European Union Negotiations Processes, Networks and Institutions (Routledge Advances in European Politics) [1 ed.] 0415341248, 9780415341240, 9780203320136

Citation preview

European Union Negotiations

The European Union negotiates in order to reach agreements on common policies and to implement them. This book presents a detailed analysis of EU negotiation processes, drawing out distinguishing features and examining how they differ from traditional international negotiations. The authors focus on the EU both as a negotiation arena and as an actor negotiating in the international arena. The book includes chapters on transparency, reform and compliance bargaining and explores the role of the EU in enlargement negotiations, in the launch of the Doha Round and in its relations with the US. This title will interest students and researchers studying EU policy making, external relations and governance. It will also provide useful perspectives and conceptual tools for those who participate in EU negotiations. Ole Elgström is Professor of Political Science at Lund University, Sweden. He has published a number of articles on negotiation and mediation in the EU and is currently directing a research project entitled ‘New ways to influence in international politics: New roles for the European Union?’. Christer Jönsson is Professor of Political Science at Lund University, Sweden. He is the author of Communication in International Bargaining (1990) and co-author of Organizing European Space (2000). He has served as President of the Nordic International Studies Association (1996–99).

Routledge Advances in European Politics

1 Russian Messianism Third Rome, revolution, Communism and after Peter J. S. Duncan 2 European Integration and the Postmodern Condition Governance, democracy, identity Peter van Ham 3 Nationalism in Italian Politics The stories of the Northern League, 1980–2000 Damian Tambini 4 International Intervention in the Balkans since 1995 Edited by Peter Siani-Davies 5 Widening the European Union The politics of institutional change and reform Edited by Bernard Steunenberg 6 Institutional Challenges in the European Union Edited by Madeleine Hosli, Adrian van Deemen and Mika Widgrén 7 Europe Unbound Enlarging and reshaping the boundaries of the European Union Edited by Jan Zielonka 8 Ethnic Cleansing in the Balkans Nationalism and the destruction of tradition Cathie Carmichael 9 Democracy and Enlargement in Post-Communist Europe The democratisation of the general public in fifteen Central and Eastern European countries, 1991–1998 Christian W. Haerpfer

10 Private Sector Involvement in the Euro The power of ideas Stefan Collignon and Daniela Schwarzer 11 Europe A Nietzschean perspective Stefan Elbe 12 The European Union and e-Voting Addressing the European Parliament’s internet voting challenge Edited by Alexander H. Trechsel and Fernando Mendez 13 European Union Council Presidencies A comparative perspective Edited by Ole Elgström 14 European Governance and Supranational Institutions Making states comply Jonas Tallberg 15 European Union, NATO and Russia Martin Smith and Graham Timmins 16 Business, the State and Economic Policy The case of Italy G. Grant Amyot 17 Europeanization and Transnational States Comparing Nordic central governments Bengt Jacobsson, Per Lægreid and Ove K. Pedersen 18 European Union Enlargement A comparative history Edited by Wolfram Kaiser and Jürgen Elverst 19 Gibraltar British or Spanish? Peter Gold 20 Gender Politics and Society in Spain Monica Threlfall, Christine Cousins and Celia Valiente 21 European Union Negotiations Processes, networks and institutions Edited by Ole Elgström and Christer Jönsson 22 Evaluating Euro–Mediterranean Relations Stephen C. Calleya

23 The Changing Face of European Identity A seven-nation study of (supra)national attachments Edited by Richard Robyn 24 Governing Europe Discourse, governmentality and European integration William Walters and Jens Henrik Haahr 25 Territory and Terror Conflicting nationalisms in the Basque country Jan Mansvelt Beck

European Union Negotiations Processes, networks and institutions

Edited by Ole Elgström and Christer Jönsson

First published 2005 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group This edition published in the Taylor & Francis e-Library, 2004. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © 2005 Ole Elgström and Christer Jönsson for selection and editorial matter; individual contributors their contributions All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data European Union negotiations: processes, networks and institutions/edited by Ole Elgström and Christer Jönsson p. cm. Includes bibliographical references and index. 1. European Union–Foreign relations. 2. Negotiation–European Union. 3. Administrative procedure–European Union. 4. Policy networks–European Union. I. Elgström, Ole, 1950– II. Jönsson, Christer, 1944– JZ1570.E95 2005 341.242⬘2–dc22 2004010175 ISBN 0-203-32013-1 Master e-book ISBN

ISBN 0–415–34124–8 (Print Edition)

Contents

List of illustrations Notes on contributors Preface and acknowledgements 1 Introduction

ix x xii 1

CHRISTER JÖNSSON AND OLE ELGSTRÖM

PART I

The European Union as a negotiation arena 2 Negotiations in networks

11 13

CHRISTER JÖNSSON AND MARIA STRÖMVIK

3 Consolidating ‘unobjectionable’ norms: negotiating norm spread in the European Union

29

OLE ELGSTRÖM

4 Negotiating transparency: the role of institutions

45

BO BJURULF AND OLE ELGSTRÖM

5 Reform negotiations: the case of the CAP

63

MALENA ROSÉN AND MAGNUS JERNECK

6 Compliance bargaining

79

JONAS TALLBERG AND CHRISTER JÖNSSON

7 Negotiations in European Union committees ANDERS SANNERSTEDT

97

viii

Contents

PART II

The European Union as an actor in international negotiations 8 The European Union as an international negotiator

115 117

OLE ELGSTRÖM AND MARIA STRÖMVIK

9 Setting the global trade agenda: the European Union and the launch of the Doha Round

130

ANDERS AHNLID

10 The European Union as actor in international relations: the role of the external environment for EU institutional design

148

JOAKIM REITER

11 The European Union and the United States of America: the politics of ‘bi-multilateral’ negotiations

164

MICHAEL SMITH

12 The Cotonou agreement: asymmetric negotiations and the impact of norms

183

OLE ELGSTRÖM

13 Breaking the waves: the European Union’s enlargement negotiations with EFTA and central and eastern Europe

200

LYKKE FRIIS

14 Conclusion

212

OLE ELGSTRÖM AND CHRISTER JÖNSSON

Index

225

Illustrations

Tables 6.1 Infringement cases per member state by stage, 1978–2000 6.2 Infringement cases closed in 1999 by stage in the procedures 7.1 Autonomy of members of expert groups, Council working groups, and comitology committees, measured in several ways 7.2 Furthering national interests in expert groups, Council working groups and comitology committees 7.3 What characterizes negotiation behaviour in conflict situations? 7.4 What happens in the group when conflict between different interests occurs? 7.5 When conflict of interest occurs in your group, what is the outcome? 7.6 The spirit of ‘saklighet’ 7.7 Sweden’s coalition patterns in different types of committees 7.8 How common is it that the Commission acts in the following roles at the meetings in your group? 7.9 How common is it that the Presidency acts in the following roles at the meetings in your group? 7.10 Importance of contacts 7.11 Sources of information 11.1 Bilateral, multilateral and bi-multilateral negotiations in EU–US relations: components and issues

92 93 101 102 103 104 104 105 107 109 110 111 111 175

Figures 8.1 Four different types of negotiation situations 10.1 Stylized model of propositions on the role of international cooperative arrangements for EU actorness 10.2 The WTO and the OECD 10.3 The IMF and the FSF

122 156 158 161

Notes on contributors

Anders Ahnlid is presently Minister for Trade and Economic Affairs at the Embassy of Sweden to the USA. Previously he has held, inter alia, the position as head of the Department for International Trade Policy at the Ministry for Foreign Affairs in Stockholm. He formed part of Sweden’s negotiating team in the Uruguay Round of multilateral trade negotiations. Ahnlid is the author of several articles on trade policy and trade negotiations. Bo Bjurulf is Associate Professor at the Department of Political Science, Lund University. His initial area of research was a simulation study of the effects of different voting rules when applied to the same system of preferences. He has published articles on formal modelling, electoral behaviour and the transparency process in the European Union. Ole Elgström is Professor of Political Science at Lund University. He has published articles on negotiation and mediation in the European Union in several academic journals and is the editor of European Union Council Presidencies (Routledge 2003). He is a board member of the ECPR (European Consortium for Political Research) Standing Group on the European Union. Lykke Friis is Director of European Affairs at the Confederation of Danish Industries. Previously she has been Research Director at the Danish Institute of International Affairs (DUPI). She holds a Doctorate from the University of Copenhagen and has written extensively on EU enlargement. Her most recent publication is ‘The Danish EU Presidency: “Wonderful Copenhagen”’ in the Journal of Common Market Studies. Magnus Jerneck is Professor of Political Science and the Director of the Centre for European Studies at Lund University. His major fields of interest are international relations, European affairs, regionalism and democratic theory. Christer Jönsson is Professor of Political Science at Lund University. His research interests include international negotiation, diplomacy and the role of transnational networks in international cooperation. He is the author of Communication in International Bargaining (1990) and the co-author of International Cooperation in Response to AIDS (1995) and Organizing European Space (2000). In addition to

Contributors

xi

numerous articles in academic journals, he has contributed the chapter ‘Diplomacy, Bargaining and Negotiation’ to the Handbook of International Relations (2002). Joakim Reiter is working in the European Commission, DG TRADE. He has previously served as Special Advisor to the Minister of Industry and Trade in Sweden, and has worked at the Swedish National Board of Trade and the Ministry for Foreign Affairs. He holds an MA in Political Science from Lund University and an MSc Econ. from the London School of Economics and Political Science. Malena Rosén is a PhD candidate at the Department of Political Science at Lund University. Her research is focused on the Swedish EU debate, with particular emphasis on the communication between the Swedish government and the voters preceding the referenda on Swedish EU and EMU membership, respectively. Other research interests include negotiations and decision making in the European Union. Anders Sannerstedt is Associate Professor at the Department of Political Science, Lund University. His research has focused on comparative government and on parliamentary negotiations (for example,‘Negotiations in the Riksdag’, in L-G. Stenelo and M. Jerneck (eds) The Bargaining Democracy, Lund University Press, 1996). Michael Smith is Professor of European Politics and Jean Monnet Chair in the Department of Politics, International Relations and European Studies at Loughborough University, UK. He has written widely on EU–US relations, on EU external relations and on negotiation in the European Union, and was editor (with Ole Elgström) of the special issue of the Journal of European Public Policy on ‘Negotiation and Policymaking in the EU’ (2000). Maria Strömvik is a PhD candidate at the Department of Political Science, Lund University. Her thesis focuses on competing explanations to the development of the European Union’s collective foreign policy. She was a visiting scholar at the Institute of International Relations at the University of British Columbia 2000–01, and worked as an analyst at the Swedish Ministry for Foreign Affairs between 2002 and 2004. Jonas Tallberg is Associate Professor at Stockholm University and Research Fellow at the Swedish Institute of International Affairs. He has published numerous articles on European governance in international journals and is the author of European Governance and Supranational Institutions: Making States Comply (Routledge 2003).

Preface and acknowledgements

This book is an outgrowth of a research project at the Department of Political Science, Lund University. Several of the authors are original members of the project group; others are part of the extended network established as a result of the project. Labelled ‘The European Union: Negotiations in Networks’, the project received financial support from the Swedish Council for Research in the Humanities and Social Sciences, which is gratefully acknowledged. From the outset, ideas and drafts were discussed in an international reference group, consisting of Guy Olivier Faure, Morten Kelstrup, Sven Norberg and Michael Smith. The whole project profited from their valuable advice. To them, as well as the numerous others who have commented on drafts and shared ideas, the authors extend their deep thanks. Chapter 4 of this volume, which appeared under the same title in the Journal of Common Market Studies (Vol. 42, No. 2, 2004), is reprinted with the kind permission of Blackwell Publishers. Chapter 12 has been published in the European Foreign Affairs Review (Vol. 5, Issue 2, 2000, pp. 175–95) under the title ‘Lomé and postLomé: Asymmetric negotiations and the impact of norms’. The chapter is reprinted with the kind permission of Kluwer Law International. The book is primarily intended for three different audiences. First, it might be used as a main text or supplementary reading in university courses on European policy making or international relations. Second, it addresses issues of concern to researchers on international negotiations as well as EU governance. Third, and not least, it provides useful perspectives and conceptual tools for officials who participate in EU negotiation processes as representatives of governments or other principals. Ole Elgström Christer Jönsson Lund, May 2004

1

Introduction Christer Jönsson and Ole Elgström

The European Union has been described as a ‘negotiated order’ (Smith 1996), a ‘permanent negotiation institute’ (Bal 1995: 1) and a ‘negotiation marathon’ (Kohler-Koch 1996: 367). The EU policy process is dependent on negotiations as a mode of reaching agreements on, and implementing, common policies. This volume represents an effort to characterize these negotiation processes. EU negotiations differ from traditional international negotiations in several respects. At the same time, they are not immediately comparable to negotiations taking place within states. What, then, are the distinguishing features of these ‘hybrid’ EU negotiations? This is the first overarching question addressed in the volume. The continuity and institutionalization of negotiation processes as well as the involvement of a variety of actors besides governments, often linked in informal networks, are illuminated in the contributions. A second pervasive theme concerns the variety of negotiations in the European Union. All negotiation processes do not look alike. What, then, are the determinants of this variety? First, the European Union remains a constantly changing phenomenon and should be understood as a process rather than a frozen organization. The policy process thus looks different depending on when you look at it. Another common notion in the EU literature is that the policy process varies with issueareas. The European Union has been characterized as a ‘prismatic political system’, in which rays of activity and authority are scattered or focused differently in various policy areas (Laffan et al. 2000: 199). Moreover, a distinction must be made between negotiations among various actors within the European Union and negotiations where the European Union figures as a negotiating party externally. The contributions to this volume allude to, yet go beyond, these common distinctions based on time, issue-areas or the European Union as negotiation arena as opposed to negotiating actor. For instance, the authors explore the impact of different stages in the policy process, distributive as compared to norm negotiations and the nature of the external negotiating partner. Negotiations, networks and institutions are the key concepts around which this volume is organized. Negotiation can be regarded as one identifiable mode of joint decision making, to be distinguished from coalition, when the choice is made by numerical aggregation (such as voting), and adjudication, when the choice is made hierarchically by a judge who aggregates conflicting values and interests into a

2 Christer Jönsson and Ole Elgström single decision. In negotiations the parties are left to themselves to combine their conflicting points of view into a single decision (Zartman 1977: 621–3). To be sure, the European Union, to a greater extent than any other international organization, allows for decision making by qualified majority and endows the European Court of Justice with considerable supranational authority. Yet, regardless of formal decision rule, an element of negotiation usually precedes social decisions in general, and EU decisions in particular. Voting and court processes tend to be the ultima ratio in EU policy processes; hence, negotiations are often conducted in the shadow of the vote or in the shadow of the court. Any bargaining situation is characterized by the coincidence of cooperative and conflictual elements as well as interdependent decisions. ‘Without common interest there is nothing to negotiate for, without conflict nothing to negotiate about’ (Iklé 1964: 2). Interdependence entails the need for mutual rather than unilateral action and renders the best course of action by individual actors dependent on the behaviour of others. Far from all bargaining situations, thus understood, lead to negotiations. The parties may exchange non-verbal signals, ‘each aware that his own actions are being interpreted and anticipated, each acting with a view to the expectations that he creates’ (Schelling 1960: 21). When negotiations do occur, they may, in concordance with the definition of bargaining situations, be understood as ‘a process in which explicit proposals are put forward ostensibly for the purpose of reaching an agreement on an exchange or on the realization of a common interest where conflicting interests are present’ (Iklé 1964: 3–4). The most general lesson from the rich negotiation literature is that process matters. Outcomes are not only determined by the character of the issues or the attributes or power of the actors, but are affected by the nature of the negotiation process. The words bargaining and negotiation are frequently used interchangeably in the literature. To the extent that a distinction is made, it goes in either of two directions. On the one hand, bargaining can be seen as the broader concept, including the exchange of verbal as well as non-verbal communication, formal as well as informal exchanges. Negotiation, in this view, is a subclass of bargaining, referring to a formalized process relying on verbal communication. On the other hand, bargaining can be understood as bazaar-like haggling in contrast to joint problem solving. While both connotations can be found in the following chapters, the latter is most prominent, since the European Union is often adduced as an example of problem-solving negotiations as opposed to bargaining (cf. Elgström and Jönsson 2000). While much of the theoretical literature on negotiations – especially that inspired by game theory – is premised on bilateral encounters, negotiations in the European Union are typically multilateral, raising the problem of managing complexity (Winham 1977a, 1977b) for practitioners and theorists alike. ‘The more the messier’ (Zartman 1994: 3) applies not only to the parties to multilateral negotiations, but to issues and roles as well. Methods to reduce complexity therefore become important in multilateral negotiations, not least in the EU. Such methods include coalition building, issue aggregation and disaggregation, and role

Introduction

3

differentiation. Coalitions reduce the number of negotiating actors to a manageable amount (see Dupont 1996). Two diametrically opposite techniques to handle the complexity following from a multitude of issues are issue disaggregation – also referred to as issue decomposition and sequencing (Hopmann 1996: 81; Hampson with Hart 1995: 45–7) – which involves negotiating each issue separately and sequentially; and issue aggregation or issue linkage, which involves combining sub-issues that would be non-negotiable if treated separately into package deals or tradeoffs (see, e.g., Hopmann 1996: 81–4). Multilateral negotiations involve not only many parties and many issues, but also have a multi-role character (Zartman 1994: 5). Mediator roles can be assumed by parties or groups of parties to multilateral negotiations. Leadership roles are considered especially significant in multilateral negotiation (see Underdal 1994; Malnes 1995; Sjöstedt 1999). The following chapters illustrate the various ways of reducing complexity in multilateral EU negotiations. We tend to think of multilateral negotiations in the international arena as processes involving government representatives of a number of states – in the European Union the member states. Yet effective international cooperation often requires more than state-to-state interaction. And EU negotiations tend to be ‘polylateral’ in the sense that they involve interaction between official entities (such as a member states, the European Council, and the Commission) and unofficial, non-state entities, with reasonable expectations of systematic relationships (Wiseman 1999). This means that negotiations typically involve elements not only of intergovernmental relations but also transgovernmental relations – interactions between governmental sub-units across state boundaries – and transnational relations – interactions between non-state actors across state boundaries. In the wake of formal structures – especially complex, ‘polylateral’ ones like the European Union – informal and less visible structures typically emerge, which are often labelled networks. Such informal networks provide access points for a variety of non-state actors. Network analysis replaces the traditional ‘closed-system’ view of self-contained organizational units with an ‘open-system’ perspective. Moreover, it directs our attention to data concerning relations rather than actor attributes: to contacts, ties and connections rather than attitudes, opinions and behaviour of individual actors. ‘Informal organization’ rather than formal relations of authority within and between units becomes the primary focus. One of the main advantages of networks over formal organizations is that they allow informal interorganizational coordination, which avoids problems associated with hierarchy and representation, while facilitating the development of trust and frank exchanges of information (Chisholm 1989). The informality of networks rests on the personal relationships that develop as a result of frequent interaction. Networks imply personal contacts within the context of organizational contacts. As political scientists, the contributors to this volume are interested in ‘policy networks’. A distinction is often made between two types of policy networks. On the one hand, some networks display a high degree of commonality, consistency of values and permanence; students of national politics usually refer to these as ‘policy communities’, whereas students of international politics prefer the

4 Christer Jönsson and Ole Elgström term ‘epistemic communities’. On the other hand, networks may be more temporary coalitions of heterogeneous actors around a specific issue; the term ‘issue networks’ is used domestically, ‘advocacy coalitions’ internationally. Both varieties can be found in EU negotiations. Negotiations in the European Union are normally not occasional and brief encounters, but continuous processes among individuals who form part of informal networks. Moreover, these negotiations are embedded in an institutional framework. Whereas students of international institutions have paid little attention to negotiation processes, and students of international negotiations have largely ignored institutions, the contributors to this volume explore the role of institutions in EU negotiations. An institution can be understood broadly as a relatively stable collection of social practices consisting of easily recognized roles coupled with underlying norms and a set of rules or conventions defining appropriate behaviour for, and governing relations among, occupants of these roles (cf. Young 1989: 32; March and Olsen 1998: 948). These norms and rules ‘prescribe behavioral roles, constrain activity, and shape expectations’ (Keohane 1988: 383). Institutions may or may not involve organizations, groups of individuals who pursue a set of collective purposes. Organizations are entities that normally possess physical locations, offices, personnel, equipment and budgets (Stone Sweet et al. 2001: 6; Young 1989: 32). According to this distinction, the market is an institution, the firm an organization. Marriage is an institution, the family its organizational manifestation. By the same token, the acquis communautaire can be said to represent the institutional heritage of the European Union, while the Council of Ministers and the Commission are organizations. This distinction is not always upheld, and the terms ‘institution’ and ‘organization’ are frequently used interchangeably. And when we speak of the European Union as an institution, we usually have both its ideational aspects and its organizational manifestations in mind. Yet the distinction between the two components should be kept in mind. EU negotiations can be characterized as institutionalized, insofar as norms and rules impinge on the negotiations which involve EU organizational units and persisting negotiation patterns can be found over time. For example, protection of the minority is an EU norm, which is reflected, inter alia, in the voting rules of the Council of Ministers. The principle of juste retour governs EU negotiations to a considerable extent (cf. Peterson 1991: 283). Because this norm permeates negotiations, small states tend to have more influence than their size would indicate. Formal EU decision rules affect negotiation as well. For one thing, many issues can ultimately be decided by voting rather than by consensus, as in most other multilateral international settings. Negotiation theorists have coined the acronym BATNA (Best Alternative to a Negotiated Agreement). Negotiators are assumed to use the consequences of no agreement as the yardstick against which possible negotiation outcomes are evaluated: only if a negotiated solution is better than their BATNA will they agree (Fisher and Ury 1981; cf. Raiffa 1982). Applied to the European Union, for states and coalitions with great voting strength, majority decisions are always a feasible alternative to a negotiated agreement. This can be used as a tactical instrument, and anticipation of voting down the line may

Introduction

5

influence current negotiations. For states and coalitions with less voting strength, it means an imperative to negotiate seriously to modify the majority proposal, as it is better to agree to a proposal that is acceptable, but not ideal, than to be outvoted. Merely obstructing the majority proposal makes sense only if the ultimate decision rule is unanimity. Thus, ‘majority rule lowers the costs of cooperation which stem from the danger of non-decision’ (Kohler-Koch 1995: 176). Institutions do not emerge full fledged and immutable but evolve through processes of institutionalization. This involves the development of practices and rules in the context of using them and has earned a variety of labels, including structuration and routinization, which refer to the development of codes of meaning, ways of reasoning, and accounts in the context of acting on them. (March and Olsen 1998: 948) Institutionalization coordinates and patterns behaviour, channelling it in one among all possible directions. We may differentiate three levels of institutionalization: (1) a set of shared symbols and references, (2) a set of mutual expectations, agreedupon rules, regulations and procedures, and (3) formal organizations (cf. Ruggie 1998: 55). EU negotiations, while being influenced by existing institutions, at the same time contribute to the institutionalization of European cooperation at these three levels. The chapters of this book explore the distinguishing features and the variety of EU negotiations while paying attention to informal networks as well as institutions and institutionalization. The volume is divided into two parts. The first contains chapters dealing with the European Union as a negotiation arena. In other words, it focuses on the diversity of negotiations between member states and other actors within the European Union. The second part redirects the limelight to the European Union as an actor in international negotiations. A question asked both by its negotiating partners and by scholars is whether the European Union can be regarded as a credible unitary actor. The chapters in this part of the volume discuss the drawbacks and advantages of the European Union’s multi-state character and explore the diversity of EU behaviour in different bilateral and multilateral negotiations. Chapter 2, by Christer Jönsson and Maria Strömvik, outlines the background and characteristics of EU networks and addresses the questions of why informal networks are so prevalent in the European Union, what the significant characteristics of EU networks are and what effects networks have on negotiations in the European Union. The authors argue that the fluidity and complexity of the formal European framework contribute to the evolution of informal networks; that EU networks are issue-based, transcend organizational boundaries and often have supranational bodies in ‘linking-pin’ roles; and that networks facilitate negotiations in several interrelated ways. In Chapter 3, Ole Elgström focuses on a certain type of negotiation processes: norm negotiations. He elucidates the processes by which norms are spread, or are

6 Christer Jönsson and Ole Elgström prevented from spreading, within the EU policy-making machinery. Norm resistance exists even in cases of ‘unobjectionable’ norms that are more or less impossible to oppose openly. The unwillingness to adopt a new norm then leads to norm negotiations between norm entrepreneurs and opponents. As the objectors cannot openly reject the norm as such, they either use exclusionary arguments or try to negotiate exceptions, transition periods or fuzzy definitions. In the following chapter, Bo Bjurulf and Ole Elgström direct the searchlight to institutional aspects. Their aim is to demonstrate empirically the importance of institutions in EU negotiations. On 30 May 2001 a new regulation was introduced concerning public access to EU documents. This may appear puzzling to adherents of intergovernmental bargaining theory, as several powerful states opposed increased transparency. The authors submit that the final outcome of these negotiations can only be understood if the specific EU institutions are taken into account. Chapter 5, by Malena Rosén and Magnus Jerneck, investigates the peculiarities of EU reform negotiations. The authors argue that this type of negotiation ordinarily possesses two specific characteristics: the frequency of inert supranational compromises and the extensive use of linkages and package deals. The inclination to reform a policy is often ad hoc in character and a result of external factors or political opportunities. The Common Agricultural Policy (CAP), used by the authors as an empirical example, represents a difficult case of reform negotiations, characterized by a high level of politicization. The existence of a strong policy community, working against dramatic reforms, is linked to a lack of supranational consensus culture in the policy area. A high degree of institutionalization of structures and norms also helps to explain the difficulty in finding a supranational compromise on agricultural reform. The next chapter investigates a topic that has rarely attracted scholarly attention: negotiations in the implementation phase of the decision-making process. Noting that bargaining and compliance remain the objects of separate strands of research in EU studies, Jonas Tallberg and Christer Jönsson spell out the key characteristics of compliance bargaining as it takes place in the European Union. Two principal arguments are advanced. First, the dynamics of compliance bargaining differ in distinct ways from pre-decisional negotiations. Second, the institutional structure of the European Union provides for a form of compliance bargaining that has proven particularly effective as a way of addressing state violations. The distinction of different stages in the EU policy process reappears in Chapter 7, where Anders Sannerstedt analyzes similarities and differences between three major types of committees in terms of negotiation characteristics: expert groups, Council working groups and comitology committees. These are found in different phases of the decision-making process – in the preparatory, decision-making and implementation phase, respectively. The comparison between these three committee types is based on a survey among Swedish committee members. There are, argues Sannerstedt, striking similarities between the negotiations taking place in the three types of committees flowing from a common negotiation culture.

Introduction

7

The second part starts, in Chapter 8, with a general discussion of the European Union as an actor in international negotiations, written by Ole Elgström and Maria Strömvik. According to received wisdom, the European Union is a difficult negotiating partner. As its negotiation mandate usually stems from complicated bargaining between the member states, EU negotiators tend to be inflexible and unwilling to make concessions. The authors challenge this conventional view, contending, first, that EU negotiation behaviour to a large extent depends on contextual factors; second, that the European Union’s structural features need not always result in a disadvantage for the Union; third, that the European Union, by virtue of its immaterial assets, is increasingly well suited to play an active and problem-solving role in many international negotiations. In Chapter 9, Anders Ahnlid analyzes the behaviour of the European Union in a particular arena, the World Trade Organization (WTO). The European Union has taken on the leadership role, previously performed by the United States of America, in the WTO. The European Union was the main instigator of the ‘Doha Development Agenda’, launched in 2001. At the same time, the protectionist nature of the Common Agricultural Policy would seem to reduce the ability of the European Union to provide effective leadership. How could the European Union pursue credible and effective leadership in favour of free trade, given the protectionist nature of its agricultural policy? Tracing the lengthy negotiating process that led to the ‘Doha Development Agenda’, this chapter discusses the prerequisites for EU leadership in the world trading system. EU ‘actorness’ is also the topic of Chapter 10, by Joakim Reiter. The conventional wisdom is that the European Union’s external activity tends to reflect the internal distribution of competence. However, this explanation fails to account for the diversity in EU ‘actorness’ both between and within specific issue areas. Reiter compares the European Union’s activities in trade negotiations and financial negotiations. He claims that the internal policy coordination process, and in particular the informal negotiating networks among member states and the Commission, largely reflect the constitutive elements and institutions of the external environment. In Chapter 11, Michael Smith proceeds from the evident significance, yet problematical nature, of bilateral EU–US negotiations within the global political economy. Advocating a negotiation perspective in the study of EU–US relations, he coins the term ‘bi-multilateral’ negotiations and raises a number of potential research questions derived from this perspective. In particular, he identifies different modes of bi-multilateral negotiations and offers several illustrative cases. Ole Elgström, in Chapter 12, analyzes the so-called Cotonou agreement reached in 2000 between the EU and 71 Third World countries. The chapter explains the contents of the agreement and relates it to previous agreements (the Lomé agreements) between these parties, highlighting changing elements of the bargaining situation. His two basic contentions are, first, that these negotiations reflect the impact of norms and actor identities, not only traditional bargaining power; second, that internal negotiations within the EU affect the outcome more than the negotiations between the parties. Thus, an analysis based solely on material power

8 Christer Jönsson and Ole Elgström resources could not explain either the shifting negotiation outcomes over time or the specific contents of the Cotonou agreement. Chapter 13 by Lykke Friis focuses on another important type of external negotiations: enlargement negotiations. Enlargement has tended to take place in waves. Several countries negotiate in parallel and join at the same time. Despite this tendency, empirical studies and theoretical conceptualizations fail to take into account how the various negotiations affect each other. This chapter sets out to investigate the dynamics between the various negotiating tables: How does the fact that the EU is always negotiating with several applicants at the same time affect the overall negotiation outcome? Can a candidate ‘break the wave’ and negotiate independently, or is it rather swept away by the wave? The chapter approaches these questions by comparing the European Union’s two latest accession rounds, EFTA and eastern enlargement. Its core conclusion is that parallel negotiations entail a nested game situation, where both the European Union and the applicants have a clear interest in linking the various tables together.

References Bal, L. J. (1995) ‘Decision-Making and Negotiations in the European Union’, University of Leicester, Centre for the Study of Diplomacy, Discussion Paper, No. 7. Chisholm, D. (1989) Coordination Without Hierarchy: Informal Structures in Multiorganizational Systems, Berkeley, CA: University of California Press. Dupont, C. (1996) ‘Negotiation as Coalition Building’, International Negotiation, 1: 47–64. Elgström, O. and Jönsson, C. (2000) ‘Negotiation in the European Union: Bargaining or Problem-Solving?’, Journal of European Public Policy, 7: 684–704. Fisher, R. and Ury, W. (1981) Getting to Yes: Negotiating Agreement Without Giving In, Boston, MA: Houghton Mifflin. Hampson, F. O. with Hart, M. (1995) Multilateral Negotiations: Lessons from Arms Control, Trade, and the Environment, Baltimore, MD and London: Johns Hopkins University Press. Hopmann, P. T. (1996) The Negotiation Process and the Resolution of International Conflict, Columbia, SC: University of South Carolina Press. Iklé, F. C. (1964) How Nations Negotiate, New York: Praeger. Keohane, R. O. (1988) ‘International Institutions: Two Approaches’, International Studies Quarterly, 32: 379–96. Kohler-Koch, B. (1995) ‘The Strength of Weakness: The Transformation of Governance in the EU’, in S. Gustavsson and L. Lewin (eds) The Future of the Nation State, Stockholm: Nerenius & Santérus. Kohler-Koch, B. (1996) ‘Catching Up with Change: The Transformation of Governance in the European Union’, Journal of European Public Policy, 3: 359–80. Laffan, B., O’Donnell, R. and Smith, M. (2000) Europe’s Experimental Union, London and New York: Routledge. Malnes, R. (1995) ‘“Leader” and “Entrepreneur” in International Negotiations: A Conceptual Analysis’, European Journal of International Relations, 1: 87–112. March, J. G. and Olsen, J. P. (1998) ‘The Institutional Dynamics of International Political Orders’, International Organization, 52: 943–69. Peterson, J. (1991) ‘Technology Policy in Europe: Explaining the Framework Programme and Eureka in Theory and Practice’, Journal of Common Market Studies, 29: 269–90.

Introduction

9

Raiffa, H. (1982) The Art and Science of Negotiation, Cambridge, MA: Harvard University Press. Ruggie, J. G. (1998) Constructing the World Polity: Essays on International Institutionalization, London and New York: Routledge. Schelling, T. C. (1960) The Strategy of Conflict, Cambridge, MA: Harvard University Press. Sjöstedt, G. (1999) ‘Leadership in Multilateral Negotiations: Crisis or Transition?’, in P. Berton, H. Kimura and I. W. Zartman (eds) International Negotiation: Actors, Structure/ Process, Values, New York: St. Martin’s Press. Smith, M. (1996) ‘The European Union and Concepts of Negotiated Order in Europe’, paper presented for the British International Studies Association Annual Conference, Durham, 16–18 December. Stone Sweet, A., Sandholtz, W. and Fligstein, N. (eds) (2001) The Institutionalization of Europe, Oxford: Oxford University Press. Underdal, A. (1994) ‘Leadership Theory: Rediscovering the Art of Management’, in I. W. Zartman (ed.) International Multilateral Negotiations: Approaches to the Management of Complexity, San Francisco, CA: Jossey-Bass. Winham, G. R. (1977a) ‘Negotiation as a Management Process’, World Politics, 30: 87–114. Winham, G. R. (1977b) ‘Complexity in International Negotiation’, in D. Druckman (ed.) Negotiations: Social-Psychological Perspectives, Beverly Hills, CA and London: Sage. Wiseman, G. (1999) ‘“Polylateralism” and New Modes of Global Dialogue’, University of Leicester, Centre for the Study of Diplomacy, Discussion Paper No. 59. Young, O. R. (1989) International Cooperation: Building Regimes for Natural Resources and the Environment, Ithaca, NY: Cornell University Press. Zartman, I. W. (1977) ‘Negotiation as a Joint Decision-Making Process’, Journal of Conflict Resolution, 21: 619–38. Zartman, I. W. (1994) ‘Introduction: Two’s Company and More’s a Crowd: The Complexities of Multilateral Negotiation’, in I. W. Zartman (ed.) International Multilateral Negotiations: Approaches to the Management of Complexity, San Francisco, CA: Jossey-Bass.

Part I

The European Union as a negotiation arena

2

Negotiations in networks Christer Jönsson and Maria Strömvik

Introduction The term multi-level governance is frequently used by specialists on the European Union to characterize the peculiarities of the EU policy process. The use of the term governance in the EU setting has grown out of the notorious difficulty of characterizing the European Union as a political entity in unequivocal and familiar terms. Descriptions in terms of either an unusually ambitious intergovernmental organization or a supranational state in the making fail to give an adequate and comprehensive picture of the Union. In fact, the term governance has been chosen by EU analysts in order to avoid associations with statehood. ‘Government without statehood’ is precisely the title of the concluding chapter of a much-used textbook (Wallace 1996). The proliferation of the term ‘governance’ in the vernacular of both national and international politics, among scholars and practitioners alike, has not entailed conceptual precision. It is obvious, though, that governance is a broader notion than government. In fact, ‘governance without government’ (cf. Rosenau and Czempiel 1992) has become a common catchword. One way of understanding this phrase is to see governance systems as social institutions in terms of rules, roles and practices, and governments as organizations or material entities established to administer some, but far from all, governance systems (Young 1994: ix–x, 1997: 4–5). Efforts at defining governance are typically rather vague and inclusive, such as ‘the sum of the many ways individuals and institutions, public and private, manage their common affairs’ (Our Global Neighbourhood 1995: 2); or ‘the process whereby elements in society wield power and authority, and influence and enact policies and decisions concerning public life, and economic and social development’ (Emmerij et al. 2001: 188, quoting International Institute of Administrative Sciences). Similarly, EU scholars understand governance in terms of ‘coordinating multiple players in a complex setting of mutual dependence’ and ‘the patterns that emerge from governing activities’ among these actors (Kohler-Koch 1995: 188). Governance is considered ‘independent of the existence of a central authority and beyond the territorial congruence of those who govern with those who are subject to governance’ ( Jachtenfuchs and Kohler-Koch 1995: 5).

14 Christer Jönsson and Maria Strömvik The prefix ‘multi-level’ indicates that the EU policy process takes place at several different levels, from the local to the supranational. This understanding of the European Union can be seen as a ‘response to the inability of the state-centric approach to recognize or explain the independent influence of supranational institutions and the mobilization of domestic actors directly in the European arena’ (Marks et al. 1996: 41). Having ‘been used in several hundred articles and papers and some dozens of books over the past few years’, the term multi-level governance describes the dispersion of authoritative decision making across multiple territorial levels – subnational, national and supranational (Hooghe and Marks 2001: xii, 2). According to one recent effort at summarizing the literature on multilevel governance (Hooghe and Marks 2001: 3–4), the concept catches three essential elements of European integration: (1) decision-making competencies are shared by actors at different levels rather than monopolized by national governments; (2) collective decision making among states involves a significant loss of control for individual national governments; and (3) political arenas are interconnected rather than nested, which means that subnational actors operate in both national and supranational arenas. The separation between domestic and international politics, which is fundamental in state-centric perspectives, is rejected by the notion of multi-level governance. We will argue that the label multi-level governance and the reality it tries to mirror draw our attention to negotiations as key processes and to networks as key structures of governance. Negotiation processes at various levels are central to governance whenever the actors are not ordered hierarchically in a system of superordination and subordination. And diffuse formal political structures give rise to the kind of informal structures that political scientists label issue-based policy networks. Whereas the other chapters in this volume will explore different dimensions of EU negotiation processes, we will focus on the role of networks in these negotiations. More specifically, we will address three broad questions: (1) Why are networks so prevalent in the European Union? (2) What are the significant characteristics of EU networks? (3) What effects do networks have on negotiations in the European Union? But let us preface our discussion of these specific questions with some general remarks on the network concept itself and its application to multi-level governance in the European Union.

