b
Large
Small Medium
Large
40
63
88
38
68
63
10 20 30 29
30 14 23 30
25 25 38 0
13 25 25 56
26 15 26 27
32 16 26 19
57 50
30 27
50 50
33 33
36 36
39 29
40
58
88
38
71
47
30
40
88
13
56
37
Small = London and North-East; Large = Scotland. Small = Wales and North-East; Large = North-West, London, and South-East.
Table 8.8. Team-related matters: party-level features (%) Party
Good teamwork with MEPs from same party (% yes) Nature of teamwork—Geographical Functional Mix Sharing of office/staff (% fully/partially) Concern over reselection (% yes) Impact of reselection on MEP relations (% yes) Good relations with MEP’s party in region (% yes) Good relations with MEPs from other parties (% yes)
190
No. of MEPs of same party in region
No. of MPs of same party in region
Lab.
Con.
Oth.
1
2
3+
0
1–10
11+
81
64
33
7
100
72
0
48
80
43 5 43 36
24 28 24 34
7 13 7 5
0 0 0 0
27 56 27 25
36 14 36 37
0 0 0 0
14 29 10 28
37 11 40 30
40
46
18
18
25
46
0
40
40
50
34
9
6
25
44
0
32
38
52
60
67
57
64
58
20
76
54
52
44
33
29
55
47
0
52
46
Life Under List: Representing a Region Table 8.9. Team-related matters: individual-level features (%)
Good teamwork with MEPs from same party (% yes) Nature of teamwork—Geographical Functional Mix Sharing of office/staff (% fully/partially) Concern over reselection (% yes) Impact of reselection on MEP relations (% yes) Good relations with MEP’s party in region (% yes) Good relations with MEPs from other parties (% yes)
Length of service in EP
Former MP
New
Incum
Yes
No
56
72
56
63
14 22 19 29 35 29
44 8 36 24 38 38
22 22 22 42 50 50
27 15 27 25 34 30
72
40
56
60
44
44
56
42
The two respects in which the MEPs from smaller regions stand out is with regard to nature of intra-party teamwork—which seems to place greater stress on functional teamwork than is the case in larger regions (especially when measured by electorate, where 25 per cent of MEPs in smaller regions engage in functional teamwork)—and the tendency to share office resources (56 per cent, when measured by electorate).7 Party-level features are dealt with in Table 8.8, revealing that Labour MEPs appear to have fared best on most dimensions: 81 per cent of them claimed to have good intra-party teamwork and 36 per cent of them shared office resources—suggesting that the party’s stress on the ‘Labour team’ in its 1999 election campaign paid some dividends—much of which is geographical (43 per cent)—perhaps reflecting a greater ability to accommodate their fellow party MEPs by establishing agreed geographical bailiwicks; and 52 per cent of Labour MEPs claimed good relations with MEPs from other parties. When it comes to relations with the party in the wider region, it was the MEPs from smaller parties that appeared to have had most success (67 per cent): something that makes
191
Life Under List: Representing a Region
intuitive sense given that the MEP was often the most senior politician in the region. There are three other interesting findings in this table. First, having two MEPs from the same party in a region as expected did seem to produce better intra-party teamwork (100 per cent), with a greater emphasis on functional divisions of labour. Second, relations with the party in the region were better where it had only a few MPs (76 per cent). However, a third finding was that where the party had many Westminster MPs in the region, teamwork between its MEPs seemed to be enhanced (80 per cent). Reselection concerns also appeared largely as expected, with the concerns greatest among Conservative MEPs (46 per cent), reflecting the fact that their fates were determined by regional party hustings with (unlike for Labour) no inbuilt incumbency protection measures. Nevertheless, despite the higher concerns felt by Conservative MEPs, it was Labour MEPs whose intra-party relations were most affected by the reselection process (50 per cent), perhaps reflecting the greater degree of competition between incumbents produced by that party’s selection procedures (for details, see Butler and Westlake 2000: ch. 4). There was also a clear relationship between the number of MEPs in the region and reselection concerns (46 per cent in those cases where there were 3 or more MEPs). Table 8.9 reports on individual-level variations in the nature of MEP teamwork. Incumbents were more inclined to claim that intra-party teamwork was working well (72 per cent), reflecting the fact that they often had the biggest input into determining the nature of the regional carve up with fellow MEPs.8 Incumbents also placed greater stress on geographical teamwork (44 per cent), whereas the newer MEPs saw more of a role for functional teamwork (22 per cent), and were more likely to share office resources (29 per cent). New MEPs seemed to have better relations with the party in the wider region (72 per cent).9 As we might expect, former MPs were more likely to emphasize geographical teamwork (27 per cent), and less likely to share office resources (25 per cent). Finally, reselection concerns were also as anticipated. Incumbents were more anxious (38 per cent), as were former
192
Life Under List: Representing a Region
MPs (50 per cent)—the latter possibly reflecting their prior experience as sole stewards of their parliamentary constituencies.
8.4. Conclusion Chapter 6 left us with something of a conundrum in the shape of the representative behaviour of British MEPs. Our theoretical expectations were that the shift from SMP to a closed-list electoral system should produce a change in how British MEPs represent voters, yet in the cross-national analysis of survey trends British MEPs remained as stubborn outliers. The purpose of this chapter has been to make use of additional material derived from detailed face-to-face interviews with a large sample of British MEPs, interviews with some of their parliamentary assistants, and visits to and observation of MEPs’ regional offices across Britain, with the aim being to assess the degree to which the summary evidence provided by quantitative surveys accurately reflects the true state of affairs. The analysis in this chapter was arranged around the assessment of two main hypotheses on (a) the representative role of British MEPs and (b) their team-related activities (including an additional hypothesis on pressures resulting from the candidate selection process). The first hypothesis was in large part confirmed. Even though the MEPs continue to emphasize voter representation in their responses to our interview questions (and evidently also in their responses to the survey instrument analyzed in Chapter 6), when we tease the matter out in more detail it is clear that there has been a decline in constituency representation, not least as revealed by the shrinking size of their mailbags, but also based on the more considered responses of MEPs to questions about how things compared pre- and post-1999, as well as the reactions of their staff to how things have changed. And certainly there are very few MEPs (a mere 8 per cent of our sample) who claim that constituency work has actually increased as a result of the reform.
193
Life Under List: Representing a Region
The evidence presented in Tables 8.4–8.6 provides important clues of where the electoral reform has had greatest impact. It is worth summarizing some of the highlights, as follows: (a) the larger the region the greater the impact of the new electoral system on MEPs’ representative roles, (b) Conservative MEPs show greatest reluctance to adapt, (c) the more MEPs a party has in a region the lower its emphasis on voter representation, and (4) first-time MEPs are more likely to embrace and work with the new system than are incumbents. Similarly, our second hypothesis—predicting a greater emphasis on teamwork—was also in large part confirmed. The requirements of the new multi-member closed-list electoral system have led inevitably to a growing emphasis on intra-party teamwork, as well as a recognition of the need, at least on occasions, to cooperate across party lines. Once again we can present some of the highlights of our findings (from Tables 8.7–8.9), as follows: (a) teamwork tends to be more prominent among MEPs in larger regions (although, functional teamwork features more prominently among MEPs in smaller regions, as well as in those cases where there are no more than two MEPs from the same party in a region), (b) Labour MEPs (in part, perhaps, reflecting the ‘team-based’ nature of their 1999 election campaign) are more likely than their Conservative counterparts to emphasize teamwork, and (c) functional rather than geographical teamwork is more prominent among firsttime MEPs (as well as among those who had not previously been MPs). As suggested in Section 8.1, the irony of the closed-list system is that while, as an electoral system, it may incentivize teamwork among a party’s MEPs, the selection procedures that inevitably accompany it can have the opposite effect—or, at least in the heat of the selection process, can set limits on the degree of cooperation between a party’s MEPs. This formed the basis of our final hypothesis (2a), which was also confirmed by our interview evidence. As we saw in Tables 8.8 and 8.9, there were some important differences based on the party affiliation of MEPs (with Conservative MEPs showing greatest concern over reselection, but their Labour
194
Life Under List: Representing a Region
counterparts reporting reselection to have had the greatest impact, as well the political experience of individual MEPs (with incumbents feeling most concern). The evidence from our interviews with the bulk of British MEPs points to quite considerable change in their representative activities. Life for British MEPs is undoubtedly quite different under List than it had been under SMP. It may well be true (as demonstrated by the cross-national survey evidence presented in Chapter 6) that British MEPs continue to emphasize the representation of individual voters to a greater degree than might be anticipated by the electoral system dynamics generally associated with a closed-list system, but what our interview evidence has shown is that when we explore the underlying attitudes to the representative process and the changing practices associated with it, there can be little doubt that change has occurred, and in directions that are very much consistent with our expectations.
Notes 1. To preserve the anonymity of our interviewees, we provide at most their party label (as Conservative, Labour, or smaller party). However, in those instances where there is a risk of revealing an MEP’s identity we do not report the party label. 2. The irony of using Sainte-Laguë for allocating seats across regions while using the less proportional d’Hondt formula for allocating seats across parties was not lost on some scholars (e.g. Curtice and Range 1998). 3. Of course, for some MEPs the shift to multi-member regions did have its plus points: ‘From the point of view of one’s own time, of course, it is much easier because it provides the eternal excuse that if you can’t be at an engagement in XXXX then you must be at an engagement in XXXX; actually you are at home doing the gardening or whatever it is’ (Conservative MEP). 4. An MEP is counted as having a publicly advertised constituency office if one with an address in their constituency was listed in the List of Members (published by the parliament or available on the Europarl website) or was found on a party or individual website.
195
Life Under List: Representing a Region 5. Not all our interviewees gave clear or coherent explanations for why they felt their role differed from that of MPs, or other MEPs, which explains the missing cases here and in later tables. 6. Furthermore, those MEPs with a significant job in the EP (e.g. a committee chair) as a reflection of their status may place greater stress on parliamentary-oriented activities (as opposed to constituencyoriented activities) than their counterparts. For instance, when asked the unprompted question on the more important feature of an MEP’s role, one MEP responded: ‘The most important feature? Well, the role I have got. I am chairman of the XXX committee. It’s a specific position, and it kind of overrides everything else.’ In later versions of this chapter we will include this dimension in our analysis. 7. Interestingly, regional size variations produce no major differences with regard to reselection concerns. 8. In a number of our interviews it was obvious that the richer regional pickings often went to the incumbents. 9. In general, the new MEPs seem more inclined to try and accommodate the regional party. For instance, often the incumbents retained the constituency office they used under the previous SMP mandate, whereas many of the new MEPs established new offices that were perhaps more appropriately located in terms of the wider interests of the party, including in some cases the sharing of office space with MPs and other elected representatives in the region.
196
9 Conclusion: The Failure of Parliamentary Representation in Europe
This book has ranged over a considerable number of topics, themes, and sources of empirical data in its analysis of the state of parliamentary representation in Europe. It is now time to bring all of these disparate threads together, and to assess what, taken collectively, they tell us. The purposes of this concluding chapter are two-fold. First, we will briefly summarize the major findings and central arguments of the book. Second, we will then assess what we believe to be their main implications. Three sets of implications will be developed: for the EP, and its place within the institutional structures of the EU; for the study of electoral systems; and for the future of political representation in the EU, and how electoral institutions might be adjusted to facilitate more effective representation.
9.1. What Have We Learned? We began this study with two observations: that the EP offers enormous potential as a research site for students of electoral systems and their impact on political representation; and that many member states of the EU have chosen to employ systems for their European elections that probably make the EP less effective than it might be at providing the political representation that the
197
Conclusion: The Failure of Parliamentary Representation in Europe
public most want or are most likely to respond positively towards. The greater bulk of this book has been devoted to elaborating on, illustrating, and extending these two central ideas. Thus, we began the detailed work of our study, in Chapter 2, by exploring public attitudes to political representation in the EU. The core argument developed in this chapter was that while public knowledge of such matters is certainly low, and their attitudes are not particularly well-formed (nor, for that matter, very wellresearched), it does not follow that we have no idea at all as to the forms of representation that the public want, or would likely respond favourably to. Public attitudes to the EU, including the EP, have become markedly negative over the past 10–15 years; much of this appears linked to a prevailing public sense that the EU is a remote entity about which they have little knowledge or understanding. And there is strong evidence, albeit fragmentary in nature, of a public desire for representatives with whom they can identify and see as linking the structures of the EU to themselves and their communities. Chapter 3 was concerned with exploring the developing state of scholarly knowledge about electoral systems, and their consequences for political representation. As we showed, while the development of much of this field of enquiry has been impressive, there remains no clear academic consensus when evaluating the role of electoral system differences in shaping many aspects of representative attitudes and behaviour, notably in regard to what we term ‘geographical representation’. We then moved on, in Chapter 4, to demonstrate that—notwithstanding recent legislation on UEP for EP elections—there are still substantial differences among the electoral systems used to choose the members of the EU’s elected chamber. These differences exist with regard to all three features of electoral system design: district magnitude, EF, and ballot structure. Drawing on the discussion in Chapter 3, we classified the different electoral systems used into three main types: ‘open’, ‘ordered’, and ‘closed’. And we further showed a clear relationship between electoral system variation and the types of individuals elected to the EP, a relationship that confirmed
198
Conclusion: The Failure of Parliamentary Representation in Europe
the findings of previous research. Despite a decline in the ratio of MEPs to citizens across the EU, the EP has been becoming steadily more representatives of its voters. Consistent with the adoption of proportional electoral systems, there is a good story to tell about the EP’s record with regard to ‘descriptive representation’, as shown by the representation of smaller parties, gender balance, and the entry of small numbers of ethnic minority MEPs. While there remain variations in the representativeness of national delegations to the EP that are partly attributable to electoral system differences, the EP has increasingly come to look much more like the European populations that it exists to represent. Chapter 5 began the process of examining the attitudes of European parliamentarians to political representation and the potential role of electoral systems in shaping those attitudes. We explored MEPs’ views on the representative role of their institution and on their own role as individual representatives. We found, particularly in regard to the latter, that electoral systems were among a number of factors (along with individual, political, and national differences among representatives) that help explain these attitudes. This analysis then led on to our more detailed examination of electoral system effects on geographical representation in Chapter 6. Here, we showed that there is also a relationship between electoral system design and the representative style of MEPs once elected. Our detailed cross-national examination of survey data revealed that MEPs from more ‘open’ electoral systems do tend to pay more attention to geographical representation, serving the interests of individual voters (‘constituents’) and, in Carey and Shugart’s terminology (1995), ‘chasing personal votes’ in election campaigns. The relationships were strong and significant, although, as anticipated, not entirely straightforward. And consistent with the views of such scholars as Richard Katz and Pippa Norris, the relationships were also heavily conditioned by cultural factors, nowhere more distinctly than in Britain whose MEPs manifested strong constituency orientations even after the replacement of their SMP electoral system with closed regional lists.
199
Conclusion: The Failure of Parliamentary Representation in Europe
These issues were explored from another angle in Chapters 7 and 8. Here, we looked in detail at the British case and the impact of the dramatic electoral reform instituted in 1999. We found, once again, that a shift to PR had altered who was elected, and in the expected direction: Britain’s cohort of MEPs became much more descriptively representative, in all major respects, after 1999. But we also found that the change in electoral institutions had influenced how those elected understood their role as representatives, and the manner in which they sought to carry it out. After electoral reform there was a decline in the importance that British MEPs accorded to representing individual citizens, a decline in their active attention to constituency-related matters (something that was particularly noticeable in the larger British regions) and a growing tendency for them to concentrate on work inside the EP and on relations with their party colleagues. Although the degree of importance accorded to geographic representation by British MEPs remains high in comparison to their counterparts from other member states, there is little doubt that it has declined in comparison to what occurred prior to electoral reform.
9.2. Implications for the EP The EP is an institution that has experienced probably more change and development than any other EU body in the past 15–20 years. As the EP has grown in importance, so subsequently there has been a substantial growth and development of scholarly understandings of the chamber. We now know a great deal about the powers of the EP (e.g. Tsebelis and Garrett 2003; Tsebelis et al. 2001); about the history of the EP’s development (Rittberger 2005); about EP elections (van der Eijk and Franklin 1996; Blondel, Sinnott and Svensson 1998); and about the parliament’s internal politics (Kreppel 2002; Scully 2005; Hix, Noury and Roland 2006). But this study has been able to generate some further significant lessons for students of the EP.