Network: concept and EU practice With growing popularity, the term ‘network’ has been used in several different ways. For example, physical networks are composed of constructions, lines and channels for the transportation of goods, people and information; socio-cultural networks unite individuals, and therefore also fields of knowledge and social environments; and organizational networks bind together different sites and entities of political and economic life. The various understandings share but one basic idea: the minimalist, technical definition of a network as a set of interconnected nodes. In physical networks, like railroads, airways, electric grids and telecommuni-

Negotiations in networks

15

cations, the nodes are stations, the only places in which access to the network is granted. In social networks the nodes are human beings, either individuals or collectives – in the EU policy process, organizations of various kinds. Networks direct our attention to relational data: to contacts, ties and connections. Thus, network analysis replaces the conventional ‘closed-system’ view of self-contained organizational units with an ‘open-system’ perspective. Moreover, it chooses interaction between different organizations as the unit of analysis. A network perspective also challenges the traditional view of political space. The prevalent state-centric understanding of international and domestic politics depicts political space in territorial terms. Territory is understood as a contiguous part of the earth’s surface. The territory is distinguished from its environment by a boundary, and the boundary marks the difference between inside and outside, between belonging and exclusion. A network, by contrast, represents space as discrete points (nodes) bound together by lines (links). A network discriminates between nodes that are hooked up to the net and those that are not. This implies differences in the view of distance and proximity as well. Accessibility and reachability are no longer contingent on one’s physical location on a surface, but on one’s position in relation to different nodal networks. In physical transportation networks, air routes and express trains connect cities in ‘nodal landscapes’, turning areas that are far from airports, without railway stations, or located along railroad tracks on which express trains pass into ‘remote areas’. Similarly, proximity in interorganizational networks differs from physical proximity: well-developed transportation and communication systems have reduced distance-related friction by allowing access to, and contacts between, nodes that may be far from one another physically. A specialist in a government department in, say, Sweden, who is a member of an EU network, may feel closer to, and have more frequent interaction with, colleagues in Britain and Greece than his/her Swedish colleague sitting at a desk in Stockholm. A network approach also implies assigning more significance to individuals than in the traditional state-centric perspective. Participants in interorganizational networks are of course not organizations in their entirety, but certain individuals occupying certain roles in the constituent organizations. While individual network participants are agents of organizations, it is also a crucial feature of networks that interorganizational relations are strengthened by interpersonal links (cf. Marsh and Rhodes 1992: 261–2). The interface between organizations consists primarily of ‘boundary-role occupants’. As brokers between their own organization and its environment, boundary-role occupants must not only represent the organization to its environment, but also represent the environment to their constituents. Personal relationships develop as a result of frequent interaction. Networks, in short, imply personal contacts within the context of organizational contacts. These personal relationships provide the basis for the characteristics of networks as a mode of social organization: their informality and relative lack of hierarchy. Formal coordination mechanisms normally introduce an element of hierarchy, which invites controversy. To wit, everyone wants coordination but no one wants to be coordinated. Therefore, non-hierarchical coordination, relying on informal

16 Christer Jönsson and Maria Strömvik channels based on personal relationships, has certain advantages. Donald Chisholm (1989), in a seminal study, emphasizes the virtues of such informal coordination structures: they are adaptable, provide for unhampered information, avoid representation problems, and generate trust: Informal systems of coordination . . . tend to be flexible and adaptive. The disruptive effects of innovation in a formal hierarchy, because of its tightly coupled interdependencies, are avoided in the more loosely coupled, flat, informal system of coordination. Such informal systems are problem oriented and pragmatic. They are self-organizing in the sense that they respond to the effects of experience rather than to the a priori demands of organizational designers. Against the canons of classical management theory, they appear to be uncoordinated; but . . . this is a false impression. Because they are flat, they cannot and do not coordinate by hierarchy. But they are marked by extensive lateral coordination, which occurs at virtually every level of activity – producing an overall system that is quite resistant to serious disruption. (Chisholm 1989: 12) Whereas formal channels tend to be ineffective when information is sensitive or politically charged, informal channels facilitate the free flow of information. Problems of representation are inescapable facts of formal structures that inevitably limit the number of organizations represented, while informal coordination among multiple independent, partly overlapping organizations provide more points of access to the decision-making process. Informality engenders mutual trust, which makes it possible for one agent to make the first move with reasonable expectations of being repaid in the future. In the same vein, students of the European Union claim that informal EU networks allow for wide and flexible participation, reduce frictions and produce results that the formal system would not be able to achieve (Middlemas 1995: xvi). Rather than being neatly separated, the various levels or arenas in EU multi-level governance are linked by transnational networks (Kohler-Koch 1996: 368). The importance of interpersonal links is highlighted by one Commission cabinet official: The European Community is a very strange animal. One would expect it to be run on a very businesslike basis, like a major multinational company. Instead of that, it is run like a small, local theatrical society in that it is relationships between people at different levels that allow it to work. (quoted in Peterson 1995: 78) A European Union composed of ‘an elaborate set of networks’, in the words of former Belgian Foreign Minister Willy Claes, helps Europe ‘to reconcile its undeniable diversity with its equally undeniable common interests and aspirations’ (quoted in Peterson 1995: 98). In sum, observers tend to agree on the benefits of informal EU networks in terms of effectiveness. Networks facilitate coordination

Negotiations in networks

17

among diverse actors and eschew the state-to-state confrontations typical of many international organizations.

Why networks in the European Union? If the European Union is indeed a ‘hothouse’ for different types of networks (Peterson 1995: 69), what are the reasons for this unusual degree of informality and broad participation? First, the Treaty of Rome created a social and political space that intentionally emphasized and encouraged transnational economic interests and exchange across state boundaries (Stone Sweet and Sandholtz 1998: 2). There is thus an element of conscious organizational design. More important still, the fluidity and complexity of the formal European framework that has evolved since the Treaty of Rome encourages the emergence of informal structures. The proliferation of networks in the European Union can be seen as a condition as well as a consequence of the fluidity and complexity of the formal European framework. The formal organizational apparatus of the European Union is extraordinarily complex and cumbersome, with diffuse and overlapping competencies and responsibilities. In the same way that scientists have claimed that in theory bumblebees cannot fly, organization theorists might argue that the institutional design of the European Union cannot work. Yet bumblebees fly, and the European Union works. In the case of the latter, this has a lot to do with the informal networks that have evolved in the shadow of the formal bodies. According to one observer (Héritier 1997), ‘subterfuge’ – the creative use of informal strategies to avoid deadlock in formal policy making – has become second nature to the European Community. One senior Brussels official has argued that ‘if you were to stick to the formal procedures, it would take ten years every time’ and that ‘the more there is disagreement, the more the informal is necessary’ (quoted in Middlemas 1995: xxii). Moreover, the Commission is known to frequently pursue a deliberate networking strategy, actively encouraging informal sectoral links and empowering – or building coalitions with – transnational and subnational groups (cf. Fuchs 1994: 191; Grande 1996: 323; Héritier 1997: 178). These groups thus may become vehicles of ‘reverse lobbying’, supporting the Commission’s policies by putting pressure on governments, business associations and other actors at the national level (Schneider et al. 1994: 480, 490). From a different angle, the pattern of transnational networks can be seen as the Europeanization of national actors and networks. In this perspective, EU networks mirror the proliferation of formal and informal contacts among a variety of actors in member states. The creation of the EU as a new kind of political system can best be explained by recognizing the link between the welfare and service state, with its increasing use of public functions and instruments, and growing European and global interdependencies. The EU level was not, and is not, an isolated, marginalized arena; it became an integral part and dynamic factor in the

18 Christer Jönsson and Maria Strömvik overall development of the interaction among governments, administrations, and intermediary groups at different levels and in several configurations. (Wessels 1997: 20)

Characteristics of EU networks First, participation in EU networks rests on a combination of ‘know-how’ and ‘know-who’. As networks are issue based, expertise in a particular issue area is a prerequisite for network participants. To that extent, EU networks often have the character of ‘epistemic communities’, that is, ‘a network of professionals with recognized expertise and competence in a particular domain and an authoritative claim to policy-relevant knowledge within that domain or issue-area’ (Haas 1992: 3). At the same time, expert knowledge within a particular domain needs to be combined with knowledge of the relevant organizational and individual actors who might contribute to broad policies and individual solutions: who are they, under what constraints do they work, how can they be accessed? This combination of ‘know-how’ and ‘know-who’ increases the likelihood of successful outcomes in negotiations within networks. Second, EU networks tend to transcend organizational boundaries, involving governmental as well as non-governmental organizations (NGOs), national as well as regional and subnational organizations. There are thousands of special interest groups of various kinds in Brussels, making the number of lobbyists in Brussels roughly equal to the number of Commission officials (Grande 1996: 320). These groups represent business interests (cross-sectoral organizations such as the European Round Table of Industrialists, sectoral organizations and individual firms), labour interests, public interests (for example, environmental and consumer groups) as well as territorial interests (regions and localities) (cf. Greenwood 1997). The European Union, in short, provides an unusual abundance of access points to the policy-making process for interested actors. And several types of actors have accumulated significant political resources. There is, in other words, great potential for networks spanning a variety of organizations and individuals. Third, EU networks are not hierarchical, nor are they entirely ‘flat’ or ‘horizontal’. Students of interorganizational networks have pointed to the significance of so-called linking-pin organizations, which occupy central positions in terms of being reachable from, and able to reach, most other organizations in the network and may serve as brokers and communications channels. The Commission’s ‘linkingpin’ role has been noted by students of the European Union. While there are multiple points of access, the Commission enjoys a privileged position by being the only participant that is constituted as a community body and by serving as ‘process manager – setting the timing, prescribing consultation procedures and deciding which interest representation will be recognized’ (Kohler-Koch 1996: 368). Networks usually coalesce around the Commission, and Commission representatives are often the most permanent and central participants in negotiation processes. The European Parliament is emerging as another linking-pin in EU networks. For many NGOs the Parliament may be the most accessible counter-

Negotiations in networks

19

part, at least on issues of lesser weight. While various interest groups queue to influence Commission officials, parliamentarians may still be more accessible and open to dialogue. And the Parliament is no longer the toothless organ it has sometimes been depicted as, but has gradually been given increased competencies. Through networking and coalition building it has proved capable of influencing policy making in several issue areas.

Effects of networks on EU negotiations The existence of networks can be assumed to facilitate EU negotiations in at least five interrelated ways: (1) by avoiding state-to-state clashes and stalemates, (2) by transforming the role of the state, (3) by creating multiple or diffuse loyalties and identities, (4) by engendering trust and (5) by contributing to implementable decisions. Avoiding clashes and stalemates Multilateral negotiations among states, pursuing their egoistic interests, have been characterized as ‘a very messy affair, almost defying generalization’ (Holsti 1982: 160). The prospects for agreements when numerous states are pitted against each other are generally considered slim, and stalemates or impasses are likely to ensue. The existence of networks makes a difference. The emphasis on issue-area expertise and the inclusion of other actors than states in EU negotiations increase the scope for overlapping interests and opens up more possibilities for compromises and barters. The large number of issues under negotiation at any given moment in the European Union offers ample opportunities for package deals. On the other hand, the issue-based or sectoral nature of EU networks often makes package deals across issue areas more difficult to achieve. Transforming the role of the state The changing role of the state is a central aspect of multi-level governance and networks in the EU. By participating in informal networks, governments are gradually turning into mediators between political and economic spheres and between domestic and international activities, rather than being the omnipotent authority internally, and exclusive agent externally, of a well-defined territory and population, as the traditional sovereignty principle prescribes. We may speak of the emergence of the ‘negotiating state’ (cf. Jönsson et al. 2000: 174–6). As demonstrated graphically by the European Union, the shift from domination to negotiation involves not only increased cooperation among states but interaction with several other types of actors as well. If we situate the state in the context of negotiations in networks, sovereignty is reduced to a certain legal authority that states can use as a lever in bargaining. Sovereignty becomes ‘less a territorially defined barrier than a bargaining resource for a politics characterized by complex transnational networks’ (Keohane 1995: 177).

20 Christer Jönsson and Maria Strömvik In a network perspective, thus, we need to consider the relative bargaining assets and weaknesses of states vis-à-vis other actors. Government representatives tend to occupy central positions in policy networks and cannot be neglected or bypassed by other actors when negotiating an issue. In addition, the state still has a certain advantage in the collection and assemblage of information. On the other hand, states are territorially bound, whereas their business and NGO counterparts are transnational in character. The democratic European states also have lengthy decision processes, whereas many of their counterparts can make speedy decisions, unrestrained by public opinion or constitutional checks and balances. In sum, the EU pattern of states as negotiating entities, participating in transnational networks, yields a more multifaceted and fine-grained picture of the adaptability and changing role of states than the simplistic alternatives of demise or survival characterizing much of the current debate. Creating multiple/diffuse loyalties and identities While negotiators in multilateral talks are normally state representatives and nothing else, negotiators in networks tend to gradually identify themselves with the networks in addition to their primary identification with their constituency. While representing their home states, participants in issue-based EU networks share similar expertise and a sense of companionship. Several studies confirm that a sense of collective responsibility and supranational loyalty follows from participation in EU working groups and committees (see Egeberg 2001: 20–1). Engendering trust Networks tend to engender trust among their participants. The informality of EU networks combines with the ‘shadow of the future’ (Axelrod 1984) – the realization among participants that they will be in continuous negotiations, with no endgame and no exit option in sight – to develop a greater sense of trust than in most other multilateral negotiations. Trust can be understood as an agent’s theory of how another agent or group of agents will behave in the future. It implies a willingness to enter into dependent positions, to increase one’s vulnerability to others whose behaviour is not under one’s control (cf. Gambetta 1990). Trust makes it possible to move beyond specific reciprocity, ‘situations in which specified partners exchange items of equivalent value in a strictly delimited sequence’ (Keohane 1986: 4), and engage in diffuse reciprocity, when the participants do not insist on immediate and exactly equivalent reciprocation of each and every concession, on an appropriate ‘quid’ for every ‘quo’. Contributing to implementable decisions By involving most interested parties in the policy-making process, networks may contribute to more effective, implementable decisions. In comparison to other international negotiation forums, the European Union produces a higher share of

Negotiations in networks

21

outcomes that are, in fact, implemented. While this owes much to effective enforcement mechanisms and compliance bargaining once violations occur (cf. Tallberg and Jönsson, Chapter 6, this volume), networks contribute by forging links and sharing information among actors that are crucial not only to policy making but also to implementation. Thus, social pressures and full information combine to make failures to implement negotiated policies less likely in the European Union.

Illustrative cases By way of illustration, let us briefly sketch three cases of negotiations in networks, taken from the areas of social policy, telecommunications and common foreign and security policy (CFSP). They concern efforts to initiate ‘community action for the elderly’, the deregulation of telecommunications and the decision in June 2003 to launch an EU military mission to the Democratic Republic of Congo, respectively.1 Social policy represent a ‘most likely’, ‘low-politics’ case of networking, CFSP a ‘least-likely’, ‘high-politics’ case, with telecommunications in an intermediary position. In other words, whereas informal networks are generally expected when social-policy issues of marginal political consequence are discussed, negotiations on sensitive security issues are usually assumed to display intergovernmentalism rather than network activity. Pronounced redistributive aspects and entrenched state interests would seem to make the deregulation of telecommunications a rather unlikely candidate for informal networks. Community actions for the elderly In November 1990 the Council decided on a three-year programme, ‘Community Actions for the Elderly’, and designated 1993 as the ‘European Year of Older People and Solidarity between the Generations’. The programme had limited objectives and a modest budget, but was obviously politically sensitive. Since the signing of the Rome Treaty, Community social policy had been almost exclusively related to the free movement of persons, and attempts to expand its scope had always been highly controversial. In the 1980s the European Parliament passed a series of resolutions calling for Community actions for the elderly, which went unheeded by the Commission. Yet in 1983 the Parliament managed to add an allocation for measures involving older people to the final budget. After three years of modest budget increases, these funds almost doubled in 1987 and again in 1988, when the Parliament specified a portion of the funds for preparations for a European Year of Seniors. This ‘coup’ meant that the Commission finally had to take action – the money had to be spent – and draft a programme text. On closer scrutiny, what might on the surface seem to be a tug-of-war between the Parliament and the Commission turns out to involve network relations with NGO representatives in pivotal positions. In the European Parliament issues of ageing belonged to the Social Affairs Committee but were dealt with in greater depth in the Intergroup on Ageing, which consisted of MEPs from different

22 Christer Jönsson and Maria Strömvik political groups who shared an interest in this specific topic. Intergroups are not officially recognized or registered by the Parliament; they have no funding and only limited working facilities. Thus, in the Intergroup on Ageing the secretariat was provided by an NGO, Eurolink Age. With offices in London and Brussels, Eurolink Age received about 80 per cent of its funds from the Commission. The appointed Commission official in the Division Responsible for the Family and the Elderly within DG V (employment, industry relations, social affairs) maintained continuous contacts with Eurolink Age and other NGOs. Thus, the Intergroup on Ageing, Eurolink Age and DG V were symbiotically linked through mutual resource dependencies. The Intergroup relied on information and ideas as well as secretariat services from Eurolink Age, which, in turn, was dependent on funding from the Commission. The idea to ‘sneak’ the issue of ageing into the budget seems to have been the result of close collaboration between a number of MEPs and Eurolink Age. The allotted funds were subsequently spent on Eurolink Age, which, in turn, used parts of the money to set up the secretariat of the Parliament’s Intergroup on Ageing. Eurolink Age also worked hard to establish closer links between DG V and the Parliament. The linking-pin position of Eurolink Age was further illustrated once the text of the programme left DG V and met resistance in Leon Brittan’s cabinet. In the end, after Eurolink Age had activated its British contacts among MPs and within the government to argue in favour of the programme and send the message that Britain would not object in the Council, Brittan accepted the text, which was subsequently adopted without opposition in the Council. Deregulating the telecommunications market The efforts to formulate a Community telecommunications policy in the early 1980s are frequently adduced as an example of creative network building (cf. Schneider et al. 1994; Fuchs 1994; Sandholtz 1993). Traditionally, telecommunications had been the exclusive domain of national monopolies: Post, Telephone and Telegraph administrations (PTTs). The Commissioner responsible for Industrial Affairs, Etienne Davignon, was concerned about European firms losing ground in the information technology (IT) area. Reframing telecommunications as part of an overall industrial policy, he created an IT Task Force in 1979, which was independent of the DGs and reported directly to the Commissioners. The Task Force commissioned a number of technical and economic studies from prominent consulting firms, which enabled the Commission to enter into highly technical discussions with industry and telecoms administrations. In its quest to deregulate and liberalize the telecommunications market, which resulted in a Proposal for an Action Programme in 1984 and a Green Paper in 1987, the Commission relied heavily on forging links with two sets of actors who could counterbalance the hitherto predominant PTTs: major producers of telecoms equipment and users of advanced telecoms services. The manufacturers needed Community-scale markets to remain competitive with US and Japanese rivals, and recognized the need for regional technical standards; major users of

Negotiations in networks

23

telecoms services, especially international banks and multinational corporations, complained about the high rates and inflexibility of national PTTs and saw how their US counterparts had acquired a competitive edge through the deregulation of telecommunications in the United States. In 1981–2 Davignon invited leading representatives of Europe’s largest IT firms to come together under Community auspices to form the ‘Big 12 Round Table’, thereby helping to engineer an industrial consensus. Similarly, large users were relatively easy to mobilize politically because of their small number. Organizations like the International Telecommunications Users Group (INTUG) and the European Council of Telecommunications Users Associations (ECTUA) were supported, and used strategically, by the Commission. In short, the 1980s saw a dramatic shift in political constellations in the telecommunications issue area. One well-established policy community favouring domestic monopolies and bilateral cartel arrangements, centred around the national PTTs, was replaced by an advocacy coalition favouring competition, based on manufacturer and user interest. The Commission played a prominent role as ‘process manager’ in this evolution, building new coalitions and networks and making use of ‘reverse lobbying’ to sway member states’ policy makers. The process has also been noted for the measure of individual, entrepreneurial leadership provided by the Commissioner Davignon (Sandholtz 1993: 256). EU military mission to the Democratic Republic of Congo Negotiations and policy networks within the common foreign and security policy have received far less attention than their counterparts in Community affairs. While there has certainly been a recognition of the existence of policy networks within the second pillar as well (Forster and Wallace 1996: 433), some have doubted their possibility to affect the area of high politics. It has been argued that ‘when issues preoccupy high-level decision-makers, particularly foreign ministers, the ability of policy networks to shape decisions is strictly limited’ (Peterson and Bomberg 1999: 248). There are, however, reasons to believe that the development of an increasingly intense and complex institutional structure for the CFSP in Brussels over the last few years has been accompanied by an increased importance of policy networks, often revolving around the Council Secretariat. Without these networks, the CFSP structure would in many cases fail to generate Union policies towards the outside world, in particular on occasions when time is of the essence and extremely quick decisions are needed. Mapping the participants and the structure of EU policy networks within the CFSP is perhaps more difficult than in most other policy areas, due to the sensitive nature of the issues involved and the relatively fewer participants of a nongovernmental nature. To the extent that there has been an identification of something like a linking-pin role within CFSP networks, it has been assumed to be occupied by the Commission (Peterson 1998: 16). However, within the CFSP, in particular the security and defence policy domain, the Commission’s role is still relatively limited. Instead, other Brussels-based actors have begun to assume roles

24 Christer Jönsson and Maria Strömvik that are at least similar to those of the Commission’s linking-pin position within the first pillar. The combined post of the High Representative for the CFSP and the Secretary General of the Council Secretariat is of particular importance in this respect. On many occasions, it seems that the first occupant of the post, Dr Javier Solana, and the administrative structures supporting him, are now participating in the policy process in a role resembling a forceful additional EU member. The Council Secretariat’s formal and informal relations with another increasingly important Brussels-based body, the Political and Security Committee (PSC), and both these bodies’ contacts with national foreign ministries, probably constitute the main arena for the policy formulation on EU crisis management. The importance of these relatively new bodies, and their relations with each other, may be briefly illustrated by the speedy process leading up to the EU decision in June 2003 to launch an EU military mission to the Democratic Republic of Congo. By early May 2003 the ongoing interethnic conflict in northeastern Congo intensified in and around the town of Bunia. MONUC, the UN peacekeeping mission in the area, had neither the political mandate nor the weaponry required to halt the situation. The UN peacekeeping troops from Uganda fled the region on 7 May, after which the fighting intensified during the following weeks. By mid-May, the UN Secretary General made a request to the Security Council for urgent help from the international community. Kofi Annan’s call met with apparent disinterest from the United States of America, the United Kingdom and the North Atlantic Treaty Organization (NATO), a few of the most obvious actors with a rapid reaction capacity. On 20 May, however, media reported that the EU’s high representative Javier Solana had announced that the UN ‘has another place where it can go to draw forces, which is the European Union’ (SHAPE News, 20 May 2003). At that time, however, many EU member states, as well as the United Nations, were lukewarm about the idea of an EU operation in the Congo. France was prepared to carry out the operation on its own, and had already initiated bilateral contacts with the UN. The United Kingdom and many other EU members were outwardly silent, while Germany openly came out strongly against it. The German view, as quoted in the media, was that ‘Congo was a long way from Europe, a difficult military challenge and a conflict on which there was no agreed EU foreign policy’ (SHAPE News, 27 May 2003). The UN also seemed hesitant to involve the militarily inexperienced European Union in a mission that demanded quick and forceful action. The idea of an EU mission carried several highly sensitive connotations of a political nature. Should the European Union militarily venture ‘out of area’ at a time when the diplomatic climate between the United States of America and some EU members, following the Iraqi war, was characterized by serious diplomatic tensions? The idea of an EU mission also evoked certain question marks concerning the EU’s actual capacity to carry out a mission in Africa. The mission would require a ‘robust mandate’2 and involve considerable risks. Military officials warned that the EU did not yet have sufficient experience for such a delicate mission (SHAPE News, 28 May 2003; Deutsche Welle, 2 June 2003). A decision on an EU

Negotiations in networks

25

mission would mean, as one analyst put it, that the Union would be ‘going into a military and political minefield’ (AFP, 11 June 2003). Despite all this, it took only two and a half weeks following Solana’s statement on 20 May until all member states had agreed on the EU’s first-ever military operation outside Europe and, equally, its first operation without the use of NATO assets. On 5 June the Council adopted the joint action on Operation Artemis. The operation was to be led by France and to provide time for the UN to reinforce its own presence in the Congo by September. The following day the first troops arrived in Bunia. This decision would supposedly never have come about without the active participation of several Brussels-based bodies that knew how to push the process forward and how to maximize influence in the highly complex and extremely speedy negotiations. One of the main actors was the EU’s high representative. In particular following the Council meeting on 19 May, when he was given the task of studying the feasibility of a possible EU operation, Solana actively pushed the process forward. By publicly citing his contacts with the UN Secretary General and evoking the latter’s demand for an EU-led mission, he made a refusal politically difficult for reluctant member states. In addition, there were network links between members of Solana’s staff, UN staff involved in peacekeeping operations and interested member states. It was also Solana, with his administrative support in the Council Secretariat, who provided the first feasibility reports to the PSC. Along with France, it was furthermore Solana, rather than the Greek Presidency, who handled the contacts with the UN Security Council. Whereas the formal preparatory work was carried out mainly in the ‘member state forums’ (the PSC and the EU Military Committee) as well as in Paris, the Council Secretariat played a vital formal and informal role in the process. Most EU documents, including the joint actions, were partly or wholly drafted by the Council Secretariat. Sometimes on demand, and sometimes on their own initiative, the EU Military Staff and the Situation Centre, both located in the Secretariat, continuously provided various types of intelligence information. Furthermore, Mr Aldo Ajello, the EU Special Representative for the Great Lakes Region, who had for a long time established a vast network of contacts, was providing the High Representative and the PSC with first-hand knowledge and information throughout the process, in particular during the implementation phase. In turn, it was this information that was communicated back to the member states’ capitals, thereby contributing to the respective national positions. It was, in sum, to a large extent the Council Secretariat that possessed the ‘know-how’ and, particularly in the case of the high representative and the EU’s special representative, also the ‘know-who’.

Conclusions Networks may take various forms and have variegated effects. In this chapter, we have explored the background and characteristics of EU networks and have argued that they serve as important lubricants in complex negotiations. We have primarily focused on the advantages of informal networks. Of course, there are

26 Christer Jönsson and Maria Strömvik drawbacks associated with the informality and lack of transparency of negotiations in networks. These will be treated further in the concluding chapter of this volume. Our illustrative cases demonstrate that networks are not limited to ‘low-politics’ issue areas. In fact, the role of informal networks in the ‘least likely’ case of the highly sensitive decision to send an EU military mission to Congo indicates their omnipresence. Our cases also illuminate the linking-pin role of supranational bodies. The role of ‘process manager’ is assumed not only by the Commission, as conventional wisdom has it, but also by the European Parliament and the Council Secretariat. In sum, a fuller understanding of negotiations in the EU requires that informal structures be investigated alongside formal ones.

Notes 1 The description of the elderly and Congo cases are based on interviews with participants, whereas the telecommunications case draws on secondary sources. 2 Sanctioned by a UN Security Council mandate founded in Chapter VII of the UN Charter, thereby allowing the use of force.

References AFP (Agence France-Presse), www.afp.com Axelrod, R. (1984) The Evolution of Cooperation, New York: Basic Books. Chisholm, D. (1989) Coordination Without Hierarchy: Informal Structures in Multiorganizational Systems, Berkeley: University of California Press. Deutsche Welle, www.dw-world.de. Egeberg, M. (2001) ‘An Organisational Approach to European Integration: Outline of a Complementary Perspective’, Oslo: Department of Political Science and ARENA, University of Oslo (mimeo). Emmerij, L., Jolly, R. and Weiss, T. G. (2001) Ahead of the Curve? UN Ideas and Global Challenges, Bloomington and Indianapolis: Indiana University Press. Forster, A. and Wallace W. (1996) ‘Common Foreign and Security Policy’, in H. Wallace and W. Wallace (eds) Policy-Making in the European Union, Oxford: Oxford University Press. Fuchs, G. (1994) ‘Policy-Making in a System of Multi-Level Governance: The Commission of the European Community and the Restructuring of the Telecommunications Sector’, Journal of European Public Policy, 1: 177–94. Gambetta, D. (1990) (ed.) Trust: Making and Breaking Cooperative Relations, Oxford: Blackwell. Grande, E. (1996) ‘The State and Interest Groups in a Framework of Multi-Level DecisionMaking: The Case of the European Union’, Journal of European Public Policy, 3: 318–38. Greenwood, J. (1997) Representing Interests in the European Union, London: Macmillan. Haas, P. M. (1992) ‘Introduction: Epistemic Communities and International Policy Coordination’, International Organization, 46: 1–35. Héritier, A. (1997) ‘Policy-Making by Subterfuge: Interest Accommodation, Innovation and Substitute Democratic Legitimation in Europe: Perspectives from Distinctive Policy Areas’, Journal of European Public Policy, 4: 171–89.

Negotiations in networks

27

Holsti, K. J. (1982) ‘Bargaining Theory and Diplomatic Reality: The CSCE Negotiations’, Review of International Studies, 8: 159–70. Hooghe, L. and Marks, G. (2001) Multi-Level Governance and European Integration, Oxford: Rowman & Littlefield. Jachtenfuchs, M. and Kohler-Koch, B. (1995) ‘The Transformation of Governance in the European Union’, paper presented at the Fourth Biennial International Conference of the European Community Studies Association, Charleston, SC. Jönsson, C., Tägil, S. and Törnqvist, G. (2000) Organizing European Space, London: Sage. Keohane, R. O. (1986) ‘Reciprocity in International Relations’, International Organization, 40: 1–27. Keohane, R. O. (1995) ‘Hobbes’s Dilemma and Institutional Change in World Politics: Sovereignty in International Society’, in H.-H. Holm and G. Sørensen (eds) Whose World Order? Uneven Globalization and the End of the Cold War, Boulder, CO: Westview. Kohler-Koch, B. (1995) ‘The Strength of Weakness: The Transformation of Governance in the EU’, in S. Gustavsson and L. Lewin (eds) The Future of the Nation State, Stockholm: Nerenius & Santérus. Kohler-Koch, B. (1996) ‘Catching Up with Change: The Transformation of Governance in the European Union’, Journal of European Public Policy, 3: 359–80. Marks, G., Scharpf, F. W., Schmitter, P. C. and Streeck, W. (1996) Governance in the European Union, London: Sage. Marsh, D. and Rhodes, R. A. W. (1992) ‘Policy Communities and Issue Networks: Beyond Typology’, in D. Marsh and R. A. W. Rhodes (eds) Policy Networks in British Government, Oxford: Clarendon Press. Middlemas, K. (1995) Orchestrating Europe: The Informal Politics of European Union 1973–1995, London: Fontana Press. Our Global Neighbourhood: The Report of the Commission on Global Governance (1995), Oxford and New York: Oxford University Press. Peterson, J. (1995) ‘Decision-Making in the European Union: Towards a Framework for Analysis’, Journal of European Public Policy, 2: 69–93. Peterson, J. (1998) ‘Introduction: The European Union as a Global Actor’, in J. Peterson and H. Sjursen (eds) A Common Foreign Policy for Europe: Competing Visions of the CFSP, London and New York: Routledge. Peterson, J. and Bomberg, E. (1999) Decision-Making in the European Union, London: Macmillan. Rosenau, J. N. and E.-O. Czempiel (eds) (1992) Governance Without Government: Order and Change in Word Politics, Cambridge: Cambridge University Press. Sandholtz, W. (1993) ‘Institutions and Collective Action: The New Telecommunications in Western Europe’, World Politics, 45: 242–70. Schneider, V., Dang-Nguyen, G. and Werle, R. (1994) ‘The Structure of Policy Networks: A Comparison of the “Chemicals Control” and “Telecommunications” Policy Domains in Germany’, Journal of Common Market Studies, 32: 473–98. SHAPE News, http://www.nato.int/shape/news/2003/06/index.htm Stone Sweet, A. and Sandholtz, W. (1998) ‘Integration, Supranational Governance, and the Institutionalization of the European Polity’, in W. Sandholtz and A. Stone Sweet (eds) European Integration and Supranational Governance, Oxford: Oxford University Press. Wallace, W. (1996) ‘Government without Statehood: The Unstable Equilibrium’, in H. Wallace and W. Wallace (eds) Policy-Making in the European Union, Oxford: Oxford University Press.

28 Christer Jönsson and Maria Strömvik Wessels, W. (1997) ‘The Growth and Differentiation of Multi-Level Networks: A Corporatist Mega-Bureaucracy or an Open City?’, in H. Wallace and A. R. Young (eds) Participation and Policy-Making in the European Union, Oxford: Clarendon Press. Young, O. R. (1994) International Governance: Protecting the Environment in a Stateless Society, Ithaca, NY: Cornell University Press. Young, O. R. (1997) ‘Rights, Rules, and Resources in World Affairs’, in O. R. Young (ed.) Global Governance: Drawing Insights from the Environmental Experience, Cambridge, MA: MIT Press.

3

Consolidating ‘unobjectionable’ norms Negotiating norm spread in the European Union Ole Elgström

Introduction Certain norms reach an ‘unobjectionable’ status. Such predominant norms are considered impossible to oppose openly; at least by most people in a certain geographical context and during a certain time period. Policies that are based on these norms are, by implication, impervious to normatively based criticism. Instead, critics have to rely on indirect attack. In present-day European political debate, democracy and human rights promotion, environmental preservation, free competition, gender mainstreaming and increasing transparency are examples of ‘unobjectionable’ norms. At the same time, we all know that these laudable goals are not always transformed into community legislation, let alone implemented in concrete policy decisions. There is a lively debate on ‘the democratic deficit’ of the European Union, including criticism of excessive bureaucratic secrecy. Women still have lower average wages than their male colleagues, despite EU legislation. Thirty years of EU environmental action programmes notwithstanding, change at the discursive level ‘has not affected the Community practices to any great extent’ (Kronsell 1997: 49). There is within the EU machinery itself a resistance to change that complicates the achievement even of overtly consensual policy objectives. The puzzle is: how is it that norms that no one openly objects to are often not translated into concrete policies? The aim of this chapter is to elucidate the processes by which norms are spread, or are prevented from spreading, within the EU policy-making machinery. I argue that norm resistance exists even in cases of ‘unobjectionable’ norms. The unwillingness to adopt a new norm leads to norm negotiations between proponents (norm entrepreneurs) and opponents. As the objectors cannot openly reject the norm as such, they either use exclusionary arguments (‘the principle is not relevant in this issue area’) or try to negotiate exceptions, transition periods or fuzzy definitions (to hinder efficient implementation). In these efforts, they can rely on the use of competing, process-oriented norms like time or cost effectiveness.

30 Ole Elgström The argument is constructed in the following way. I start by defining what I mean by norms and ‘unobjectionable’ norms, and analyse the characteristics of norm negotiations. Thereafter, I briefly discuss the process by which norms spread and become or do not become ‘unobjectionable’, making a distinction between the emergence and the acceptance of norms. In the main theoretical section I distinguish between two phases: the legalization phase and the implementation phase of the norm diffusion process within the EU decision-making apparatus. Each phase is characterized by a special type of negotiation: text negotiations and implementation negotiations, respectively. I delineate the different tactics and strategies a norm objector may use. Two empirical cases are used to illustrate the processes outlined in the theoretical discussion: first, the introduction of gender equality norms in EU development policy; second, an account of how transparency norms have spread within the European Union after Maastricht and Amsterdam.

Norms and ‘unobjectionable’ norms Norms are collective understandings of the proper behaviour of actors with a given identity and in a given context (cf. Finnemore and Sikkink 1998: 891; Legro 1997: 33). Norms, in this usage, entail moral considerations (they are about how an actor should behave), are shared and social, and concern behaviour (Finnemore 1996: 22; Florini 1996: 364; Kacowicz 2000: 6). Consistent with constructivist thinking, norms are claimed to have independent causal effect (Finnemore 1996: 15; Spruyt 2000: 68). Some norms attain a predominant status. They become taken for granted (‘the only game in town’, to quote Linz and Stepan 1996) and are generally considered inviolable. They are perceived as being so morally superior that it is considered almost taboo to criticize them. The reasons why some norms reach such a position are discussed in the next section. ‘Unobjectionable’ norms are, however, not eternally untouchable: their status is linked to a certain social context or geographical space and is time bound. Norms are continuously reconstructed and are therefore always open to change. From a negotiation perspective, norms in general have some specific traits. They are extremely difficult to compromise. It is normally considered unthinkable to make concessions regarding central and nationally widespread norms in international negotiations, both for moral reasons and because the domestic consequences of giving in on highly cherished values are often believed to be disastrous. Norm negotiations thus often acquire an ‘all-or-nothing character’. In this respect, norm negotiations are very different from distributive or redistributive negotiations, where compromises are normally expected. Package deals including norm concessions are, for the same reason, often out of the question. If parties hold opposing norms, ‘reasoned consensus’ (Risse 2000: 7) is difficult to attain: actors cannot be expected to concede on their principles, even when faced with the most convincing and eloquent persuasive argument. They may, however, be prepared to concede on matters of implementation (‘how and when’) and on the scope or applicability of the norm.

Consolidating ‘unobjectionable’ norms

31

‘Unobjectionable’ norms are often of a very broad nature, outlining basic moral prescriptions. After having been established, they are often specified and ‘translated’ (Czarniawska and Sevón 1996) into more limited issue areas. Negotiations regarding the introduction of an ‘unobjectionable’ norm into a new organizational context often include two different dimensions. First, as predominant norms are not adapted to this specific context, there may be an interpretative negotiation among norm advocates. If the general norm includes variants, negotiations may concern which of these ought to apply in the new setting: how radical the interpretation that is going to be adopted. There may also be competing scientific approaches to the issue at hand; this can lead to negotiation among counter discourses (cf. Bäckstrand 2001, chapters 6–7, with examples from environmental diplomacy). The debate concerns how to frame the issue, and about what formula should inform the document to be decided upon. Second, opponents of the new norm try to prevent the application of the novel standards in competitive negotiations. ‘Unobjectionable’ norms are by definition impossible to criticize as such; ‘negative’ arguments therefore have to focus on other aspects and these will be delineated below. The two types of negotiations are in reality often linked, as norm opponents can side with those norm promoters who advocate a less radical and farreaching interpretation of the norm.

The emergence and acceptance of norms In the early stages of norm diffusion processes, norm entrepreneurs (i.e. agents who have strong notions of desirable behaviour and actively promote this moral prescriptive) are generally deemed to be crucial (Finnemore and Sikkink 1998: 896). In the first stage, norm emergence, moral entrepreneurs call attention to issues and frame the problem at hand in a way that attracts attention to their preferred norm. They normally have to fight competing norms ‘in a highly contested normative space’ (Finnemore and Sikkink 1998: 897). For example, adherents of the right to abortion in the US (propagating the ‘abortion should normally be permitted’ norm) clash with anti-abortionists (advocating the ‘abortion should normally be forbidden’ norm). If and when a certain contested norm ‘wins’ such debates – and this might be a long or a short process – it has a chance to spread further and become ‘unobjectionable’ (as the right to abortion norm is in some countries). Typically, persuasion is the instrument that norm entrepreneurs have at their disposal. Moral entrepreneurs need organizational platforms, either specifically constructed for the purpose of promoting a norm, or from existing international organizations, like the United Nations (UN). A high legitimacy of platforms and knowledge claims enhances the chances for a norm to survive and prosper (cf. Bernstein 2000: 482). If the norm has the backing of scientists, who can frame an issue and argue for a certain solution in scientific terms, this is a powerful legitimizing mechanism (cf. Haas 1992; and, from a discourse perspective, Bäckstrand 2001). Norm supporters need to secure the support of international organizations and states. Once they have succeeded in gaining a foothold in a few countries – often

32 Ole Elgström with the help of like-minded national mass movements and coalition partners within state administrations – these countries act as fellow entrepreneurs and try to persuade other states to follow. Obviously, it matters which states are first to adopt the norm (Finnemore and Sikkink 1998: 901; Florini 1996: 374). When a certain threshold is reached, and there is a critical mass behind the new norm, a ‘norm cascade’ may follow and the norm becomes dominant. The stage of acceptance has been reached. States may accept the new norm because decision makers internalize it or because they calculate that the costs would be unacceptable if they did not do so. Emulation, praise and shaming are all essential elements in this process (Checkel 1997, 1999; Finnemore and Sikkink 1998: 902–3), as is the use of traditional material levers (Payne 2001). According to Thomas Risse et al. (1999), recalcitrant states typically follow a shared pattern by first denying the validity of the norm, then accepting the norm in principle but denying allegations of violating it; thereafter they officially accept the norm as a result of combined external and internal pressure and start changing their laws; finally, the norm is internalized. The more compatible a norm is with other international norms – the higher its ‘adjacency, precedence and fit’, in the words of Finnemore and Sikkink (1998: 908) – and the more widely endorsed a norm is (in treaties and in international discourse), the more likely it is to become influential and possibly predominant (Bernstein 2000: 483; Florini 1996: 374; Legro 1997: 34). Influential norms become associated with other positive values. For example, environmental concern has been linked to both security and economic profitability; gender equality to participation and democracy. In some cases, critics may argue that conflation of a laudable norm and other dominant international norms respresents a compromise, diluting the essence of the basic norm. The victory of what Steven Bernstein (2000) calls ‘liberal environmentalism’ (combining green concerns with economic expansion and liberalized trade) over ‘ecodevelopment’ (which privileges environmental protection over growth and views growth through economic liberalization as invariably perverse) is one example. Keck and Sikkink (1998: 27) argue that the intrinsic characteristics of some norms, that is, the qualities of the norm itself, contribute to its success. They claim that norms involving (a) prevention of bodily harm for vulnerable groups and (b) legal equality of opportunity are particularly effective. In other words, some norms are claimed to carry a particularly high degree of ‘emotive appeal’ (Baumgartner and Jones 1991). When (if ever) a norm becomes highly institutionalized and taken for granted, it is what I call an ‘unobjectionable’ norm. At this stage, it is considered socially appropriate to follow the norm and immoral not to do so. ‘Unobjectionable’ norms have to be continually consolidated. They are normally incorporated into a variety of legal texts and implemented in different organizational contexts. In these ways, the degree of internalization is gradually increased and diffused from political circles to different bureaucratic environments. There is also a tendency to increase the scope or domain of a general norm by applying it to other issue areas. General demands for gender equality first focused on female suffrage but have spread to demands for equal representation in both political and non-political areas. Protests against discrimination of homosexuals have spread to demands for equal rights as

Consolidating ‘unobjectionable’ norms

33

regards marriage and adoption. External events often act as precipitants in the diffusion and consolidation of norms. In the environmental arena, we have seen how cases of river pollution and oil tanker accidents have hastened the ‘greening’ of the European Union, as have debates about holes in the ozone layer and climate change (cf. Kronsell 1997: 106).

Norm spread and negotiation When a general norm (for example, ‘the principle of sustainable development should inform all decision making’ or ‘concern for gender should be integrated into policy making across all policy sectors’) has been politically accepted, it still has to be transformed into legal texts and, finally, implemented in terms of concrete policy. Analytically, two stages exist: the phases of ‘legalization’ and ‘implementation’. In both these stages, I argue, the new norm usually encounters resistance. Therefore, negotiations are normal and crucial elements in both phases. Resistance against ‘unobjectionable’ norms exists for two main reasons. First, many different cultures and ideologies co-exist in a broad political context (like the European Union). The new norm is normally more highly prioritized in some of these than in others. Some states are ‘vanguards’, others are ‘laggards’. So while decision makers and officials in countries that are ideational laggards cannot, for reasons of political correctness, openly oppose the norm, they are still not enthusiastic (the norm is not yet internalized) and may try to dilute the effects of the new norm in actual policy making. Second, officials in any bureaucracy are firmly embedded in an organizational culture, by education and by intraorganizational socialization. New, politically decided norms do not always fit the dominant norm structure within a bureaucracy. Therefore, the existing value system often acts as a stabilizer, preventing rapid change (cf. Goldmann 1988; Legro 1997: 35). Furthermore, counter-norms may exist in circumscribed areas: norms that are also unobjectionable and are felt by defenders to be threatened by an unrestrained spread of the relevant norm. Protection of the individual’s personal integrity is thus a norm that in certain cases may compete with the transparency norm. There might also be norm-based objections that pertain to the consequences of the norm if it is applied too widely or too strictly: effectiveness norms are of this kind. For example, ‘too much openness’ is claimed to result in non-optimal decision making by actors who resist increased transparency. In general, these types of norm-based objections tend to result in demands for restrictions in the scope of the new norm. Some exceptions are normally allowed even by the adherents of ‘unobjectionable’ norms (for example, in Sweden abortions are allowed only under very special circumstances after the 18th week of pregnancy; information that may endanger state security or business secrets are protected even in the highly open Swedish transparency legislation), but objectors tend to strive for a considerable increase in the number and significance of exceptions. Norm negotiations are of two kinds, I argue. In the legalization phase, norm resistance is typically played out in text negotiations. In any legal text, norms have to be defined and interpreted in the concrete policy context. Often, the scope of new

34 Ole Elgström rules has to be delimited and rules of application decided upon. The text is penetrated paragraph by paragraph and the exact wording is carefully examined. This creates possibilities for norm opponents to influence the impact of the norm by acting for: ● ● ● ●

vague definitions and fuzzy language limited application areas numerous exceptions long transition periods.

The fact that norm advocates are often internally divided, and that this may lead to interpretative negotiations, can be utilized by norm objectors. They can support the not-so-radical norm entrepreneurs in the internal struggle to have their less farreaching standpoints reflected in the text. At this stage, persuasive efforts and debate are clearly linked to both norm internalization and to cost–benefit analyses among reluctant actors. Once a legal text exists, the new norm has to be integrated into concrete policy and decisions to become a part of every-day decision making. In the highly legalistic European Union, the fact that a norm has a legal base is in itself very important. Now, moral entrepreneurs can refer to the legal document when advocating further norm spread. Without a legal base, many officials could easily ignore the new norm – when such a base exists, this is no longer a viable option. The European Union may prescribe that reference to the norm be included in policy documents. Many factors thus speak in favour of an ongoing institutionalization process, whereby a growing number of officials take the new norm for granted. Increasing internalization of the norm is thus to be expected. As pointed out by Jeff Legro (1997: 34), the robustness of a new norm influences its penetrative power. A widely endorsed and highly specific norm that has long durability and has weathered many challenges is more likely to be fully implemented. Still, norms differ in this respect and, therefore, implementation negotiations must be expected. Norm opponents may argue that the new norm is not applicable in concrete cases or that it is irrelevant in a certain context. ‘The less specific the norm, the easier it is for actors to justify violations in the fuzzy language of a norm’s prescriptions’ (Shannon 2000: 304). They may also claim that it is not advisable to implement the norm as it is in conflict with highly valued process-oriented norms, such as efficiency. The cost of implementation may be too high or it may take too long a time. Persuasion and pressure are prominent features at this stage as well. Next, two empirical cases are described in order to illustrate the emergence, acceptance, legalization and implementation phases in the EU setting. First, attention is turned to the introduction of gender equality norms into EU development aid policy. Thereafter the processes behind the ongoing spread of openness norms in the organizational units of the EU is highlighted. The cases are similar in many respects, but the process of norm diffusion has gone further in the ‘gender and development’ case, where legal texts exist and the process has entered the implementation phase. In the transparency case, the solid agreement on openness

Consolidating ‘unobjectionable’ norms

35

as a general guiding principle conceals widely differing opinions when it comes to deciding the scope and applicability of the norm: traditions of secrecy are still strong in many member state bureaucracies. The process has still not reached the implementation stage.

Gender and development Emergence and acceptance of gender equality norms In the 1990s the European Union adopted a new and broader agenda for equal opportunities between women and men (Pollack and Hafner-Burton 2000). From a traditional focus on equal treatment and equal pay at the workplace, the European Union decided to add positive actions on behalf of women to its tool-kit to promote gender equality. Furthermore, ‘gender mainstreaming’ quickly became a new catchword, meaning the integration of gender analysis into all kinds of EU policies and programmes. The old emphasis on women’s rights in their roles as workers was gradually diffused to other issue areas. One of these was development aid policy (for a detailed description of this case, see Elgström 2000). Behind the introduction of gender norms and the broadening of the agenda stood a plethora of norm entrepreneurs. Supranational actors like the Equal Opportunity Unit of the Commission and the Women’s Rights Committee of the European Parliament were joined by various non-governmental organizations (NGOs) and national gender experts to form a powerful transnational advocacy coalition (the term is from Keck and Sikkink 1998; see also Pollack and HafnerBurton 2000: 434). In the specific arena of development cooperation, national experts from a number of mostly northern member states (the Netherlands, Great Britain, Denmark and Sweden) played the role of forerunners. They had the support of development NGOs, both broadly focused (like VOICE and Eurostep) and gender-oriented (e.g. Women In Development Europe, WIDE), and a very small number of ‘women in development’ officials (two!) in Commission Directorate General (DG) I and VIII. The broadening and widening of the gender agenda furthermore benefitted from a number of crucial internal and external events. The Maastrich Treaty gave the European Parliament expanded powers. As the Parliament is known as an energetic engine behind gender equality, this contributed to the spread of gender norms. In the specific case at hand, the Fourth World Conference on Women in Beijing in 1995 was a catalytic event. The Beijing conference first of all firmly introduced the concept of gender mainstreaming into the dominant frame. Second, EU decision makers believed that an EU policy on gender and development was needed in Beijing in order to avoid being shamed by this gathering of norm promoters (Elgström 2000: 463). In brief, gender equality had become a predominant norm in the European Union in the early 1990s. A well-organized network of norm entrepreneurs, including important institutional actors as well as NGO representatives and member state advocates, succeeded in transferring gender norms into other settings,

36 Ole Elgström including foreign aid. As hypothesized, major external events helped to widen the scope of gender mainstreaming efforts. Legalization of gender equality norms in development aid policy In a Council conclusion of May 1993, the Commission was asked to produce a policy document on gender and development. The decision was prompted by the perceived need for a platform at the approaching Conference on Women and coincided with the renaming of the DG gender desks, from ‘women in development’ to ‘gender and development’. The signal was clear: gender concerns were from now on part and parcel of development policy. As it was evident from the start that the understaffed gender desks needed help in accomplishing the task of drafting a gender resolution, the first move by the Commission was to call upon the support of member state officials. A gender expert group, already existing but almost dormant, was reinvigorated. The expert group became an arena for interpretative negotiations. Vanguard experts from countries like the United Kingdom, the Netherlands, Denmark and Sweden advocated mainstreaming (that gender analysis should form part of all EU development projects and policies), while experts from other countries still hung on to the ‘old-fashioned’ (according to the radical experts) focus on specific aid projects to further the role of women in society. The radical group was also more inclined to bolster positive action (for example, allocating jobs or positions by quota) than the traditional group. The hesitancy showed by the less radical country experts was partly due to the fact that they realized that their home governments would never accept too radical formulations, although they themselves often sympathized with feminist ideology. As this political fact was acknowledged by all delegations, the result (formally, a Commission communication to the Council, delivered in September 1995) was a rather general document with uncontroversial formulations in order to get all states on board. It was nevertheless a document that clearly introduced ‘gendered thinking’ into the development aid arena and made mainstreaming a key concept, very much in line with the end results of the Beijing conference. In this sense, the expert committee acted collectively as a norm entrepreneur. The text negotiation that had started in the expert committee continued, more intensely, in the Council working group that took over. Here, the ideological differences were much more pronounced, especially to begin with, than in the expert group, which had consisted entirely of gender experts. Some of the diplomats represented in the working group made open fun of the gendered language used by the experts. Many member state representatives were not convinced that gender mainstreaming was really a necessary or important part of foreign aid. Hence, it is possible to speak of resistance to the novel ideas. On the other hand, all delegations acknowledged the necessity of producing an EU document on the issue – the general norm was ‘unobjectionable’. Certain countries were more active than others in the working group. States with developed gender policies in their own aid policy were at the forefront. Thus, the Danish, Swedish, Dutch and

Consolidating ‘unobjectionable’ norms

37

British representatives played leading roles in the discussions and tried, together with the Commission, to carry through the ideas contained in the communication. The result was text negotiation, which mirrored a concern among member states about basic normative tenets, regarding the phrasing and wording of individual paragraphs. For example, France refused to support a sentence stating that the Council ‘fully’ endorsed the main recommendations of the Commission communication. The word ‘fully’ was finally dropped. Many countries attempted to introduce concepts and norms that were of particular concern to them. A dispute also broke out over positive action: the traditionalists wanted to downplay this type of intervention, arguing that discriminating special treatment should only be used in exceptional cases. In the end, positive action was given a less prominent place than mainstreaming, but was still to be given ‘special attention’. In brief, text negotiations in the expert group and in the working group produced a path-breaking document on gender and development. The new gender norm was irresistible in this context, too. But the norm objectors succeeded in watering down the document, in de-emphasizing the most radical concepts and in making the resolution much more general and vague than the norm proponents would have wished for. Implementation negotiations in the comitology The Council resolution meant that its major norms were now formally accepted by the member states and the European Union. But formal norm adoption does not necessarily result in effective implementation of these norms. In this case, the implementation phase has proved to be as problematic as it is crucial. On the one hand, the existence of a legal base has created a momentum of institutionalization: most EU officials feel obliged to take existing legal norms seriously. They follow a logic of appropriateness (March and Olsen 1989). On the other hand, gender norms have encountered continued resistance in the EU bureaucracy. ‘Gender concerns’ more often result in a superficial reference in a text than in penetrative analysis and field implementation. Some structural traits of the Development General Directorate (formerly DG VIII) have served to inhibit implementation. The Directorate has traditionally been devoted primarily to large, infrastructure projects. The organizational culture has put a premium on substantial resource transfers, and so-called horizontal questions (such as social development in general) have had a low status. Furthermore, the mainly geographical structure of the Directorate has been an obstacle to norm diffusion. Gender and other horizontal issues have been placed in special administrative units, removed from day-to-day project administration. Besides such structural impediments, dominant attitudes have also countered norm consolidation. Gender is a peripheral, low priority issue for many officials. Horizontal issues are ‘unfamiliar’ – as the educational backgrounds of the officials have not stressed such concerns – and create extra work for them in their efforts to ‘get things done’. Gender awareness – norm internalization – is very low. Therefore, quite a few officials view gender as a ‘temporary fad’, which must be

38 Ole Elgström given official recognition but is not really integrated into routine analysis, as mainstreaming implies. They also utilize exclusionary arguments, claiming that norm concerns are not appropriate in the specific setting they are handling (‘gender analysis has no place in planning for road construction’). Norm entrepreneurs have therefore had to engage in continuous implementation negotiations. These have mostly taken place in the EDF (European Development Fund) committee, a comitology committee for development issues. Some member states have repeatedly commented on the lack of gender analysis in the documents submitted to the committee. This has created some irritation, but has not been without effect: sections on gender are now routinely incorporated in texts. Iterated complaints of norm non-enforcement can result in norm institutionalization. DG gender experts also seek to increase gender awareness within the organization by norm teaching. They constantly try to explain to sceptical colleagues why gender analysis is vital even in technical or infrastructure projects. In brief, the implementation phase is characterized by an ongoing mixture of persuasion and negotiation. The adoption of the gender resolution has facilitated norm consolidation, and more and more documents include gendered analysis. Gender equality advocates still face substantial opposition, however, and have to continue their negotiation efforts in the EDF committee and within the administrative machinery itself.

Transparency The emergence and acceptance of openness norms ‘Transparency’ and ‘openness’ are complex concepts involving many different dimensions. These include the right of citizens in a democracy to be informed, to be able to access proposals and other documents that have been decisive in decision-making processes. They may include less complex and more consumerfriendly language in these documents. They may include the right for citizens to know how their representatives behave and vote in legislative bodies. Finally, they may imply a right for citizens to participate in the decision-making process. In general, a British–Continental secrecy culture has confronted a Nordic openness culture. In the continental administrative tradition, the basic norm is that all documents in the public sphere are secret, unless the author/producer of the relevant information has given his or her permission for them to be released. According to openness norms, all information is open and available to the public, unless there is a clear legal base for protecting individual or public interests. Behind the openness norm is the belief that democracy demands that decision making be transparent and that the citizen be able to decide what information he or she needs to make informed decisions. The Danish no and the French petit oui to the Maastricht Treaty of 1992 led to a perceived need to increase the legitimacy of the European Union. Increased transparency soon became a major instrument to achieve this end. External events thus had a momentous effect on norm development, strengthening the openness

Consolidating ‘unobjectionable’ norms

39

norm – to the extent that it was made almost ‘unobjectionable’ – in its competition with secrecy. But the introduction of explicit openness norms in the Treaty of Amsterdam was not only the consequence of large-scale societal developments: the activities of influential norm entrepreneurs helped to frame the debacle of 1992 as a ‘transparency question’ and to place openness solidly on the Amsterdam agenda. With the entry of Finland and Sweden in 1995, a Nordic openness bloc was created, including Denmark and the Netherlands as well. All these countries were strong supporters of openness norms (albeit of different shades, see below) and acted as change agents within the European Union. The fact that Holland held the Council Presidency during the final negotiations leading to the Amsterdam Treaty certainly helped in sharpening key formulations. Later on, Britain and Ireland joined the openness coalition as a consequence of their own internal transparency processes, further strengthening the bloc of norm entrepreneurs. In the end, Article 1 of the Amsterdam Treaty (‘an ever closer union among the peoples of Europe, in which decisions are taken as openly as possible and as closely as possible to the citizen’) meant that the general openness norm was now politically accepted by the European Union. It had become flagrantly politically incorrect to argue for less transparency in EU policy-making processes. However, the new predominant norm had still to be translated into specified legal texts to really influence EU bureaucratic behaviour. The legalization of the openness norm According to the Amsterdam Treaty, a decision on rules regarding the application of the Treaty text had to be taken before May 2001 (‘General principles and limits on grounds of public or private interest governing this right of access to documents shall be determined by the Council’; Treaty of Amsterdam, Article 191a). This regulation is supposed to guide EU officials in their interpretation and use of the general openness norm. Therefore, it is of major importance for what actual practice will develop in the EU institutions. Not surprisingly, all actors concerned realized this. The result has been intense text negotiations. Analytically, it has first of all been possible to ascertain the existence of interpretative negotiations among the openness norm entrepreneurs themselves. Some countries, for example Sweden, advocated the protection of all information until the political decisions have been taken; documents should not be released before the politicians have had the opportunity for internal deliberation. Others pressed for the right of citizens to monitor – and possibly participate in – the whole decision-making process, including the preparatory phase. The Commission regulation proposal excluded ‘internal documents, which express individual opinions or reflect free and frank discussions . . . as part of internal consultations and deliberations’ (Preamble, point 9; see Statewatch 2000a). This exception can be interpreted as a way of meeting exclusionary arguments from the first-mentioned school of norm adherents. Second, norm opponents entered into fierce competitive bargaining with norm entrepreneurs. As they could no longer openly advocate secrecy norms, they now

40 Ole Elgström relied on attempts to introduce exceptions and vague formulations that would give hesitant officials a huge leeway in implementing the norm. In fact, the Commission proposal was heavily criticized by ombudsman Jacob Söderman for ‘a list of exemptions from access without precedent in the modern world’ (Statewatch, 2000a) – indicating that norm opponents had been able to infiltrate the informal negotiations leading up to the proposal. Mr Söderman suggested that the actual application of the regulation would result in a very wide room for interpretation by EU officials: ‘In practice . . . citizens would not so much enjoy rights as be dependent on the goodwill of officials’ (Statewatch 2000b). Another critical area, emphasized by the European Parliament – a very active player in the competitive negotiations – was whether to stress the right of access to documents or the right of access to information. Getting access to a written memorandum does not guarantee that you will understand the text or its consequences. Therefore, norm entrepreneurs in the European Parliament asked for increased clarity and simplicity in EU documents, improved use of the Internet and a larger number of open meetings. The fear is that norm opponents may otherwise succeed in what is believed to be their main aim: to prevent public insight into internal bureaucratic considerations (as they see it), even if a legal text exists. Norm resisters used efficiency norms as their main argument in text negotiations. The primary motive behind secrecy has always been efficiency. All good arguments will not be aired if openness prevails, it is claimed, and this will decrease the quality of decision making. Open debates take more time than protected bureaucratic decision making and also use up resources that could be better spent. In a political environment that is extremely sensitive to accusations of mismanagement and inefficiency, as the European Union, such concerns carry a potential appeal. The agreement reached in May 2001 was a compromise that reflected both the views of those committed to increased transparency and the concerns of more reluctant actors. The Swedish Presidency – one of the main entrepreneurs behind the transparency norm – was in the awkward position of being in a minority in the more conservative Council. In such a situation a country holding the Presidency is expected to give in and refrain from pursuing its own interests. However, Sweden skilfully used the radical, openness-oriented position of the European Parliament to broker a compromise that defended Swedish national interests in the area. Concessions were made in areas of major concern for other member states (not least the great powers of Germany and France) – notably restricted access to documents in the second pillar – to produce a deal that in general considerably increased the openness of European institutions. The MEPs hesitantly agreed to a solution that in their view was unlikely to be bettered in either of the next two (Belgian and Spanish) Presidencies. The proof of the pudding, though, is still in the eating. As the regulation concerning the application of the openness principle has come into place, actual implementation and field application will test the strength of the new norm. According to my theoretical framework, implementation negotiations are predicted

Consolidating ‘unobjectionable’ norms

41

in all organizational structures that will monitor and assess norm application. Norm opponents will continue to resist a thorough reshaping of the old, more secrecy-oriented regime. Norm entrepreneurs will have to vigilantly follow implementation and carry on their persuasive efforts.

The two cases: a discussion Both cases have demonstrated the prevalence of norm negotiations during efforts to consolidate ‘unobjectionable’ norms. Norm entrepreneurs have had to engage in bargaining with norm opponents. The reluctant actors have in both cases tried to dilute the effects of the norm in text negotiations. They have advocated vague formulations that would leave implementers a high degree of discretion. Intensive discussions have been held around what exceptions to allow. In the transparency case, even most norm advocates admit that some restrictions are necessary, because complete openness could adversely affect other values. The norm opponents, however, have argued for a much wider range of restrictions and exemptions, which in reality would allow for highly secretive practices. In the gender case, reluctant actors have maintained that gender concerns are irrelevant in their spheres of development aid (exclusionary arguments), but have been hard pressed to find convincing substantive arguments for their positions. In the end, they have often referred to the costs and trouble involved in taking equality measures into account. Much of the debate in both cases has thus concerned effectiveness and efficiency norms. Critics have argued that ‘too much’ transparency or gender concern leads to lower goal fulfilment and/or higher costs. Norm entrepreneurs have countered by claiming that openness and equality in fact improve goal attainment: transparency creates a better ideational and factual foundation for good decisions; involving women in decision making ensures that not only male experiences are taken into account. The gender equality norm today holds a strong position in legal texts and in policy discourse. But it is still not completely consolidated. Implementation of gender mainstreaming seems to lag behind. In the field, serious consideration of gender aspects is often missing from project planning. The legalization phase of the transparency process has only recently been concluded. Regulations on what rules to apply, how and when, were decided upon in May 2001. It remains to be seen how the new rules are interpreted and implemented. In both fields, norm negotiations will continue.

Conclusion In this chapter I have argued that not even ‘unobjectionable’ norms are automatically translated into concrete policy. The reason, I maintain, is that norm opposition continues even after high-level political consensus has made a norm impossible to criticize openly. The new norm is less enthusiastically endorsed in some quarters and reluctant norm recipients try to dilute the consequences of the norm. Therefore, norm negotiations are to be expected.

42 Ole Elgström In contrast to negotiations about divisible objects, norm negotiations of the type studied here are not mainly about finding compromises on substance. Instead, they concern (a) finding acceptable wording and formulations and (b) deciding about the scope of the new norm. They are to a large extent implementation oriented in that both proponents and opponents try to influence what impact the new rules will actually have on practical decision making. The question is whether there is another theoretical approach that might better explain the patterns I have described. A realist, or intergovernmentalist, model seems to be the natural contender. Such an approach would argue that politics is not a fight about ideas, but a struggle between state interests. Power concerns and material interests, such as security or welfare, are the driving forces behind new policies. States, and primarily great powers, are the main actors. Norms and values do not have any independent causal effect in a realist account. If norms are claimed by politicians to influence policy or are used by them in a political debate, these are seen as tactically motivated moves to further material goals. In my view, such an interpretation is difficult to defend in my cases. First of all, not only states are important actors but also institutions and non-governmental organizations. Second, the most active state actors are smaller nations like Denmark and Sweden and mid-range powers like the Netherlands (although the United Kingdom has also been actively involved in both cases). These countries have acted as norm entrepreneurs and pushed other EU member states towards norm acceptance. To maintain that these actors are driven by material interests in their engagement for gender equality and transparency is scarcely credible. They may well define these values to be included in the national interest – but this would rather strengthen the constructivist argument that norms help to form state identities and interests. States resist norm implementation not because their interests are at stake but because they have doubts about the supremacy of the new norm. I do not claim, however, that all state action in the two empirical cases is driven by normative concerns: it is acknowledged that actors may ascribe to a norm either because they have internalized it or because they feel pressured to accept the norm. Norm diffusion is the result of both socialization and rational considerations. Some states probably view the norm-based policy declarations they sign as ‘symbolic decisions’, which they have no intention to implement. The fact that these acts may still in the end lead to norm spread and internalization is another story. A governmentalist bargaining model à la Moravcsik (1993) is not well suited to fruitfully analyse the cases at hand. A focus on state preferences (based on material interests), power (in terms of alternatives) and package deals is of little help in explaining the motivations and behaviour of either norm entrepreneurs or their opponents. Normative convictions have to enter the calculus. But maybe the cases chosen in this chapter are ‘easy cases’ for my argument? Maybe gender concerns and openness are seen by leading politicians as peripheral issues that do not really concern national interests? Maybe a traditional realist account would be appropriate if vital interests were threatened by a new norm? Such assertions are obviously difficult to disclaim. It is furthermore clear that, for

Consolidating ‘unobjectionable’ norms

43

example, environmental norms have on occasion been superseded by welfare concerns when there has been a clash of interests. Still, I agree with the growing number of scholars who propose that ‘norms matter’. In my cases, millions of euros are spent on gender and other human rights projects and aid has been suspended to countries that have violated these cherished norms. Swedish Prime Minister Göran Persson has declared that he will never accept an EU decision on transparency that squarely goes against Swedish tradition and values. The fact that a country holding the Presidency of the Council is prepared to confront fierce criticism by a majority of the member states demonstrates plainly the power of norms. I therefore believe that my account of the spread of and resistance against ‘unobjectionable’ norms has a wide range of applicability.

References Bäckstrand, K. (2001) What Can Nature Withstand? Science, Politics and Discourses in Transboundary Air Pollution Diplomacy, Lund: Department of Political Science (dissertation). Baumgartner, F. and Jones, B. (1991) ‘Agenda Dynamics and Policy Subsystems’, Journal of Politics, 53 (4): 1044–74. Bernstein, S. (2000) ‘Ideas, Social Structure and the Compromise of Liberal Environmentalism’, European Journal of International Relations, 6 (4): 464–512. Checkel, J. T. (1997) ‘International Norms and Domestic Politics: Bridging the RationalistConstructivist Divide’, European Journal of International Relations, 3 (4): 473–95. Checkel, J. T. (1999) ‘Norms, Institutions, and National Identity in Contemporary Europe’, International Studies Quarterly, 43 (1): 83–114. Czarniawska, B. and Sevón, G. (eds) (1996) Translating Organizational Change, Berlin and New York: Walter de Gruyter. Elgström, O. (2000) ‘Norm Negotiations: The Construction of New Norms Regarding Gender and Development in EU Foreign Aid Policy’, Journal of European Public Policy, 7 (3): 457–76. Finnemore, M. (1996) National Interests in International Society, Ithaca, NY and London: Cornell University Press. Finnemore, M. and Sikkink, K. (1998) ‘International Norm Dynamics and Political Change’, International Organization, 52 (4): 887–917. Florini, A. (1996) ‘The Evolution of International Norms’, International Studies Quarterly, 40: 363–89. Goldmann, K. (1988) Change and Stability in Foreign Policy: The Problems and Possibilities of Détente, Princeton, NJ: Princeton University Press. Haas, P. (1992) ‘Introduction: Epistemic Communities and International Policy Coordination’, International Organization, 46 (1): 1–35. Kacowicz, A. M. (2000) ‘Studying International Norms: Is Constructivism a PreRequisite?’, paper presented at the Annual Meeting of the International Studies Association, Los Angeles. Keck, M. E. and Sikkink, K. (1998) Activists beyond Borders: Advocacy Networks in International Politics, Ithaca, NY and London: Cornell University Press. Kronsell, A. (1997) Greening the EU: Power Practices, Resistances and Agenda Setting, Lund: Lund University Press (dissertation) Legro, J. W. (1997) ‘Which Norms Matter? Revisiting the “Failure” of Internationalism’, International Organization, 51 (1): 31–63.

44 Ole Elgström Linz, J. and Stepan, A. (1996) ‘Toward Consolidated Democracies’, Journal of Democracy, 7 (2): 14–33. March, J. G. and Olsen, J. P. (1989) Rediscovering Institutions: The Organizational Basis of Politics, New York: Free Press. Moravcsik, A. (1993) ‘Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach’, Journal of Common Market Studies, 31: 473–524. Payne, R. A. (2001) ‘Persuasion, Frames and Norm Construction’, European Journal of International Relations, 7 (1): 37–61. Pollack, M. A. and Hafner-Burton, E. (2000) ‘Mainstreaming Gender in the European Union’, Journal of European Public Policy, 7 (3): 432–56. Risse, T. (2000) ‘Let’s Argue! Communicative Action in World Politics’, International Organization, 54 (1): 1–39. Risse, T., Ropp, S. C. and Sikkink, K. (eds) (1999) The Power of Human Rights: International Norms and Domestic Change, Cambridge: Cambridge University Press. Shannon, V. P. (2000) ‘Norms are What States Make of Them: The Political Psychology of Norm Violation’, International Studies Quarterly, 44 (2): 293–316. Spruyt, H. (2000) ‘The End of Empire and the Extension of the Westphalian System: The Normative Basis of the Modern State Order’, International Studies Review, 2 (2): 65–92. Statewatch (2000a) Text of article by Jacob Söderman, ‘The EU’s bid for opacity’, in Wall Street Journal Europe 24/2/2000, Statewatch 10 (1). Statewatch (2000b) Letter by Jacob Söderman to the President of the European Parliament, Statewatch 10 (x).

4

Negotiating transparency The role of institutions Bo Bjurulf and Ole Elgström

Aim and focus On May 30 2001 a new regulation on public access to EU documents (EP and Council 2001) was presented. It was heralded in the European press as a breakthrough for transparency in the arcane EU decision-making apparatus. The regulation contained rules and norms that had long been advocated by openness proponents. This outcome of the negotiation process poses a puzzle for many observers. A clear majority in the Council of Ministers, including the most powerful member states, supported a more limited transformation of existing rules, corresponding to the tradition of circumscribed openness that permeates their own societies. How did this majority finally agree to a text that was widely perceived to differ from their national interests? According to Realist accounts of international negotiations, the distribution of power is the best predictor of bargaining outcomes (Gulliver 1979; Habeeb 1988). Similarly, intergovernmental approaches to EU negotiations emphasize asymmetric bargaining power among the member states in explaining the results of negotiations (Moravcsik 1993). The focus in both cases is limited to state actors. Although some recent research has added complexity to the traditional view by exploring the role of non-state actors and networks ( Jönsson et al. 1998; Elgström and Smith 2000; Beach 2002), in general little attention has been paid to the role of institutions in EU negotiations. In this chapter we argue that a focus on institutional arrangements enhances our understanding of negotiation processes and final negotiated outcomes. In brief, the main message of institutionalist thinking is that ‘institutions affect outcomes’: ‘political struggles are mediated by prevailing institutional arrangements’, as Simon Bulmer puts it (1994: 355). The effect of institutions – ‘legal arrangements, routines, procedures, conventions, norms, and organizational forms that shape

We thank the Swedish Research Council for the Social Sciences and the Humanities and the Bank of Sweden Tercentenary Foundation for financial support and Christer Jönsson and Jonas Tallberg for helpful comments. We are indebted to the officials who shared their experiences and observations with us.

46 Bo Bjurulf and Ole Elgström and inform human interaction’ (Norgaard 1996: 39) – is either constraining or empowering (Aspinwall and Schneider 2000: 4). From this perspective, both formal structures and informal norms influence the bargaining strategies of the actors and, indirectly, the result of the negotiations. The aim of this chapter is to demonstrate empirically the importance of institutions in internal EU negotiations. We do this by detailing actor interests and strategies in the transparency area and by tracing and analyzing the negotiation process that resulted in the Regulation on public access to EU documents. By presenting a rich and detailed case study that draws on extensive primary material, we add to existing knowledge on actual EU negotiation processes. The empirical parts of the chapter are based mainly on interviews, on official documents and on material from the Statewatch homepage. Our claim is that the institutional arrangements surrounding the process created an operational environment for the relevant actors that led them to accept the final document. We submit that the outcome of the negotiations would have been different if other institutional mechanisms had been present. The institutions that will receive particular attention are: ● ● ● ● ●

agenda-shaping rules; decision-making procedures and voting rules; informal norms, especially those surrounding the office of the Presidency; timetables and deadlines; interventions by institutional actors.

After a theoretical background outlining the institutional approach, we go on to describe the main actors and their interests as well as the major conflict dimensions that characterize the case in hand. We continue with an analysis of the negotiation process, first going into the Commission proposal and the resulting intergovernmental negotiations during the French Presidency. We then scrutinize the complex, multi-actor negotiations during the Swedish Presidency, involving the Council, the European Parliament (EP) and the Commission. Finally, we summarize the impact of the institutional arrangements.

Theoretical framework In this chapter we use a broad institutional approach, concentrating on the ways in which institutions form a political environment for member states and institutional actors. Institutions provide a framework within which actors interact, shape their expectations and limit what options they perceive as possible (cf. Aspinwall and Schneider 2000). Thus, they structure relationships and processes. Institutional features and procedures combine with preferences to produce outcome (cf. Shepsle 1989: 135–7). They may determine, inter alia, who has the right of initiative, which actors have the right to participate in a decision, how and when decisions should be made and how implementation is monitored and sanctioned. Rules thus privilege certain actors and proscribe certain alternatives from consideration (Shepsle

Negotiating transparency: the role of institutions

47

1989: 137). Institutions also provide information that reduces uncertainty about the behaviour of others, thereby leading actors towards particular calculations (Hall and Taylor 1996: 945). Formal rules do not, however, always provide clear-cut prescriptions for what action is to be taken, and these rules are not the only institutional mechanisms that influence actor behaviour. Informal norms also provide essential guidelines for actors in the EU setting (cf. Eising 2002; Schmidt 2000). Expectations as regards permissible or appropriate behaviour circumscribe the performance of EU actors, especially when they fulfil institutional roles, such as holding the Council Presidency. In this context, we do not need to take a stand on whether norm compliance stems from strategic calculations of rational actors, making them choose actions that will result in lower expected normative costs, or if it stems from a ‘logic of appropriateness’ (March and Olsen 1984), where institutional identities mould actors’ decisions on how they should behave in a certain situation. Our research aim does not force us to choose between various institutionalist approaches (cf. Aspinwall and Schneider 2000; Hall and Taylor 1996; Peters 1999). All institutionalist designs agree that institutions have the capacity to shape, and create regularities in, individual behaviour. They provide a context, or political environment within which individual decisions are set. In our effort to explain concrete actor strategies and behaviour in one particular case, there is no need to take a decision on whether the actor interests we observe are exogenously driven, products of historical circumstances or constituted by institutional identities. Therefore, we have chosen a broad, eclectic institutionalist framework, including informal norms as well as formal rules. Our approach focuses on the particular ways in which actor strategies are shaped by different institutional settings and is close to the empirical case at hand. We follow the advice of Mark Aspinwall and Gerald Schneider to let ‘empiricism . . . play a more prominent role in the process of EU theorizing’ (2000: 15). By tracing an actual decision-making process in detail, we want to demonstrate convincingly the real-life impact of institutional arrangements.

Conflict dimensions, actors and interests Increased transparency is proclaimed to be an essential aspect of EU democratization. Greater openness and easier access to documents would, it is argued, help EU citizens and media to monitor and evaluate decision-making processes, thereby increasing public participation and strengthening understanding of EU policy making and loyalty to the Union. Still, different positions exist as to what ‘increased transparency’ should actually mean and what the practical implications should be. These diverse views were clearly visible in the decision-making process leading up to the transparency regulation. The controversies concerned: 1 2 3

the form of the negotiation process; the handling of sensitive documents; the degree of secrecy during the preparatory decision-making stage;

48 Bo Bjurulf and Ole Elgström 4 5 6 7

the degree of harmonization within the European Union and possibility of national variations; the degree of control to be exercised by the original source of a document; the degree of service-mindedness in relation to ordinary citizens; and the degree of legal oversight.

These conflicts divide the member states and other actors in complex and overlapping ways. To a certain extent, they mirror different traditions of openness and secrecy. According to one tradition, documents should be open if they are not classified as secret with reference to specific laws. In other camps, the predominant view is that all documents are in principle to be protected if the original source of the document does not permit access. In the first controversy, the Council’s diplomatic secrecy orientation is contrasted with the European Parliament’s attitude that democratic processes must be open. The second problem concerned how to handle sensitive security documents and created frictions between openness-oriented member states and the Parliament, on the one hand, and states with a strong security interest, on the other. The third conflict puts the European Parliament’s wish for open committee meetings and information in the first phases of the decision-making processes against the Council’s and the Commission’s wish to protect the early stages of decision making in order to ensure an open discussion between bureaucrats and politicians. The fourth division entailed a collision between those who wanted to have the same rules across the European Union – the Commission, the European Parliament and many member states – and those who wanted to continue with traditional, national rules (including Sweden with its offentlighetsprincip). The fifth problem concerned, inter alia, whether EU institutions should be treated as states, with a right to keep control over documents even when distributed among member states. The sixth dimension contrasts a service-minded attitude, typical of a politician’s perspective, with the bureaucratic interest in controlling the resources spent on service; and the seventh, finally, involved to what extent the European Parliament would have the right to monitor rule implementation. Four major players are identified: the Commission, the secrecy-oriented majority of the member states, the openness-oriented minority of member states and the European Parliament. The Commission’s main interests were to protect the ‘space to think’ during preparatory stages of the decision-making process and to diminish the potential administrative burdens and costs associated with stringent openness rules. The Commission wanted to ensure that internal documents would be shielded from insight to allow its officials freedom to engage in open and unfettered discussion and to put forward ‘trial balloons’ in internal consultations. Also, the potential costs of the regulation worried the Commission: ‘Allow access and everyone will want to see everything. This will entail indexing, and staff to deal with requests . . .’ (Birkinshaw 1996: 49). There was a fear that officials would be swamped by too many applications for documents. Therefore, rules were needed that, for example, detail in what form requests may be made. The Council majority in general preferred a situation where documents could be

Negotiating transparency: the role of institutions

49

easily protected. It was led by France and Germany but generally included also the southern member states, Belgium and Luxembourg. It should be noted, however, that for many of these states transparency was not a prioritized issue. As their interest was limited, these actors could be expected to agree to any compromise that was not ‘too’ radical in its openness approach. The majority was primarily interested in the confidentiality of the proceedings in the Council negotiation process. From this perspective, intergovernmental negotiations presume high levels of trust in and sensitivity to the dynamics of the process. ‘Through secret negotiations, parties can obtain relevant information they might otherwise not have received. . . . The trust between the negotiating parties can also be disrupted if the substance of the talks is disclosed at too early a stage’ (Stenelo 2000: 71, 73). Effective negotiations require confidentiality. The great powers – France, Germany and the United Kingdom – were even more concerned with the handling of sensitive documents related to national security and defence questions. Increasing contacts with the North Atlantic Treaty Organization (NATO) accentuated this problem. To exempt such documents from regulatory rules was the number one priority for the states with strong security interests. The Council minority included primarily Denmark, Finland, the Netherlands and Sweden (in some cases also the United Kingdom and Ireland). This group of states wanted citizens to have an extensive right of access to information, with few and carefully defined exceptions. Rejections of a request for a document should always be properly justified, and the EU institutions should in general adopt a serviceminded attitude in their contacts with applicants. The scope of application of the new rules should be wide, including for example all incoming documents. The openness advocates demanded the setting up of registers to allow the public to keep track of existing and incoming documentation. Furthermore, legislative Council meetings should be open — but not the intermediary stage of negotiation leading up to the vote. The European Parliament had an interest in gaining more information about Council negotiations, in order to become a more efficient partner in decision making, and therefore wanted openness also in preparatory deliberations, in the Commission as well as in Council committees. Increased transparency would make it easier for the European Parliament to spot cracks in the Council’s negotiating position but also allow parliamentary rapporteurs to produce better informed documents. The Parliament also upheld the citizens’ right to information and their right to be consulted during the decision-making process. The European Parliament has a strong interest in legislative oversight, including the right to monitor how the new rules are implemented and to participate in any future rule revision.