200
Conclusion: The Failure of Parliamentary Representation in Europe
A first lesson that we can identify is something of a negative one. It was clear from the discussion in Chapter 2 that while knowledge about the EP has developed greatly in many areas, our understanding of public attitudes towards the chamber remains distinctly limited. While we were able to make a strong case regarding the forms of representation that the public desire and would respond positively towards, the development of this case required us to draw on quite a fragmentary body of evidence. There remain few detailed studies of public attitudes towards specific EU institutions (cf. Caldeira and Gibson 1995; Gabel 2003). One defence of this state of affairs might be that existing evidence—as was mentioned in Chapter 2—points to very high levels of public ignorance of the EP and other EU institutions: it would seem difficult to develop an analysis of public attitudes when such attitudes do not really exist! But the obvious answer to this is simply that most survey respondents do, nonetheless, manage to reply to questions about the EU, and it is important for scholars to probe much further, to understand in more detail what lies behind such replies. One area where such work is certainly needed is to explore further what people might want, or do want, from their political representatives in the EU. A second lesson that we can draw from this study is really an old lesson for political science, but re-learned in the context of the EP. A ubiquitous feature of the comparative study of parliaments and parliamentarians is an emphasis on the ‘electoral connection’ (Mayhew 1974): the notion that a considerable amount of the behaviour of elected representatives, and even much of the organization of the institutions in which they serve, can most readily be understood as following from the imperatives created by representatives’ desire to secure a continuing electoral mandate. Notions of an electoral connection in the EU have hitherto received little attention (see Carruba 2001). But our study indicates that they ought to receive much greater attention. Our evidence indicates strongly that how MEPs understand and approach their role is shaped quite significantly by electoral imperatives, and that a change in these imperatives, caused by a change in the
201
Conclusion: The Failure of Parliamentary Representation in Europe Table 9.1. Electoral system design and elements of voter linkage: evidence from the 2004 EP elections Types of electoral system Open Ordered Closed
Candidate called to home (%)
Voter received election leaflets (%)
Voter had necessary information (%)
13.6 2.2 3.7
75.4 67.7 59.3
68.9 55.4 58.8
Note: Electoral system categorisations based on Figure 4.1. For details on question wording, and country scores, see Table 2.5. Source: Flash Eurobarometer: Post European Elections 2004 Survey, June 2004.
manner in which re-election is achieved, can alter the attitudes and behaviour of these representatives. Furthermore, while the EP does struggle for attention, the behaviour of MEPs is not wholly without public impact. We show this in Table 9.1, when we compare public reactions to the 2004 EP election campaign across our three categories of electoral system—open, ordered, and closed. The table displays the mean national average ratings for each of the three classes of electoral system on three questions, and we can see some strong differences. The public in countries using open electoral systems clearly believed that they had received more attention from candidates than did those from the other types of system; at the same time, they also felt that they were better informed about the election by the end of the campaign. There is strong evidence that, although it may operate differently from how it does in most EU member states, there is nonetheless an important electoral connection that links representatives and represented in the EP. The final lesson that we would draw about the EP is a reasonably positive one. As we observed in the opening chapter of the book, it is certainly possible to criticize the EP’s role in EU politics: there is little or no sign that the strategy of empowering the institution to help legitimize the EU as a whole is actually working. But as we have seen, the EP has done better as a representative body in some respects than is generally recognized: in terms of descriptive representation, certainly, the chamber has increasingly
202
Conclusion: The Failure of Parliamentary Representation in Europe
come to look like the societies that it is representing. And where the EP has done less well as a representative body, this would seem to be at least in part due to a factor—the electoral systems under which most MEPs are elected—that could potentially be changed.
9.3. Implications for the Study of Electoral Systems Much like the dramatic change and development of the EP visà-vis the EU as a whole, the past 20 years have also witnessed dramatic developments in the study of electoral systems. Lijphart’s observation two decades ago that the ‘study of electoral systems is undoubtedly the most underdeveloped subject in political science’ (1985: 3) is now quite patently outdated. The period since then has witnessed considerable growth and development of a cumulative body of research. But significant gaps remain, and none is more significant than that relating to research on the representative link between politicians and their publics. As one recent authoritative review of the state of electoral systems research contends, this area is one of those most in need of development; moreover, as Matthew Shugart went on to observe, research hitherto ‘has been hampered by a sometimes nebulous characterization of the dependent variables, a lack of data, and even worse, a lack of clear understanding of what the rules being investigated across countries are’ (2005: 36). Our study has made significant progress in addressing all three of Shugart’s concerns. First, from the outset we established a clear and simple characterization of the dependent variable: namely, the degree of contact of elected politicians with their constituents (‘geographical representation’). This has served as the key dependent variable throughout the main analysis conducted in the book, and particularly that in Chapters 6–8. Second, we were able to deploy two substantial and highly informative new sources of data: these were the cross-national surveys of MEPs (and particularly that conducted by us in 2006, which was able to draw on a wide range of electoral system variation from 25 EU
203
Conclusion: The Failure of Parliamentary Representation in Europe
member states), and the qualitative evidence gathered from our series of face-to-face interviews with the bulk of British MEPs in the years after the 1999 British electoral reform. Finally, in this study we have gone to considerable effort to gather full and accurate information on the various electoral systems in use across the 25 EU member states for electing their MEPs, and we have then drawn on this information to categorize clearly the various electoral systems into three groups, in terms of their varying degrees of ‘openness’. Having been able to avoid these weaknesses in previous work, we are therefore in a strong position to draw out major electoral system-related conclusions from the study. These are ‘cultural’ factors that are clearly important in shaping how representatives understand and seek to carry out their role: the quantity and form of geographical representation conducted by representatives is likely to be influenced strongly by prevailing norms in the political culture about how important such matters are. But cultural influences do not preclude fairly direct electoral system effects. And as we saw in the case of British MEPs, exogenous institutional changes may themselves prompt changes in ways of thinking and behaving that may plausibly, in the longer-term, shape a new cultural understanding of political representation.