The negotiation process: institutionally constrained bargaining The EU transparency process started in earnest with a non-binding declaration on the right of access to information in the Maastricht Treaty. The Amsterdam process explicitly introduced principles of openness into the Treaty. Article 255

50 Bo Bjurulf and Ole Elgström required implementing rules to be laid down within two years (before 1 May 2001) on the basis of a proposal from the Commission and by co-decision between the European Parliament and the Council. The Commission’s proposal (Commission 2000) was presented on 26 January 2000. The Commission proposal: resigning its agenda-setting powers? According to formal rules, the Commission enjoys the monopoly to initiate legislative proposals to the Council and the European Parliament in the first pillar (Pollack 1997; Schmidt 2000). When the Commission has submitted a proposal, the Council has to decide by unanimity if it wants to enforce changes. As it is often easier, under qualified majority voting (QMV), to accept the proposal by a majority vote than to alter it, the Commission is claimed to have considerable agenda-setting powers (Schmidt 2001: 126; cf. Crombez 2000; Steunenberg 2000; Tsebelis and Garrett 2000). A skilful and far-sighted Commission can therefore, it is argued, produce proposals that are close to its own interests and are simultaneously acceptable to a majority in the Council. This obviously demands correct information about member state preferences. In the case in hand, the Commission proposal can be regarded as a compromise – but a compromise that no one applauded. The Commission included in its text elements that were supposed to appeal to the transparency advocates. Thus, incoming documents were included in the scope of application and a right of partial access to sensitive documents was granted. Such provisions were looked upon with suspicion by the Council majority. On the other hand, the list of restrictions and exceptions was long and detailed, and the proposal vigorously protected the confidentiality of the proceedings and the ‘space to think’ – aspects strongly emphasized by the secrecy-oriented member states. These elements also mirrored the self-interests of the Commission itself, which were further accentuated by the inclusion of two novel grounds on which access may be refused: ‘the effective functioning of the institutions’ and ‘the stability of the Community legal order’. These lofty formulations, which could be interpreted as a carte blanche for officials to decide when to agree to an application, provoked the transparency group. Thus, neither of the two factions of member states was satisfied with the proposal. The Commission strategy was to produce a compromise that was designed to appease the openness proponents but still retained essential features of the traditional secretive handling of documents. The Commission trusted that even transparency-inclined states wanted to defend the ‘space to think’ and the confidentiality of the process, two aspects that were mainly attacked by non-state transparency advocates. It probably believed that the concessions to openness described above would satisfy the minority without being so radical as to alienate the majority. In fact, representatives of the openness coalition also warmly welcomed these elements. What the Commission failed to predict was that its rather crude defence of its own self-interests, when linked to the imprecise limitation clauses it introduced, still made its proposal unacceptable to the

Negotiating transparency: the role of institutions

51

openness-oriented minority. In the words of one of our Commission interviewees: ‘We tried it the Nordic way, but that didn’t work.’ As predicted, the Commission initiative was an attempt to set the agenda and to include firmly in the text provisions protecting vital Commission interests. The institutionally determined right of initiative of the Commission forced other actors to abstain from official activism until the proposal was presented (though they did not abstain from unofficial campaigning; also, other actors, such as the Ombudsman and pro-transparency non-governmental organizations (NGOs) – who were not consulted by the Commission – were very active in persuasive efforts directed towards the Commission and the European Parliament). Commission proposals can in general be interpreted as single negotiating texts (SNT) (cf. Raiffa 1982) – texts that other actors have to react to and that will automatically give the Commission a privilege of formulation. The Commission failed, however, in this case to find a middle ground that was broadly acceptable to a Council majority. Its proposal was too ‘radical’ to please the secrecy-oriented members’ and the omission to deal explicitly with defencerelated sensitive documents irritated the great powers. Had the approach to these elements been satisfactory, the Commission might have found itself with a majority behind its proposal. Nor was the document creative enough to cater for a majority consisting of openness proponents and undecided middle-of-the-roaders. As it was, the field was left open to Council-led counter-initiatives, headed by the incoming Presidency, France. In the end, the Commission did not use its potential power position and persist in its own proposal. Later in the process it gave in to and supported Presidencysuggested alterations without much struggle. The control over the single negotiating text was increasingly transferred to the Presidency, and in the process the text was substantially revised. In this way, our case supports the conclusion of Susanne Schmidt that in empirical cases the Commission often resigns its formal powers, taking instead an ‘approach of least resistance’ and losing its initiative to the Presidencies (Schmidt 2001: 129–31). Negotiations under the French Presidency: defining a majority position The formal functions of the Council Presidency, as defined in the treaties, are limited. It is primarily supposed to convene and administratively manage all Council and working group meetings. Over the years the Presidency has, via evolving praxis, been given several new, and increasingly important, tasks. Today, the Presidency has, besides its administrative and coordinating tasks, three major additional functions: setting political priorities, acting as an ‘honest broker’ and representing the Council (Elgström 2003a). The Presidency may shape the agenda by means of its Presidency programme and by deciding what dossiers to include in, or exclude from, the various Council agendas. Although external events and Commission proposals largely predetermine the agenda, there is still room for manoeuvre for the chair. In this sense, the Presidency competes with the power of initiative of the Commission. The office

52 Bo Bjurulf and Ole Elgström also competes with the Commission in performing the function of a mediator in Council negotiations. There is a strong informal norm that says that the Presidency is expected to act as an impartial cobbler of compromises. ‘Presidency proposals’, based on ‘confessionals’, tour de tables and informal probing, are part of the negotiation game. The Presidency is thus expected to be an active participant with the goal of leading the processes forward and reaching decisions. The behaviour of the EU actors is, we argue, shaped and constrained by this institutional set-up. France approached its Presidency in the autumn of 2000 with two main goals for the transparency dossier: first, it wanted to achieve an agreed text before the end of its term and thereby be able to claim a success, something that Presidencies dearly aspire to do in as many areas as possible. Second, it wanted to produce a document that mainly reflected the majority position in the Council (that is, a text that also mirrored French interests). The first goal was unrealistic. There were too many disagreements within the Council that had to be debated and worked out in order to reach a solid majority. Also, and equally important, it turned out that the European Parliament, whose assent to any Council proposal was necessary due to the co-decision procedure, could not reach a common standpoint until very late during the French Presidency. Internal quarrels concerning what parliamentary committee should bear the main responsibility for the dossier led to a long drawn-out process. The second goal implied that France deliberately tried to chisel out a position that could achieve a Council majority. In this process, only those concessions to the transparency minority that were considered absolutely necessary were included. The French leadership does not seem to have been overly anxious about the informal consensus norm, which states that the Council should try to avoid taking votes and that minority interests should therefore be taken into account. Likewise, and contrary to our theoretical expectations, the given co-decision procedure does not seem to have influenced the French to seek contact with, or even to seek to meet some of the wishes of, the European Parliament. Rationally, we would have expected the Presidency to foresee and try to meet at least a minimum of Parliament objections, as this was absolutely necessary if a joint decision was finally to be reached. The minority strategy was clear: taking into account the institutional realities (that is, the French Presidency period and the Presidency prerogatives) the minority decided to adopt delaying tactics. The transparency group knew that the French would be followed by the Swedes – a member of the coalition. So the members decided to fight for their positions and clarify their standpoints while probing for possible compromise solutions during the upcoming Presidency. The result was a number of working party meetings at which the member states presented their objections to, or agreement with, the Commission proposal and feelers were sent out to provide for later concessions (see Council 2000a). The seeds of a Council compromise on service-level rules (controversy 6 above) were sown. The transparency group was critical of the Commission proposal, which it considered provided for extremely cumbersome and unhelpful procedures that

Negotiating transparency: the role of institutions

53

would severely restrict the access to documents. Based on a Finnish proposal, Germany and Belgium opened the door to a higher service level. This was an easy concession, as the consequences, in the form of more work and resources spent on service, were to be borne primarily by the Commission officials. After long discussions, disagreements on the list of exceptions had gradually diminished, though some bones of contention remained. In the 22 December French Presidency proposal (Council 2000b), which reflected the majority position at the end of the French period at the helm, the question of originator control (controversy 5 above) was the one issue where fewest developments had been made during the French leadership. In Sweden, the sender of a document (‘the originator’) cannot influence whether the document will be classified as open or not. In the Continental tradition, originator control was seen as natural. Differences also existed as to whether states and organizations outside the European Union – for example, NATO – should be able to exert such control and on how the rights of third parties, inter alia NGOs, would be affected. All paragraphs in the December document on this issue were bracketed. To summarize, the French Presidency used its institutional powers, informally as well as formally shaped, to form the process. Other actors took the leading position of the Presidency into account and reacted to institutional realities, including the sequence of upcoming Presidencies. The Presidency completely took over the initiative from the Commission, which came to occupy a very low-key position. Internal parliamentary quarrels contributed to postponing any decision. Additional institutional constraints: interventions from the European Court of Justice and the High Representative The influential role played by the European Court of Justice and other EU courts in creating a rule-based context for policy making has been emphasized repeatedly in the EU literature (cf. Hix 1999; Tallberg 1999). Decisions by the courts set the parameters for future initiatives and shape actor expectations. In the case at issue, the courts were arenas for judicial processes that interpreted existing transparency rules simultaneously with the Council decision-making process. In one case (Court of First Instance T–14/98), the Court of First Instance ruled that a decision to classify a document as secret did not preclude that less sensitive parts might be open to the public, thus granting citizens the right to partial access. The same court also decided that a refusal of a request for documents must include the reasons why the documents were not released (Court of First Instance T–124/96). Perhaps most importantly, the European Court of Justice declared that the Council had to take the citizens’ interests into account when deciding whether to disclose documents or not. This decision established a ‘harms’ test principle’, according to which the negative effects of granting access must be weighed against the public interest in openness. By all these decisions, the courts helped the transparency coalition in the Council. Expectations of predicted court decisions ‘solved’ a number of outstanding contentious issues in the Commission proposal and created a fait accompli for the member states.

54 Bo Bjurulf and Ole Elgström In late July 2000 the High Representative of the Council, Mr Solana, issued a decision (High Representative 2000) regulating the openness of documents that are sensitive for security or defence reasons, and in effect severely curtailing the possibility to gain access to such documents. The decision followed a debate in the Working Party on Information on the legal possibilities to introduce special rules on the security area into the regulation. For the EU great powers, this was a key priority, not least because of an increasing interaction with NATO, but most member states admitted that special rules were necessary for political reasons. The Finns, generally transparency oriented, had declared that special provisions were indeed legal, but that they had to be decided by co-decision. The ‘Solana coup’, quickly confirmed by the Council (but with Finland, the Netherlands and Sweden voting against and Denmark making a reservation), was heavily criticized from many sides. The Parliament, which in principle understood that some special rules were needed, condemned the one-sided nature of the decision – the EP had not been consulted – and also decried the vagueness of the text formulations, fearing that the scope could be widened to include also sensitive documents in nonmilitary domains. In the French Presidency document, the essential provisions of the Solana decision were included to give them a legal basis. The high priority given to the issue by some leading Council members, and the negative attitudes created by the coup in the Parliament, was later used by the Swedish Presidency as a tool in its efforts to find compromise proposals. The Swedish Presidency: using co-decision to one’s own ends When Sweden took over the Presidency in January 2001, its predecessor, France, had presented a compromise proposal, in the form of a Presidency note, on 22 December (Council 2000b). This document mirrored, as we have demonstrated, the preferences of the Council majority. No official Council position existed. The Parliament had adopted amendments to the Commission proposal in November. Sweden itself occupied a somewhat awkward position. The nation holding the Presidency, according to strong informal norms, is expected to act as an impartial mediator and to ‘pay the price of the Presidency’, i.e. to make concessions in order to broker a compromise agreement. It is, above all, not expected to pursue its own national interests forcefully (Elgström 2003b). At the same time, Sweden was an ardent advocate of the transparency minority position. To complicate things even more, Sweden also had some particular concerns with the outcome of the process: the Swedish freedom of information act, Tryckfrihetsförordningen, established already in 1766, has constitutional status, and any perceived hollowing of these ingrained national rules would meet with fierce domestic resistance. Sweden had therefore early in the process tried to argue that the appropriate legal form of the EU transparency document was a ‘decision’ (aimed only at the institutions) rather than a ‘regulation’ – as the latter has direct effect in the member states. When this failed (the Council legal service immediately proclaimed that the correct legal form

Negotiating transparency: the role of institutions

55

was a regulation), Sweden had to fight a lonely battle to include formulations that would ensure its right to continue using the traditional national rules. Defending its traditional rules was to Sweden an interest of highest priority – as constitutional interests often are. The Swedish strategy depended upon the institutional context. First, it could rely on its Presidency position, which gave Sweden the possibility to put forward compromise proposals and to initiate informal negotiations. Secondly, it could utilize the co-decision procedure to its advantage. Co-decision necessitated a joint Council–Parliament agreement. As the proposed EP amendments reflected a rather radical transparency position, they were a far cry from the Council majority proposal. This situation allowed Sweden to put pressure both on the European Parliament and on the secrecy-oriented majority in the Council. In the words of the minister in charge: ‘We will [in the General Affairs Council] point out that codecision with the European Parliament must entail concessions also from the Council members’ (Swedish EU Committee 2001). The Swedish strategy was thus to use the wide gap between Council and Parliament to its own benefit by acting as a mediator between the two legislative actors. As the Swedish standpoints were in between the two positions, although far closer to the Parliament’s side, it could manoeuvre to seek a compromise that suited its own (and the Council minority’s) preferences. In brief, Sweden set out to be a ‘biased mediator’ – at the same time as it strongly underlined its impartial status to the other actors so that it was not seen as an outright norm-breaker. The trialogue – a new institutional mechanism Sweden could also make use of another, relatively new, working mechanism, the trialogue (cf. Shackleton 2001). If Parliament and Council do not agree, a conciliation process is set in motion. The trialogue in conciliation is a meeting ground where ‘ideas can be exchanged without formal decisions being taken, each side accepting to refer back possible solutions to their delegation’ (Shackleton 2001: 9). It is a highly restricted forum with a limited number of participants: the Presidency for the Council, two or three members of the Parliament delegation and a representative from the Commission. The Council has consistently argued that trialogues are recommendable already in the first and secondary readings, before the actual conciliation process starts, ‘in the interests of a clearer understanding of the respective positions and a more precise identification of areas of difficulty’ (Shackleton 2001: 10). From a power perspective, early trialogues are ‘good for the Council, which could treat the Parliament as a sixteenth country, but bad for the Commission that lost influence in the first and second readings’ (interview with member of the European Parliament). In contrast to trialogue in conciliation, where the Parliament has adopted amendments to a Council common position and the Council is obliged to respond to these, there is no common position at first reading, which means that the Council has to persuade Parliament to adopt in plenary amendments that the

56 Bo Bjurulf and Ole Elgström Council can accept. In other words, ‘both institutions are negotiating without either having finalised a point of departure for negotiation’ (Shackleton 2001: 11). Trialogue at this stage is even more informal than during conciliation. Representing the Council in such a small and informal group puts the Presidency in a favourable position: it can present its version of Council positions to the European Parliament, and report back its version of the EP response. The room for tactical manoeuvring is therefore substantial. In the case under discussion, Sweden pressed for a trialogue to start at this early stage. The Parliament, which otherwise is in principle reluctant to engage in nontransparent negotiations with the Council (see Shackleton 2001: 10), agreed. The reason was primarily to be found in the sequence of Presidencies: the Parliament knew that Sweden would be followed by Belgium and Spain, which were both unlikely to improve the terms offered by the openness-oriented Swedes. The European Parliament also felt the pressure of the looming deadline. An intensive round of contacts between the institutions followed. The Commission continued with its low-profile behaviour, being more of a listener than an active participant. This may have been an intended consequence behind the other institutions’ decision to enter into an early trialogue: ‘The Commission is the most conservative actor and by using the trialogue and an agreement in first reading you could side step the Commission’ (interview with member of the European Parliament). In many ways, the institutional peculiarities of the European Parliament made the negotiation process difficult. To begin with the Parliament was represented by a rather large number of individuals. Several committees had been involved in the internal EP process and all these wanted to be represented in negotiations with the Council. As described by Statewatch (2001): Not until 2 April, after the fifth ‘trilogue’ meeting, did the EP actually appoint a formal delegation, give it a mandate and agree it should formally report back to the main committee. Over the five months of secret negotiations the composition of the EP delegation varied. The system with trusted, powerful committee rapporteurs influenced the contents of the parliamentary positions and how they were presented. The rapporteurs were all highly engaged and transparency oriented (and all came from countries with a tradition of openness). As they dominated the debate in Parliament and the vast majority of the MEPs were, according to our interviewees, largely uninterested in the file, the arguments against radical openness were never really developed in the Parliament. The result was a rigid negotiation position. In the early round of the trialogue, few concessions were made, also because the European Parliament used the most radical transparency proponents as their spokespersons. It was not until the more flexible chair of the Citizens’ Freedoms and Rights Committee, Graham Watson, was given the role as sole ‘chief negotiator’ for the Parliament that the process started to move.

Negotiating transparency: the role of institutions

57

The end game: biased Presidency mediation Many other member states looked upon the Swedish activities with suspicion. The majority was worried that the Presidency, with its well-known inclination for openness, would sell out the majority interests. They were not convinced that their views, for example on potential concessions, were correctly represented in the trialogue, or that Sweden correctly depicted the offers from the European Parliament. Still, they felt the time pressure of the deadline, set in the Amsterdam Treaty to be 1 May, and had formally to respect the prerogatives of the Presidency, which at all times vigorously proclaimed its impartiality. Concretely, the Swedish negotiation tactics consisted of proposing a package deal, whereby the Parliament was promised Council concessions as regards service level (inter alia reducing the maximum time to respond to citizen requests to just 15 days) and parliamentary oversight (giving the European Parliament the right to monitor implementation) against EP concessions on the handling of sensitive security-related documents. The Swedish Permanent Representative, Gunnar Lund, entered into intense, very private, discussions with EP representative Watson. The negotiations were at times interrupted by consultations with their principals. Sweden argued that this was as much as the Council would ever get from the European Parliament, emphasizing the Parliament’s concessions. In a similar vein, Watson maintained that this was the best possible outcome for the Parliament, given the circumstances. In the end, this hectic endgame resulted in a compromise package ‘that was effectively presented [to the Parliament and the Council] on a take it or leave it basis’ (Shackleton 2001: 11). Both Council and Parliament, having weighed the pros and cons of the Lund–Watson proposal, decided to accept. The relative lack of interest of the Council majority helped to facilitate a compromise: no actor in this coalition cared so intensively for any individual aspect of the regulation – with the exception of sensitive documents – that the concessions were seen as reasons for blocking the compromise. The provisions protecting ‘secret documents’ were by far the most burning issue for the great powers, which had no problem in trading higher service levels for security guarantees. In this process, the concerns of the Commission were of little relevance: ‘They [the Council representatives] didn’t give two hoots if the servants in the Commission got problems’ (interview). The European Parliament also had to give way as regards its wish to achieve a more open decision-making process, allowing the Commission the ‘space to think’ and the Council the ‘process confidentiality’ they demanded. The inclusion of the harms’ test made this concession easier. All in all, both the Council and the Parliament had to make substantial concessions, although the European Parliament is often claimed to have moved furthest from its previous standpoints. ‘If you compare their [the EP] proposals from November with the final result you will see who backed down’ (interview). The Council minority, including the Swedish Presidency, had reasons to be very satisfied with the result. Its major concerns were all met, although not always to the extent that it had aspired to. The Presidency had been able to use its mediator

58 Bo Bjurulf and Ole Elgström position, and the possibilities created by the co-decision procedure and the trialogue, to construct a package which it itself could accept with satisfaction. The one disappointment was the lofty, and contradictory, formulations regarding the applicability of the Regulation in the member states. Without the support of the Parliament, Sweden did not succeed in guaranteeing a continued acceptance of its traditional principles.

The impact of institutional arrangements In this section the importance of institutional arrangements is summarized by scrutinizing each of the institutions that we initially ascertained would be our main foci of attention. Agenda-shaping rules In theory, the institutionally-based formal power of initiative confers substantial influence to the Commission. Its documents normally function as single negotiating texts that often survive Council discussions almost unscathed. However, the Commission ceded its powers in not insisting on its own proposal, leaving the initiative and the control of the text to the other institutional player with formal agenda-shaping power, the Presidency. Other actors’ perceived opportunities were clearly shaped and constrained by Presidency initiatives. Because the rules privilege Presidency proposals, other state actors were mainly assigned reactive roles. Not least during the final stages of the bargaining process, other member states limited their activities to responding to Presidency proposals. Decision-making procedures and voting rules The fact that the Amsterdam Treaty established co-decision as the given procedure in deciding upon transparency rules turned out to be of major significance. Co-decision obliged the Council to find a joint solution with the Parliament. As we have demonstrated, this created opportunities for the openness-oriented Presidency to manipulate the process, pitching Council interests against parliamentary concerns and advocating a compromise that in fact was close to its own preferences. The emerging trialogue mechanism further strengthened the position of the Presidency by providing leeway for tactical manoeuvring. In the end, this formed the provisions of the final agreement. As regards voting rules, a vote in the Council was not on the agenda during the French Presidency, as there was no point in forcing through a majority common position in the Council when it still had to deal with the Parliament afterwards. The shadow of the vote did, however, arguably play a role in the endgame: when the Swedish Presidency had persuaded the less engaged member states to rally around its take-it-or-leave-it compromise proposal, thus creating a majority behind it, the remaining opponents faced the unpleasant choice of either sticking to their principles and being defeated in a vote or joining a consensus on a non-palatable but minimally acceptable solution. They chose the latter alternative — and thus ended up with a solution they would not have accepted under unanimity.

Negotiating transparency: the role of institutions

59

Informal norms Strong norms constrained and created opportunities for actor behaviour. The norm stating that Presidencies are expected to act as mediators (cf. Schmidt 2001) empowered the Swedish Presidency and gave it an institutional ground to stand on when delivering its compromise proposals. At the same time, the Presidency is also subject to the impartiality norm; it should be a broker, but an honest broker. Faced with the potential costs of being accused of breaking the impartiality norm, the Swedish Presidency used considerable ingenuity to avoid such reactions. It did not succeed entirely; in fact, Swedish activities during the transparency negotiations evoked considerable criticism. The constraining force of the norm is still evidenced by the efforts made to keep the appearance of being impartial. The consensus norm in the Council supported broad understandings and, in the end, encouraged the recalcitrant minority to accept the proposed Presidency compromise. During the process several member states offered unilateral concessions in order to reach a broad agreement. The intensity of such efforts was probably diminished by the knowledge of the Parliament’s rather radical negotiation position: why use energy to bring about a Council compromise if it was to be unravelled in conciliation? Timetables and deadlines The given sequence of Presidencies influenced the decisions taken by the Parliament to a) enter into an early trialogue, which it was otherwise unwilling to do, and b) approve the many concessions it had to make in the final round of negotiations. Knowing that transparency-reluctant Belgium and Spain stood in line after Sweden entered into the cost–benefit analysis of the European Parliament, helped to tip the balance in favour of acceptance. Furthermore, the existence of a formal deadline, set by a European Council, created a formidable pressure for compromise that certainly helped to bring about the lastminute agreement. Interventions by other institutional actors In the highly legalized EU system, decisions by the European Court of Justice and other courts have an enormous influence. In the case under discussion, certain decision alternatives were effectively ruled out by decisions taken, or to be taken, by EU courts. Such decisions created faits accomplis that formed the basis for future negotiations and in these ways constrained actor decision making. We have also demonstrated how the initiative of a ‘new’ institutional actor, the High Representative, unexpectedly altered the parameters of the bargaining process and influenced the decision-making calculi of the actors.

Conclusion: institutions as structuring elements of EU negotiations This chapter has filled a gap in research on EU negotiations by highlighting the role of institutions. We have provided empirical evidence that institutional factors

60 Bo Bjurulf and Ole Elgström do indeed have considerable impact upon EU negotiations. We have demonstrated how institutional rules and norms affected the transparency negotiation processes and how the outcome of the negotiations can be explained by reference to particular institutional arrangements. Had other arrangements prevailed, the outcome would have been considerably different. Consider, first, an environment where decisions were taken by unanimity and where supranational institutions were irrelevant, as implicit in an intergovernmental approach to international negotiations. In such a case, the defenders of the status quo would have a definite advantage. They would have permitted increased transparency only in areas where this was in their perceived interest. The end result would have been a lowest common-denominator solution. Consider, secondly, a situation with consultation as the decision-making mechanism instead of co-decision, the dominant situation in the 1980s. The European Parliament would in such a scenario have had little power, and the Presidency would not have been able to use its position with more radical actors on both sides. The outcome would have been decided in negotiations between the member states, where the secrecy-oriented majority in the Council – which in most conflict dimensions included the major EU powers – would have opposed and prevented any kind of radical change. In both cases, the outcome would have been similar: a solution biased in favour of the member states with the greatest bargaining power. Efforts to understand EU negotiation processes without taking institutions into account are likely to go astray. By carrying out an in-depth case study of an actual negotiation process, we have highlighted certain significant institutional features that structure EU negotiation processes and have tried to illuminate the dynamics of such processes. We have, inter alia, demonstrated how the introduction of conciliation committees and trialogues may influence the institutional balance of power. Our case also showed how an informal norm, such as Presidency impartiality, constrained and shaped actor behaviour, but to various degrees depending on which member state held the office. It provides additional input into the debate on the agenda-shaping role of the Commission and illustrates the importance of the Presidency both as initiator and as broker. A focus on institutional constraints and empowerment is, we submit, a fruitful approach to understanding the sometimes puzzling patterns and manifestations of concrete EU negotiation processes.

References Aspinwall, M. D. and Schneider, G. (2000) ‘Same Menu, Separate Tables: The Institutionalist Turn in Political Science and the Study of European Integration’, European Journal of Political Research, 38: 1–36. Beach, D. (2002) Bringing Negotiations Back into the Study of European Integration, University of Southern Denmark: Department of Political Science (dissertation). Birkinshaw, P. (1996) Freedom of Information: The Law, the Practice and the Ideal, London: Butterworth. Bulmer, S. J. (1994) ‘The Governance of the European Union: A New Institutionalist Approach’, Journal of Public Policy, 13 (4): 351–80.

Negotiating transparency: the role of institutions

61

Commission (2000) Proposal for a Regulation of the European Parliament and of the Council regarding public access to European Parliament, Council and Commission documents. COM(2000)30 final/2, 21 February. Council (2000a) Note from the Presidency to Working Party on Information. SN 2970/3/ 00, REV3, 18 August. Found on Statewatch homepage: http://www.statewatch.org/ secreteurope.html Council (2000b) Presidency Note to delegations. 14938/00, 22 December. Found on Statewatch homepage: http://www.statewatch.org/secreteurope.html Court of first instance, Case T–124/96, Interporc Im- und Export GmbH against the Commission. http://www.curia.eu.int/jurisp/ Court of first instance; Case T–14/98, Heidi Hautala against the Council. http:// www.curia.eu.int/jurisp/ Crombez, C. (2000) ‘Codecision: Towards a Bicameral European Union’, European Union Politics, 1 (3): 363–8. Eising, R. (2002) ‘Policy Learning in Embedded Negotiations: Explaining EU Electricity Liberalization’, International Organization, 56 (1): 85–120. Elgström, O. (ed.) (2003a) ‘Introduction’, in Comparing European Union Council Presidencies, London: Routledge. Elgström, O. (ed.) (2003b) ‘“The Honest Broker”? The Council Presidency as a Mediator’, in Comparing European Union Council Presidencies, London: Routledge. Elgström, O. and Smith, M. (eds) (2000) ‘Negotiation and Policy-Making in the European Union’, special issue of the Journal of European Public Policy, 7 (5). EP and Council (2001) Regulation of the European Parliament and of the Council regarding public access to European Parliament, Council and Commission documents. No. 1049/2001, 30 May. OJ L 145/43, 31 May. Gulliver, P. H. (1979) Disputes and Negotiations: A Cross-Cultural Perspective, New York: Academic Press. Habeeb, W. M. (1988) Power and Tactics in International Negotiations, Baltimore, MD and London: Johns Hopkins University Press. Hall, P. A. and Taylor, R. C. R. (1996) ‘Political Science and the Three New Institutionalisms’, Political Studies, XLIV: 936–57. High Representative (2000) Decision by the General Secretary/High Representative of the CFSP, EGT 2000 C239/1, 27 July. Hix, S. (1999) The Political System of the European Union, Houndmills and London: Macmillan. Jönsson, C., Bjurulf, B., Elgström, O., Sannerstedt, A. and Strömvik, M. (1998) ‘Negotiations in Networks in the European Union’, International Negotiation, 3: 319–44. March, J. G. and Olsen, J. P. (1984) ‘The New Institutionalism: Organizational Factors in Political Life’, American Political Science Review, 78 (3): 734–49. Moravcsik, A. (1993) ‘Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach’, Journal of Common Market Studies, 31 (4): 473–524. Norgaard, A. (1996) ‘Rediscovering Reasonable Rationality in Institutional Analysis’, European Journal of Political Research, 29: 31–57. Peters, B. G. (1999) Institutional Theory in Political Science, London and New York: Continuum. Pollack, M. (1997) ‘Delegation, Agency, and Agenda Setting in the European Community’, International Organization, 51 (1): 99–134. Raiffa, H. (1982) The Art and Science of Negotiation, Cambridge, MA and London: Harvard University Press. Schmidt, S. K. (2000) ‘Only an Agenda-Setter? The European Commission’s Power over the Council of Ministers’, European Union Politics, 1 (1): 37–61.

62 Bo Bjurulf and Ole Elgström Schmidt, S. K. (2001) ‘A Constrained Commission: Informal Practices of Agenda-Setting in the Council’, in G. Schneider and M. Aspinwell (eds) The Rules of Integration: Institutionalist Approaches to the Study of Europe, Manchester and New York: Manchester University Press. Shackleton, M. (2001) ‘Codecision Since Amsterdam: A Laboratory for Institutional Innovation and Change’, paper presented at the ECSA Seventh Biennial International Conference, Madison, WI, 31 May–2 June 2001. Shepsle, K. A. (1989) ‘Studying Institutions: Some Lessons from the Rational Choice Approach’, Journal of Theoretical Politics, 1 (2): 131–47. Statewatch (2001) ‘European Parliament Votes for “Deal” with Council’, Statewatch, 11 (2): 21–3. Stenelo, L.-G. (2000) ‘The Bargaining Democrat’, in L. Karvonen and K. Ståhlberg (eds) Festschrift for Dag Anckar on his 60th Birthday on February 12, 2000, Åbo: Åbo Akdemi University Press. Steunenberg, B. (2000) ‘Seeing What You Want to See: The Limits of Current Modelling on the European Union’, European Union Politics, 1 (3): 368–73. Swedish EU Committee (2001) Protokoll från EU-nämnden, 2000/01: 22, 15 March. http:// www.riksdagen.se/debatt/0001/eunprot/22/22S00004.ASP Swedish EU Committee (2001) Speech by Minister Britta Lejon, Minutes from the Swedish EU Committee 2000/01: 22, 15 March 2001. Tallberg, J. (1999) Making States Comply: The European Commission, the European Court of Justice and the Enforcement of the Internal Market, Lund: Department of Political Studies. Tsebelis, G. and Garrett, G. (2000) ‘Legislative Politics in the in the European Union’, European Union Politics, 1: 9–36.

Interviews Hans Brunmayr, Council Secretariat, 8 February, 2001 and 18 June 2001 Lars Danielsson, Swedish State Secretary, Prime Minister’s office, 21 August 2001 Heidi Hautala, MEP, Rapporteur, 19 June 2001 Pascal Hector, German Permanent Representation to the EU, 3 July 2001 Claas Knoop, German Permanent Representation to the EU, 3 July 2001 Joost Korte, Commission General Secretariat, 8 July 2001 Pedro Lourtie, Portugese Permanent Representation to the EU, 7 July 2001 Gunnar Lund, Swedish Permanent Representation to the EU, 4 July 2001 Hanja Maij-Weggen, MEP, rapporteur, 6 December 2000 Cecilia Malmström, MEP, rapporteur, 6 December 2000 Jan R. Olsson, Swedish Permanent Representation to the EU, 2 July 2001 John Palmer, European Policy Centre, 2 July 2001 Mary Preston, Commission, 18 June 2001 Michael Shackleton, European Parliament, 2 July 2001 Astrid Thors, MEP, rapporteur, 19 June 2001

5

Reform negotiations The case of the CAP Malena Rosén and Magnus Jerneck

Introduction This is a study of reform negotiations of the Common Agricultural Policy (CAP). The CAP is a policy area well known for its reluctance to change. Ever since its inception as a fully common policy in 1968, with a strong legal base, it has been subject to repeated reform attempts (Roederer-Rynning 2002, 2003). The CAP could thus be assumed to demonstrate clearly the difficulties of reforming the European Union’s negotiated order (cf. Elgström and Smith 2000; Green Cowles and Smith 2000). The case under investigation is the Agenda 2000 negotiations (1997–9), which aimed to prepare the European Union for eastern enlargement and a new WTO Round (cf. Schwaag Serger 2001). The objectives of the CAP were set out in the Rome Treaty (Article 39). These objectives were clearly marked by the state of agriculture in the immediate postwar Europe, where food shortages were a reality for many Europeans. Securing the supply of agricultural products therefore became the main priority, which would be achieved through artificially high prices within the Common Market in order to stimulate production. In order to secure farmers’ incomes, the Community also committed itself to buying surplus production, which could not be sold at the prices set. However, the development of modern agricultural technology, in combination with the stimulation for increased production generated by the high prices, soon reversed the situation for agricultural products: too much, instead of not enough, was produced in the Community. This was already the situation when the CAP became a fully common policy in 1968. The Agriculture Commissioner at the time, Mansholt, launched the first reform proposal of the CAP that same year. In 1972 a watered-down version of his proposal was put into effect. However, the CAP remained essentially unreformed (Ingersent et al. 1998: 2). This was a pattern which would repeat itself in the years to come. One solution to get rid of the mounting surpluses was to subsidize exports, selling the products cheaply outside the Common Market. But with growing surpluses this solution became extremely costly, and with increasing trade liberalization it also became a recurrent issue of dispute between the Community and other actors on the world market. Pressures for change, from within the Community as well as from trade partners externally, increased during the 1980s and 1990s.

64 Malena Rosén and Magnus Jerneck A long series of reform attempts followed, none of which brought about fundamental change in the policy. The most wide-ranging reform thus far was the MacSharry reform in 1992, the central elements of which were phased reductions in support prices, mainly for cereals, supplemented with direct aid payments and with support partly decoupled from production (Grant 1997: 77–8; Ingersent et al. 1998: 3–4). By substituting direct payments for support prices as the principal instrument, the agricultural support became more transparent – something farmers’ organizations and member states in favour of the status quo feared would increase the demands for reform among Europe’s consumers and member states who considered the policy area to be far too expensive.

EU reform negotiations Our point of departure is that EU reform negotiations possess at least two specific characteristics, which by their combination and extent differentiate these negotiations from negotiations on both national and ‘purely international’ levels in a reform perspective: the frequency of inert supranational compromise and the extensive use of linkages (package deals and issue linkages). Stenelo (1991: 26) argues that supranational decision making stimulates the propensity to compromise, and that these supranational compromises are more complex than their national counterparts. The supranational compromise is preceded by compromises reached at lower levels, in different political cultures, and this makes it inflexible or inert. The inertia is self-reinforcing in that the parties are aware of the difficulties in ‘simultaneously making mutual retreats from common positions’ (Stenelo 1991: 230–1, our translation). As new compromises are built upon previous ones and linked to each other, further compromises are becoming increasingly difficult to reach. The room for compromises decreases, since every new compromise presumes the breaking up of an old one, which is usually associated with considerable political cost since more than one level is involved (Stenelo 1991: 232, 238). Inertia also brings with it positive consequences. It effectively hinders hastily agreed decisions of an ad hoc and short-time character. Supranational compromises can give rise to a supranational consensus culture, in which national interests are merged into common interests. In stable supranational decision-making structures this can contribute to the generation of trust among the parties, which in turn increases the propensity for diffuse reciprocity in the negotiations (Stenelo 1991: 229–30). Linkages between different compromises are common in EU decision-making processes and characterize ‘the Community method’ of decision making. The higher the political level, the more frequent is the use of issue linkages and package deals (Spence 1995: 385). The purpose of linking is to enhance the prospects of creating win–win solutions, thereby increasing the propensity for integrative solution. Negotiations are influenced by type of policy. Lowi’s fourfold classification of policies into constituent, redistributive, distributive and regulatory types (1972)

Reform negotiations: the case of the CAP 65 has been applied to the European Union in attempts to distinguish between different clusters of policy making (Wallace 1996: 446; Elgström and Jönsson 2000). Constituent policies are about the rules and priorities of the system itself; redistributive policies encompass the transfer of financial resources from some actors to others; distributive policies are about allocating community funds within sectors; and regulatory policies refer to the adoption of common regulations of the activities of public and private sectors among the member states. In short, constituent and redistributive policies seem to be most prone to difficult negotiations and are therefore likely to be most resistant to reform. In the process of preparing compromises and linkages, coalition building, leadership and mediation play important roles. A coalition is defined ‘as a set of actors that coordinate their behaviour in order to reach goals they have agreed upon’ (Elgström et al. 2001). With two or more coalitions negotiating over multiple issues, there is increased likelihood of divisions within the parties. The division of a party may shift with the issue of negotiation, so that an actor finds its opponent on one issue to be its ally on another. Compared to reinforcing cleavages, which usually prevail in bilateral negotiations and which are believed to result in polarized positions, cross-cutting cleavages increase the prospects of package deals, which further promotes the likelihood of integrative, win–win solutions (Hopmann 1996: 254–5). The reform-minded actors in the context of the European Union are probably favoured by cross-cutting cleavages, since internal divisions increase the propensity for package deals and, probably, also for agreement on reform. With ‘unitary’ actors, however, the positions are more cemented. A few actors defending the status quo suffice to hinder reform. Leadership and mediation are central in the working out of agreements – or compromises – between different coalitions. A leader is an actor actively driving the negotiations forward, concerned primarily with collective goals rather than pure self-interest (Malnes 1995), while a mediator is an actor who intervenes when a conflict has occurred, in order to secure the reaching of an agreement (cf. Stenelo 1972: 37). A leader or mediator performing its task well will clearly enhance the prospects of agreement on reform. In the European Union, the Commission and the Presidency are the actors most often assigned such roles. Finally, we assume that some contextual factors will facilitate reform negotiations, while others will render them more difficult – i.e. that there will be different types of reform negotiations. Two contextual factors will be discussed from a reform perspective: the type of networks at hand and political pressures (cf. Elgström and Jönsson 2000; Peterson 1995; Wallace 1996).