9.4. So What Would Be the Ideal Electoral System for the EP? The academic debate over which is the ‘best’ electoral system has tended to be rather inconclusive. While some seem prepared to express an opinion, favouring one electoral system over all others (for a sample, see Blais and Massicotte 1996; Lijphart 1994; Reilly 2001; Reynolds 1999; Sartori 1997; Taagepera and Shugart 1989), there is plenty of disagreement over which electoral system is best (although, in recent years, mixed-member systems do appear to feature quite highly in the rankings of many scholars; see Bowler and Farrell 2006); and some scholars are quite insistent
204
Conclusion: The Failure of Parliamentary Representation in Europe
that there is no such thing as an ‘ideal’ electoral system. As Katz (1997: 308) puts it, the decision over which electoral system to adopt is contingent on the circumstances of the case under consideration. The reasons for such differences of opinion are perfectly understandable. It is very difficult to make categorical statements about the best electoral institutions for all circumstances, given the various permutations that can occur in electoral system design, and the wide range of consequences that flow from these permutations, whether for the partisan, ethnic or socio-economic composition of the parliament, the makeup of the government, the nature of policy output, styles of representation, overarching system stability, engagement of voters, and so on. But given the much more specific focus of this study on the effects of electoral system design for the representative process in the EU, we can afford to be more definitive in our judgement of what would constitute an ideal electoral system for electing MEPs. Our proposals are much in line with the principal features of the 1998 Anastassopoulos report (see Chapter 4, p. 73), especially with regard to the following two key features. First, we would propose the abolition of closed-list systems. Ideally we would prefer to see MEPs elected via an open list system (such as Finland’s or Italy’s) or even the STV used in Ireland and Malta; however, we would still see it as an improvement to move to an ordered list system (e.g. such as used to elect Belgian and Dutch MEPs).1 Second, we would propose that larger member states (those any larger than the Netherlands) should regionalize their electoral procedures, abolishing national lists and thereby bringing their European representatives ‘closer’ to their citizens. We realize that this latter reform would have some implications for aggregate proportionality (given the relationship between district magnitude and proportionality). But this leads us to our third recommendation, namely that, wherever possible,2 in those cases where MEPs are elected to multi-member regions, rather than nationally, the determination of seat proportions for the parties should be based on their national vote, with seats subsequently being awarded to candidates based on regional votes
205
Conclusion: The Failure of Parliamentary Representation in Europe
(personal and/or party)—in other words, the basis of the electoral system used to elected Italian MEPs. That these proposals have already been mooted (and, in the latter case, actually applied) supports our contention that they are realistic. Nevertheless, we can expect criticism, particularly from those scholars (and MEPs) who may object that this would undermine the efficiency and effectiveness of the Parliament. There is undoubtedly a trade-off between a strong ‘geographical’ or constituency-orientation on the one hand and the emphasis placed by MEPs on national and party issues and more specifically a strong committee-orientation on the other. The argument commonly made is that forcing MEPs to pay more attention to issues relating to their voters can distract them from their legislative and committee roles. We would make three points in response to this. The first is simply that even under a fully open electoral system there is still plenty of scope for MEPs to place emphasis on their activities in the Parliament. Enhancing the incentives for representatives to accord attention to their constituents does not in any way mandate total dereliction of their other parliamentary duties. A second point we would make is that even if there is something of a trade-off, this may be no bad thing: MEPs should be kept well informed about the needs and desires of their voters if they are to behave inside the parliament in a manner that gives due consideration to those whom they represent. Third, it is quite arguable that many such inefficiencies are in fact a ‘good’ for the representative process. In this we find ourselves in agreement with Andreas Follesdal and Simon Hix (2006: 555) who suggest that a ‘worry about the efficiency loss of politicization . . . seems illfounded’. To these immediate responses to potential criticisms of our proposals, we would add one final, concluding thought. What is the major problem facing the EU as an entity today? It is probably not that the Union suffers from a lack of sufficiently detailed scrutiny of proposed European legislation or of the actions of EU executive agencies. (To say this is in no way to decry the value of such scrutiny when it is effectively conducted.) Far
206
Conclusion: The Failure of Parliamentary Representation in Europe
more pressing, we would suggest, is the fact that while the EU has in many respects been an extraordinarily successful political project—quite arguably one of the most extraordinary and most successful in human history—it has lost the support of many of Europe’s people along the way. Much of the European public feel distant and alienated from the EU; the main representative body in the Union, the EP, has not thus far been able to bridge this gap effectively. While it may not offer a complete answer to the problem, bringing Europe’s elected representatives rather closer to those they represent could well be one of the most practical steps that could be taken in order to better and more effectively represent Europe’s citizens.
Notes 1. In this respect, we concur with Shugart’s normative position that more open electoral system design is ‘most appealing’ (2005: 44). 2. We realize that STV systems are likely to be excluded from this option.
207
This page intentionally left blank
Appendices: Principal Data Sources Used in This Study
The empirical analysis developed in this book draws on several sources of data. Details about each source are given where relevant in the text; here, we summarize the major sources used.
The 1996 MEPMNP Survey The 1996 survey data analysed in this book were produced as part of the broader European Representation Study, which included surveys of MEPs (alongside parallel surveys of Members of National Parliaments). These surveys were coordinated by Jacques Thomassen and Bernhard Wessels. Fieldwork for the MEP study was conducted in May–June 1996, with the survey returning 314 responses, a return rate of 50.2 per cent. For further details regarding this survey and the data-set produced, see Katz and Wessels (1999).
The MEP2000 Survey The MEP2000 Survey was co-authored by Simon Hix and Roger Scully, on behalf of the European Parliament Research Group, and funded by a grant to Simon Hix under the ‘One Europe or Several?’ research programme of the Economic and Social Research Council of the United Kingdom (Grant: L213252019). The survey was administered as a mail survey in September 2000; response rates by nation were as detailed in the table below. Further details about the survey, the complete data-set and an English version of the entire questionnaire used are available from the website
209
Appendices: Principal Data Sources Used in This Study
No. respondents
No. MEPs
Response rate (%)
2 6 7 7 22 27 8 4 23 5 15 11 17 10 35
21 25 16 16 87 99 25 15 87 6 31 25 64 22 87
9.5 24.0 43.8 43.8 25.3 27.3 32.0 26.7 26.4 83.3 48.4 44.0 26.6 45.5 40.2
199
626
31.8
Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden UK Total
of the European Parliament Research Group: http://www.lse.ac.uk/ Depts/eprg/.
The 2006 MEP Survey The MEP2000 Survey was co-authored by David Farrell, Simon Hix, and Roger Scully in collaboration with Katarina Thomson and Mark Johnson from the National Centre for Survey Research. Core funding for the survey was provided by the Economic and Social Research Council of the United Kingdom (Grant RES-000-22-1554) as well as under the EU’s Framework 6 programme (resulting from David Farrell’s involvement in the CIVICACTIVE project directed by Richard Sinnott at University College Dublin). In addition, we are grateful for the financial support provided by Claes de Vreese (University of Amsterdam) who supplied some of his budget from a Dutch National Science Foundation grant, and also from Susan Banducci (University of Twente). The survey was administered via the Internet in spring 2006. The survey was arranged across several discrete ‘pages’, and there was some attrition of respondents between the first and final pages. The number of respondents who completed at least some substantial part of the survey was as follows.
210
Appendices: Principal Data Sources Used in This Study
Austria Belgium Cyprus Czech Rep. Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Portugal Slovakia Slovenia Spain Sweden UK Total
No. respondents
No. MEPs
Response rate (%)
7 13 5 11 6 3 5 23 34 2 8 7 29 4 6 3 1 9 22 9 3 5 11 9 37
18 24 6 24 14 6 14 78 99 24 24 13 78 9 13 6 5 27 54 24 14 7 54 19 78
38.9 54.2 83.3 45.8 42.9 50.0 35.7 29.5 34.3 8.3 33.3 37.2 37.2 44.4 46.2 50.0 20.0 33.3 40.7 37.5 21.4 71.4 20.4 47.4 47.4
272
732
37.2
Further details about the survey, the complete data-set and an English version of the entire questionnaire used are available from the website of the European Parliament Research Group: http://www.lse.ac.uk/Depts/ eprg/.
Interviews From late 2001 through to early 2003 we carried out detailed face-toface interviews with 61 British MEPs, 73 per cent of the (then) British contingent to the EP. Funding for this part of the project was provided by the ESRC (R000239231). The breakdown of these interviews by party was: Conservatives (24), Labour (21), Liberal Democrats (8), and other parties (8). The breakdown by electoral region was: North West (8), Scotland (8), East (6), East Midlands (6), London (6), West Midlands
211
Appendices: Principal Data Sources Used in This Study (6), South East (5), Yorkshire (5), North East (4), Wales (4), and South West (3). In addition, we visited MEPs regional offices and interviewed their key staff. In all, twenty such visits were made to offices (some of which were joint offices of two or more MEPs from the same party) across the UK. The breakdown by electoral region was: East Midlands (2), London (2), North East (1), North West (2), Scotland (5), South West (1), West Midlands (2), and Yorkshire (3).