Agenda 2000: pressure for reform At the summit of the European Council in Madrid in December 1995, the Commission was asked to present, as soon as possible, a document on the challenges that the Union would be likely to face in the immediate future, with emphasis on the preparations for eastern enlargement. This document was to contain an

66 Malena Rosén and Magnus Jerneck evaluation of each applicant’s readiness for EU membership, as well as a detailed analysis of the finances of the European Union in the first years of the new millennium (2000–6) and assessments of the implications of enlargement for major policy areas. The Commission presented its first Agenda 2000 proposal in July 1997. Here, the Commission outlined ‘the broad perspectives for the development of the Union and its policies beyond the turn of the century, the horizontal issues related to enlargement, and the future financial framework beyond 2000 taking account of the prospect of an enlarged Union’ (Commission 1997: 3). The need for strong finances put pressure on reform of the two most costly policy areas in the EU budget – the Common Agricultural Policy and the Structural Funds. Especially the CAP, the most expensive of the two, could arguably ‘pose the greatest political, economic, budgetary and administrative obstacles to the accession process’, and was seen to be in urgent need of reform (Centre for European Policy Studies 1999: i). There was also an external dimension to the need for reforming the CAP: to fulfil the Uruguay Round Agreement and to prepare the Union for the next WTO round (Commission 1997: 28–9; cf. Schwaag Serger 2001). The negotiation process: pivotal events The Agenda 2000 process stretched over almost two years. It was initiated with the Commission’s proposal in July 1997 and ended with the negotiations in the Agriculture Council and the final decision by the European Council in March 1999. Initial positions: responses to the Commission’s proposal, July 1997 The issues in the first proposal which became most important in the negotiation process included substantial price cuts coupled with direct income aid in the beef and cereals sectors, and to a lesser extent in the milk sector; decentralization of policy implementation, with more margin being left to member states and regions; and the unsettled question of whether or not to make moderate reforms in the milk sector (Commission 1997). In statements preceding the official presentation, Agriculture Commissioner Fischler made it clear that the Commission would propose quite substantial price cuts in the beef and crop sectors, bringing the prices closer to world market levels. This would be done in the spirit of the MacSharry reform, for example by compensation in the form of direct payments. In both these sectors, large and costly surpluses were expected in the immediate future, and the need for reform was accepted by most member states, even if their views differed on the acceptable levels of the price cuts and the compensation (Agra Europe, 6 June 1997). The third major sector of the CAP, the dairy sector, was a controversial issue within the Commission even before the launch of the proposal. While most Commissioners were keen to begin reform of this essentially unreformed sector of

Reform negotiations: the case of the CAP 67 the CAP, Fischler deemed it less urgent. He argued that less costly surpluses were expected in this sector, but his hesitation to propose a reform of the dairy sector was probably also influenced by the fact that there was considerable disagreement over the sector among the member states, especially regarding the continuation of the milk quota system. A reform proposal would likely give rise to difficult negotiations (Agra Europe, 6 June 1997). In the eventual proposal, the Commission suggested price cuts along with direct payments, while the issue of a more thorough reform of the sector remained open. From the outset, the issue of milk reform was one of the major bones of contention in the negotiations. Germany and Austria, both with large domestic interests in keeping milk quotas, were spearheading the campaign for maintaining the status quo in the Agriculture Council. There was, however, internal disagreement on the issue, mainly in Germany. In the short term, a radical reform would be more costly than preserving the status quo, since compensations would have to be paid during a transition period. However, in a long-term perspective a reform would save money for net contributors, such as Germany and Austria. The German Ministry of Finance therefore viewed favourably a radical reform of the dairy sector. France strongly opposed any removal of the quota system. Unlike Germany and Austria, France would profit from saving the quotas, as a net beneficiary of the CAP and only a minor overall net contributor. The United Kingdom, Sweden and Denmark argued in favour of a total removal of the quota system at its expiration in April 2000. These three member states – the first two net contributors to the EU budget – took the most reform-minded position in the Agriculture Council. The reform-minded group sought close cooperation with the Commission, trying to influence it to propose major reform. Overall, although most member states were in favour of keeping the quota system, at the outset there was no clear majority for any of the discussed options, which in addition to the two options already presented also included two-tier pricing and an increase in quotas. However, the French used the linkage between milk reform and the EU budget in a strategic way in the negotiations. A radical reform of the milk sector would initially lead to increased costs. France suggested early on in the negotiations that the member states should agree upon a ceiling on CAP expenditure, which the net contributors favoured. But by agreeing to this, they also contributed to reducing the scope for more radical reform – which was in the interest of France. Middle game: detailed CAP proposal in March 1998 The detailed CAP proposal presented by the Commission in March 1998 ‘marked the beginning of the real negotiations’. With this it became clearer what the Commission really wanted. Compared to the first proposal, the price cuts remained about the same for cereals and beef, albeit with lower compensations. In the milk sector, the Commission proposed a larger price cut than previously (15 per cent instead of 10 per cent) and a two per cent increase in quotas, of which half would go to mountainous regions and the other half to new entrants to farming. The

68 Malena Rosén and Magnus Jerneck Commission avoided setting a deadline for the quota system; instead, it confined itself to stating that the current system would not be sustainable in a long-term perspective. The issue of a new division of functions between Brussels and the member states remained on the agenda, with the Commission proposing that member states be allowed more flexibility in allocating aid payments (Commission 1998). These new – or more detailed – proposals were, according to a majority of the Agriculture Ministers, ‘even less acceptable’ than the previous proposals (Agra Europe, 3 April 1998). The suggested allocation of the increased milk quotas received vociferous opposition from several member states. Italy, dissatisfied with what it considered as too low a level for its quota and under pressure from its farmers, who had been staging massive protests against the quota regime, aligned with the reform coalition on the issue, and demanded an end to the quota system. Italy’s reasoning was clearly that rather than allow the ‘injustice’ in regard to its share of the quotas to continue, it would prefer the system to be brought to an end. With Italy joining the three member states opposed to the quota regime, this group now constituted a blocking minority in the Council. The London Group, as these four member states became known, was favourably viewed by the Commission, since its stance increased the scope for talks on reform – and thereby also the role of the Commission in the negotiations. In the summer, the London Group presented a common reform proposal to the Council and the Special Committee on Agriculture (SCA).1 The proposal suggested a 30 per cent price cut and phased increases in milk quotas, eventually leading to the abolition of the quota regime. France continued to oppose reform of the milk sector as well as of most other sectors. After bilateral meetings between Prime Ministers and Agriculture Ministers, France united with Ireland in an overall protest against the CAP proposals. France also joined an alliance with Germany on the issue of milk quotas. Germany, however, continued to be split over several of the proposals. Germany’s ambivalence was not only a result of splits within the Kohl government – between the Ministries of Finance and Agriculture – but was also affected by the upcoming election in September. A spokeswoman from the Social Democratic Party stated that if it won the election, the SPD would support radical reform of the CAP. Despite the United Kingdom’s rather efficient Presidency in regard to the CAP negotiations, which made progress on several dossiers, the uncertain state of Germany’s position, depending on the eventual outcome of its election, led to a standstill in the negotiations that no actor was able to break. In May, Agra Europe reported that: there is growing recognition among [agricultural] ministers on the need to negotiate a rapid conclusion to CAP reform in tandem with the rest of the EU’s (non-agricultural) Agenda 2000 package, for which the Europe’s foreign ministers have overall responsibility. Without agreement on CAP reform, farm ministers recognise that they run the risk of being ‘sidelined’. (Agra Europe, 29 May 1998)

Reform negotiations: the case of the CAP 69 However, not even the risk of losing influence to the Foreign Ministers pushed the Agriculture Council, now led by Austria, towards any substantial progress during the second half of 1998. Instead, a discussion document, issued by the Commission in October, contributed to the deadlock even after the German election, which was won by the SPD. The document discussed a series of options for changing the funding of the European Union, intended to please the net contributors. The Commission suggested that up to 25 per cent of direct aid payments under the CAP be paid directly from national budgets rather than the EU budget. As stated by Agra Europe, ‘it seems unlikely that member states will show a great deal of willingness to agree to CAP reforms while they remain unsure about how the policy will be financed’ (Agra Europe, 9 October 1998). France, already reluctant to agree on reform of the CAP, declared it would oppose any measure of co-financing. End game: the final decisions in March 1999 The Farm Ministers’ proposals were less ambitious than the Commission’s because they phased the price cuts. The Agenda 2000 package that was finally approved by the European Council was even more watered down, calling for smaller cuts in support prices and delaying the implementation of dairy reform (USDA 1999: 15). The Agriculture Council was scheduled to hold its last negotiating Council in February in order to resolve all details before the summit in March. However, the February Council stalled on the issue of re-balancing EU funding. Germany, Austria, Sweden and the Netherlands had all declared their support for the Commission’s proposal on co-financing, and most other member states accepted it – France constituting the exception. With the question of co-financing unresolved, France refused to sign a package of agricultural reforms until a financial framework for an overall EU budget was settled by the heads of government (Agra Europe, 26 February 1999). However, France accepted that ‘the expenditures for agriculture were increasing too much with the reform proposals by the Commission’ (Ferrer and Emerson 2000: 8–9). Therefore, it proposed lesser price cuts, abandonment of milk reform and degressivity, meaning that payments would be cut over time in order to bring the costs down. An impasse in the Franco–German relation followed France’s proposal, which Germany regarded as less favourable than co-financing. The frictions between the two member states increased when France, in a press statement, accused Germany, now holding the Presidency, of acting in the national interest and not as a neutral Presidency. The Franco–German impasse was not broken until the beginning of March, after bilateral negotiations which resulted in ‘a highly diluted proposal’ and demonstrated that Germany ‘did not regard substantial agricultural reforms as necessary’ (Ferrer and Emerson 2000: 9). The London Group contested the joint Franco–German proposal, and the negotiations stalled once again. In an attempt to break the deadlock, the German Presidency made a final move and proposed a package which was similar to the original Commission proposal.

70 Malena Rosén and Magnus Jerneck The German proposal, however, reduced overall expenditure for the period until 2006 by delaying the introduction of milk reform and introducing transition periods for the remaining reforms (Ferrer and Emerson 2000: 9). The Agriculture Council finished its negotiations on 11 March, after a marathon stop–start Council which lasted two and a half weeks and resulted in a non-voted ‘agreement’ (Agra Europe, 12 March 1999). The Agriculture Council de facto left the situation unsolved and open to change at the Berlin summit. The Presidency’s claim that an agreement was reached was clearly not shared by France and some other member states, which hinted that all was open for negotiation in Berlin (Ferrer and Emerson 2000: 9). The French view was well illustrated by its Agriculture Minister, who stated: ‘I can’t consider that a decision has been taken because there was no vote.’ Fischler, on the other hand, claimed that ‘radical reform’ had been agreed in the Agriculture Council (Agra Europe, 12 March 1999). At the European Council meeting Germany hoped to exercise influence by putting the CAP as the first item on the agenda to be solved within the Agenda 2000 package. At the summit the frictions between France and Germany were once again evident. With Schroeder, Germany had become somewhat more open to reform, while the French President Chirac, former Agriculture Minister and dependent on farmers’ votes for his re-election as President, still wished a minimum of reform to be adopted. In the end, Chirac’s knowledge of the policy area and his strong-arm tactics: forced his colleagues to lever open the CAP consensus reached on March 11, at which point the German Presidency worked to accommodate his demands in order to avoid the political embarrassment of negotiations collapsing. The total package of Agenda 2000 needed unanimous support from all delegations. (Agra Europe, 26 March 1999) It was clear that bringing the CAP back into line would probably involve some substantive reopening of the compromise. Agricultural reforms were therefore cut down considerably at the last minute, as a result of French insistence. The final decision, which included lesser price cuts than those agreed by the Agriculture Ministers and left the question of degressivity to future negotiations, was facilitated by the breaking up of the London Group. From the outset there had been a great risk of ‘prisoners’ dilemma’ in the coalition, implying that member state(s) leaving the coalition could make individual gains. Italy, which had been campaigning for an increase of its quota for quite some time, was finally bought out by Germany and France. The Agriculture Ministers’ decision on initiating reform of the milk sector in 2003, and possibly ending the quota system, was replaced by the European Council’s decision to delay dairy reform until 2006 (Europeiska rådet 1999: 8). However, as a result of the hard bargaining by the London Group, it was decided that an overview of the quota system would be conducted in 2003. After all, the group had clearly had an impact on the negotiations, and it had

Reform negotiations: the case of the CAP 71 demonstrated more internal coherence than those opposed to its radical proposal had thought possible at the start. With the pronounced aim of reforming the CAP in order to prepare the policy for the integration of new members and for the Union to be able to comply with the Uruguay Round Agreement and enter a new trade round – likely to demand increased liberalization in agricultural trade – the outcome of the reform negotiations must be considered unsuccessful.2 In the terminology of Hall (1993) and Daugbjerg (1999), the reform can be characterized as a second-order change: whereas objectives, instruments and their settings were altered, the basic policy paradigm remained the same.

Supranational compromises and linkages The CAP/Agenda 2000 negotiations demonstrated that supranational compromises and linkages – and their combination – had a considerable impact on the negotiation process and on the possibilities to reach an agreement on major reform of the policy area. This seems to be true for previous reform negotiations within the CAP as well. The supranational compromises, on which the CAP is built, are likely to be very complex and multi-layered since the CAP is the oldest common policy area in the European Union. When new compromises are built upon previous ones and linked to each other, further compromises become increasingly difficult to reach. Breaking up previous compromises is associated with considerable political costs. The CAP is also a policy area characterized by strong national interests; this increases these costs as well as the difficulties of reaching agreement between member states that have to take their domestic political situations into careful consideration. In the negotiations that we have analyzed, the national interest was, as many times before, most vigorously defended by France. The bilateral relation between France and Germany, expressed as a necessity of a Franco–German compromise, has traditionally been considered among the factors most decisive to the outcome of CAP negotiations. Webber (1998: 1), who has analyzed agricultural crisis politics in the European Union in the late 1960s and early 1990s, argues that: where both governments opposed a given project, it was blocked; where both supported it, it was approved; where the two governments were divided, the outcome – until such time as they found a modus vivendi – was crisis and deadlock. The tensions between these two member states, mainly during the German Presidency, add weight to Webber’s argument. The negotiations remained stalled until the two had reached a bilateral compromise. At the same time, other member states – mainly the reform coalition – exercised great influence on the negotiations. Nevertheless, in all cases, domestic considerations seem to have been the dominant influence on the member states’ actions, which made a new supranational compromise hard to agree on.

72 Malena Rosén and Magnus Jerneck The Agenda 2000 negotiations did not demonstrate much evidence of a supranational consensus culture, one of the potential positive consequences of the supranational compromise. Many accounts of the CAP characterize the Agriculture Council as ‘club-like’, in that the ministers often regard each other as allies and their colleagues from the Finance Ministries as the opponents. This is partly true, and more so for some of the member states (e.g. Germany). Still, national interests prevail over any major common interest, or supranational consensus, in the policy area. The reasons are probably twofold: first, the CAP is a redistributive, and to a certain extent also distributive, policy, which accounts for more than 50 per cent of the EU budget, implying that large sums of money are at stake for the member states, either as beneficiaries or contributors. The negotiations within Agenda 2000 had a clear redistributive character, with the possibilities of reaching a new supranational compromise at the summit accordingly limited. Second, despite being common in name, the CAP has been based on certain measures, such as the use of green money,3 which has contributed to giving it a national rather than a supranational character. The prevalence of national interests was demonstrated in the stand taken by the Agriculture Council, which – despite the ministers’ awareness that they risked losing influence to the Foreign Ministers who were negotiating on the rest of the Agenda 2000 issues – postponed an agreement in their policy area. The negotiations involved extensive linking within the sector. However, previous linking reduced the scope for integrative solutions, serving to restrict new creative linking. In regard to cross-sector linking, or package deals, the CAP often exerts more influence on other policy areas than vice versa. The Berlin summit, which spent most of its time dealing with this policy area, was able to solve other contested issues only after an agreement had been reached on the CAP (Ferrer and Emerson 2000: 9). But the fact that no clear decisions had been agreed in the Agriculture Council left the CAP more open to negotiations between the heads of governments at the summit, as part of the package they had to agree upon. Thus the CAP became more intensively linked to other policies than would have been the case if the Agriculture Ministers had been able to make all ‘final’ decisions. This will, of course, in its turn serve as a hindering factor in future reform negotiations.

Coalitions, leadership and mediation Coalitions building on cross-cutting cleavages across member states and institutions – most importantly managed by the Commission in the beginning of the process – as well as coalitions based on ‘unitary’ actors were present during the negotiation process. The loose coalition campaigning for status quo, with France, Germany and Ireland at its core, represented both types: France and Ireland being ‘unitary’ actors and Germany plagued by internal divisions. The reform group – with the third large member state, the United Kingdom – had both sector-specific and overall budgetary interests in a major reform of the CAP. The members of this coalition were thus fairly coherent internally, and they sought close cooperation

Reform negotiations: the case of the CAP 73 with the Commission, hoping to influence it to propose major reform. The Mediterranean member states formed a third large coalition, which has not been mentioned here since its main interests related to sectors that figured marginally in the negotiations (with exception of the milk sector, where most of these member states wanted an increase of the quota). In the status quo coalition, there were clearly problems in the Franco–German alliance, which was weaker than at any time previously, even though this alliance has always been based on conflictual cooperation (Webber 1998: 35). The problems became acute during the German Presidency. Still, they were insufficient to break up the alliance this time, which was probably closely related to the fact that Germany needed agreement with France in order not to risk the final agreement of Agenda 2000 – and the political embarrassment that would follow for the Presidency. The negotiation process lacked a clear leader. In several other EU policy areas, France and Germany are often described as being in the driver’s seat, moving integration forwards. In the case of the CAP, however, neither one has acted as a constructive leader; rather as laggards because of their domestic political considerations. Germany had an important opportunity to demonstrate leadership or mediating skills, holding the Presidency during the endgame of the process. But the new German government was mainly concerned with national interests at the summit, which restricted the scope for possible reforms. By giving clear priority to its own national interests, the German Presidency probably also lost other member states’ confidence in its role as leader and/or mediator. The Commission was the main mediator in the process. The London Group, by taking an extreme position compared to the rest of the member states, increased the scope for talks on reform and accordingly also the role of the Commission in the negotiations. It is interesting to note that the Commission was considered as the main mediator in the process, rather than any of the Presidencies. However, the Commission’s internal divisions on the CAP, particularly prior to the launch of the first proposal, and the strong national interests within the policy area put limitations on its potential mediating role.

Networks Policy networks are defined as more or less stable sets of public and private organizational actors, linked to each other by communication and by the exchange of resources, such as information and expertise (Jönsson et al. 1998: 326). They constitute informal structures which emerge within and as complements to formal organisations, especially when these are complex. Not all policy networks are alike – rather, they demonstrate differences along several dimensions. Policy networks with stable membership, high insularity and strong dependencies among their members – termed policy communities – are found at one end of a continuum, with policy networks characterized by fluid membership, high permeability and weak dependencies among their members, socalled issue networks, at the other end (Peterson 1995: 77).

74 Malena Rosén and Magnus Jerneck Daugbjerg has analyzed the impact of network characteristics on the possibilities to reform a policy area, using Swedish and EU agricultural policies as examples (Daugbjerg 1997, 1999). In line with other studies on policy networks (e.g. Peterson 1995), Daugbjerg argues that policy communities demonstrate a large degree of consensus within the network and usually possess considerable power over the decision-making process within a given policy sector. Members of a policy community are likely to favour the status quo, for two reasons: first, there is a risk of unintended political and economic consequences of reform; second, a major reform is likely to affect the existing distribution of power and influence, so that the members holding central and powerful positions in the policy community may lose their power and influence when the policy is redesigned (Daugbjerg 1999: 413–14). They have, in other words, no incentive to break up the existing supranational compromise. In issue networks, on the other hand, ‘each member is prepared to break up the compromise and further its own interests when an opportunity arises’ (Daugbjerg 1997: 130). With fluid membership, there are no shared values and overarching consensus within an issue network, and some members are likely to have an interest in breaking up the existing order. We can thus assume that a policy area dominated by an issue network is less resistant to reform. The network characteristics of the CAP, however, come close to the definition of policy community. The network is not only dominated by the farm lobby; it lacks any strong interest countervailing the farmers, such as consumer interest (Daugbjerg 1997: 135). There is thus a large degree of consensus within the European agricultural network, which gives its members both the capability and the will to defend the status quo. However, as Daugbjerg argues, in order to have reform demands removed from the agenda, members of policy communities may agree to give limited concessions to reformers. There is also the possibility that consensus within the network may eventually break down and lead to policy changes (Daugbjerg 1999: 414).

Political pressures The wider political environment is always more or less important for a decisionmaking system. Reform is directly dependent on it, since the need for reform arises due to internal and/or external pressure for change. Such pressures thus constitute windows of opportunity for reform as they can create critical moments which, if exploited, can be turned into critical junctures (Laffan 2000). Strong pressure increases the propensity for making changes to previous supranational compromises. Critical junctures could thus be argued to occur when the parties consider the political costs in case of non-reform larger than the costs associated with the reaching of a new supranational compromise. In the case of CAP/Agenda 2000, the ‘shadow of enlargement’ and the need to comply with international trade agreements clearly constituted a window of opportunity for major reform. This was, however, offset by other factors, mainly the domestic financial and economic situations in the member states, which at the time prioritized the fulfilling of the euro convergence criteria. The catchwords in

Reform negotiations: the case of the CAP 75 the preparations for the single currency were budgetary rigour and efficient expenditure, and in particular the main net contributors to the EU budget were campaigning for a reduced burden – an issue that was at the forefront of their domestic political agendas at the time (Galloway 1999: 12). Considering the fact that major reform of the CAP would initially be more costly than preserving the status quo, interest in such a reform was low from the outset in several member states. External pressure is more likely than internal pressure to lead to reform of the CAP. The most thorough reform of the CAP so far, the MacSharry reform, was negotiated under heavy external pressure. At the time of the reform negotiation in the early 1990s, the whole Uruguay Round was deadlocked due to the positions of the European Union and the United States of America on agriculture (Tangermann 1998: 19–25). Under such strong external pressure, there is less scope for taking domestic considerations of certain member states into account. External pressure can also serve as a unifying factor, increasing the internal coherence of the European Union. Even if Agenda 2000 ostensibly needed to prepare the European Union for the next trade round, scheduled to begin in December 1999, the pressure was not acute, since the eventual agreement was years away at the time (cf. Schwaag Serger 2001: 127–8). Also, the parts of the Uruguay Round Agreement that required the European Union to cut its subsidized exports were transformed into an ‘internal’ question – the European Union could comply with the agreement without major changes, at the cost of lost export incomes.

Conclusions In the case of the CAP/Agenda 2000, a relatively high level of politicization, a redistributive policy type and the existence of a policy community gave rise to difficult reform negotiations. The political pressure for reform was considerable, whereas the external pressure was not considered acute. The lack of any considerable supranational consensus culture and the impact of national interests make it difficult to reach agreement on new supranational compromises. Thus, problems arise when common decision making is not accompanied by a fully developed supranational consensus culture. The difficulties are reinforced by cemented positions and inflexible coalition patterns, as well as by lack of leadership and/or mediation. When the negative consequences of supranational decision making prevail, the result is path dependency, and the need for reform is met with piecemeal adjustments, adding amendments to the existing negotiated order. This does not pose any major problem as long as there is no acute or severe pressure for reform. When such pressure arises, however, the European Union is often ill adapted to make the necessary adjustments. The specific negotiating environment of the EU has generated some positive consequences, resulting from stable, permanent supranational decision-making structures. Studies on day-to-day negotiations, or policy-shaping decisions, have demonstrated the existence of a supranational consensus culture in the European Union, in which national interests have been merged into common, supranational, interests. Through repeated mutual interaction in the permanent negotiating

76 Malena Rosén and Magnus Jerneck machinery of the European Union, the members become socialized into accepting new codes and behavioural norms, and ‘new loyalties and identifications are formed’ (Elgström and Jönsson 2000: 5–6). The existence of such a supranational consensus culture on lower political levels and on issues with a low degree of politicization might eventually ‘spill over’ to higher political levels and to issues of higher politicization. In regard to politicized reform negotiations, where a lack of supranational consensus culture is demonstrated, the reaching of new supranational compromises can be facilitated by coalitions building on cross-cutting cleavages, as well as by mediation and strong leadership. However, this was not the case in the CAP/Agenda 2000 negotiations. As a political phenomenon the European Union is by no means unique in demonstrating reluctance to change. Political institutions, as well as the policies they generate, often demonstrate considerable resistance to reform, despite internal or external pressure for change. Both actors and institutions of a particular political system tend to act and function in ways that contribute to conserving the existing order. Regarding the actors, a reform proposal usually mobilizes those who risk losing existing benefits to a larger extent than those who might gain from changes. It can be unclear who will gain from a reform and in what ways, or the gains can be distributed over a large group with very small gains for each individual. The losses can also be perceived as more ‘real’ than the potential gains. Therefore, those who act in order to prevent reform tend to be more active than those wishing to see a reform adopted. As stated by March and Olsen (1996: 258): institutions are usually associated with routinization and repetition, persistence and predictability, rather than with political change and flexibility, agency, creativity and discretion. Surviving institutions seem to stabilize their norms, rules, and meanings so that procedures and forms adopted at birth have surprising durability. Once established, institutions seem to take on lives of their own, ‘sometimes enduring in the face of apparent inconsistency with their environments, sometimes collapsing without obvious external cause’ (March and Olsen 1996: 255). Institutionalization thus contributes to cementation of the existing order. In grand budgetary bargains in the European Union there is a high degree of path dependency in institutional development and the policies these institutions generate. Yet a transformation can take place in times of crisis. In the case of the CAP, however, thus far every critical moment or window of opportunity for major reform of the prevailing policy paradigm has been lost.

Notes 1 The SCA is the equivalent of COREPER in the agricultural area. Its members are senior officials from Agriculture Ministries.

Reform negotiations: the case of the CAP 77 2 For more detailed accounts of the insufficiency of the reform, see e.g. CEPS Working Party Report No. 25, 1999 and Ferrer and Emerson 2000. 3 Green money was an artificial unit of currency from which CAP prices were translated into national currencies. This contributed to undermine the operation of the CAP as a common market. Grant argues: ‘For member states, the green currency mechanism became a means of pursuing national agricultural and food policies while remaining within the common market’ (1997: 85).

References Agra Europe, various issues 1997–9. Centre for European Policy Studies (CEPS) (1999) The Challenges of Enlargement to EU Agriculture, Working Party Report No. 25, Brussels. Commission (1997) Agenda 2000: For a Stronger and Wider Union, DOC/97/6. Commission (1998) Proposals for Council Regulations (EC) Concerning the Reform of the Common Agricultural Policy, COM (1998) 158 (final), Brussels 2000–03–18. Daugbjerg, C. (1997) ‘Policy Networks and Agricultural Policy Reforms: Explaining Deregulation in Sweden and Re-Regulation in the European Community’, Governance, 10 (2): 123–41. Daugbjerg, C. (1999) ‘Reforming the CAP: Policy Networks and Broader Institutional Structures’, Journal of Common Market Studies, 37 (3): 407–28. Elgström, O. and Jönsson, C. (2000) ‘Negotiations in the EU: Bargaining or ProblemSolving?’, Journal of European Public Policy, 7 (5): 684–704. Elgström, O. and Smith, M. (2000) ‘Introduction: Negotiation and Policy-Making in the European Union – Processes, System and Order’, Journal of European Public Policy, 7 (5): 673–83. Elgström, O., Bjurulf, B., Johansson, J. and Sannerstedt, A. (2001) ‘Coalitions in European Union Negotiations’, Scandinavian Political Studies, 24 (2): 111–28. Europeiska rådet (1999) Ordförandeskapets slutsatser: Europeiska rådet i Berlin den 24 och 25 mars 1999, SN 100/1/99 REV 1. Ferrer, J. N. and Emerson, M. (2000) Good Bye, Agenda 2000 – Hello Agenda 2003: Effects of the Berlin Summit on Own Resources, Expenditures and EU Net Balances, Brussels: CEPS Working Document No. 140. Galloway, G. (1999) ‘Agenda 2000 – Packaging the Deal’, in G. Edwards and G. Wiessala (eds) ‘The European Union Annual Review 1998/1999’, Journal of Common Market Studies, 37. Grant, W. (1997) The Common Agricultural Policy, London: Macmillan. Green Cowles, M. and Smith, M. (2000) ‘Risks, Reform, Resistance, and Revival’, in M. Green Cowles and M. Smith (eds) The State of the European Union, Vol. 5, Oxford: Oxford University Press. Hall, P. A. (1993) ‘Policy Paradigms, Social Learning, and the State: The Case of Economic Policymaking in Britain’, Comparative Politics, 25 (3): 275–96. Hopmann, P. T. (1996) The Negotiation Process and the Resolution of International Conflicts, Columbia, SC: University of South Carolina Press. Ingersent, K. A., Rayner, A. J. and Hine, R. C. (1998) The Reform of the Common Agricultural Policy, London: Macmillan. Jönsson, C., Bjurulf, B., Elgström, O., Sannerstedt, A. and Strömvik, M. (1998) ‘Negotiations in Networks in the European Union’, International Negotiation, 3: 319–44. Laffan, B. (2000) ‘The Big Budgetary Bargains: From Negotiation to Authority’, Journal of European Public Policy, 7 (5): 725–43.

78 Malena Rosén and Magnus Jerneck Lowi, T. (1972) ‘Four Systems of Policy, Politics and Choice’, Public Administration Review, 32 (4): 298–310. Malnes, R. (1995) ‘Leader and Entrepreneur in International Negotiations: A Conceptual Analysis’, European Journal of International Relations, 1 (1): 87–112. March, J. G. and Olsen, J. P. (1996) ‘Institutional Perspectives on Political Institutions’, Governance, 9 (3): 247–64. Peterson, J. (1995) ‘Decision-making in the European Union: Towards a Framework for Analysis’, Journal of European Public Policy, 2 (1): 69–93. Roederer-Rynning, C. (2002) ‘Farm Conflict in France and the Europeanisation of Agricultural Policy’, West European Politics, 25 (3): 105–24. Roederer-Rynning, C. (2003) ‘From “Talking Shop” to “Working Parliament”? The European Parliament and Agricultural Change’, Journal of Common Market Studies, 41 (1):113–35. Schwaag Serger, S. (2001) Negotiating CAP reform in the Europan Union: Agenda 2000, Lund: Swedish Institute for Food and Agrucultural Economics. Spence, D. (1995) ‘Negotiations, Coalitions and the Resolution of Inter-State Conflicts’, in M. Westlake (ed.) The Council of the European Union, London: Cartermill. Stenelo, L.-G. (1972) Mediation in International Negotiations, Lund: Studentlitteratur. Stenelo, L.-G. (1991) ‘Demokratisk politik och internationalisering’, in J. P. Olsen (ed.) Svensk demokrati i förvandling, Stockholm: Carlssons förlag. Tangermann, S. (1998) ‘An Ex-post Review of the 1992 MacSharry Reform’, in K. A. Ingersent, A. J. Rayner and R. C. Hine (eds) The Reform of the Common Agricultural Policy, London: Macmillan. USDA (United States Department of Agriculture) (1999) ‘The European Union’s Common Agricultural Policy: Pressures for Change’, in International Agriculture and Trade Reports, Washington DC, October 1999. Wallace, W. (1996) ‘Government Without Statehood: The Unstable Equilibrium’, in H. Wallace, and W. Wallace (eds) Policy-Making in the European Union, Oxford: Oxford University Press. Webber, D. (1998) The Hard Core: The Franco-German Relationship and Agricultural Crisis Politics in the European Union, European University Institute, Working Paper RSC No 48/6.

Interviews (carried out by Malena Rosén) Anders Buch Kristensen, Agricultural Councillor, Danish Representation at the European Union, Brussels, 24 February 2000. Håkan Emsgård, Deputy Director, Swedish Foreign Ministry, Stockholm, 11 February 2000. Thomas Hagman, Principal Administrative Officer, Swedish Ministry of Agriculture, Stockholm, 10 February 2000. Anders Klum, Chairman of the Special Committee on Agriculture (SCA), Stockholm, 11 February 2000. Christophe Manson, Agricultural Attaché, French Representation at the European Union, Brussels, 22 February 2000. John Muldowney, Agricultural Councillor, Irish Representation at the European Union, Brussels, 18 February 2000. Erik Rudal, Agricultural Attaché, Swedish Representation at the European Union, Brussels, 18 February 2000.

6

Compliance bargaining Jonas Tallberg and Christer Jönsson

Introduction The notion that negotiations proceed through several identifiable stages is common in the negotiation literature. Some theorists postulate an early contentious stage and a later problem-solving phase, with varying numbers of stages in between (cf. Pruitt and Rubin 1986: 137); others distinguish a diagnostic, a formula and a detail phase (Zartman and Berman 1982). In any event, it is the process leading up to the signing of an agreement that has been in focus. Attempts to extend negotiation analysis beyond the negotiating table have primarily dealt with the pre-negotiation stage (Saunders 1985; Stein 1989). By contrast, postagreement bargaining, following from the conclusion of an agreement, has received scant attention (but see Spector and Zartman 2003). In this chapter we will analyse one particular form of post-agreement bargaining in the European Union, which we label compliance bargaining, understood as a process of bargaining between the signatories to an agreement already concluded, or between the signatories and the international institution governing the agreement, and which pertains to the terms and obligations of this agreement. Elsewhere ( Jönsson and Tallberg 1998) we have developed an analytical framework for understanding the forms, sources and effects of compliance bargaining, which will inform our analysis of this important yet largely neglected phenomenon in the European Union. Bargaining and compliance remain the objects of separate strands of research in EU studies. The literature on bargaining and negotiations in the European Union is heavily focused on the intergovernmental bargaining preceding history-making deals in European cooperation (e.g. Laursen and Vanhoonacker 1992; Moravcsik 1998), the inter-institutional negotiations leading up to the adoption of secondary legislation (e.g. Mesquita and Stokman 1994; Tsebelis and Garrett 2000) and the European Union as a negotiating party in external relations (e.g. Friis 1996; Meunier 2000). In all cases, the bargaining processes explored by existing scholarship are pre-decisional, that is, they consist of the exchange of proposals for purposes of reaching agreement on new rules. By contrast, post-decisional processes of bargaining over compliance with these rules have so far been neglected in the literature. The pre-decisional bias in existing research is well illustrated by two recent special issues on negotiation and bargaining in the European Union, which

80 Jonas Tallberg and Christer Jönsson exclusively feature articles on the processes preceding new deals (International Negotiation 1998; Journal of European Public Policy 2000). The literature on compliance and enforcement in the European Union may similarly be divided into three branches: public policy research on the implementation of EU policy (e.g. Siedentopf and Ziller 1988; Börzel 2000), legal and political research on the European Commission’s execution of its function as ‘guardian of the treaties’ (e.g. Audretsch 1986; Tallberg 2003) and legal and political research on the interaction between the European Court of Justice (ECJ) and national courts in the decentralized enforcement of EU law (e.g. Slaughter et al. 1998; Alter 2001). In all three branches of research, compliance tends to be conceptualized slightly differently, but in none of them are processes of bargaining at the centre of this understanding, though parts of both the public policy literature and the literature on Commission enforcement are sensitive to the negotiations involved in securing rule conformance. In this chapter we isolate the primary characteristics of compliance bargaining as it takes place in the European Union – the structure of bargaining, the distribution of bargaining power, and the effects of bargaining on compliance, rule interpretation and the allocation of gains. We advance two principal arguments. First, the dynamics of compliance bargaining differ in distinct ways from pre-decisional negotiations over new EU rules. Most notably, compliance bargaining is a bilateral process between the Commission and non-complying member states, rather than a multilateral process of intergovernmental and inter-institutional negotiations, and it takes place in the shadow of the law rather than in the shadow of the vote. To incorporate compliance bargaining in our conception of EU policy making therefore requires more than just extrapolating pre-decisional patterns of negotiation into the post-decisional phase. Second, the institutional structure of the European Union provides for a form of compliance bargaining that has proven particularly effective as a way of addressing state violations. The Commission’s position as third-party prosecutor within the formal framework of infringement and sanctioning procedures – laid down in EU treaties – produces a strategic context in which non-complying states generally find themselves at a bargaining disadvantage. While pervasive, rule violations also tend to be a temporal phenomenon in the European Union, not least because of institutionally defined processes of compliance bargaining. The chapter is divided into three parts. In the next section, we present a condensed version of the framework for analysing compliance bargaining introduced in our previous work. The body of the chapter consists of the empirical section on compliance bargaining in the European Union. In this section, the structure of bargaining, the distribution of bargaining power and the effects of bargaining in the post-decisional phase are contrasted with the well-known characteristics of predecisional negotiations. We conclude with a brief summary of our argument.

Compliance bargaining in international relations Why does international compliance bargaining occur? What forms may it take? And what are the effects of compliance bargaining? Let us briefly address these basic questions (cf. Jönsson and Tallberg 1998: 378–87).

Compliance bargaining

81

Origin Compliance bargaining typically stems from actions of non-compliance and the ambiguity of most international agreements. Clear and manifest violations rarely stand unchallenged, but tend to provoke reactions from the parties to an agreement. Compliance bargaining primarily serves as a way for the other parties to persuade the signatory in breach that the costs of non-compliance outweigh possible benefits of continued violation, and for the alleged violator to justify its action and avoid sanctions. Non-compliance as such may be a product of either a lack of willingness or a lack of capacity to comply. Lack of technical, economic and administrative capacity is a well-documented and widespread source of noncompliance, making some states more prone to violate international agreements than others (Chayes and Chayes 1995). Similarly, states’ interests are unequally reflected in the rules and codes agreed upon (Downs et al. 1996). Diverging interpretations of treaty language provide a fertile ground for bargaining about what actions do and do not constitute compliance. Ambiguous formulations may stem from the fact that these are necessary in order for the parties to reach a minimum level of consensus and come to an agreement at all. Broad and general language may also offer a ‘veil of uncertainty’, which permits a number of parallel interpretations and visions as to the future development of a cooperative endeavour (Young and Osherenko 1993). Imprecision and ambiguity can serve the function of insurance policy or escape clause, when gains and costs from an agreement are unpredictable (Lebow 1996). Yet another reason may be the inability of drafters to foresee all possible applications and to plan for all contingencies, with an ensuing mismatch between the coverage and formulations of the treaty and the practice it seeks to regulate (Chayes and Chayes 1995). Forms One basic distinction can be made between self-help and third-party compliance bargaining. These two forms constitute ideal types, and concrete conflicts and bargaining situations may exhibit features of both or may oscillate between the two. Self-help bargaining refers to bargaining between the parties to the treaty, now taking place in the post-agreement phase. Lack of a common authority to enforce rules is the defining characteristic of the setting within which self-help bargaining occurs. Third-party bargaining has as its defining and unique characteristic the existence of an international institution which interacts with the signatories of an agreement in the interpretation of compliance and the settling of disputes. Third-party bargaining, in turn, may be of two kinds, depending on whether the international institution acts as ‘judge’ or ‘prosecutor’. The traditional conception of international institutions as third parties is that of a judge. Member states in conflict over treaty compliance and interpretations bring the case before a dispute settlement body. Bargaining, in this context, takes place between the disputing states within the framework of the dispute settlement process. International institutions as judges are a common form of third-party enforcement in international trade. The dispute settlement mechanisms of the World Trade Organization (WTO) and the North American Free Trade Agreement (NAFTA) are cases in point. An alternative conception of international institutions as third parties is that of a prosecutor. Institutions as prosecutors do not issue interpretations as much as they

82 Jonas Tallberg and Christer Jönsson act independently and strike down on member states suspected of violating the treaty. Bargaining, in this context, primarily takes place between the international enforcement institution and the signatory suspected of non-compliance. Institutions as prosecutors are a less common form of third-party enforcement, with the prime examples being the European Commission of the European Union and the International Atomic Energy Agency (IAEA), which both have enforcement powers that can be used against member states in breach of treaty provisions. Under third-party enforcement generally, bargaining results from the combination of a ‘sanctioning ladder’ and the interest of all parties to settle disputes at an early stage, rather than letting cases or conflicts run their full course. The term sanctioning ladder is used to denote the consecutive steps which may be taken to induce compliance, and which typically are characterized by a progressive increase of pressure and costs of non-compliance. Effects Why does compliance bargaining matter and why do we need to pay attention to it as an empirical phenomenon? Compliance bargaining may alter outcomes and affect future rounds of bargaining in three principal ways: (1) by influencing the level of compliance, (2) by defining what constitutes compliance and non-compliance, and (3) by affecting the distribution of gains in future bargaining. First, and most fundamentally, compliance bargaining influences the level of compliance. In the search for mutually acceptable solutions, it might put an end to actions perceived to be in breach by one of the parties. From the perspective of the guardians, compliance bargaining serves to induce and persuade violators to step into line, to the extent that it raises the cost of non-compliance. From the perspective of the violators, compliance bargaining serves to test the limits of the other parties’ tolerance of deviant behaviour. The second effect of compliance bargaining is to provide definitions of what constitutes compliance and what actions are or are not in line with a treaty. Compliance as defined in post-agreement bargaining may not correspond to compliance as perceived by the parties when entering into the agreement. In other words, states settle for agreements and negotiation outcomes whose terms and distribution of gains they believe they understand and foresee, but which are often substantially altered when compliance is ultimately defined through postagreement bargaining. The third essential effect of compliance bargaining is its influence on how gains are distributed in future rounds of bargaining. In a context where states interact on a regular basis, other states are more likely to enter into future agreements with a state, and on more favourable terms, if it carries a reputation for keeping commitments. Therefore, a good reputation is crucial to the realization of future benefits from cooperation (Keohane 1984; Chayes and Chayes 1995). Compliance bargaining reinforces and contributes to the distribution of positive and negative reputational effects, a distribution that ultimately rests on how well states comply. Consequently, compliance bargaining does not alter the distribution of gains only in agreements already entered into, but also in those to come.