Appendix Tables for Chapter 4 Table A.4.1. Disproportionality trends in EP elections, 1979–2004 (Gallagher index)
Austria Belgium Britain Cyprus Czech Rep. Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands N. Ireland Poland Portugal Slovak Rep. Slovenia Spain Sweden
212
1979
1984
1989
1994
1999
2004
6.56 22.31
6.86 18.54
5.74 17.50
8.99 23.90
3.10 5.50 7.81
7.85
6.35
7.65
6.30
3.41
6.22 3.59
8.30 5.34 2.17
5.70 2.17 3.59
9.64 9.67 4.54
4.80 4.58 4.85 5.65
10.09 1.05
13.90 1.05
8.80 1.24
9.70 1.16
7.00 1.30
12.66
12.65
14.50
8.62
10.75
5.78 13.26
3.97 15.44
4.66 12.88
3.55 10.23
2.50 17.62
4.79
3.09
6.26
4.35
3.50
3.47 2.77
4.72 4.52 8.24 10.32 6.97 4.48 13.12 6.55 10.21 4.00 3.15 3.75 9.77 1.50 13.21 9.15 11.59 10.54 3.88 17.78 3.65 2.83 8.64 12.43 2.56 3.34
Appendices: Principal Data Sources Used in This Study Table A.4.2. Effective number of (Parliamentary) parties trends in EP elections, 1979–2004 (Laakso–Taagepera index)
Austria Belgium Britain Cyprus Czech Rep. Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands N. Ireland Poland Portugal Slovak Rep. Slovenia Spain Sweden
1979
1984
1989
1994
1999
2004
6.00 1.56
7.29 2.00
7.41 2.00
9.62 1.69
3.47 9.92 3.13
6.10
5.45
5.12
5.82
5.33
3.85 2.67
2.58 2.80 3.00
4.50 3.74 2.91
5.00 2.94 3.24
5.33 6.79 3.14 3.49
3.69 4.22
3.28 4.22
3.81 4.95
3.16 6.02
3.69 7.99
2.57
2.57
2.57
3.60
3.60
3.11 3.00
3.57 3.00
3.49 3.00
4.68 3.00
5.24 3.00
3.68
3.14
2.68
3.52
3.00
3.00 5.23
3.45 8.02 4.12 3.60 4.18 4.91 3.00 4.46 4.14 3.96 2.94 2.50 3.76 5.78 3.53 4.32 3.00 1.92 5.58 3.00 6.12 2.50 4.91 3.76 2.41 5.91
213
Appendices: Principal Data Sources Used in This Study Table A.4.3. Proportion of women MEPs elected to the EP, 1979–2004
Austria Belgium Britain Cyprus Czech Rep. Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands N. Ireland Poland Portugal Slovak Rep. Slovenia Spain Sweden
214
1979
1984
1989
1994
1999
2004
8.33 14.10
16.67 14.81
16.67 14.81
32.00 19.05
38.10 32.00 24.14
31.25
37.50
37.50
43.75
37.50
22.22 14.81
20.99 19.75 8.33
23.46 30.86 4.17
28.74 34.34 16.00
43.75 40.23 36.36 16.00
13.33 12.35
13.33 9.88
6.67 12.35
26.67 12.64
33.33 9.20
16.67
33.33
50.00
33.33
0.00
20.00 0.00
28.00 0.00
28.00 0.00
32.26 0.00
35.48 0.00
12.50
12.00
20.00
15.00
32.81
34.37 40.91
38.89 29.17 25.33 0.00 20.83 35.71 33.33 35.71 43.59 31.31 29.17 33.33 38.46 19.23 22.22 38.46 50.00 0.00 44.44 33.33 12.96 25.00 35.72 42.86 33.33 57.90
Appendices: Principal Data Sources Used in This Study Table A.4.4. Ratio of MEPs per voting age population in the EP, 1979–2004
Austria Belgium Britain Cyprus Czech Rep. Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands N. Ireland Poland Portugal Slovak Rep. Slovenia Spain Sweden
1979
1984
1989
1994
1999
2004
283,358 519,615
290,653 537,401
295,678 546,026
288,452 503,496
278,460 293,739 514,904
232,827
237,791
245,221
246,512
250,600
434,328 572,802
455,317 548,790 324,596
473,434 564,492 347,807
448,503 610,310 338,385
259,527 461,293 614,009 359,045
145,920 520,906
160,893 548,627
163,563 577,845
175,483 556,008
190,957 566,417
35,457
35,739
36,337
37,339
37,001
391,990 343,163
419,046 354,678
444,859 373,502
374,796 383,434
382,645 396,733
322,081
342,467
344,026
486,013
486,630
513,584 302,918
336,067 314,677 574,461 79,318 345,145 286,619 145,634 301,500 532,290 623,054 414,119 335,260 240,888 639,158 154,996 204,161 35,720 60,857 450,699 357,556 543,978 361,266 300,747 232,703 622,655 359,022
Sources: Times Guides to the European Parliament (various); www.europarl.eu.int/election/default. htm.
215
Appendices: Principal Data Sources Used in This Study
Appendix Table for Chapter 6 Table A.6.1. Country codes for the modified Shugart index
Austria Belgium Britain Cyprus Czech Republic Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Northern Ireland Poland Portugal Slovakia Slovenia Spain Sweden
216
B
V
M
Total
2 2 1 2 2 2 2 2 1 1 1 1 3 2 2 2 2 3 2 3 1 1 2 2 1 2
2 2 1 2 2 3 3 4 1 1 1 1 4 3 2 3 3 4 2 4 1 1 2 2 1 2
1 1 1 1 1 2 2 2 1 1 1 1 2 2 1 2 2 2 1 2 1 1 1 1 1 1
5 5 3 5 5 7 7 8 3 3 3 3 9 7 5 7 7 9 5 9 3 3 5 5 3 5
Bibliography
Ames, B. (1995). ‘Electoral Strategy under Open-List Proportional Representation’, American Journal of Political Science, 39: 406–33. Anderson, C. (1995). ‘Economic Uncertainty and European Solidarity Revisited’, in C. Rhodes and S. Mazey (eds.), The State of the European Union 3. London: Longman. (1998). ‘When in Doubt, Use Proxies: Attitudes towards Domestic Politics and Support for European Integration’, Comparative Political Studies, 31: 569–601. Andeweg, R. (2005). ‘The Netherlands: The Sanctity of Proportionality’, in M. Gallagher and P. Mitchell (eds.), The Politics of Electoral Systems. Oxford: Oxford University Press. Bardi, L. (1987). ‘Preference Voting and Intra-party Competition in Euroelections: Prospects for Harmonization’, European University Institute Working Papers. (1990). ‘The Harmonisation of European Electoral Law’, in S. Noiret (ed.), Political Strategies and Electoral Reforms: Origins of Voting Systems in Europe in the 19th and 20th Centuries. Baden-Baden, Germany: Nomos Verlagsgesellschaft. Bean, C. (1990). ‘The Personal Vote in Australian Federal Elections’, Political Studies, 38: 253–68. Beetham, D. (1991). The Legitimation of Power. London: Palgrave Macmillan. Black, T. R. (1999). Doing Quantitative Research in the Social Sciences: An Integrated Approach to Research Design, Measurement and Statistics. London: Sage. Blais, A. and Massicotte, L. (1996). ‘Electoral Rules’, in L. LeDuc, R. Niemi, and P. Norris (eds.), Comparative Democratic Elections. London: Sage.
217
Bibliography Blondel, J., Sinnott, R., and Svensson, P. (1998). People and Parliament in the European Union: Participation, Democracy, and Legitimacy. Oxford: Clarendon Press. Bogdanor, V. (1981). The People and the Party System: The Referendum and Electoral Reform in British Politics. Cambridge: Cambridge University Press. (1985). ‘Conclusion’, in V. Bogdanor (ed.), Representatives of the People? Parliamentarians and Constituents in Western Democracies. Aldershot, Hants, UK: Gower. Boston, J., Levine, S., McLeay, E., and Roberts, N. (1999). ‘The New Zealand Parliamentarians: Did Electoral System Reform Make a Difference?’, Legislative Studies Quarterly, 12: 63–75. Bowler, S. and Farrell, D. (1993). ‘Legislator Shirking and Voter Monitoring: Impacts of European Parliament Electoral Systems upon LegislatorVoter Relationships’, Journal of Common Market Studies, 31: 45–69. (1999). ‘Parties and Party Discipline within the European Parliament: A Norms-Based Approach’, in S. Bowler, D. Farrell, and R. Katz (eds.), Party Discipline and Parliamentary Government. Columbus, OH: Ohio State University Press. (2006). ‘We Know Which One We Prefer But We Don’t Really Know Why: The Curious Case of Mixed Member Electoral Systems’, British Journal of Politics and International Relations, 8: 445–60. and Pettitt, R. (2005). ‘Expert Opinion on Electoral Systems: So Which Electoral System is “Best”?’, Journal of Elections, Public Opinion and Parties, 15: 3–20. Brinegar, A., Jolly, S., and Kitschelt, H. (2004). ‘Varieties of Capitalism and Political Divides over European Integration’, in G. Marks and M. Steenbergen (eds.), European Integration and Political Conflict. Cambridge: Cambridge University Press. Butler, D. and Westlake, M. (2000). British Politics and European Elections 1999. Basingstoke, UK: Macmillan. Cain, B., Ferejohn, J., and Fiorina, M. (1987). The Personal Vote: Constituency Service and Electoral Independence. Cambridge, MA: Harvard University Press. Caldeira, G. A. and Gibson, J. L. (1995). ‘The Legitimacy of the Court of Justice in the European Union: Models of Institutional Support’, American Political Science Review, 89: 356–76.