Compliance bargaining

83

Compliance bargaining in the European Union While largely neglected in the existing literature, processes of bargaining constitute the predominant mode of settling disputes over compliance in the European Union. The characteristics of compliance bargaining are significantly different from the characteristics of pre-decisional negotiations over new legislation and treaty rules. In this section, we isolate the structure of EU compliance negotiations, the sources of bargaining power for the major actors involved and the effects of these processes on compliance with EU law. We demonstrate how the structure of bilateral interaction between the Commission as prosecutor and member governments as defendants produces distinct supranational bargaining advantages, which are instrumental to the European Union’s demonstrated effectiveness in ending violations. Structure: bilateral supranational bargaining With the shift from pre-decisional negotiations over new rules to post-decisional bargaining over compliance, the structure of interaction in the EU policy process is fundamentally transformed. Multilateral intergovernmental and inter-institutional negotiations in EU decision making are replaced by bilateral supranational bargaining in EU enforcement. Despite the existence of institutional procedures for interstate enforcement and bargaining, negotiations over compliance almost exclusively take place between the Commission, acting as third-party prosecutor, and national governments, representing states suspected of non-compliance. In stylized terms, pre-decisional politics in the European Union is characterized by two interlocking bargaining processes: intergovernmental negotiations between state representatives in the Council and inter-institutional negotiations between the Commission, the European Parliament and the Council. It is the nesting of these two games that leads one observer to characterize EU decision making as a ‘multilateral inter-bureaucratic negotiation marathon’ (Kohler-Koch 1996: 367). Less stylized accounts of pre-decisional politics typically stress the multi-level character of EU negotiations, the involvement of non-governmental actors, and the existence of informal negotiation networks. In the ‘internal’ negotiation of European Union business, there is a wide range of institutional, governmental, non-governmental and quasi-governmental participants; when the ‘external’ implications of negotiation are added to the mix, there is a heterogeneous and at times almost bewildering array of actors in negotiation. (Elgström and Smith 2000: 675) Post-decisional negotiations over compliance, by contrast, feature a structure of interaction that is essentially bilateral, with the supranational Commission and the offending member state as the two exclusive parties. In the words of H. A. H. Audretsch, ‘[compliance in the European Union] is generally achieved in an

84 Jonas Tallberg and Christer Jönsson amicable way through negotiations between the supervising body and the state concerned’ (Audretsch 1986: 410). Expressed in generic terms, compliance bargaining in the European Union is essentially third-party and prosecutor based, with the Commission pursuing cases against non-complying member states in a hierarchical judicial system, where the ECJ has the ultimate power to adjudicate disputes and interpret existing rules. The structure of compliance bargaining in the European Union is the product of both institutional design and state preference. As noted in the previous section, the specific character of compliance bargaining in a given institutional setting tends to be shaped by existing dispute-settlement mechanisms. In the European Union, the founding treaties provide for two alternative avenues for settling disputes over compliance. On the one hand, governments may engage each other directly under Article 227, which offers a procedure by which one member state may sue another for non-compliance and have the case decided by the ECJ. On the other hand, governments may leave the task of ensuring compliance to the Commission, which enjoys the authority under Article 226 to initiate infringement proceedings against non-complying member states, and ultimately refer these cases to the ECJ. Whereas the design of dispute settlement procedures thus provides for both interstate and supranational enforcement, the historical record demonstrates an overwhelming preference on the part of national governments to let the Commission carry the burden, rather than continue the state-to-state bargaining of the pre-decisional phase. Less than a handful of cases have been initiated by member governments under Article 227. By contrast, the Commission has initiated more than 16,000 cases between 1958 and 2000 (Krislov et al. 1986; European Commission annual monitoring reports). The literature on EU enforcement points to a set of advantages of the supranational infringement procedure over against the interstate version, making the former a more attractive alternative to member governments. First, enforcement through the Commission saves the member states the costs of litigation. Second, the supranational procedure reduces the risk of spirals of retaliation, since, under the interstate procedure, ‘a snowball effect might be the consequence of a complaint; where one State starts, another follows!’ (Audretsch 1986: 237). Third, the Article 226 procedure satisfies the preference for diplomatic courtesy among the member states. And fourth, the supranational procedure provides a process of enforcement that is easier for member governments to accept, since the proceedings are initiated by ‘an institution representative of the whole, and hence objective both by its status and its task’ (Pescatore 1974: 82). For these reasons, compliance bargaining in the European Union almost exclusively takes place within the framework of the supranational infringement procedure. The seemingly formal and inflexible framework of the procedure provides ample room for bargaining and compromises between the Commission and the member states. At the heart of the procedure are four consecutive stages where conflicts may be resolved. In the first, the informal phase, the Commission notifies the member state in question of its alleged act of non-compliance and the state is given the opportunity to respond. If the matter is not settled in the informal

Compliance bargaining

85

phase the procedure continues by way of formal means. In the second phase, the Commission formally initiates an infringement proceeding by sending a ‘letter of formal notice’ to the member state, whereby the Commission informs the state of its grounds for complaint and invites it to submit its views. The third phase consists of the Commission giving a ‘reasoned opinion’. Whereas the formal notice described the subject matter of the violation, the reasoned opinion presents the Commission’s legal arguments. If the member state fails to comply with the reasoned opinion and continues its action in breach of EU law, the case enters into the procedure’s fourth and final stage of referral to the ECJ. To the extent that a government persists in its violations even after an ECJ decision, the Commission may restart the entire process by initiating a new and identically structured proceeding through the Article 228 sanctioning procedure. Formal and informal bargaining aimed at finding mutually acceptable solutions takes place at all stages of the infringement procedure. Bargaining is most intense in the early stages of the procedure, however, and close to non-existent once a case has been referred to the ECJ. These processes of bargaining over compliance within the formal framework of the infringement procedure point to the complementarity of negotiation and adjudication as means for settling disputes. Francis Snyder (1993: 30) notes: We usually think of negotiation and adjudication as alternative forms of dispute settlement. It may be suggested, however, that in the daily practice and working ideology of the Commission, the two are not alternatives but instead are complementary. The main form of dispute settlement used by the Commission is negotiation, and litigation is simply a part, sometimes inevitable but nevertheless generally a minor part, of this process. Compliance bargaining within the supranational infringement procedure consists of bargaining in a wide sense, and involves both direct and indirect verbal and behavioural communication. In letters and face-to-face meetings, the Commission attempts to persuade member states to comply by explaining their violations under EU law, by communicating the threat that the Commission may bring the case to the next step in the procedure and by reminding states that economic sanctions may be imposed on them if they fail to comply with ECJ judgments. Moreover, the Commission attempts to raise the cost of non-compliance by mobilizing social and political pressure, thus exploiting states’ concern with reputational repercussions. To this end, the Commission presents official reports on rule violations, publicly announces its initiation of infringement proceedings through press releases, and reports non-compliance patterns at Council meetings. Member states, for their part, attempt to explain to the Commission the political, economic, social or administrative reasons and rationales behind the measures under review. Member states may also present alternative interpretations of the problem at hand, suggest compromise solutions or signal the unilateral decision not to budge and let the case run its course. Since the late 1980s compliance bargaining in this wider sense has been

86 Jonas Tallberg and Christer Jönsson supplemented with direct and institutionalized negotiations – so called ‘package meetings’ – in matters pertaining to internal market and environmental regulation. The European Commission (1993: 13) describes package meetings as ‘an instrument of partnership between the Commission and the Member States which is designed to arrive at non-contentious solutions to existing litigation concerning national compliance with Community law’. Package meetings are conducted with one member state at a time and consist of exchanges on all cases currently under review by the Commission, whether at the informal or the formal stage of the infringement procedure. During the 1990s the Commission conducted package meetings with each member state about every second year. The term ‘package meeting’ is to some extent misleading, since these negotiation sessions in fact consist of a series of meetings. Each series typically includes a preparatory meeting, a package meeting, a follow-up meeting and a concluding stocktaking meeting. All meetings are preceded as well as followed by communication between the parties, and the entire process is generally described as one of ‘dialogue’ (European Commission 1994; interview, Swedish government official, 6 December 1996). Like all forms of bargaining, compliance bargaining between the Commission and member governments consists of both cooperative and conflictual elements. Both parties share a preference for amicable solutions, yet they disagree as to whether or not the member state is in breach of EU law. On the conflictual side, the parties obviously disagree about the conformance of state actions with existing EU rules. Member states generally maintain that their actions are justifiable and in line with existing legal provisions. National governments typically consider the Commission’s arguments to be based on flawed legal assumptions and interpretations, or complain that the Commission fails to comprehend the constraints of domestic political processes and capacity limitations. The Commission, for its part, bases the initiation of infringement proceedings on the fundamental notion that member states are in breach of EU law. In its role as guardian of the treaties, the Commission is obliged to fight such non-compliance by way of the legal and political means at its disposal. At the same time, both parties share a preference for amicable solutions, and neither member governments nor the Commission desire infringement proceedings. As one Commission official put it, ‘legal proceedings are not good for anyone’ (interview, 24 September 1996). From the perspective of member governments, infringement proceedings in general, and ECJ judgments in favour of the Commission in particular, are highly uncomfortable and tarnish the states’ reputations as cooperative partners. ‘The ultimate fate of having their failure to fulfil an obligation under the EEC Treaty formally established by the European Court is one the Member States are evidently anxious to avoid’ (Dashwood and White 1989: 411). As a result, ‘Member States endeavour, by means of a variety of objections, as far as possible to avoid Judgments being given against them in proceedings for failure to fulfil a Treaty obligation’ (Everling 1984: 221). The Commission’s desire to close infringement proceedings and put an end to violations by way of amicable solutions stems from its dual role in European integration and its limited resources. The Commission’s other role as policy

Compliance bargaining

87

initiator and prime promoter of European integration entails that the institution must seek to ensure the continued cooperation of member states for integration to proceed. As a consequence, political considerations might induce the Commission to make concessions to national interests and pressure of the States. The Commission knows after all that the progress of integration depends to a high degree upon the willingness and the cooperation of the States. It is extremely important for the Commission to ensure this cooperation permanently. (Audretsch 1986: 420) Equally important, however, the Commission is subject to considerable resource constraints. With limited staff and an ever-increasing workload, the option of informal and amicable solutions is very attractive in comparison to the resourceintensive alternative of continued infringement proceedings (European Commission 1989; interview, Commission official, 24 September 1996). Summing up, the structure of bargaining in post-decisional enforcement is one of bilateral negotiations between the supranational Commission and member states suspected of non-compliance. Both share a preference for amicable solutions, and the Article 226 infringement procedure offers a formal framework for bargaining to this end. Power: bargaining in the shadow of the law The shift from pre-decisional negotiations to post-decisional compliance bargaining fundamentally alters the power relations between the major actors in the EU policy process. Traditional determinants of bargaining power, such as voting strength, preference intensity, decision rules and legislative procedures, are replaced by sources related to the parties’ relative control over the interpretation of EU law, command of the legal procedures and capacity to shape cooperative outcomes. In post-decisional compliance bargaining, the Commission has the upper hand, whereas member governments generally – irrespective of the power positions they occupy in EU decision making – find themselves at a disadvantage. In the negotiation of new EU rules, the relative bargaining power of the parties is typically shaped by the institutional provisions governing the adoption of these rules. The Commission and the Parliament enjoy relatively stronger bargaining positions in relation to the Council in the adoption of secondary legislation than in the chiselling out of new treaty rules. Likewise, the institutions hold a stronger hand in some legislative procedures than in others, best exemplified by the Parliament’s pronounced role in the co-decision procedure, as opposed to the consultation and cooperation procedures. The relative bargaining power within the Council, between the member states, is similarly shaped by the rules of adoption. In the restricted number of cases where new rules must be adopted by unanimity, ‘the pattern of preference intensity or “asymmetric interdependence” dictates the relative value each state places on an agreement, which in turn dictates

88 Jonas Tallberg and Christer Jönsson its respective willingness to make concessions’ (Moravcsik 1998: 60). However, in the large majority of cases, where rules are adopted through qualified majority voting, the bargaining positions of the individual states are intimately connected to their relative voting power, placing large states at an advantage. Depending on the applicable decision rules, Council bargaining takes place either ‘in the shadow of the veto’ or ‘in the shadow of the vote’. In post-decisional compliance bargaining, by contrast, the most distinct source of variation in bargaining power is the parties’ relative control over the interpretation of EU law. Expressed in other terms, the compliance game is played ‘in the shadow of the law’, far removed from the vetoes and votes of EU decision making. With a monopoly on the interpretation of EU law, the Commission and the ECJ enjoy extremely favourable positions in relation to EU governments, whose bargaining moves are typically designed to probe the limits of acceptable behaviour, given their inability to overthrow supranational legal interpretations. The Commission’s bargaining power is vested in its unilateral control over the infringement procedure and its discretion in decisions about initiation, intensification and closure of cases. Ostensibly, the infringement procedure is a strictly judicial process in which the Commission initiates and pursues infringement proceedings against member states that have committed clear and objectively identifiable violations of EU law. Beneath the surface of neutral and objective law, however, Article 226 proceedings are highly political and subject to substantial discretion on the part of the Commission. The Commission exercises its supervisory function ‘not only from a purely technical, but also from a political point of view – not incidentally, but permanently’ (Audretsch 1986: 408). The discretion is particularly notable with regard to three essential aspects of the infringement procedure: the decision on whether to initiate infringement proceedings or not, the decision on what time limits governments must comply within before the Commission moves the case to the next step in the procedure and the decision on when and how infringement proceedings are closed (Evans 1979; Audretsch 1986; Dashwood and White 1989). In other words, it is the Commission alone that decides about the initiation, intensification and closure of suits against non-complying states. Member governments’ lack of means for monitoring the Commission in its enforcement function further enhances the institution’s autonomy, by making it difficult for other parties to judge whether the Commission acts in accordance with its formal enforcement obligations (Tallberg 2003). In fact, the Commission’s discretion constitutes a requirement for compliance bargaining as such. In essence, it allows the Commission room for manoeuvre within the confines of its formally delegated role. The Commission is thereby in a position where it may arrive at ‘compromise solutions in a flexible way through negotiations, conciliatory measures, and mutual concessions’ (Audretsch 1986: 449). Expressed differently, had the discretion of the Commission been very limited, or even non-existent, there would have been little scope for bargaining, since the institution would have had little to offer member states in return for conciliatory moves. Bargaining, by definition, requires a zone of acceptance within which both parties may consider compromise solutions if there is to be any agreement at all.

Compliance bargaining

89

The bargaining power related to the Commission’s unilateral control over the legal procedures is further reinforced by the institution’s strategic exploitation of the member governments’ desire to avoid reputational and financial costs. Rather than avoiding steps that can throw negative light on member states, the Commission has made it its strategy to embarrass laggards into action through means of shaming (Tallberg 2003). For purposes of gradually increasing the member states’ discomfort and raising the reputational cost of non-compliance, these efforts are reinforced with each step in the infringement procedure. ‘When the State appears to persist in the violation, an attempt will be made to raise the cost of violation or to lower its profit. The threat of political and social pressure tends to raise the cost’ (Audretsch 1986: 410–11). Whereas, previously, the European Union’s supranational institutions lacked the power to impose actual financial sanctions on non-complying states, the Commission and the ECJ can now use the threat of such economic countermeasures in their attempts to persuade member governments to comply. Since the entry into force of the Maastricht Treaty in November 1993, the Commission and the ECJ may, under the revised Article 228, impose penalty payments on member states that refuse to comply with ECJ judgments. In cases where member states have not taken the measures to comply within the assigned time period, the Commission may propose a penalty to be approved by the ECJ in a second decision. The Commission has actively used this weapon as a bargaining resource in negotiations with governments over non-compliance. Since the mid-1990s the Commission has included the threat of economic sanctions as an ultimate possibility in its communication with member states in association with infringement cases. The credibility of the Commission’s threats to impose reputational and financial costs is partially contingent on the likelihood that the ECJ will share its interpretation of the alleged infringement cases. As it is, it almost always does. The historical record testifies to a strong tendency on the part of the ECJ to rule in favour of the Commission in cases that go all the way to court decisions. About 90 per cent of all judgments in infringement proceedings are in favour of the Commission (e.g. Audretsch 1986; European Commission 1996b). There is no doubt that the ECJ’s judgment record constitutes a significant deterrent, which provides strong incentives for member states to reach a solution at an earlier stage of the infringement procedure. If we turn to the other party in EU compliance negotiations, the member states’ bargaining power generally rests with the Commission’s overarching desire to ensure the progression of integration, and the governments’ crucial role in delivering this outcome. Theoretically, national governments could have exploited their position as ‘masters of the treaty’ and threatened to repeal the enforcement authority granted to the Commission and the ECJ, when pressured by the institutions in sensitive cases of non-compliance. But, as Mark Pollack notes, ‘the threat of treaty revision is essentially the “nuclear option” – exceedingly effective, but difficult to use – and is therefore a relatively ineffective and noncredible means of member state control’ (Pollack 1997: 118–19). Instead, the member governments’ delegation of enforcement powers to the Commission and the ECJ is best considered an act of self-commitment and is intended to secure the credibility of

90 Jonas Tallberg and Christer Jönsson their mutual policy obligations (Garrett 1992; Tallberg 2002a). Any tampering with these powers would therefore undermine the purpose of delegation in the first place. More important in terms of bargaining leverage is the member states’ capacity to inflict contained damage in specific policy issues, either by prevailing in their noncompliance or by refusing to cooperate in pre-decisional policy formation. It must not be forgotten that member states are unilaterally in control of the decision whether or not to comply, and the Commission is physically unable to exert compliance by way of force. While subjecting themselves to the risk of negative reputational and economic consequences, member governments may nevertheless choose to challenge the Commission with the option of sustained non-compliance. Such challenges speak directly to the Commission’s interest in preserving the authority of EU law and to the institution’s lack of interest in committing more of its scarce resources to the same case. Alternatively, member governments may draw bargaining advantages from their capacity to inflict policy-phase retaliation on the Commission. The Commission requires and desires the continued and constructive cooperation of member states in the decision-making phase if integration is to progress. While disconnected in theory, the Commission’s policy enforcement and policy initiation functions are not as easily separated in practice. The Commission’s approach towards a member state in the field of enforcement evidently runs the risk of affecting its relations to this member state in pre-decisional policy formation. As Audretsch (1986: 277) succinctly puts it: ‘Suing a Member State for (alleged) failure, and trying to obtain its fiat for a political compromise are often hard to combine.’ The essence of this linkage sanction is therefore the possibility that governments will obstruct Commission initiatives in the pre-decisional phase of EU policy making in retaliation for infringement proceedings in the postdecisional phase. Combined, these aspects translate into a dependency of the Commission on state collaboration and a resulting bargaining power on the part of member governments. As opposed to pre-decisional negotiations, where the structural power of the member states is translated into voting strength, post-decisional compliance bargaining mediates existing differences between large and small states. As we will show in the next section, little speaks in favour of a systematic variation in the capacity of national governments to get their way in compliance bargaining with the Commission. Instead, states tend to find themselves equally constrained by the supranational institutions’ monopoly on the interpretation of EU law. Effects: post-decisional outcomes The effects of bargaining constitute the final dimension on which post-decisional compliance negotiations differ markedly from pre-decisional rule negotiations. Multilateral and carefully calibrated agreements, the result of long and arduous negotiations, are typically assumed to be final. Yet, as a result of compliance bargaining, they may be replaced by bilateral deals that define actual expected

Compliance bargaining

91

behaviour, distribute true gains and burdens, and express genuine acts of cooperation. In fact, the very phenomenon of compliance bargaining points to the non-conclusive nature of the agreements that emerge from the negotiations conducted within the European Union’s legislative processes or at intergovernmental conferences (IGCs). No EU rules are adopted lightheartedly and without preceding multilateral negotiations between the member states, between the institutions and with concerned non-state actors. Even the formal part of the most frequently used legislative procedure takes about two years from initiation to final decision. The purpose of these pre-decisional negotiations is to achieve a convergence of views on the particular proposal and to arrive at a decision that is interpreted as legitimate and effective, both by the key negotiating parties and by the concerned societal interests. Typically, the texts eventually adopted constitute compromises, whose every detail has been reviewed a number of times by the national, supranational and sub-national actors involved in the process. As products of negotiation, these agreements also reflect the relative bargaining power of the parties, and the expectation of a particular distribution of costs and benefits from the implementation of the new rules. Depending on its outcome, compliance bargaining serves to safeguard or reconstruct the original agreement. Compliance bargaining works to preserve the rules formally adopted when inducing states to follow up on their policy commitments, when upholding a definition of acceptable behaviour that accords with the parties’ pre-existing understanding and when confirming the distribution of gains reflected in the original compromise. By contrast, compliance bargaining reconfigures the formal deal when producing less-than-perfect levels of compliance, when redefining the range of acceptable behaviour and when shifting the original distribution of costs and benefits through separate bilateral deals. In sum, compliance bargaining in the European Union may be said to produce the actual outcome, as opposed to the formal agreement. Whether compliance bargaining safeguards or reconstructs the original agreement can only be determined on a case-by-case basis. In aggregate terms, however, the empirical record demonstrates that bargaining, threats and persuasion within the framework of the infringement procedure constitute an effective way of solving non-compliance cases. While bargaining, by definition, is a process of give and take, member states tend to do a lot of the giving and not nearly as much of the taking in EU compliance bargaining. The distribution of bargaining power in favour of the Commission produces a strategic context in which states generally find sufficient reasons to agree to settlements, even if these are biased towards the preferences of the Commission. The effects of compliance bargaining on rule conformance are best demonstrated by the sharp decrease in the number of ongoing violations from one step of the infringement procedure to the next. Indeed, existing data show that non-compliance is largely a temporal problem and that few cases can be classified as intractable (Tallberg 2002b). The stages of the infringement and sanctioning procedures function as a political enforcement ladder that progressively increases the pressure and the costs

92 Jonas Tallberg and Christer Jönsson of non-compliance, thereby encouraging governments to find bargaining solutions acceptable to the Commission. At the first, informal, stage, the Commission engages in unofficial contacts with member states once suspected violations have been detected, thereby solving about one-third of all cases (Steiner and Woods 1996). A significant share of the remaining cases are settled at the second and third stages of the infringement procedure, where the Commission raises the reputational costs of non-compliance, threatens court referrals and warns member states of the risk of economic sanctions. Of the total number of proceedings initiated between 1978 and 2000, only 38 per cent reached the stage of reasoned opinions and only 11 per cent were referred to the ECJ, as demonstrated by Table 6.1. If ‘the initial stages, both formal and informal, between the Commission and the States, are designed to achieve compliance by persuasion’ (Steiner and Woods 1996: 413), then this has proven to be a very effective design indeed. Framing these steps of the procedure as a political stage preceding court action, the Commission emphasizes that it: endeavours to make the fullest use of the pre-litigation stage of the infringement proceedings to persuade the offending Member State to remedy its deficiency or to negotiate a settlement. As the Court has held, referral of an action to it is the last resort, the ultima ratio enabling the Community interests enshrined in the Treaty to prevail over the inertia and resistance of the Member States. (European Commission 1996b: 9) Once a case has been referred to the ECJ, the room for political solutions is significantly reduced and bargaining is replaced by the legal process of court proceedings. The cases that are closed before judgments are delivered are generally the result of member states getting cold feet, rather than amicable settlements. The Table 6.1 Infringement cases per member state by stage, 1978–2000 State Belgium Denmark Germany Greece France Ireland Italy Luxembourg Netherlands Portugal Spain UK EU 12

Formal notice

Reasoned opinion

ECJ referral

1,443 779 1,352 1,611 1,772 1,143 1,970 1,083 1,021 1,230 1,048 1,125

652 103 517 659 721 405 1,010 409 288 483 349 304

45.2% 13.2% 38.2% 40.9% 40.7% 35.4% 51.4% 37.8% 28.2% 39.3% 33.3% 27.0%

236 22 136 198 234 121 396 130 72 63 74 50

16.4% 2.8% 10.1% 12.3% 13.2% 10.6% 20.1% 12.0% 7.1% 5.1% 7.1% 4.4%

15,577

5,900

37.9%

1,737

11.2%

Source: European Commission annual monitoring reports.

Compliance bargaining

93

relatively restricted number of cases where the ECJ indeed delivers opinions contribute to the elucidation of EU law by producing court interpretations that complete the initial contracts. In the few cases where member governments persist in their violations even after court decisions, the process of compliance bargaining resumes with the initiation of a new proceeding under the Article 228 sanctioning procedure. While structured exactly as the original infringement procedure (letter of formal notice, reasoned opinion, court referral), this procedure provides for economic sanctions as a last resort if member states are found guilty of non-compliance in a second judgment. In the few protracted cases that remain after the ECJ’s first judgment, the sanctioning procedure constitutes an effective tool for finally reaching solutions. The Commission made use of the new sanctioning power for the first time in January 1997, and during the period 1997–2000 it proposed penalties in 21 cases, the amounts ranging from 6,000 to 264,000 euro per day (European Commission 1999, 2000, 2001). Member states have been quick to back down in the face of the sanctioning threat. In only one case were the Commission and the ECJ forced to actually impose the proposed penalties (Greek waste dumping in Crete). Table 6.2 provides a snapshot of the effects of political and judicial action on state compliance. The table expresses the percentage of the total number of cases closed in 1999 that were solved at each of the stages in the infringement and sanctioning procedures. Of 1,900 cases closed, as many as 95 per cent were solved through Commission communication with member governments before referral to the ECJ. As testified by Table 6.1, all member states display the same preference for backing down or finding amicable solutions at the early stages of the infringement procedure. Yet, member states differ in terms of their bargaining profiles. In particular, states vary as to when in the infringement procedure they tend to settle cases. Some, in particular Denmark, but also the United Kingdom, the Netherlands and Spain, go to great lengths to close cases as early as possible in the procedure. Others, most notably Italy and Belgium, but also Greece and France, tend to persist in their violations and end up having a higher share of cases referred Table 6.2 Infringement cases closed in 1999 by stage in the procedures Stage Before formal notice Before reasoned opinion Before ECJ referral Before ECJ judgment Before second formal notice Before second reasoned opinion Before second ECJ referral Before ECJ sanctioning judgment Total Source: European Commission 2000.

Number

Percentage

763 593 435 40 46 12 10 1

40.2 31.2 22.9 2.1 2.4 0.6 0.5 0.1

1,900

100.0

94 Jonas Tallberg and Christer Jönsson to the ECJ. Germany, Luxembourg and Ireland represent the average EU bargaining profile. Package meetings as a form of direct and institutionalized negotiation over compliance are hailed by the Commission as a particularly effective way of settling non-compliance cases: ‘These meetings ensure that the situation is constantly under review and allow the Commission to bring extra pressure to bear on the competent national departments’ (European Commission 1992: 2). The effects of these direct negotiations are reflected in the Commission’s reports. In 1994, for instance, 74 of the 217 cases terminated in the area of free movement of goods were settled through package meetings, and in 1995 the corresponding figures were 60 out of 238 (European Commission 1995: 26, 1996a: 23). In sum, compliance bargaining between the Commission and member states, within the framework of the infringement and sanctioning procedures, has demonstrated a notable capacity to induce state conformance with EU rules and ‘to put an end to infringements . . . without actions necessarily having to be brought before the Court of Justice’ (European Commission 1992: ii). Compliance bargaining thereby serves to realize the principled agreements concluded in the Council or at IGCs. The infringement procedure has become an instrument of policy attainment.

Conclusion In this chapter we have explored the phenomenon of compliance bargaining in the European Union. The EU literature on negotiation and compliance has largely neglected the processes of bargaining that are central in the post-decisional phase of policy making. Typically, rule-conforming behaviour is not the result of a static decision to comply, but rather the product of dynamic processes of interaction between violators and guardians of the law. Drawing on the general analytical framework outlined in our previous work, this chapter has isolated the primary characteristics of compliance bargaining in the European Union – the structure of bargaining, the distribution of bargaining power and the effects of bargaining. More specifically, we have advanced two arguments. First, the dynamics of compliance bargaining in the European Union are decisively different from those of pre-decisional negotiations. Second, the particular institutional structure of the European Union provides for a form of compliance bargaining that is effective in addressing violations of EU law.

References Alter, Karen J. (2001) Establishing the Supremacy of European Law: The Making of an International Rule of Law in Europe, Oxford: Oxford University Press. Audretsch, H. A. H. (1986) Supervision in European Community Law, Amsterdam: NorthHolland. Börzel, Tanja A. (2000) ‘Why there is no “Southern Problem”: On Environmental Leaders and Laggards in the European Union’, Journal of European Public Policy 7 (1): 141–62.

Compliance bargaining

95

Chayes, Abram and Chayes, Antoina Handler (1995) The New Sovereignty: Compliance with International Regulatory Agreements, Cambridge: Harvard University Press. Dashwood, Alan and White, Robin (1989) ‘Enforcement Actions under Article 169 and 170 EEC’, European Law Review 14: 388–413. Downs, George W., Rocke, David M. and Barsoon, Peter N. (1996) ‘Is the Good News about Compliance also Good News about Cooperation?’, International Organization 50 (3): 379–406. Elgström, Ole and Smith, Michael (2000) ‘Introduction: Negotiation and Policy-Making in the European Union – Processes, System and Order’, Journal of European Public Policy 7 (5): 673–83. European Commission (1989) Sixth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law (1988), COM (89) 411 Final (21.12.1989). European Commission (1992) Ninth Annual Report on Commission Monitoring of the Application of Community Law (1991), COM (92) 136 Final (12.5.1992). European Commission (1993) Making the Most of the Internal Market: Strategic Programme. Communication from the Commission to the Council, COM (93) 632 Final (22.12.1993). European Commission (1994) The Community Internal Market: 1993 Report. European Commission (1995) The Single Market in 1994: Report from the Commission to the Council and the European Parliament, COM (95) 238 Final (15.6.1995). European Commission (1996a) The Single Market in 1995: Report from the Commission to the Council and the European Parliament, COM (96) 51 Final (20.2.1996). European Commission (1996b) Thirteenth Annual Report on Monitoring the Application of Community Law (1995), COM (96) 600 Final (29.05.1996). European Commission (1999) Sixteenth Annual Report on Monitoring the Application of Community Law (1998), COM (1999) 301 Final (09.07.1999). European Commission (2000) Seventeenth Annual Report on Monitoring the Application of Community Law (1999), COM (2000) 92 Final (23.06.2000). European Commission (2001) Eighteenth Annual Report on Monitoring the Application of Community Law (2000), COM (2001) 309 Final (16.07.2001). Evans, A. C. (1979) ‘The Enforcement Procedure of Article 169 EEC: Commission Discretion’, European Law Review 4: 442–56. Everling, Ulrich (1984) ‘The Member States of the European Community Before Their Court of Justice’, European Law Journal 9: 215–41. Friis, Lykke (1996) When Europe Negotiates: From European Agreements to Eastern Enlargement, Ph.D. dissertation, University of Copenhagen. Garrett, Geoffrey (1992) ‘International Cooperation and Institutional Choice: The European Community’s Internal Market’, International Organization 46: 533–60. International Negotiation (1998) Special issue on the theme ‘Negotiating the European Union’, 3 (3). Journal of European Public Policy (2000) Special issue on the theme ‘Negotiation and PolicyMaking in the European Union’, 7 (5). Jönsson, Christer and Tallberg, Jonas (1998) ‘Compliance and Post-Agreement Bargaining’, European Journal of International Relations 4 (4): 371–408. Keohane, Robert O. (1984) After Hegemony: Cooperation and Discord in the World Political Economy, Princeton: Princeton University Press. Kohler-Koch, Beate (1996) ‘Catching up with Change: The Transformation of Governance in the European Union’, Journal of European Public Policy 3 (3): 359–80. Krislov, Samuel, Ehlermann, Claus-Dieter and Weiler, Joseph (1986) ‘The Political Organs and the Decision-Making Process in the United States and the European Community’,

96 Jonas Tallberg and Christer Jönsson in Mauro Cappelletti, Monica Seccombe and Joseph Weiler (eds) Integration through Law. Volume 1: Methods, Tools and Institutions, Berlin: Walter de Gruyter. Laursen, Finn and Vanhoonacker, Sophie (eds) (1992) The Intergovernmental Conference on Political Union, Maastricht: European Institute of Public Policy. Lebow, Richard Ned (1996) The Art of Bargaining, Baltimore: Johns Hopkins University Press. Mesquita, Bruce Bueno de and Stokman, Frans N. (eds) (1994) European Community DecisionMaking: Models, Applications and Comparisons, New Haven: Yale University Press. Meunier, Sophie (2000) ‘What Single Voice? European Institutions and European Union– US Trade Negotiations’, International Organization 54 (1): 103–35. Moravcsik, Andrew (1998) The Choice for Europe: Social Purpose and State Power from Messina to Maastricht, Ithaca: Cornell University Press. Pescatore, Pierre (1974) The Law of Integration, Leyden: Sijthoff. Pollack, Mark A. (1997) ‘Delegation, Agency, and Agenda-Setting in the European Community’, International Organization 51 (1): 99–134. Pruitt, Dean G. and Rubin, Jeffrey Z. (1986) Social Conflict: Escalation, Stalemate, and Settlement, New York: Random House. Saunders, Harold H. (1985) ‘We Need a Larger Theory of Negotiation: The Importance of Pre-negotiating Phases’, Negotiation Journal 1 (3): 249–62. Siedentopf, Heinrich and Ziller, Jacques (eds) (1988) Making European Policies Work: The Implementation of Community Legislation in the Member States. Volume 1: Comparative Syntheses, London: Sage. Slaughter, Anne-Marie, Stone Sweet, Alec and Weiler, Joseph H. H. (eds) (1998) The European Courts and National Courts: Doctrine and Jurisprudence, Oxford: Hart Publishing. Snyder, Francis (1993) ‘The Effectiveness of European Community Law: Institutions, Processes, Tools and Techniques’, Modern Law Review 56: 19–54. Spector, Bertram I. and Zartman, I. William (eds) (2003) Getting It Done: Postagreement Negotiation and International Regimes, Washington DC: United Nations Institute of Peace Press. Stein, Janice Gross (ed.) (1989) Getting to the Table: The Processes of International Prenegotiation, Baltimore: Johns Hopkins University Press. Steiner, Josephine and Woods, Lorna (1996) Textbook on EC Law, London: Blackstone Press. Tallberg, Jonas (2002a) ‘Delegation to Supranational Institutions: Why, How, and with What Consequences?’, West European Politics 25 (1): 23–46. Tallberg, Jonas (2002b) ‘Paths to Compliance: Enforcement, Management, and the European Union’, International Organization 56 (3): 609–43. Tallberg, Jonas (2003) European Governance and Supranational Institutions: Making States Comply, London: Routledge. Tsebelis, George and Garrett, Geoffrey (2000) ‘Legislative Politics in the European Union’, European Union Politics 1 (1): 9–36. Young, Oran R. and Osherenko, Gail (1993) ‘Testing Theories of Regime Formation: Findings from a Large Collaborative Research Project’, in Volker Rittberger (ed.) Regime Theory and International Relations, Oxford: Clarendon Press. Zartman, I. William and Berman, Maureen R. (1982) The Practical Negotiator, New Haven: Yale University Press.

7

Negotiations in European Union committees Anders Sannerstedt

Introduction The theme of this book is that decision making in the European Union can be seen as a continuous flow of partly interconnected multilateral negotiations. The European Union as a negotiation milieu can be characterized as a mix of cooperative negotiation behaviour and the negotiating parties’ efforts at furthering their respective interests. However, even as we look for a specific EU negotiating style, we have to recognize that EU negotiations vary in several respects. We discover this when we compare negotiations in different issue areas. This chapter is comparative, too, but along another dimension: we will compare some aspects of negotiating behaviour in different types of working groups and committees within the EU political system.1 Our point of departure is the distinction between three major types of working groups and committees: expert groups, Council working groups, and the implementation committees labelled ‘comitology’ (cf. Schaefer 1996). The expert groups play a part in the Commission’s preparation of policy proposals. The Council’s decisions are prepared by the Committee of Permanent Representatives (COREPER), and many different working groups prepare the COREPER decisions (cf. Bal 1995; Hayes-Renshaw and Wallace 1997: 97–100). The comitology is part of the Commission’s implementation of EU decisions. Its role is dual, though, since it was also intended to be the member states’ instrument of control over the Commission in the implementation process (Dehousse 2003). Several different types of comitology committees can be discerned (advisory, managerial, regulatory); these committees have different competences and their rules for decision making vary (van der Knaap 1996; Dogan 1997). However, in this chapter we will largely ignore these differences (even though they appear theoretically important, cf. Steunenberg et al. 1996; Pollack 2003) and treat the comitology as one distinct type of EU committee. We do this because in our study we find very few differences between various types of comitology committees. Members of these three committee types are appointed according to one and the same principle. As a rule, each member state is represented in each committee. The committees can thus be seen as arenas where the member states interact. These

98 Anders Sannerstedt committees form – through their great number – an important part of the EU political machinery (cf. Falke 1996; Dogan 1997). Most Council decisions (estimations range between 70–80 per cent) are in fact made in the working groups (Beyers and Dierickx 1998: 290–1). They have, however, been given scant attention in political scientists’ research on decision making in the European Union (cf. e.g. Schaefer 1996; Dogan 2000: 46). All these expert groups, working groups and comitology committees show variation as to their degree of institutionalization. Some are permanent, and some are temporary. Some meet often, others seldom. Some have a budget of their own, while others have not (Van Schendelen 1998). In this chapter these three types of EU committees are compared. The comparison is focused on one particular aspect of these committees – negotiations – and we compare negotiation styles and negotiation cultures. One could of course expect that such a comparison would reveal few differences, since all these committees belong to the same EU system. However, there are at least two reasons for expecting variations between the committees. First, two types of committees are linked to the Commission (expert groups and the comitology), while the third is directly involved in the workings of the Council (working groups). Differences between the way the Commission and the Council work might presumably be reflected in the negotiations of the committees as well. Second, these three types of committees represent three different stages in the decision-making process. If we make a simple distinction between three stages – a preparatory, a decision and an implementation stage – we realize that the expert groups are part of the preparatory stage, the Council working groups belong to the decision stage and the comitology is part of the implementation stage. Christiansen and Kirchner (2000: 7) make a similar distinction between agenda setting, decision making and implementation; but preparation is more than just agenda setting. Thus we have good reasons to justify a comparison between the three types of EU committees. Our study is based upon a questionnaire sent to Swedish members of these groups and committees. This data source has some obvious limitations. Swedish delegates might differ from those of other countries – Sweden being a small state and, at the time of data collection, a new member of the European Union, and a fairly reluctant one as well. This is of less importance, though, when our main purpose is a comparison between different types of committees. Ideally, an analysis would require that similar questionnaires be sent to delegates from other countries; it would also require in-depth interviews and case studies of specific decision-making processes. Since few systematic studies of these groups and committees have been conducted, it seems justifiable to offer a minor contribution – based upon the experiences and attitudes of Swedish delegates – to the study of these groups. The survey was conducted in 1998. Swedish representatives were asked about the activity in these groups and committees, with a focus on negotiation aspects. Some results from this survey have already been reported (Elgström and Jönsson 2000; Elgström et al. 2001).