218
Bibliography Caramani, D. (2005). ‘Is There a European Electorate and What Does It Look Like? Evidence from Electoral Volatility Measures, 1972–2004’, West European Politics. (forthcoming). Carey, J. and Shugart, M. (1995). ‘Incentives to Cultivate a Personal Vote: A Rank Ordering of Electoral Formulas’, Electoral Studies, 14: 417–39. Carruba, C. (2001). ‘The Electoral Connection in European Union Politics’, Journal of Politics, 63: 141–58. Carter, E. (2005). The Extreme Right in Western Europe: Success or Failure? Manchester, UK: Manchester University Press. Chan, W. (1988). ‘Constituency Representation in Korea: Sources and Consequences’, Legislative Studies Quarterly, 13: 225–42. Cotta, M. (1980). ‘Italy: How a Quick Start Became a Late Arrival’, in V. Herman and M. Hagger (eds.), The Legislation of Direct Elections to the European Parliament. Farnborough, Hants, UK: Gower. and Best, H. (2000). ‘Between Professionalisation and Democratisation: A Synoptic View on the Making of the European Representative’, in H. Best and M. Cotta. (eds.), Parliamentary Representatives in Europe, 1848–2000: Legislative Recruitment and Careers in Eleven European Countries. Oxford: Oxford University Press. Cowley, P. and Lochore, S. (2000). ‘AMS in a Cold Climate: The Scottish Parliament in Practice’, Representation, 37: 175–86. Cox, G. (1990). ‘Centripetal and Centrifugal Incentives in Electoral Systems’, American Journal of Political Science, 34: 903–35. (1997). Making Votes Count: Strategic Coordination in the World’s Electoral Systems. Cambridge: Cambridge University Press. Crewe, I. (1985). ‘MPs and Their Constituents in Britain: How Strong Are the Links?’, in V. Bogdanor (ed.), Representatives of the People? Parliamentarians and Constituents in Western Democracies. Aldershot, Hants, UK: Gower. Curtice, J. and Range, M. (1998). ‘A Flawed Revolution? Britain’s New European Parliament Electoral System’, Representation, 35: 7–15. Diamond, I. (1977). Sex Roles in the State House. New Haven, CT: Yale University Press. Diez Medrano, J. and Gutierrez, P. (2001). ‘Nested Identities: National and European Identity in Spain’, Ethnic and Racial Studies, 24: 753–78. Dyson, K. (2000). The Politics of the Euro-Zone: Stability or Breakdown? Oxford: Oxford University Press.
219
Bibliography Esaiasson, P. (1999). ‘Not All Politics Is Local: The Geographical Dimension of Policy Representation’, in W. Miller et al. (eds.), Policy Representation in Western Democracies. Oxford: Oxford University Press. Farrell, B. (1985). ‘Ireland: From Friends and Neighbours to Clients and Partisans. Some Dimensions of Parliamentary Representation under PRSTV’, in V. Bogdanor (ed.), Representatives of the People? Parliamentarians and Constituents in Western Democracies. Aldershot, Hants, UK: Gower. Farrell, D. (2001). Electoral Systems: A Comparative Introduction. London: Palgrave. (2001a). ‘The United Kingdom Comes of Age: The British Electoral Reform “Revolution” of the 1990s’, in M. Shugart and M. Wattenberg (eds.), The Best of Both Worlds? Oxford: Oxford University Press. and McAllister, I. (2006). The Australian Electoral System: Origins, Variations, Consequences. Sydney: University of New South Wales Press. and Scully, R. (2003). ‘Electoral Reform and the British MEP’, Journal of Legislative Studies, 9: 14–37. Fenno, R. (1978). Home Style: House Members in Their Districts. Boston, MA: Little Brown. Ferejohn, J. and Gaines, B. (1991). ‘The Personal Vote in Canada’, in H. Bakvis (ed.), Representation, Integration and Political Parties in Canada. Toronto: Dundurn. Follesdal, A. and Hix, S. (2006). ‘Why There Is a Democratic Deficit in the EU’, Journal of Common Market Studies, 44: 533–62. Franklin, M., Marsh, M., and McLaren, L. (1994). ‘Uncorking the Bottle: Popular Opposition to European Unification in the Wake of Maastricht’, Journal of Common Market Studies, 32: 101–17. van der Eijk, C. and Marsh, M. (1996). ‘Conclusions: The Electoral Connection and the Democratic Deficit’, in C. van der Eijk and M. Franklin (eds.), Choosing Europe? The European Electorate and National Politics in the Face of Union. Ann Arbor, MI: University of Michigan Press. and Wlezien, C. (1997). ‘The Responsive Public: Issue Salience, Policy Change and Preferences for European Unification’, Journal of Theoretical Politics, 9: 347–63. Gabel, M. (1998). Interests and Integration: Market Liberalization, Public Opinion, and European Union. Ann Arbor, MI: University of Michigan Press. (2003). ‘Public Support for the European Parliament’, Journal of Common Market Studies, 41: 289–308.
220
Bibliography and Whitten, G. (1997). ‘Economic Conditions, Economic Perceptions, and Public Support for European Integration’, Political Behavior, 19: 81–96. Gaines, B. (1998). ‘The Impersonal Vote? Constituency Service and Incumbency Advantage in British Elections, 1950–92’, Legislative Studies Quarterly, 23: 167–95. Gallagher, M. (1987). ‘Does Ireland Need a New Electoral System?’, Irish Political Studies, 2: 27–48. (1991). ‘Proportionality, Disproportionality and Electoral Systems’, Electoral Studies, 10: 33–51. Golub, J. (1999). ‘In the Shadow of the Vote? Decision Making in the European Community’, International Organization, 53: 733–64. Goodin. R. (2003). Reflective Democracy. Oxford: Oxford University Press. Gosnell, H. F. (1969). ‘Representative Democracy’, in Hanna Pitkin (ed.), Representation. New York: Atherton Press. Grofman, B. (1982). ‘Should Representatives Be Typical of Their Constituents?’, in B. Grofman et al., (eds.), Representation and Redistricting Issues. Lexington, MA: D. C. Heath. and Lijphart, A. (eds.) (1986). Electoral Laws and Their Political Consequences. New York: Agathon Press. Gros, D. and Thygesen, N. (1998). European Monetary Integration: From the European Monetary System to Economic and Monetary Union (Second Edition). London: Longman. Grunert, T. (1990). ‘The European Parliament in Search of a Uniform Electoral Procedure’, in S. Noiret (ed.), Political Strategies and Electoral Reforms: Origins of Voting Systems in Europe in the 19th and 20th Centuries. Baden-Baden, Germany: Nomos Verlagsgesellschaft. Hagger, M. (1980). ‘The United Kingdom: The Reluctant Europeans’, in V. Herman and M. Hagger (eds.), The Legislation of Direct Elections to the European Parliament. Farnborough, Hants, UK: Gower. Hazan, R. (1999). ‘Constituency Interests without Constituencies: The Geographical Impact of Candidate Selection on Party Organisation and Legislative Behaviour in the 14th Israeli Knesset, 1996–99’, Political Geography, 18: 791–811. Heath, A., Rothon, C., and Jarvis, L. (2002). ‘Centre and Periphery: Regional Variations within England’, Paper Presented at the Annual Conference of the Elections, Public Opinion and Parties Specialist Group of the Political Studies Association, Salford, UK.
221
Bibliography Herman, V. and Lodge, J. (1978). The European Parliament and the European Community. London: Macmillan. Hix, S. (2005). The Political System of the European Union (Second Edition). London: Palgrave Macmillan. Noury, A., and Roland, G. (2006). Democratic Politics in the European Parliament. Cambridge: Cambridge University Press. Hooghe, L. and Marks, G. (2005). ‘Calculation, Community, and Cues: Public Opinion on European Integration’, European Union Politics, 6: 419–43. Huber, J. and Powell, G. B. (1994). ‘Congruence between Citizens and Policymakers in Two Visions of Liberal Democracy’, World Politics, 46: 291–326. Human Rights Watch (2004). Between Hope and Fear: Intimidation and Attacks against Women in Public Life in Afghanistan. New York: HRW. (2005). Campaigning against Fear: Women’s Participation in Afghanistan’s 2005 Elections. New York: HRW. Imig, D. and Tarrow, S. (2001). Contentious Europeans: Protest and Politics in an Emerging Polity. Lanham, MD: Rowman and Littlefield. Inglehart, R. (1970). ‘Public Opinion and Regional Integration’, International Organization, 24: 764–95. Judge, D. (1999). Representation: Theory and Practice in Britain. London: Routledge. Katz, R. (1980). A Theory of Parties and Electoral Systems. Baltimore, MD: Johns Hopkins University Press. (1997). Democracy and Elections. Oxford: Oxford University Press. (1997a). ‘Representational Roles’, European Journal of Political Research, 32: 211–26. (1999). ‘Role Orientations in Parliaments’, in R. Katz and B. Wessels (eds.), The European Parliament, the National Parliaments and European Integration. Oxford: Oxford University Press. (1999a). ‘Electoral Reform and Its Discontents’, British Elections and Parties Review 9. King, A. (1981). ‘The Rise of the Career Politician in Britain—and its Consequences’, British Journal of Political Science, 11: 249–85. Kreppel, A. (2002). The European Parliament and Supranational Party System: A Study in Institutional Development. Cambridge: Cambridge University Press.
222
Bibliography Kuklinksi, J. (1979). ‘Representative-Constituency Linkages: A Review Article’, Legislative Studies Quarterly, 4: 121–40. Laakso, M. and Taagepera, R. (1979). ‘The Effective Number of Parliamentary Parties: a measure with application to Western Europe’, Comparative Political Studies, 12: 3–27. Lakeman, E. (1974). How Democracies Vote: A Study of Majority and Proportional Systems, 4th edn. London: Faber and Faber. Lancaster, T. and Patterson, W. (1990). ‘Comparative Pork Barrel Politics: Perceptions from the West German Bundestag’, Comparative Political Studies, 22: 458–77. Lane, J. (1995). ‘The Election of Women under Proportional Representation: The Case of Malta’, Democratization, 2: 140–57. Lijphart, A. (1985). ‘The Field of Electoral Systems Research: A Critical Survey’, Electoral Studies, 4: 3–14. (1986). ‘Proportionality in Non-PR Methods: Ethnic Representation in Belgium, Cyprus, Lebanon, New Zealand, West Germany, and Zimbabwe’, in B. Grofman and A. Lijphart (eds.), Electoral Laws and Their Political Consequences. New York: Agathon Press. (1994). Electoral Systems and Party Systems: A Study of Twenty-seven Democracies, 1945–1990. Oxford: Oxford University Press. (1994a). ‘The Political Consequences of Electoral Laws, 1945–85’, American Political Science Review, 84: 481–96. (1999). Patterns of Democracy: Government Forms and Performance in Thirty-six Countries. New Haven, CT: Yale University Press. Lindberg, L. and Scheingold, S. (1970). Europe’s Would-Be Polity: Patterns of Change in the European Community. Cambridge, MA: Harvard University Press. Lipset, S. M. and Rokkan, S. (eds.) (1967). Party Systems and Voter Alignments: Cross-National Perspectives. New York: Free Press. Lodge, J. (1980). ‘Germany: Modell Deutschland’, in V. Herman and M. Hagger (eds.), The Legislation of Direct Elections to the European Parliament. Farnborough, Hants, UK: Gower. (ed.) (2005). The 2004 Elections to the European Parliament. Houndmills, Basingstoke, UK: Palgrave Macmillan. Lundberg, T. (2002). ‘Putting a Human Face on Proportional Representation: Early Experiences in Scotland and Wales’, Representation, 38: 271– 83.
223
Bibliography Lundell, K. (2005). Contextual Determinants of Electoral System Choice: A Macro-Comparative Study. Åbo: Åbo Akademi University Press. Maier, M. and Tenscher, J. (eds.) (2007). Campaigning in Europe— Campaigning for Europe: Political Parties, Campaigns, Mass Media, and the European Parliament Elections 2004. London, UK: LIT Publishers. Mansbridge, J. (1999). ‘Should Blacks Represent Blacks and Women Represent Women? A Contingent “Yes”’, Journal of Politics, 61: 628–57. (2003). ‘Rethinking Representation’, American Political Science Review, 97: 515–28. Mayhew, D. (1974). Congress: The Electoral Connection. New Haven, CT: Yale University Press. McCarthy, A. (1998). ‘Fair Votes for Europe: Creating a Positive Agenda’, Renewal, 6: 34–41. McLaren, L. (2002). ‘Public Support for the European Union: Cost/Benefit Analysis of Perceived Cultural Threat?’, Journal of Politics, 64: 551–66. McLean, I. (1991). ‘Forms of Representation and Systems of Voting’, in D. Held (ed.), Political Theory Today. Cambridge: Polity Press. Mueller, W. and Saalfeld, T. (eds.) (1997). Members of Parliament in Western Europe: Roles and Behaviour. London: Frank Cass. Niedermayer, O. (1995). ‘Trends and Contrasts’, in O. Niedermayer and R. Sinnott (eds.), Public Opinion and Institutionalised Governance. Oxford: Oxford University Press. Norris, P. (1994). ‘Labour Party Quotas for Women’, in D. Broughton, D. Farrell, D. Denver, and C. Rallings (eds.), British Elections and Parties Yearbook 1994. London: Frank Cass. (1997). ‘The Puzzle of Constituency Service’, Journal of Legislative Studies, 3: 29–49. (2004). Electoral Engineering: Voting Rules and Political Behaviour. Cambridge: Cambridge University Press. (2005). Electoral Engineering: Voting Rules and Political Behavior. Cambridge: Cambridge University Press. Norton, P. (1981). The Commons in Perspective. Oxford: Martin Robertson. and Wood, D. (1993). Back from Westminster: British Members of Parliament and Their Constituents. Lexington, KY: University Press of Kentucky. Pennings, P., Keman, H., and Kleinnijenhuis, J. (1999). Doing Research in Political Science: An Introduction to Comparative Methods and Statistics. London: Sage.
224
Bibliography Pennock, J. R. (1979). Democratic Political Theory. Princeton, NJ: Princeton University Press. Phillips, A. (1995). The Politics of Presence: The Political Representation of Gender, Ethnicity, and Race. Oxford: Clarendon Press. Pitkin, H., (1967). The Concept of Representation. Berkeley, CA: University of California Press. (ed.) (1969). Representation. New York: Atherton Press. Powell, G. B. (2000). Elections as Instruments of Democracy. New Haven, CT: Yale University Press. Putnam, R. (1993). Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press. Rae, D. (1967). The Political Consequences of Electoral Laws. New Haven, CT: Yale University Press. Reilly, B. (2001). Democracy in Divided Societies: Electoral Engineering for Conflict Management. Cambridge: Cambridge University Press. Reynolds, A. (1999). Electoral Systems and Democratization in Southern Africa. Oxford: Oxford University Press. Reilly, B. and Ellis, A. (2003). Electoral System Design: The New International IDEA Handbook. Stockholm: International IDEA. (2005). Electoral System Design: The New International IDEA Handbook. Stockholm: IDEA. Richard Commission (2004). Report of the Commission on the Powers and Electoral Arrangements of the National Assembly for Wales. London: The Stationary Office. Rittberger, B. (2003). ‘The Creation and Empowerment of the European Parliament’, Journal of Common Market Studies, 41: 203–26. (2005). Building Europe’s Parliament: Democratic Representation Beyond the Nation-State. Oxford: Oxford University Press. Rogowski, R. (2004). ‘How Inference in the Social (but Not the Physical) Sciences Neglects Theoretical Anomaly’, in H. Brady and D. Collier (eds.), Re-Thinking Social Inquiry: Diverse Tools, Shared Standards. Oxford: Rowman and Littelefield. Rohrschneider, R. (2002). ‘The Democratic Deficit and Mass Support for an EU-Wide Government’, American Journal of Political Science, 46: 463– 75. Saari, D. (2001). Decisions and Elections: Explaining the Unexpected. Cambridge: Cambridge University Press.