Negotiations in EU committees

99

Aspects of EU committee negotiations In this chapter we highlight some specific aspects of the negotiating milieu in these groups and committees. We start by addressing the question of the autonomy of the members of the committees. If tightly controlled by their governments, these members can be seen as representatives of their countries. The committee then could be seen as an arena where the member states negotiate, and the committee members would tend to see each other as representatives of member states. If, on the other hand, the committee members have a high degree of autonomy, they can be seen as fairly autonomous actors in the process. We may hypothesize that autonomy varies with the stages in the decision-making process, with least autonomy in the decision stage, and most autonomy in the early preparatory phase. A second aspect concerns negotiating styles. In negotiation theory a classic distinction is between cooperative and competitive negotiating behaviour (cf. Walton and McKersie 1965). If the negotiating parties choose a cooperative negotiating style (sometimes labelled ‘integrative bargaining’), the negotiation assumes the character of a positive-sum game. Should the parties instead adopt a competitive style (sometimes labelled ‘distributive bargaining’), the negotiation is perceived as a zero-sum game. It is often asserted that if the parties meet in many repeated negotiations, they are more likely to adopt a cooperative negotiating style. However, the choice of negotiating style also seems related to the degree of conflict between the parties (Israelis and Palestinians are not likely to negotiate cooperatively, even if the parties meet in many negotiations). The distinction between cooperative and competitive bargaining should be seen as a continuum, rather than a dichotomy. Most negotiations are characterized by both cooperative and competitive elements, to varying degrees (cf. Sannerstedt 1992; Elgström and Jönsson 2000). Considering that the European Union is commonly understood as a cooperative enterprise, the hypothesis would be that a fairly cooperative negotiating style prevails in all types of committees. A third aspect concerns coalition building in the European Union. The complexity of multilateral negotiations can be reduced when the negotiating parties form coalitions (cf. Zartman 1994). From the individual actor’s point of view, entering a coalition means a gain in negotiating power (but a reduction of flexibility). In the European Union, a North–South coalition pattern sometime emerges, in contrast to a more general norm of issue-specific rather than permanent coalition patterns. A fourth aspect concerns leadership in EU negotiations. Leadership in multilateral negotiation is a means of facilitating agreement through the reduction of complexity (cf. Zartman 1994) and the avoidance of deadlock. The Presidency is expected to act as a leader in the Council working groups, while the Commission is expected to exert leadership in comitology and in expert groups. It is not well known, though, to what extent we can see overlapping leadership. Nor has the style of leadership in EU negotiations been subjected to much systematic investigation in previous research.

100 Anders Sannerstedt A fifth aspect concerns the committee members’ external contacts. Is the activity in the committee confined to interaction between committee members, or are external actors given the opportunity to influence the process? The European Union is sometimes described as ‘a closed and elitist system’ (cf. Eriksen and Fossum 2002: 404–5), but if lobbyists and experts participate, the process is not really closed (but may still be elitist). These five aspects will be discussed separately, but first a few comments on the data material seem warranted. There exists no reliable account of the myriad groups and committees in the European Union. The Swedish National Board of Trade, Kommerskollegium, had a database of Swedish members in groups and committees. This list, which comprised more than 1,000 committees, was used in our study to identify the population of potential survey respondents. However, the list was not exhaustive and some information was outdated. Many individuals served on two or three committees, and the survey was sent to 421 persons. The response rate was slightly above 65 per cent. Most of the respondents served on one of the three group types under study. Autonomy The autonomy of the individual committee member from their government can be measured in several ways. Has the government appointed them? How important is the ministry as a source of information? How frequently do they have contacts with the ministry? Has a negotiation mandate been set by the government? A comment should be made on the issue of appointment. In Sweden, a clear distinction can be made between the government and the ministries, on the one hand, and the public administration with strong central boards, on the other. Law rules public administration, without much direct governmental intervention. Therefore it makes sense to investigate whether the Swedish delegates have been appointed by the ministry or by a central board. Although the responses to this question cannot be directly translated to other member states, they follow the pattern of responses to the other questions. Indeed, a clear pattern is found (Table 7.1). Members of Council working groups have least autonomy, while members of the expert groups seem to enjoy most freedom. Indeed, a highly placed governmental official, when informed of this relative freedom of the expert groups, commented at least half-seriously: ‘We will have to correct that as soon as possible.’ A more indirect way of measuring autonomy is to ask the question: how important is it for you to represent Swedish interests? Here, the answer ‘very important’ is the common one, but when comparing different types of committees we once more find the same pattern.2 The differences in autonomy between the three types of committees yield a systematic pattern. Certainly, the validity of each question as an indicator of autonomy could be questioned. The validity of both the number of contacts with the ministry and the importance of the ministry as a source of information can perhaps be doubted. However, since we employ a whole set of indicators and get a

Negotiations in EU committees 101 Table 7.1 Autonomy of members of expert groups, Council working groups, and comitology committees, measured in several ways (high values indicate low autonomy) Expert groups

Working groups

Comitology committees

Respondents appointed by the government (%)

29

64

50

Respondents who say that the importance of ministry as source of information is high or very high (%)

23

87

43

Respondents who say that they are to a high degree restricted by a negotiation mandate from home (%)

35

77

47

Respondents who say that representing Swedish interests is very important (%)

56

81

69

5

100

11

Average number of contacts/year with ministry (median values)

very consistent response pattern, the result seems reliable. Indeed these results confirm our expectations. Members of Council working groups are clearly representatives of the government, and their work is made in close contact with the ministry back home. As a general rule in policy-making studies, freedom of manoeuvre seems highest at the initial stage of the decision-making process, and lowest at the decision stage. Cooperation and competition One important idea behind the grandiose effort at European cooperation is that national interests should be set aside and all actors should strive for all-European interests. However, our survey gives the distinct result that negotiations in groups and committees seem to be dominated by national interests. This situation is both recognized and generally accepted. To a large extent, members of groups and committees represent their home country, and furthering national interests constitutes an important part of their activity. Promoting the interests of one’s country ranks particularly high in Council working groups, considerably less so (although still high) in expert groups, with the comitology in between (Table 7.2). This seems directly related to the autonomy of the committee members: less autonomy means more focus on national interests. This could have important effects on the negotiating style in the committees. The emphasis on the promotion of national interests might result in negotiating behaviour that is competitive rather than cooperative. Yet this focus on member states’ interests does not seem to develop into a competitive negotiating style. On the contrary, 86 per cent of the respondents report that the work in the committee

102 Anders Sannerstedt Table 7.2 Furthering national interests in expert groups, Council working groups and comitology committees Questions: 1 How often do these attitudes exist in your group? National interests are most important. Per cent of the respondents who answer ‘very often’ or ‘fairly often’. 2 How often does this behaviour exist in your group? Stating one’s country’s position. Per cent of the respondents who answer ‘very often’ or ‘fairly often’. 3 When there is conflict between different interests, how accurately does the following statement describe the work in your group? The group pays consideration to specific country interests. Per cent of the respondents who answer ‘very accurate’ or ‘fairly accurate’. Expert groups

Working groups

Comitology committees

National interests are most important

71

96

90

Stating one’s country’s position

63

99

92

Specific country interests taken into consideration

63

89

65

is characterized by a cooperative atmosphere to a very high degree or a fairly high degree. Here almost no differences exist between various types of committees. Accordingly, 88 per cent report mutual trust between committee members to a very high or fairly high degree. When asked explicitly about common problem solving, however, comitology committee respondents report a lower degree of cooperation: 58 per cent say that common problem solving occurs very often or fairly often, as compared to 77 per cent in the working groups and 81 per cent in the expert groups. Strategies normally associated with competitive bargaining behaviour are seldom reported. Six per cent of the respondents say that threats are made very or fairly often. Four per cent say that misleading behaviour (e.g. bluffs) occurs very or fairly often. However, refusals to make concessions are more common (27 per cent of the respondents estimate that this happens very often or fairly often). A refusal to make concessions merely means an insistence on a position and should be seen as much less aggressive than threats or bluffs. On the other hand, unilateral concessions seem to be equally common (25 per cent of the respondents answer that this occurs very often or fairly often). Whenever there is a conflict within the committee, it seems that sincere efforts are made to negotiate a solution. The members of the committee seem keen on identifying common interests. Efforts are also made to take the interests of each country into account. When facing a conflict of interests, strong efforts are made to reach a solution that all parties can accept. This behaviour seems especially accentuated in the decision phase, i.e. in the Council working groups (Table 7.3). Efforts to negotiate conflicts and take different interests into account do not necessarily lead to unanimous agreements. The EU decision rules allow for

Negotiations in EU committees 103 Table 7.3 What characterizes negotiation behaviour in conflict situations? Per cent of the respondents who answer: ‘very often’ or ‘fairly often’. Expert groups

Working groups

Comitology committees

Efforts at finding common interests

77

84

71

Taking specific country interests into consideration

63

89

65

Great understanding of the opposing party

62

80

64

Increased openness concerning one’s own position

56

72

53

majority voting on many issues. To what extent do the negotiators strive for unanimity? Most members report that very or fairly often the committee strives for unanimity; however, the frequency is somewhat lower in the comitology committees. The negotiation process is not always smooth, of course. Deadlock is reported to occur very often or fairly often by more than a quarter of the respondents, although somewhat more seldom in the comitology committees (Table 7.4). In the end a solution is found, which almost always seems to be close to the position of the majority. Very often the solution is some type of compromise, although this happens less often in the comitology committees. Only rarely is a member allowed to slow down the process, especially in the comitology committees; laggards seem to have limited influence. It happens now and again though that some member is forced to give in (Table 7.5). The commitment to the goal of unanimity is not only lip-service. Voting down a minority rarely happens; in fact, voting is rare. Many of the respondents cannot even give a correct account of the decision rule in their committee (a question was posed in the survey) – a striking fact that indicates that consensus as a decision rule prevails. Decision rules are necessary only when the striving for consensus fails. In fact, our expectation was that the decision rule would affect negotiations in the committee substantially; decision rules imply a BATNA (Best Alternative To a Negotiated Agreement), and negotiating actors always have to consider their BATNA (Fisher and Ury 1991). But on this point our expectations were not corroborated. Obviously, voting is not considered to be a good alternative to negotiated agreement. The fact that a decision is made by consensus does not necessarily mean that everybody is satisfied with the decision. Yet no formal dissent remains. A difference exists between the working groups, on the one hand, and expert groups and comitology committees, on the other. Disagreement in a working group on an issue just means that the issue is handed over to the COREPER. Even if the differences in Table 7.5 are minor, one might conclude that the work in the comitology committees is characterized by a somewhat higher degree of

104 Anders Sannerstedt Table 7.4 What happens in the group when conflict between different interests occurs? Per cent of the respondents who answer ‘very often’ or ‘fairly often’. Expert groups

Working groups

Comitology committees

The group tries to reach unanimity

91

86

78

Deadlock occurs

30

31

24

Table 7.5 When conflict of interest occurs in your group, what is the outcome? Per cent of the respondents who answer ‘very often’ or ‘fairly often’. Expert groups

Working groups

Comitology committees

A solution close to the position of the majority

93

96

93

A compromise where everybody gets something and gives something away

79

85

70

A solution close to the position of the slowest member

22

20

13

A solution which makes everybody better off

50

28

35

Some member state is forced to give in

30

25

31

‘impatience’ than in the expert groups and the Council working groups. Unanimity is not as common a goal, deadlocks are less acceptable, compromise is less common and those who try to stop or delay the process have less to gain. This can be related to the special task of the comitology committees: implementing decisions that have already been made, maybe through a cumbersome negotiation process. Thus the time pressure is higher in the comitology; the process of hammering out details cannot be allowed to become too lengthy. When the clock is ticking, unanimity loses some of its importance and attraction. Still, voting seems rare, and knowledge of the voting rules is lacking. Dehousse (2003: 802) makes the same observation. Collective problem solving includes both a technical and a political aspect. The technical aspect concerns finding the best possible solution, and the political aspect concerns getting everybody to accept this solution. One might expect that the technical aspect is somewhat more accentuated in the expert groups, and this seems also to be the case, although the differences between expert groups and other committees are not particularly great (Table 7.6). The Swedish word ‘saklighet’, mentioned in the table, is not easily translated into English; expressions like ‘objectivity’ and ‘matter-of-factness’ do not catch the full meaning of the Swedish word. The few existing previous studies present similar results. Wessels (1998: 224–5) emphasizes that committee members have a ‘business-like approach’ and a tendency towards consensus, an esprit de corps and a ‘propensity towards informal

Negotiations in EU committees 105 Table 7.6 The spirit of ‘saklighet’ Per cent of the respondents who answer ‘very often’ or ‘fairly often’. Expert groups

Working groups

Comitology committees

The most important thing is to find the technically best solution

91

69

68

The work in the group is characterized by a spirit of expertise

89

75

87

behaviour’. Beyers and Dierickx (1998) draw a similar picture of the Council and its working groups. Beyers and Dierickx (1998: 292) also found that members of EU committees tended to view themselves less as national representatives, and more as members of an expertise-oriented collective. In this study we found a greater emphasis on members as representatives of member states. This is more in line with Beyers (1998: 380). Maurer et al. (2000: 26) even stress that ‘expert groups advise the Commission on the basis of the member States interests’. Our results indicate that this might be a slight oversimplification as far as the expert groups are concerned. We can now conclude that negotiations in EU committees take on a cooperative and problem-solving character. Mutual trust and mutual understanding seem to characterize the relations between members of a committee. The use of competitive negotiating strategies such as threats, bluffs and cheating seems rare. Unanimous solutions are sought, even when a decision could have been taken by majority vote. The other side of the coin, however, is that national interests seem to play a prominent role in the negotiations. Member states declare their positions, their interests are considered legitimate by the other parties and the negotiation process often takes the shape of finding a compromise whereby the interests of all member states can be accommodated. Thus, our picture deviates slightly from the idea of deliberative politics ( Joerges and Neyer 1997; Joerges 2002; cf. Eriksen and Fossum 2002), since a strong emphasis on national interests does not fit well with the idea of deliberation. Nor does it fit with the frequent accusation of ‘technocracy’ (cf. Joerges 2002) because of the strong emphasis on negotiation and national interests. Our results might, of course, be considered an effect of the survey method: the respondents might be tempted to present a polished and politically correct picture, where the cooperative aspect gets overemphasized. However, other observers have observed the norm of ‘consensus at (almost) any cost’ (Peterson and Bomberg 1999: 58) and, furthermore, the case studies and the interviews conducted within our research project point in the same direction: negotiations in the European Union are indeed a mix of cooperative problem solving and the promotion of national interests and national positions. The success of the European Union as a cooperative effort indicates that the member states are good at balancing national interests with cooperation.

106 Anders Sannerstedt Coalitions in EU negotiations Negotiations in EU committees are multilateral.3 Member states interact in the expert groups, the Council working groups and the comitology committees. In the study of multilateral negotiations, it is stressed that the complexity of the negotiation increases with the number of negotiating parties (Zartman 1994; Dupont 1994). A negotiator has to take into consideration many parties with different interests and different positions, and this calculus soon gets complex and uncertain. A common way to reduce the complexity and uncertainty of multilateral negotiations is to form coalitions. This is indeed one of the reasons why political parties once emerged in national parliaments. Not only does coalition building reduce uncertainty and complexity in multilateral negotiations. Forming a coalition brings political power; a large coalition has more power than a single individual. However, coalitions also mean the formation of alliances and counter-alliances and thus the accentuation of differences between various coalitions, between ‘us’ and ‘them’. This is no problem in national parliaments, where the division between majority and opposition is standard. The European Union, however, is a system of cooperation between equals, the member states. A permanent division into majority and opposition would ruin the cooperative project. The solution might be issue-specific, floating coalitions. Ideally, each member state would have the same probability of entering into a coalition with any other member state. Research on coalitions has focused on political contexts other than multilateral negotiations, with a special focus on government formation. This research has aimed at predicting coalition outcomes, from a rational-choice perspective. The theory often starts with simple assumptions about the actors’ utility, for example, that actors are power maximizers. From these assumptions, simple hypotheses can be deduced, as the well-known theorem of the minimal winning coalition (Riker 1962). When other goals are added, the actor’s calculus quickly becomes more complex. In many contexts, such as government formation, the actors tend to strive for durable coalitions. On other occasions, coalitions are issue specific; in those cases coalition patterns are more fluid. The US Congress might serve as an example. As for coalitions in the European Union, a case can be made for two contradictory hypotheses. Since the European Union is a cooperative project, one might predict unstable and fluid coalition patterns. A more permanent division into a majority and an opposition would hurt the idea of a cooperative project. However, we also encounter the idea of a North–South divide (e.g. Beyers 1998: 383). Two reasons are given for this divide. One is cultural: neighbouring countries have closer cooperation and cultural affinity. Language differences might be of minor importance only. The other reason is interest based. North and South have different interests to do with such things as economic differences. In many contexts, coalitions are public and explicit. In the European Union, however, coalition patterns are more subtle. The coalition behaviour of member states might consist of the search for common interests and some adjustment of a

Negotiations in EU committees 107 country’s position. It might be unclear whether or not a coalition has actually been formed. Very little research has been made on coalitions in the European Union. Observations are sometimes made in passing (e.g. Sbragia 1996). Getting the necessary data is cumbersome, especially if the minority eventually gives up, and the final decision is made unanimously. The classic rational-choice based coalition theory is also of limited relevance to the study of EU coalitions. We have investigated coalition patterns in EU committees with two survey questions. One is explicitly about coalitions: With what countries do you cooperate most in your committee? The other question is about networking: When you have contacts with other members of your group, which countries do they represent? The responses follow a stable pattern, and the two questions seem to measure the same phenomenon. One might expect different coalition behaviour in different phases of the decision-making process, but this proves not to be the case. No differences at all between different committee types can be found, only a striking similarity. The pattern in Table 7.7 is remarkable. Swedish committee members cooperate first and foremost with their Nordic neighbours, then with the United Kingdom, then, although to a lesser degree, with Germany and the Netherlands. Next follow France, Austria and Ireland. Cooperation with the Mediterranean countries seems quite rare. This pattern resembles the North–South divide. Both survey questions produce the same result, and there seems to be no important differences between the three types of committees. We have no information on coalition patterns in general in the committees, only of Sweden’s coalition partners. A previous investigation of the Council (Beyers Table 7.7 Sweden’s coalition patterns in different types of committees Per cent of the respondents who mention cooperation, and contacts, with representatives from different countries (abbreviated names of member states). Expert groups

Working groups

Comitology committees

Cooperation

Contacts

Cooperation

Contacts

Cooperation

Contacts

77 Den 75 Fin 60 UK 34 Neth 28 Ger 12 Fra 12 Ire 11 Aust 9 Bel 3 Por 2 Ita 2 Lux 2 Gre 0 Spa

69 Den 68 Fin 52 UK 35 Ger 29 Neth 17 Fra 12 Ire 11 Aust 9 Ita 9 Bel 3 Por 2 Gre 2 Spa 0 Lux

77 Den 70 UK 69 Fin 41 Neth 30 Ger 14 Fra 13 Ire 13 Aust 4 Bel 4 Por 1 Spa 1 Lux 0 Ita 0 Gre

77 Den 68 Fin 54 UK 44 Neth 35 Ger 24 Fra 23 Aust 18 Ire 17 Ita 15 Bel 14 Spa 13 Lux 13 Por 10 Gre

81 Den 80 Fin 74 UK 41 Neth 31 Ger 22 Aust 11 Ire 6 Bel 5 Por 4 Fra 3 Spa 2 Gre 2 Ita 0 Lux

75 Fin 74 Den 62 UK 34 Neth 21 Ger 20 Fra 16 Aust 10 Ire 7 Bel 6 Ita 5 Spa 3 Gre 3 Por 2 Lux

108 Anders Sannerstedt and Dierickx 1998) offers a similar pattern: neighbours tend to have close contacts, the three big countries and the country that holds the Presidency have more contacts than others. Leadership Leadership can be an important ingredient in multilateral negotiations for at least three reasons. First, leadership might be a way to reduce complexity. Second, when a deadlock is looming or has already occurred, leadership might be a way out of the situation. Third, leadership can be exerted to push a negotiation process towards a solution. We can distinguish between leaders and laggards (cf. Sbragia 1996). Leadership might entail that the leader gets acceptance for a policy innovation. As a leader, a party (or a coalition of parties) has several possibilities to promote progress in the negotiation. Leaders can point to a way out of deadlock, act as a mediator or use their power resources to induce threats or promises (e.g. side payments). The study of leadership is multidisciplinary. Leadership is often seen as a relation between leaders and followers. Common research foci have been typologies and strategies of leadership. The study of leadership in the specific context of multilateral negotiations is still underdeveloped. Efforts at developing typologies have, however, been made (Underdal 1994; Malnes 1995). A precondition for leadership is that the other parties accept leaders in their leadership role. Several studies have been made concerning acceptance of a mediator, but leadership and mediation are two different, although related, concepts. In any event, two factors seem to be of decisive importance. First, the formal position of the leader: in international negotiations, the chairman, or the secretary general or maybe the host country could assume leadership. Second, an actor with strong power resources is considered more likely to emerge as a leader; for instance, the United States of America has often taken a leadership role in international negotiations. Originally, the European Community had only six member states, and the need for leadership was limited. As the number of member states increased, the need for leadership grew. When conducting our survey, we assumed that two actors were likely to assume leadership roles: the Presidency and the Commission. The country that holds the Presidency of the Council is likely to be accepted as a leader in the working groups (cf. Metcalfe 1998). The Commission may act as a leader of the comitology, to some extent in the expert groups and possibly in the Council working groups as well. When there is conflict between member states, the Commission could be seen as a fairly impartial mediator. The Commission can also be expected to exert more active leadership because of its right to initiative and its responsibility to further the process of European integration. At certain points in the history of the European Union, the Commission has performed active and dynamic leadership, for instance when Jacques Delors was President of the Commission. At other times, the leadership of the Commission seems to have been more directed towards consolidation of what has already been achieved.

Negotiations in EU committees 109 The role of the Presidency of the Council seems to have increased in importance in later years, in connection with increased expectations from member states, EU institutions, and the public. Our survey confirms that the Presidency is considered important. In fact, the Presidency seems to play a role outside the Council as well (cf. Tallberg 2003). The emerging picture is that of overlapping leadership between the Commission and the Presidency in all three types of groups and committees. In our survey we assumed that leaders might facilitate negotiation in various ways, but also further their own interests by launching their own proposals, and sometimes speeding up or slowing down the process. We therefore asked several questions concerning the leadership of both the Commission and the Presidency. The data can be used to analyze the leadership style of the Commission and the Presidency. We can therefore compare a leader’s activity between the three types of committees. The Commission seems to be especially active in the comitology phase (Table 7.8), which is an expected result. However, the differences between the different types of committees are insignificant, and it is especially noteworthy that the Commission is seen as an active leader in the Council working groups as well. In general, the Commission is perceived to act more as a facilitator or a driving force than as a braking force. It should be noted, however, that the comitology has been conceived as a way for member states to control the Commission’s implementation of EU decisions. The fact that the Commission is seen as a strong leader, much stronger than the Presidency, leads to the conclusion that the comitology’s control of the Commission is rather limited. This is in line with the findings of Dehousse (2003). The Presidency’s leadership, as expected, is most accentuated in the Council working groups (Table 7.9). However, the Presidency seems fairly active in expert groups and comitology as well, which was less predictable. The role of mediator in the working groups is striking; an impressive 56 per cent say that this role is ‘very common’. Mediation thus seems to be a basic ingredient in the leadership role of the Presidency. Table 7.8 How common is it that the Commission acts in the following roles at the meetings in your group? Per cent of the respondents who answer ‘very common’ or ‘fairly common’.

Clarifier Mediator Driving force Slowing force Promoting own proposals Manipulating the agenda

Expert groups

Working groups

Comitology committees

78 45 75 33 64 25

84 37 66 12 75 7

86 63 85 23 86 21

110 Anders Sannerstedt Table 7.9 How common is it that the Presidency acts in the following roles at the meetings in your group? Per cent of the respondents who answer ‘very common’ or ‘fairly common’. Expert groups

Working groups

Comitology committees

Clarifier

47

84

48

Mediator

40

92

38

Driving force

52

92

49

Slowing force

53

28

30

Promoting own proposals

42

58

24

Manipulating the agenda

4

15

8

We also gave the respondents the opportunity to mention other leaders than the Commission and the Presidency. About half of them say that other leaders sometimes step forward. When asked to mention member states, most often the three big states are mentioned. The three major countries seem to have a special potential for acting in a leadership role (cf. Sbragia 1996). Asked what qualities are important for a leader in the European Union, respondents mention four equally often: issue-specific expert knowledge, political skill, experience and good networking. These are basically personal qualities. The nationality of leaders is rarely mentioned in this context. Contacts with other actors In this study member states emerge as the important actors in EU committees. However, committee members also have contacts with other actors. We asked some questions about these contacts and their importance (Table 7.10). The general picture is that these contacts are more frequent and considered more important in the early phase, i.e. the expert groups, while the actors tend to distance themselves from outside influence during the decision stage, i.e. in the Council working groups. Contacts with members of the European Parliament are considered fairly unimportant. The Parliament seems to play a marginal role in the work of groups and committees (cf. Dogan 1997: 32). In fact, the European Parliament has raised criticism of the way the comitology system works (Dogan 1997; Garman and Hilditch 1998). Members of Council working groups have fewer outside contacts. Contact with independent experts is, as expected, highest among members of expert groups. We asked about the nature of contacts with Commission staff. The usual answer was that these contacts concerned exchange of technical information and, as a rule, were not considered part of any coalition behaviour. It seems fair to assume that the longer you work in a committee, the wider your network becomes. However, this assumption was not verified by the responses.

Negotiations in EU committees 111 Table 7.10 Importance of contacts Per cent of the respondents who answer ‘very great importance’ or ‘fairly great importance’. Expert groups

Working groups

Comitology committees

Commission staff

79

74

93

Interest organizations

59

32

63

Independent experts

65

16

42

European MPs

11

14

11

Table 7.11 Sources of information Per cent of the respondents who answer ‘very great importance’ or ‘fairly great importance’. Expert groups

Working groups

Comitology committees

Commission

80

87

88

Swedish authorities

71

58

71

Government ministries

23

87

43 14

Swedish representation in Brussels

9

50

Interest organizations

38

18

37

Independent experts

46

5

24

Newcomers have the same contact patterns as their more experienced colleagues. However, the frequency of meetings seems to matter: if the committee meets often, the number of contacts is greater. We see (Table 7.11) that Council working groups have frequent contacts with the home government (and its representation in Brussels) but few other outside contacts. When a decision is imminent, the decision makers tend to isolate themselves from outside pressure. The task of interacting with other member states in order to reach an agreement is complicated enough; opinions from other actors would only complicate matters further.

Conclusions Our study points to both similarities and differences when the three committee types are compared. A distinct EU negotiation culture seems to be present in all types of committees. Negotiations are cooperative and founded on facts. Meeting and negotiating frequently, committee members develop mutual trust. National interests dominate the negotiations, and this is commonly accepted. Hostility is largely absent in the process, and aggressive negotiation strategies are seldom applied. Instead, there is a strong tendency towards reaching unanimous agreement.

112 Anders Sannerstedt Specific country interests get accommodated to some degree; yet, as a rule, the agreement is a compromise close to the majority position. In many committees the members are not familiar with the voting rules, a fact that indicates that decisions are made by consensus and not by vote. Still, differences exist between the three types of committees investigated. The expert groups seem less politicized and more dominated by issue-specific expertise. The members of these groups have more autonomy from their governments, and they have more external contacts, with interest groups, lobbyists and independent experts. In the Council working groups, national interests play an important role in the negotiations. Group members represent their governments, put forward the positions of the government and negotiate with representatives of the other member states. Negotiations are geared towards reaching agreement, taking national interests into account. Contacts with external actors are held at a minimum. In the comitology committees the tempo is higher. Unanimity is less crucial, and if unanimous agreement is not reached, a majority decision is more readily accepted. In sum, the differences found are a matter of degree, and our conclusions are merely tentative. Further studies of negotiations in EU committees are warranted.

Notes 1 This chapter draws on other articles produced within the research project, especially Elgström and Jönsson 2000; Elgström et al. 2001, and the unpublished papers by Johansson 1999 and Wienecke 1999. 2 Numbers of respondents are not indicated in the tables, since the number of missing responses varies slightly. Generally, the number of respondents answering the survey questions are 50–55 (expert groups), 60–65 (working groups), and 80–85 (comitology). 3 This section is based on Elgström et al. 2001.

References Bal, L. J. (1995) Decision making and Negotiations in the European Union, discussion paper, Leicester: Centre for Study of Diplomacy. Beyers, J. (1998) ‘How Supranational is Supranationalism? National and European Socialization of Negotiators in the Council of Ministers’, Acta Politica, 33: 378–408. Beyers, J. and Dierickx, G. (1998) ‘The Working Groups of the Council of the European Union: Supranational or Intergovernmental Negotiations’, Journal of Common Market Studies, 36: 289–318. Christiansen, T. and Kirchner, E. (2000) ‘Introduction’, in T. Christensen and E. Kirchner (eds) Committee Governance in the European Union, Manchester: Manchester University Press. Dehousse, R. (2003) ‘Comitology: Who Watches the Watchmen?’, Journal of European Public Policy, 10: 798–813. Dogan, R. (1997) ‘Comitology: Little Procedures with Big Implications’, West European Politics, 20 (3): 31–60. Dogan, R. (2000) ‘A Cross-Sectoral View of Comitology: Incidence, Issues and Implications’, in T. Christensen and E. Kirchner (eds) Committee Governance in the European Union, Manchester: Manchester University Press.

Negotiations in EU committees 113 Dupont, C. (1994) ‘Coalition Theory: Using Power to Build Cooperation’, in I. W. Zartman (ed.) International Multilateral Negotiation: Approaches to the Management of Complexity, San Francisco: Jossey-Bass. Elgström, O. and Jönsson, C. (2000) ‘Negotiation in the European Union: Bargaining or Problem-Solving?’, Journal of European Public Policy, 7: 684–704. Elgström, O., Bjurulf, B., Johansson, J. and Sannerstedt, A. (2001) ‘Coalitions in European Union Negotiations’, Scandinavian Political Studies, 24: 111–28. Eriksen, E. O. and Fossum, J. E. (2002) ‘Democracy through Strong Publics in the European Union?’, Journal of Common Market Studies, 40: 401–24. Falke, J. (1996) ‘Comitology and Other Committees: A Preliminary Empirical Assessment’, in R. H. Pedler and G. F. Schaefer (eds) Shaping European Law and Policy: The Role of Committees and Comitology in the Political Process, Maastricht: European Institute of Public Administration. Fisher, R. and Ury, W. (1991) Getting to Yes: Negotiating Agreement Without Giving In, 2nd edn, New York: Penguin. Garman, J. and Hilditch, L. (1998) ‘Behind the Scenes: An Examination of the Importance of the Informal Processes at Work in Conciliation’, Journal of European Public Policy, 5: 271–84. Hayes-Renshaw, F. and Wallace, H. (1997) The Council of Ministers, London: Macmillan. Joerges, C. (2002) ‘Deliberative Supranationalism: Two Defences’, European Law Journal, 8: 133–51. Joerges, C. and Neyer, J. (1997) ‘From Intergovernmental Bargaining to Deliberative Political Processes: The Constitutionalisation of Comitology’, European Law Journal, 3: 273–99. Johansson, J. (1999) ‘Förekomsten av koalitioner inom EU:s expert-, rådsarbetsgrupper och komitologikommittéer’, unpublished paper, Lund: Department of Political Science. van der Knaap, P. (1996) ‘Government by Committee: Legal Typology, Quantitative Assessment and Institutional Repercussions of Committees in the European Union’, in R. H. Pedler and G. F. Schaefer (eds) Shaping European Law and Policy: The Role of Committees and Comitology in the Political Process, Maastricht: European Institute of Public Administration. Malnes, R. (1995) ‘“Leader” and “Entrepreneur” in International Negotiations: A Conceptual Analysis’, European Journal of International Relations, 1: 87–112. Maurer, A., Mittag, J. and Wessels, W. (2000) ‘Theoretical Perspectives on Administrative Interaction in the European Union’, in T. Christensen and E. Kirchner (eds) Committee Governance in the European Union, Manchester: Manchester University Press. Metcalfe, D. (1998) ‘Leadership in European Union Negotiations: The Presidency of the Council’, International Negotiation, 3: 413–34. Peterson, J. and Bomberg, E. (1999) Decision Making in the European Union, London: Macmillan. Pollack, M. A. (2003) ‘Control Mechanism or Deliberative Democracy? Two Images of Comitology’, Comparative Political Studies, 36 (1–2): 125–55. Riker, W. J. (1962) The Theory of Political Coalitions, New Haven: Yale University Press. Sannerstedt, A. (1992) Förhandlingar i riksdagen, Lund: Studentlitteratur. Sbragia, A. (1996) ‘Environmental Policy’, in H. Wallace and W. Wallace (eds) PolicyMaking in the European Union, Oxford: Oxford University Press. Schaefer, G. F. (1996) ‘Committees in the EC Policy Process: A First Step Towards Developing a Conceptual Framework’, in R. H. Pedler and G. F. Schaefer (eds) Shaping

114 Anders Sannerstedt European Law and Policy: The Role of Committees and Comitology in the Political Process, Maastricht: European Institute of Public Administration. Steunenberg, B., Koboldt, C. and Schmidtchen, D. (1996) ‘Policymaking, Comitology, and the Balance of Power in the European Union’, International Review of Law and Economics, 16: 329–44. Tallberg, J. (2003) ‘The Agenda-Shaping Powers of the EU Council Presidency’, Journal of European Public Policy, 10: 1–19. Underdal, A. (1994) ‘Leadership Theory: Rediscovering the Arts of Management’, in I. W. Zartman (ed.) International Multilateral Negotiation: Approaches to the Management of Complexity, San Francisco: Jossey-Bass. Van Schendelen, M. C. P. M. (1998) ‘Prolegomena to EU Committees as Influential Policymakers’, in M. C. P. M. Van Schendelen (ed.) EU Committees as Influential Policymakers, Aldershot: Ashgate. Walton, R. and McKersie, R. A. (1965) A Behavioral Theory of Labor Negotiation, New York: McGraw-Hill. Wessels, W. (1998) ‘Comitology: Fusion in Action. Politico-administrative Trends in the EU System’, Journal of European Public Policy, 5: 209–34. Wienecke, U. (1999) ‘Governance by Committee: Agenda Setting, Decision Making and Implementation in the European Union’, unpublished paper, Lund: Department of Political Science. Zartman, I. W. (1994) ‘Introduction: Two’s Company and More’s a Crowd: The Complexities of Multilateral Negotiation’, in I. W. Zartman (ed.) International Multilateral Negotiation: Approaches to the Management of Complexity, San Francisco: Jossey-Bass.

Part II

The European Union as an actor in international negotiations

8

The European Union as an international negotiator Ole Elgström and Maria Strömvik

Introduction The European Union (EU) is a major player in a large number of international negotiations. It has an important role in multilateral trade negotiations and UN conferences. The European Union has signed trade, aid and cooperation agreements with a large number of regions and countries. It has recently concluded accession negotiations with a number of potential new members. The Union has concluded agreements with numerous states participating in its international crisis management operations, and has negotiated agreements with the North Atlantic Treaty Organization (NATO) on the provision of military assets. It is common to portray the European Union as a slow and difficult international negotiator. It is often held that EU negotiators tend to be inflexible and unwilling to make concessions, as their negotiation mandate has been preceded by complicated bargaining between the member states. The European Union may often have a ‘conservative bias’ (Smith 2000), as it tends to protect the lowest common denominator interests of its member states. The institutional complexity of the Union creates problems of synchronization and coordination. Bargaining with the European Union is often protracted, as it is a cumbersome process both to produce a common position between the member states and to renegotiate the common mandate if needed. The European Union is therefore often seen as a foot-dragger in international negotiations. In this chapter we want to problematize this conventional view of the European Union as a reactive, conservative international negotiator. We argue that the European Union’s structural features need not always result in a disadvantage for the Union in international negotiations. We also contend that EU negotiating behaviour to a large extent depends on contextual factors. Are negotiations symmetrical or asymmetrical? Are the intentions of the EU status quo oriented or change oriented? All these negotiation-specific factors determine what bargaining pattern we expect to find. In addition to the European Union’s structural features (actor level characteristics) and the negotiation situation (interaction level characteristics), we also ask if the changing nature and character of international negotiations in general (at the systems level) may also begin to impact on the conventional view of the European Union as a passive and inflexible negotiator.

118 Ole Elgström and Maria Strömvik In the following we start by describing the conventional view of the European Union as an international negotiator, stressing institutional-legal and other structural features as well as their effects on EU negotiation positions, attitudes and tactics. The negotiation-specific determinants of EU behaviour are analyzed in the following section. Thereafter, some new trends in international negotiations are outlined, and we raise the possibility that the European Union possesses certain traits that make it well suited to function as a facilitator, rather than a footdragger, under these circumstances.

The institutional structure To understand EU negotiating behaviour it is, first, necessary to highlight the legal–institutional structure behind it. The ‘standard assumption’ is that the European Union is handicapped by the complexity of its institutions and by its lack of clear leadership in international affairs (Meunier 2000: 105). First of all, there is the complex structure of exclusive and mixed competencies. In some areas, such as most trade issues, the European Union has been granted exclusive competence. Whether in bilateral or multilateral trade negotiations, Europe formally speaks with one voice. European Commission officials are free to conduct bargaining as they wish within the limits set by a negotiation mandate decided upon by the Council. In international environmental negotiations, where the European Union is nowadays recognized as an autonomous negotiator, albeit without voting right, the situation is more intricate. If discussions concern issues under the exclusive competence of the European Union, Commission representatives take the floor; in issues under the jurisdiction of member states, they decide; and in areas which are under the jurisdiction of both member states and the European Union as such – areas with mixed competence in EU parlance – the Union is sometimes represented by the Council Presidency, sometimes by the Commission (Sbragia 1997; Vogler 1999). This is an impenetrable system, which has been widely criticized for its lack of transparency. It is a ‘source of confusion and even bewilderment for those who have to deal with it’ (Vogler 1999: 26). The role of the Commission as an independent actor is per se a characteristic feature of the European Union. The Commission has for instance been called ‘a key strategic actor’ in the enlargement negotiations, not only because of its power of initiative but also because it produced the overall EU negotiation strategy and succeeded in selling it to the member states (Friis 1999: 33–4). The Commission, it is claimed, acted as an entrepreneurial leader in the Uruguay Round trade negotiations (Coleman and Tangermann 1999). The publication of a Commission green book on the future relations with the African, Caribbean and Pacific (ACP) countries set the agenda for the ensuing internal EU discussions and constituted the ideational basis for the resulting negotiating mandate ahead of the Cotonou agreement (Elgström 2000). Another institutional factor of relevance are voting rules. There is considerable uncertainty over what voting rules apply in different areas. Bargaining positions on traditional trade issues are in principle agreed on by majority, but the Council

The EU as an international negotiator 119 can decide on a case-by-case basis what rule to use in the ‘new’ areas of services and intellectual property. It has been claimed that when unanimity is used, the chance is higher that the European Union will adopt the position of its most conservative member (Meunier 2000). Likewise, rules on the ratification of trade agreements differ depending on whether competence is exclusive (majority voting in the Council) or mixed (where member states use their own national procedures) (Meunier and Nicolaidis 1999). National ratification tends to prolong the process and render agreements more difficult to reach.

Resources The European Union is considered a major actor, a key player, in most international negotiations. The Union builds its power position on huge economic resources, linked to a ‘powerful legal order’ and a ‘powerful set of common understandings’ (Smith 2000). The European Union is the world’s largest exporter, and the second largest importer of merchandise goods. The economic resources, in combination with its high institutional density, contribute to the Union’s high visibility and impact in many bargaining contexts. Furthermore, EU proposals often have the advantage of being firmly based on scientific and/or administrative expertise. To add to this picture, the European Union itself functions as a forceful magnet, drawing to it states with an interest in some type of relation, be it membership, association or a trade deal (cf. Keohane and Hoffmann 1990: 277; Rosecrance 1998: 16). When the demand for an agreement is asymmetric, outsiders approach the European Union as demandeurs or supplicants (Elgström 2000: 178; Friis 1999: 25). Power asymmetries have obvious consequences for negotiation outcomes: we would expect the stronger party to pressure the weaker into concessions.