225
Bibliography Sartori, G. (1997). Comparative Constitutional Engineering: An Inquiry into Structures, Incentives and Outcomes. 2nd edn. Houndmills, Basingstoke, UK: Macmillan. Sasse, C. et al., (eds.) (1981). The European Parliament: Towards a Uniform Procedure for Direct Elections. Florence, Italy: European University Institute. Scully, R. (2001). ‘Voters, Parties, and Europe’. Party Politics, 7: 515–23. (2005). Becoming Europeans? Attitudes, Roles, and Socialization in the European Parliament. Oxford: Oxford University Press. Searing, D. (1993). Westminster’s World: Understanding Political Roles. Cambridge, Mass: Harvard University Press. (1994). Westminster’s World. Cambridge: Harvard University Press. Shephard, M. (1997). ‘The European Parliament: Laying the Foundations for Awareness and Support’, Parliamentary Affairs, 50: 438–52. Shugart, M. (2001). ‘Electoral “Efficiency” and the Move to MixedMember Systems’, Electoral Studies, 20: 173–93. (2005). ‘Comparative Electoral Systems Research: The Maturation of a Field and New Challenges Ahead’, in M. Gallagher and P. Mitchell (eds.), The Politics of Electoral Systems. Oxford: Oxford University Press. and Wattenberg, M. (eds.) (2001). Mixed-Member Electoral Systems: The Best of Both Worlds? Oxford: Oxford University Press. Ellis, M., and Suominen, K. (2003). ‘An Open and Closed Case: Electoral Rules, the Personal Vote, and the “Connectedness” of Legislators in List Systems of Proportional Representation’, Paper Presented at the Midwest Political Science Association Annual Convention, Chicago, IL. Smith, M. (2003). Bringing Representation Home: State Legislators among Their Constituencies. Columbia, SC: University of Missouri Press. Steenbergen, M. and Scott, D. (2004). ‘Contesting Europe? The Salience of European Integration as a Party Issue’, in G. Marks and M. Steenbergen (eds.), European Integration and Political Conflict. Cambridge: Cambridge University Press. Studlar, D. and McAllister, I. (1996). ‘Constituency Activity and Representational Roles among Australian Legislators’, Journal of Politics, 58: 69–90. Taagepera, R. and Shugart, M. (1989). Seats and Votes: The Effects and Determinants of Electoral Systems. New Haven, CT: Yale University Press. Taggart, P. (1998). ‘A Touchstone of Dissent’, European Journal of Political Research, 33: 363–88.
226
Bibliography Tsebelis, G. and Garrett, G. (2003). ‘The Institutional Foundations of Intergovernmentalism and Supranationalism in the European Union’, International Organization, 55: 367–90. Tsebelis, G. et al., (2001). ‘Legislative Procedure in the European Union: An Empirical Analysis’, British Journal of Political Science, 31: 573–99. van den Berghe, G. (1981). ‘What Is a “Uniform” Procedure?’, in C. Sasse, et al., (eds.), The European Parliament: Towards a Uniform Procedure for Direct Elections. Florence, Italy: European University Institute. van der Eijk, C. and Franklin, M. (eds.) (1996). Choosing Europe? The European Electorate and National Politics in the Face of Union. Ann Arbor, MI: University of Michigan Press. Wahlke, J., Eulau, H., Buchanan, W., and Ferguson, L. (1962). The Legislative System: Explorations in Legislative Behaviour. New York: Wiley. Weir, S. and Beetham, D. (1999). Political Power and Democractic Control in Britain: The Democratic Audit of the United Kingdom. London: Routledge. Weisberg, H., Bowen, B., and Krosnick, J. (1996). An Introduction to Survey Research, Polling and Data Analysis. London: Sage. Wessels, B. (1999). ‘Whom to Represent? Role Orientations of Legislators in Europe’, in H. Schmitt and J. Thomassen (eds.), Political Representation and Legitimacy in the European Union. Oxford: Oxford University Press. Williams, S. (1991). ‘Sovereignty and Accountability’, in R. Keohane and S. Hoffman (eds.), The new European Community. Boulder, CO: Westview. Wood, D. and Young, G. (1997). ‘Comparing Constituency Activity by Junior Legislators in Great Britain and Ireland’, Legislative Studies Quarterly, 22: 217–32.
227
This page intentionally left blank
Index
Adams, J. 47 Anastassopoulos, G. 73, 205 Bean, C. 62 Blair, T. 146 Bocklet, R. 71 Bowler, S. 56, 57, 62 British electoral reform 10, 14, 16, 68–9, 73, 88, 140, 146–8, 151–2, 160 consequences 150–1, 155, 158, 167–8, 171–2, 175–6, 181, 186, 187, 193–4 Burke, E. 143 Cain, B. 56 Carey, J. 52, 121, 128, 199 Chirac, J. 1 Cook, R. 146 d’Estaing, G. 66 deGaulle, C. 65 de Gucht, K. 72 Dehousse, F. 65, 66, 71, 72 de Montfort, S. 163 electoral systems 10 and descriptive representation 43, 47–50, 61, 81–2, 83–5, 87, 150–1 and election campaigns 121–2 and geographical representation 5, 13, 42, 51, 52–3, 55–8, 120, 122–3, 132, 135–6, 144, 168, 173–4, 176–7, 183, 193, 199, 201 ballot structure 51–2, 53, 77, 125–7, 128–9, 152 cases: France 89; Germany 68, 89; Italy 54–5, 62, 67–8, 127; Lithuania 89, 138 classifications 43–5 consequences 4–5, 42, 50, 59
district magnitude 45–6, 51, 54, 74, 76, 127–9, 151–2, 169–70 effective number of parties 82, 86 electoral formula 45, 74 ideal 205–6 list systems 55 proportionality 82, 86, 150–1 European Parliament attitudes towards 29–32 ideal electoral system 205–6 uniform electoral procedures 9–10, 63, 64, 73–4, 87 European Union enlargement 2, 6, 24, 66 legitimacy 7–9, 15 permissive consensus 2, 22, 23 public opinion and 2, 8, 16, 20, 22–4, 26, 28, 33, 201 salience 29, 33, 35, 201 referendums 1, 23, 27–8, 33 representation deficit 9, 17, 34, 206–7 treaty reforms 7, 23, 72 Farrell, D. 56, 57, 62, 129 Fenno, R. 48, 55–6, 62 Ferejohn, J. 56 Fiorina, M. 56 Follesdal, A. 206 Fortuyn, P. 27 Gabel, M. 32 Gaines, B. 57 Gil-Robles, J. 73 Goodin, R. 61 Heath, E. 69 Hix, S. 206 Hume, J. 15 Inglehart, R. 22
229
Index Jenkins, R. 146 Judge, D. 142, 143 Katz, R. 58, 61, 104, 117, 199, 205
Patjin, S. 66 Pennock, C. 61 Pitkin, H. 47 Pompidou, G. 66 Putnam, R. 154
Lijphart, A. 60, 203 Mansbridge, J. 48–9, 61 McAllister, I. 62, 129 MEPs cross-party cooperation 187–8, 191–2, 194 opinions of 96–7, 99, 102 self-perception of role 103–7, 109–10, 114–15, 135, 155, 158, 189 teamwork 185–7, 191, 194 voter attitudes towards 34–5 voter contact with 35, 145–6, 202 see also electoral systems and geographical representation Norris, P. 199 Norton, P. 143, 144 opinion congruence 4
230
Rae, D. 52 representation British tradition 142–4, 145 definition 41 Scully, R. 18–9, 114 Searing, D. 144 Seitlinger, J. 70–1 Shephard, M. 145 Shugart, M. 52, 121, 128, 129–30, 131, 169, 199, 203, 207 Strauss, F. 68 Studlar, D. 62 Taggart, P. 26 Thatcher, M. 69 Wessels, B. 104 Wood, D. 144