Effects on EU negotiation positions, attitudes and behaviour The outcome of international negotiations involving the European Union is determined by bargaining at two levels: first, the member states have to agree among themselves on the Union’s negotiating mandate; second, the European Union has to reach an agreement with its partners at the negotiation table (cf. Elgström 2000; Friis 1999: 24). The conventional picture is that the European Union in its internal negotiations – which are of course heavily influenced by domestic negotiations within each member state (see Putnam 1988; Evans et al. 1993) – tends to end up advocating the position of its most conservative member, or the lowest common denominator of the members (see Paemen and Bensch 1995: 95; Scharpf 1988). It is a ‘convoy moving at the speed of the slowest vessel’ (Vogler 1999: 40). This final agreement is very difficult to unravel. It is the product of long, drawn-out negotiations, often including sensitive concessions in intricate patterns. To give in later on to demands from other actors on one aspect of the total deal would imply reopening the whole package. Therefore, the European Union demonstrates a ‘conservative bias’ in external negotiations (cf. Smith 2000: 811). Furthermore, negotiating with the European Union tends to be time consuming.

120 Ole Elgström and Maria Strömvik Both finding an internal compromise and changing this, if deemed necessary, take a long time. All member states – as well as the Commission – have to arrive at national positions, which means securing approval at the home arena and thereafter finding a joint standpoint, often by unanimity or consensus. To the extent that outside actors are aware of the rigidity of EU positions, and of the time it may take to renegotiate a deal, in itself creates a pressures for others to concede, especially if they value a quick solution. The inflexibility of its positions may thus give the European Union a bargaining advantage, notably in distributive, conflictual types of negotiations. The institutional complexity of the European Union can create coordination and synchronization problems (Forster 2000). In the words of Michael Smith, ‘there is a layering of institutions and policy-making contexts, which creates problems of linkage’ (2000: 808). Member states have to coordinate their actions, the Council has to coordinate its policies with the Commission. Coordination also takes place across layers, ‘spheres of action and interaction between policy-making actors that encompass varying scopes and extents of authority and policy-making competence’ (Smith 2000: 809). In such coordination negotiations, regional and sub-regional organizations participate together with states and supranational actors. There is also the challenge of ‘consistency’: to pursue consistent policies across all policy areas that can affect other countries (‘do EU policies in the agricultural sector contradict its foreign aid policy?’). The question is often posed whether the European Union ‘speaks with one voice’ or not (Allen 1998; Meunier 2000). In foreign policy proper, the European Union is represented by the acting Presidency, but the Commissioner for external affairs and the High Representative for the common foreign and security policy also act on behalf of the Union. Henry Kissinger’s frustrated plea for a single telephone number to one European foreign policy supremo remains topical. The existence of several representatives with often unclear competencies makes the European Union ill equipped to act swiftly in the crucial final hours of a negotiations when the decisive compromise is constructed (Meunier 2000: 106). It is an immobile, rigid bargaining organization and this is often an impediment to effective bargaining with others, especially in problem-solving negotiations where flexibility is of the essence (cf. Paemen and Bensch 1995: 94–5; Vogler 1999: 41; Forster 2000: 789). To change bargaining position, negotiators at the table often have to go back to their principals, who have to find a consensus on a new mandate. Furthermore, it is difficult to conduct negotiations with actors whose precise authority is unclear: will a deal struck with Commission officials at the table be honoured by the member states? In other words, there is a risk of involuntary defection (Putnam 1988). In brief, the existing literature paints a picture of the European Union as a rigid, conservative external negotiator with coordination and synchronization problems. Its capacity to contribute innovative, problem-solving initiatives is seen as limited. It often acts as a foot-dragger in international negotiations. This is obviously one of the reasons behind the capability–expectation gap (Hill 1993, 1998): the European Union is claimed to have great difficulties in living up to the very high expectations

The EU as an international negotiator 121 that other actors hold. These expectations may include both hopes that the European Union might exert a leadership role in multilateral negotiations and hopes for substantial and quick results in concrete negotiations. In the following sections, we problematize this traditional view. To a certain extent, we also question its main conclusions. We start by illuminating the contextual nature of European Union negotiating behaviour. Depending on the type of negotiations, the European Union may be more or less flexible and innovative.

Negotiation context matters In an interesting article Sophie Meunier (2000) argues that voting rules and negotiating competence determine both the probability that the negotiating parties reach an agreement and the substantive outcome of the negotiations. She further develops this theme by proposing that the institutional design of the European Union plays a bigger role when the Union acts as a defender of the status quo in a negotiation than when it acts as a reformer, seeking a change in the policy of the opponents. In the first case, unanimity voting and restricted delegation ‘make the EU a tough bargainer: the negotiating opponent cannot obtain more than what the most conservative EU state is willing to concede’, while qualified majority voting and extensive delegation mitigate the extremes and render an agreement more likely but deprive the European Union of some bargaining leverage, thus leading to a more favourable deal for the opponent. In the second case, involving a reformist European Union, institutional design plays a much lesser role (Meunier 2000: 104, 131). One main lesson of Meunier’s argument is that the specific negotiation situation matters. The basic orientation of the European Union (is it a reformist or a status quo actor?) is one such factor. But there are others, not mentioned by Meunier. The balance of power in a given situation is arguably a key variable. A basic distinction can be drawn between symmetric and asymmetric negotiations. For example, trade negotiations with the United States of America or Japan – bilateral or within the World Trade Organization (WTO) framework – are basically symmetric, while enlargement negotiations and bargaining with the less developed countries for aid or trade agreements are basically asymmetric, in the sense that the European Union’s alternatives to an agreement are much better and its need for a deal is much smaller than its opponents’ (cf. Habeeb 1988). If we combine this with the reformist–status quo dimension, we get the following four alternatives (Figure 8.1). Power asymmetry obviously increases the problems opponents have in getting the European Union to change its position when it defends the status quo (box 3). But even powerful actors experience severe problems when they engage in such efforts (box 1); US attempts to influence EU agricultural policy is one prominent illustration. In asymmetric negotiations, the chance for the European Union to have its way when it challenges existing policy (box 4) is greatly enhanced, compared to a situation where it wishes to challenge another dominant power’s policy (box 2). EU negotiations with the ACP countries regarding a revised Lomé

122 Ole Elgström and Maria Strömvik Relationship

Symmetric

Asymmetric

Status quo-oriented

1

3

Change-oriented

2

4

EU-orientation

Figure 8.1 Four different types of negotiation situations.

convention in the late 1990s is a good example (Elgström 2000). In this case, the European Union wanted to totally restructure its aid and trade relationship with its former colonies. Despite strong initial resistance, the European Union succeeded in driving home important changes: for the first time, the ACP countries explicitly committed themselves to introduce a reciprocal trade regime, albeit in the future; WTO compatibility has become a keyword; future trade arrangements will be based on regional integration initiatives – all key demands from the European Union. It should be added, however, that weaker negotiation partners are seldom totally without influence. Even in cases where the European Union defends an existing situation (box 3) – as in many respects they do in accession negotiations – the demandeurs have some potentially successful tactics at their disposal. Applicant countries can thus rely on the highly symbolic importance attached to enlargement. Both member states and the Commission wish the European Union to be seen as successful and open to all democratic European nations and therefore have difficulties accepting non-agreement in accession negotiations. Applicants can also refer to concessions given in previous enlargement negotiations, and they can use some members who are sympathetic to their demands as ‘Trojan horses’ (cf. Friis 1999: 35–8). The conventional view of the European Union as a conservative, difficult negotiator is primarily applicable to situations like the ones in boxes 1 and 3. Here, the Union has strong interests to defend, often of a material nature and based on powerful domestic constituencies. The result is resistance to change and unwillingness to make significant concessions. When Europe is a revisionist actor (boxes 2 and 4), the situation is different. This is in particular the case when the European Union is trying to promote milieu goals in international negotiations, that is, when it attempts to shape or reshape its external environment by influencing international normative structures (cf. Wolfers 1962: 73–6; Keukeleire 2000). An actor may try to change, or stabilize, the norms and value systems that dominate its international environment. In these instances, the European Union acts as a norm

The EU as an international negotiator 123 entrepreneur (Finnemore 1996) or ‘normative power’ (Manners 2002). For example, the European Union is active in seeking to spread such norms as democracy, human rights, free trade, regional cooperation and peaceful conflict resolution. To do so, the Union primarily relies on soft power (Nye 1990), economic or financial instruments and persuasion (‘dialogue’ in EU parlance), but these tools are often linked to traditional ways of exerting power. The inclusion of conditionality in foreign aid negotiations is one prominent example. Efforts to pursue milieu goals are more common in asymmetric negotiations (box 4). The promotion of democracy, good governance, human rights and liberal economic ideas are crucial ingredients in negotiations with candidate member states and regional groupings in the Third World. In these negotiations the European Union is an active norm entrepreneur. It may be argued that the Union is especially suited for this role: through its own achievements it can act as a role model for other regional groupings, with respect to both economic integration and peaceful change by democratic means. It has also a tradition of using soft power instruments instead of utilizing force and may therefore be viewed as a benevolent great power. Many outsiders dearly wish to become insiders – aspirations that give the European Union an excellent opportunity to promote its values by a gentle mixture of persuasion and pressure. In fact, the democratization and stabilization of central and eastern Europe is often perceived as a success story for an active EU foreign policy (Keukeleire 2000: 22). Furthermore, in their analysis of EU decision making during the war in Kosovo, Lykke Friis and Anna Murphy (2000) have demonstrated that the European Union, under certain conditions, is also capable of making quick decisions. In crises, the Union may initiate what Friis and Murphy call ‘turbo-charged negotiations’. In the Kosovo crisis, the European Union rapidly came up with the idea of a ‘stabilization pact’ for the Balkans. The move was not innovative – the idea was actually retrieved from the existing toolkit of conflict management solutions – but proved that the European Union can sometimes be a flexible facilitator in international negotiations. We can thus conclude that the European Union is not uniformly a passive, inflexible negotiator. Under some circumstances, the Union may take on a leadership role, trying actively to promote values and to spread its preferred policies. In the next section this argument is extended: we contend that several international bargaining settings are today characterized by uncertainty and complexity and that state interests in these contexts are fluid and malleable. The negotiations are carried out in informal networks and include a multitude of different actors and different issues at several levels. In such situations, the European Union may be increasingly adapted to act as a process facilitator and an agent of change.

The changing global context By taking into account the changing nature of international negotiations, we open up for questions about the consequences for the European Union as an international negotiator. By placing the Union in a changing global governance

124 Ole Elgström and Maria Strömvik perspective, some lessons drawn both from the EU governance literature and from the more general global governance literature can be combined. This, in turn, can generate a few propositions about an ongoing change in the role of the European Union in international negotiations. The question regarding the depth or extent of globalization has lately emerged as one of the most frequently debated issues in social sciences. There is little consensus on the nature of the changes, but many analysts would subscribe to the view that even if the changes might not fundamentally transform the international system, some trends are nonetheless visible. These trends are often captured in terms of changing ‘extensity, intensity, and institutionalization’ of global politics (Held et al. 1999: 52). The concrete signs of these changes are manifold, but for the purpose of this paper some deserve specific mention. Aspects of change that are of immediate interest here are connected to the changing nature of international communication and networks at the global level, and to the changing contents of global issues. The vast literature on global governance has highlighted the fact that we are increasingly witnessing a ‘multitude of jurisdictions that is proliferating on the world stage’ (Rosenau 1995: 18). New issues are making their way to the global agenda, resulting in a more complex system of international decision making. Previously narrow trade issues are for instance expanding to involve new issues, such as intellectual property rights, defense of cultural diversity, international labour rights and environmental concerns. Thereby, international negotiations are becoming increasingly complex and demand new state strategies for handling multiple issues in many negotiating situations. Furthermore, international negotiations are no longer the exclusive domain for diplomats from foreign ministries, and officials from domestic ministries are increasingly participating in various international contexts. As the boundaries of what constitutes the domestic and the international become increasingly diffuse, new problems concerning the distribution of authority within governments arise (Held et al. 1999: 53–4). Also, as the diplomats are complemented with, or substituted by, officials from domestic ministries, less experienced in an international milieu, states may encounter problems while these new negotiators learn to handle the complexity of the new situation (cf. Keohane 1989: 188). These problems are facing all states, and the capacity to adjust to these new circumstances becomes one of the determining factors for success in international negotiations. Within this description of changing global conditions, the European Union first of all is in itself a case in point. The Union is often characterized as a system of multilevel governance (Hix 1999; Jachtenfuchs and Kohler-Koch 1995; Rosamund 2000). Decision making in the European Union involves a large number of actors of different types (states, regions, supranational institutions, non-governmental organizations etc.) interacting at different levels. Authority is dispersed, although member states are still the most influential players. The policy-making environment is complex, uncertain and fluid. In such a system, decisions are often the result of multi-level network negotiations. Many compromises are formed in informal ways, involving mediation by institutional actors or other linking-pin organizations (cf. Jönsson et al. 1998).

The EU as an international negotiator 125 Second, it is important to bear these characteristics in mind when turning to the question of the European Union as an international negotiator. Even if such a question by necessity treats the Union as an actor (although not a unitary one) rather than as a system of governance, some of these features might very well prove to give the Union the upper hand in many international negotiations. We thus propose that some of the European Union’s structural features – not highlighted by traditional accounts of the Union as an international negotiator – may actually give the European Union comparative advantages at multilateral bargaining tables.

An unusually skilled negotiator? In the light of these changes the European Union might already be at an advantage, irrespective of the specific constellation representing the Union in international negotiations. As all new EU member states have testified, the state administrations have had to adjust considerably to the new circumstances. With participation in committees, working groups and expert groups within the EU bureaucracy, national officials from most ministries in all member states are constantly finding themselves in an international, multi-level, network-type working environment (cf. Beckman and Johansson 1999: 115). The top-level politicians from all member states meet regularly in Brussels and other places, to the extent that, as the former Swedish foreign minister Lena Hjelm-Wallén once commented, they often see each other more frequently than they see their own national colleagues in their home capitals ( Johansson 1999: 87). The Commission officials are perhaps the most experienced of all these groups in relation to handling an extremely complex negotiation environment, given their position and daily work at the heart of the Union. The implications are that all possible participants in any negotiating delegation from the European Union are already highly trained officials, used to managing complex international negotiations within the European Union. The Union thereby has an unusually large pool of experienced and skilled negotiators, used to an everyday environment that can most aptly be described as an extreme version of the emerging global situation. By constantly participating in the ‘multilateral inter-bureaucratic negotiation marathon’ (Kohler-Koch 1995: 181) that the European Union constitutes, they are for instance no strangers to the creative use of informal strategies to avoid deadlocks in complicated negotiations (cf Jönsson et al. 1998: 326). The European Union can thus be seen as an international actor already quite adjusted to the changing global conditions. European Union officials are also trained in a relatively non-hierarchical environment compared to officials trained within state bureaucracies. Whereas negotiations within national governments are hierarchical in nature, international negotiations typically lack strong central agenda-setting and coordinating actors. A non-hierarchical negotiating climate is also a reality in many negotiations within the European Union. None the less the EU system has developed de facto practices of leadership in most issue areas over time. The roles of agenda setter and facilitator

126 Ole Elgström and Maria Strömvik most often fall either on the Commission or on the Presidency, or on a combination of both. There are however also considerable possibilities for other active participants to influence both the agenda and the final solution by acting as problem solvers. These conditions produce experienced EU officials, familiar with thinking about informal strategies on agenda setting and solutions to complex, multi-level and multi-issue negotiations. In international negotiations – which increasingly display these same characteristics, often in an even more nonhierarchical structure – this experience can very well prove to be an increasingly important asset. The relatively non-hierarchical and network-dense features that the European Union displays can furthermore be seen as a fruitful ground for new and innovative ideas, generated to solve common problems. Three analysts of the European Union have pointed out that the Union’s own novel and unproven way to monetary union is ‘typical of an emerging policy system characterized by innovation and experimentation’ (Laffan et al. 2000: 162). If this argument is transposed to the international level, that is, if novelties and innovative solutions can emerge easier in fluid, informal network-based systems than in more traditional hierarchies, the European Union’s negotiators should be well-equipped to understand and influence this situation. Not only are they familiar with the system, but they might also be able to fall back on novel solutions that have emerged within the EU system. In certain situations these tendencies ought to be particularly visible. One such situation is when the international negotiations are concerned with regulatory issues. Negotiations on new international regulations can be assumed to display a higher degree of network bargaining than negotiations on issues of a distributive character ( Jönsson et al. 1998: 333). This in itself should give the European Union an advantage in such negotiations, according to the reasoning above. We might also add two more conditions that further reinforce the strength of the European Union in these situations. If we consider the fact that the European Union already displays the most dense web of regulations between states in the international system today, we can assume that the experiences and new ideas generated in this process can be used by EU negotiators at the international level. Moreover, as it constitutes one of the largest markets in the world, the European Union has, by virtue of its economic power, strengths and advantages in setting the agenda for regulatory negotiations. There are already signs that EU directives on internal market issues are increasingly being adopted by many third states, including the United States of America and Japan. The European Union has thereby already started to serve as an agenda setter for global standards (Laffan et al. 2000: 120; cf. Rosecrance 1998: 22). As such, the European Union has furthermore served as a facilitator in the sense that the Union, as a result of its own internal lessons, has a tendency to formulate directives in a less specific form than the domestic US equivalents. The more generic form of EU directives results in less need to renegotiate the directives as technological changes occur. Thereby, according to some observers, the spreading of EU directives worldwide can very well be seen as a solution to governance problems facing both policy makers and

The EU as an international negotiator 127 the business community (Laffan et al. 2000: 120). In other words, the European Union might increasingly gain agenda-setting powers by virtue of its own network characteristics, its immaterial resources and its de facto role as a forerunner in the creation of new international solutions to new problems created by increasing interdependence. In brief, we argue that some of the European Union’s unusual structural features combine with novel characteristics of international negotiations, brought about by a changing global context, to produce a bargaining environment where the European Union may possess certain advantages. The Union may therefore increasingly play an active and creative problem-solving role in many international contexts. EU decision-makers are used to handling complex multi-level negotiations in their internal arena. Itself a network organization, the European Union is particularly well equipped to grasp and utilize the potentials of network negotiations. The European Union has an image as a ‘soft power’ and is, to a lesser extent than the United States of America, associated with great power arrogance and coercive behaviour. As international negotiations are increasingly characterized by complexity and multi-level games, and as actors can no longer rely on traditional power assets, we predict that the European Union’s potential as a flexible and adroit international actor will increase. However, as observers of the European Union’s international role constantly point out, the Union is not always unified enough to come across as a coherent actor. A lack of political will is often said to exist when the European Union does not arrive at strong negotiation positions or even fails to come up with a common stance. But in the changing global environment, even the political will might slowly be changing as well. Individual small and medium-sized states find it increasingly difficult to influence world politics unless they join forces with similarminded states. The awareness of the benefits from a collective stance is growing, and the European Union is already there to provide the most efficient venue to combine voices ahead of international negotiations. This realization may certainly start to affect the individual member’s willingness to make concessions at the preceding intra-EU bargaining stage.

References Allen, D. (1998) ‘Who Speaks for Europe? The Search for an Effective and Coherent External Policy’, in J. Peterson and H. Sjursen (eds) A Common Foreign Policy for Europe?, London: Routledge. Beckman, B. and Johansson, K. M. (1999) ‘EU och statsförvaltningen’, in K. M. Johansson (ed.) Sverige i EU, Stockholm: SNS Förlag. Coleman, W. D. and Tangermann, S. (1999) ‘The 1992 CAP Reform, the Uruguay Round and the Commission: Conceptualizing Linked Policy Games’, Journal of Common Market Studies, 37 (3): 385–405. Elgström, O. (2000) ‘Lomé and Post-Lomé: Asymmetric Negotiations and the Impact of Norms’, European Foreign Affairs Review, 5 (2): 175–95. Evans, P. B., Jacobson, H. K. and Putnam, R. D. (1993) Double-Edged Diplomacy, Berkeley: University of California Press.

128 Ole Elgström and Maria Strömvik Finnemore, M. (1996) National Interests in International Society, Ithaca and London: Cornell University Press. Forster, A. (2000) ‘Evaluating the EU–ASEM Relationship: A Negotiated Order Approach’, Journal of European Public Policy, 7 (5): 787–805. Friis, L. (1999) An Ever Larger Union? EU Enlargement and European Integration, Copenhagen: Danish Institute of International Affairs. Friis, L. and Murphy, A. (2000) ‘Turbo-Charged Negotiations’, Journal of European Public Policy, 7 (5): 767–86. Habeeb, W. M. (1988) Power and Tactics in International Negotiations: How Weak Nations Bargain with Strong Nations, Baltimore and London: Johns Hopkins University. Held, D., McGrew, A., Goldblatt, D. and Perraton, J. (1999) Global Transformations. Politics, Economics and Culture, Stanford: Stanford University Press. Hill, C. (1993) ‘The Capability–Expectations Gap, or Conceptualising Europe’s International Role’, Journal of Common Market Studies, 31(3): 305–28. Hill, C. (1998) ‘Closing the Capabilities–Expectations Gap?’, in J. Petersen and H. Sjursen (eds) A Common Foreign Policy for Europe? Competing Visions of the CFSP, London: Routledge, pp. 18–38. Hix, S. (1999) The Political System of the European Union, London: Macmillan. Jachtenfuchs, M. and Kohler-Koch, B. (1995) ‘The Transformation of Governance in the European Union’, paper prepared for presentation at the Fourth Biennial Conference of the European Community Studies Association, Charleston, SC, 11–14 May. Johansson, K. M. (1999) ‘Sverige i EU: s institutioner’, in K. M. Johansson (ed.) Sverige i EU, Stockholm: SNS Förlag. Jönsson, C., Bjurulf, B., Elgström, O., Sannerstedt, S. and Strömvik, M. (1998) ‘Negotiations in Networks in the European Union’, International Negotiation, 3: 319–44. Keohane, R. O. (1989) International Institutions and State Power: Essays in International Relations Theory, Princeton: Princeton University Press. Keohane, R. O. and Hoffmann, S. (1990) ‘Conclusions: Community Politics and Institutional Change’, in W. Wallace (ed.) The Dynamics of European Integration, London and New York: Pinter. Keukeleire, S. (2000) ‘The European Union as a Diplomatic Actor’, DSP Discussion Papers, No. 71, Leicester: Centre for the Study of Diplomacy. Kohler-Koch, B. (1995) ‘The Strength of Weakness: The Transformation of Governance in the EU’, in S. Gustavsson and L. Lewin (eds) The Future of the Nation State, Stockholm: Nerenius and Santérus. Laffan, B., O’Donnell, R. and Smith, M. (2000) Europe’s Experimental Union. Rethinking Integration, London and New York: Routledge. Manners, I. (2002) ‘Normative Power Europe: A Contradiction in Terms?’, Journal of Common Market Studies, 40 (2): 253–74. Meunier, S. (2000) ‘What Single Voice? European Institutions and EU–US Trade Negotiations’, International Organization, 54 (1): 103–35. Meunier, S. and Nicolaidis, K. (1999) ‘Who Speaks for Europe? The Delegation of Trade Authority in the EU’, Journal of Common Market Studies, 37 (3): 477–501. Nye, J. S. (1990) ‘Soft Power’, Foreign Policy, 80: 153–71. Paemen, H. and Bensch, A. (1995) From the GATT to the WTO: The European Community in the Uruguay Round, Leuven: Leuven University Press. Putnam, R. D. (1988) ‘Diplomacy and Domestic Politics’, International Organization, 42 (3): 427–61. Rosamund, B. (2000) Theories of European Integration, London: Macmillan.

The EU as an international negotiator 129 Rosecrance, R. (1998) ‘The European Union: A New Type of International Actor’, in J. Zielonka (ed.) Paradoxes of European Foreign Policy, The Hague: Kluwer. Rosenau, J. (1995) ‘Governance in the Twenty-first Century’, Global Governance, 1 (1): 13– 44. Sbragia, A. M. (1997) ‘Shaping the Institutional Architecture of the EU: The Influence of Global Politics’, paper delivered at the International Political Science Association XVII World Congress, Seoul, Korea. Scharpf, F. W. (1988) ‘The Joint-Decision Trap: Lessons from German Federalism and European Integration’, Public Administration, 66: 239–78. Smith, M. (2000) ‘Negotiating New Europes: The Roles of the European Union’, Journal of European Public Policy, 7 (5): 806–22. Vogler, J. (1999) ‘The European Union as an Actor in International Environmental Politics’, Environmental Politics, 8 (3): 24–48 Wolfers, A. (1962) Discord and Collaboration, Baltimore: The Johns Hopkins Press.

9

Setting the global trade agenda The European Union and the launch of the Doha Round Anders Ahnlid

Introduction Emerging as the largest actor in the world trading system, the European Union1 has sought to assume a leadership role in the World Trade Organization (WTO). After eight ‘rounds’ of trade negotiations initiated by the United States, the European Union became the main instigator of the ninth, the Doha Development Agenda, which was launched in 2001. The European Union met with a fair degree of success in its quest for a new round despite the persistent protectionist nature of the Common Agricultural Policy (CAP) and new demands emanating from civil society on trade policy in areas such as the environment, labour rights and global justice. How could the European Union pursue credible and effective leadership in favour of free trade, given the protectionist nature of its agricultural policy and the new conditions for trade policy making? To what extent was its ability to lead limited by these factors? Drawing on the 1996–2001 EU campaign to start a new WTO Round, this chapter discusses the prerequisites for EU leadership – in terms of launching and setting the agenda of negotiations – in the world trading system. In doing so, Patterson’s expansion of Putnam’s two-level game model into ‘a three-level interactive game’, in which negotiations at the domestic, Community and international levels affect policy options at each of the other levels, is used to frame the analysis (Patterson 1997: 141). The intricate interplay between the three levels of the negotiating game is central to an understanding of the outcome – i.e. the Doha Development Agenda. It is argued that the ability of the European Commission to form appropriate strategies is key to EU willingness and ability to exert leadership. A successful strategy has to strike a balance between what is needed to satisfy EU member states domestically, on the one hand, and what is feasible at the multilateral level, on the other hand. The chapter will trace how domestic preferences of EU member states – through the institutional machinery set up by the Treaty on European Union – were transformed into a common EU position. The rules of the treaty on decision making and external action largely influenced EU policy. The chapter also examines how the European Union interacted with other WTO members to get the new round launched. At the international level the EU stance

Setting the global trade agenda 131 was influenced by positions of third countries and the institutional set-up of the WTO, including its consensus rule for decision making.

The origin of the EU proposal for a new round The EU objective to launch a new negotiating round was formulated after a long period of US dominance in the trading system. As post-war leader, the United States bore responsibility for the launch and completion of eight consecutive rounds of multilateral trade negotiations, the last of which, the so-called Uruguay Round, resulted in the creation of the WTO in 1995. The Uruguay Round marked the end of tacit US acceptance of the CAP, as agriculture was incorporated in the rulebook for the first time. The European Community (EC) initially resisted this move, and was also reluctant to accept the US proposal to expand the agenda of the Uruguay Round to new issues, such as services, investment and intellectual property rights. The EC was ‘not particularly interested in supporting new initiatives in GATT [the General Agreement on Tariffs and Trade], and had none of its own to propose’ (Croome 1995: 11). Yet the United States of America largely obtained the broad negotiating agenda that it wanted, not least since the EC and others did not want the United States to turn to bilateral negotiations or even unilateral action to achieve its objectives, as threatened. EC foot-dragging in agriculture stalled work across the whole Uruguay Round agenda from 1986 until late 1992, when a US–EC deal on agriculture was struck (Croome 1995). Thereafter the EC changed its attitude and took on a leading role in most remaining areas of negotiation. The EC pursued offensive interests concerning industrial tariffs, services and institutional issues, with the United States being less keen to obtain far-reaching results. The EC had an interest in these issues in their own right, but it also needed good outcomes here in order to ‘balance’ the perceived ‘loss’ in agriculture. This was crucial when EC member states were moving to adopt the Uruguay Round result under the ‘single undertaking’ rule of the game, which meant that a party had to either accept or reject the package as a whole. The single undertaking was invented for use in the Uruguay Round to facilitate linkages and trade-offs between negotiating issues and to avoid free riding. It involved both challenges and opportunities for the EC. It had implications for EC internal decision making. Some issues on the agenda, such as aspects of trade in services, investment and intellectual property rights, did not fall under common community competence. Therefore, the single undertaking meant that the EC had to decide on the final overall result by consensus and not by qualified majority, which was the rule for traditional trade policy decisions. This accentuated the need for the EC to obtain a ‘balanced’ result. At the same time the single undertaking improved the bargaining position of the EC, as it permitted requests for substantial ‘concessions’ by others to ‘balance’ EC ‘sacrifices’ in agriculture. Thus, the single undertaking helped in reaching an outcome of the sort the EC needed. The EC obtained a comprehensive and far-reaching result for industrial goods and a decent, but far from complete, deal in services, where defensive US interests

132 Anders Ahnlid now prevailed (Ahnlid 2000). The EC also achieved a binding dispute settlement system to stall unilateral US action. Finally, the creation of the WTO was considered important by the EC. At the same time, the so-called Cairns group of agricultural exporting countries did not find the deal on agriculture sufficiently far-reaching. In order to agree to the overall package, these countries requested a commitment to start negotiations on further liberalization of trade in agricultural products by 2000. In return, the EC obtained agreement to start new negotiations on trade in services the same year. Thereby, the so-called in-built agenda of additional negotiations was created and formed part of the Uruguay Round results. EC ‘concessions’ in agriculture, combined with active involvement and leadership on other issues, helped leading GATT signatories to strike the most ambitious multilateral trade agreement ever towards the end of 1993. An internal deal, worked out between the Commission and EU member states, involving some sidepayments, paved the way for the decision by the EU Council to formally accept the Uruguay Round result. After ratification by parliaments at the member state level, the European Union became a founding member of the WTO in 1995. The institutional set-up of the negotiations and the EC internal decision-making procedure spurred the EC shift from foot-dragger to leader in the Uruguay Round. The result, including the institutionally embodied ‘in-built agenda’, came to cast an important shadow over the coming EU bid for leadership in the trading system. During its first years members sought new means of moving the WTO forward. In particular the USA favoured sectoral negotiations over yet another timeconsuming and resource-intensive round. Traditional rounds were deemed to be out of date, and not suitable to meet the rapidly changing requirements of modern business. Instead priority deals were to be hammered out in one or two sectors at a time, and consecutive results were to be confirmed at the semi-annual ministerial meetings, mandated by the WTO agreement. The European Union initially agreed to proceed along those lines, albeit less convinced than the USA of the merits of doing so. At the first WTO ministerial meeting, in Singapore 1996, the new approach produced results. The meeting paved the way for an ambitious agreement on trade in information technology products as well as successful deals regarding financial and telecommunication services. However, when strong protectionist domestic interests in key countries came into play, such as in maritime services in the USA, the method proved less successful. In addition, negotiations on agriculture and services loomed in 2000. The European Commission, which negotiated on behalf of the European Union and was mandated to advance free trade in the EU common interest,2 concluded that progress could not be made on the basis of the in-built agenda alone. EU member states would not be willing to accept politically costly ‘concessions’ in agriculture in isolation. The negotiations on services would not suffice as sole counterweight to agriculture. The European Union set out to add other – and new – items to the agenda, in order to further balance the negotiating table. A first step was taken at

Setting the global trade agenda

133

the ministerial meeting in Singapore, when analytical work was initiated on investment, competition, trade facilitation and transparency in government procurement.

Forging consensus at the EU level At the EU level the conflicting national preferences of the 15 member states met with the Commission’s mandate to pursue and safeguard the common EU interest. Applying the logic of Putnam’s two-level negotiating game, the member states seek to maximize their own ability to satisfy domestic pressures at the EU level, while minimizing the adverse consequences for the European Union as a whole (Putnam 1988: 430). However, as pointed out by Patterson, ‘the unique structure’ of the European Union necessitates an expansion of the two-level game to ‘a three-level interactive game in which negotiations at the domestic, Community, and international levels affect policy options at each of the other levels’ (Patterson 1997: 141). In this game the European Commission is given the leading role by Article 133 of the EU treaty. The Commission holds the right of initiative in trade policy and is the sole negotiator for the European Union in areas where Community competence prevails. However, before entering into negotiations with third parties, the Commission has to obtain a mandate from the Council, where foreign or trade ministers represent their governments. The Council also makes the final decision on any negotiating result. Throughout the negotiating process the Commission is mandated to consult with member states in the so-called 133 Committee. As explained by Patterson, the European Union can be viewed as ‘acting as a federallike polity seeking to maximize benefits and minimize costs’ (Patterson 1997: 141). The intricate three-level negotiating game on the new trade round started in the spring of 1996. It spanned two preparatory WTO ministerial meetings in Singapore and Geneva as well as a fiasco at the Seattle meeting in 1999 before a new round, largely along the lines suggested by the European Union, was launched in Doha, Qatar, in 2001. Conflicting domestic interests Internal unity on the Commission proposal for a new trade round was a first and necessary condition for success. When Trade Commissioner Brittan informally floated the idea, domestic interest groups mobilized to influence member states’ governments, so that their domestic constituencies came to be represented at the negotiating table in Brussels, where the common EU position was to be formed. These interests comprised both groups that had traditionally been involved in trade policy making – such as business and labour – and a variety of newly mobilized civil society groups such as environmentalists, human rights and development and anti-globalization activists. The latter had got a boost by the failure of the negotiation on the Multilateral Agreement on Investment (MAI) in the Organization for Economic Cooperation and Development (OECD). They generally held that globalization and international institutions such as the MAI and WTO

134 Anders Ahnlid threatened the fulfilment of key environmental, human rights and development objectives. Their mobilization added a new dimension to domestic trade policy making in most EU member states. The relative strength of protectionist and liberally-oriented interests, as well as the weight of different civil society groups, varied between the member states. Yet, a clear difference remained – albeit not as clear-cut as previously – between more free-trade-oriented northern member states (‘the Northern Liberals’ in the internal jargon) and southern member states that were more prone to advocate protectionist policies (‘Club Med’). The enlargement in 1995, when Austria, Finland and Sweden joined the European Union, strengthened the northern flank. Dominating free-trade-oriented interest groups in countries such as Germany and the United Kingdom led the ‘Northern Liberal’ group to back the Commission proposal to launch a new WTO Round. In Germany liberal interests managed to outweigh defensive agricultural interests. The influential British manufacturing industry was free-trade-oriented, and that line was strengthened by the services industry. Being small and open economies, with dominating interest groups dependent on trade, the Netherlands, Sweden and Denmark pursued a consistent free-trade approach. In most ‘Northern Liberal’ member states business and labour interests joined forces in promoting free trade. An elaborate social security system, including policies for socially acceptable structural adjustment, underpinned the business–labour coalition. The influence of trade unions led several ‘Northern Liberal’ governments to combine a policy for free trade with a social agenda, including promotion of core labour standards worldwide. Development-oriented interests, keen to open up the EU market to imports from developing countries, and concerns for the environment generally carried weight in these countries’ trade policies as well. Nationally aggregated trade policy preferences in Portugal, Spain, France, Greece and Italy were generally determined by a strong interest in agricultural protection. Protectionist interest groups in textiles and clothing and other sectors contributed to a cautious trade policy approach, even if free-trade-oriented industries were on the rise. ‘Club Med’ countries initially maintained reservations about starting a comprehensive trade round, fearing what such a round might imply for agriculture. France, where the farm vote carried considerable weight through the largest right-wing party, Rassemblement pour la Republique (RPR), led the sceptics. The Socialists and Communists further strengthened the voice of French trade unions, which were wary of free trade. The more liberally inclined interest of business was not a sufficient counterweight at the national level (Balladur 1995: 141). In addition, France saw the rise of a vocal anti-globalization movement and added weight to the positions taken by ‘Club Med’ through traditionally successful brinkmanship that often seemed to give France disproportionate influence over EU trade policy. When the Commission meets member states The Commission took member states’ preferences into account as it moved forward. The proposal to launch a new trade round could be derived from the

Setting the global trade agenda

135

Commission’s treaty-bound mandate to advance free trade in the common interest of the Union. As explained by Brittan, the ‘central purpose’ of EU trade policy, was ‘the pursuit of assured, effective market access – world-wide, including in Europe’ (Brittan 1996b). Economically that meant exploiting gains from trade through both exports and imports. At the same time, the Commission’s ambition was constrained by societal interests at the member state level. Protectionist forces, mainly in agriculture and textiles, as well as civil society concerns over environment and workers’ rights, had to be taken into account if any proposal for a new round, let alone a final result, were to be accepted by the Council. Against this background, reciprocal trade negotiations provided an attractive way forward. Such negotiations would make it possible to obtain economically beneficial results despite political constraints in the form of protectionist special interests. For the Commission, that made the launch of a new WTO Round the best available means to pursue its free-trade objective. A new round was seen as: a vehicle of immense significance. Launching the Round enables us – no, forces us – into a pattern of cooperation as we seek to complete the negotiations. Of course there will be rows on agriculture and other subjects too. But there will be a structure to our work that actually enables . . . important trade projects to move forward. (Lamy 2001b) The societal demands for action outside the realm of traditional cross-border trade, where Community competence and qualified majority decisions prevail, complicated the task for the Commission. Joint EU action on non-cross-border aspects of trade relating to the environment, services, investment, competition or intellectual property rights would require consensus among the member states. Still the Commission did not hesitate to suggest that areas outside Community competence be included in the new round. It had already been done in the Uruguay Round. And beside substantive reasons for such action, the Commission – which was given the right to negotiate in these areas as well – found it to be in its interest, since it helped expand the Commission’s own turf. In the absence of any clear US initiative and in order to pave the way for internal consensus on a broad new trade round, the Commission defined the basis for EU external trade policy in terms of global leadership. Brittan declared that it was ‘time for Europe to abandon its defensiveness about liberalization’ (Brittan 1996a). The looming in-built agenda negotiations on agriculture was a key reason why the European Union needed a ‘comprehensive approach’. They also necessitated a rapid launch of a new round, the decision on which, according to Brittan, could not be delayed beyond the end of the century. The WTO should pursue both its traditional work programme and expand into new fields, such as investment, competition and the environment. All areas should be knitted together in a ‘single undertaking’, permitting linkages between issues and paving the way for an acceptable final result. The expansion of the scope of the WTO was needed in

136 Anders Ahnlid order to secure support from EU member states in which ‘concessions’ in agriculture were politically costly. According to Brittan, a ‘comprehensive round’ offered ‘opportunities which the narrow sectoral approach never can’ (European Commission 1999b). Thus, the Commission informally put forward the idea to launch a new round – the Millennium Round – in the spring of 1996 (Brittan 1996b). The Financial Times (6 August 1996) observed that the Commission now advocated ‘liberalisation with a convert’s passion’. The need to use international trade negotiations to advance EU economic interests, and the fact that contemporary trade policy making falls under mixed competence, largely determined the scope and content of the Commission initiative on a new round. These factors led the Commission to base its proposal, first, on a notion of strict reciprocity and, second, on the principle of a ‘single undertaking’. Such an approach would make it possible to obtain a final result that would both be economically beneficial to the Union and form the basis for a consensus decision among member states. After the informal floating of the idea, the formal EU proposal to launch a new WTO Round was hammered out in an internal negotiation process in the 133 Committee that concluded in the autumn of 1999. ‘Northern Liberal’ member states generally supported the Commission proposal, which was seen as a vehicle for free-trade results. Predictably, however, the idea prompted opposition from protectionist interests, mainly in agriculture but also in other sectors, such as textiles and clothing. This made ‘Club Med’ members adopt a cautious, if not negative, attitude. These member states feared that the launch of a new round would accelerate the opening up of the agricultural sector and legitimize external pressure in favour of such steps. Initially views also differed on the inclusion of new issues – such as investment and competition policy – in the negotiations and on the way environmental concerns and core labour standards should be taken up in the WTO. The agricultural sceptics generally showed interest in these areas, while countries like the United Kingdom and the Netherlands were less keen. The latter feared that stark EU requests in areas such as environment and core labour standards would postpone or render impossible far-reaching results on market access, which was their priority. A third group of countries, including Germany, Sweden and Denmark, emphasized both market access results and progress in new areas. These countries – together with France – also attached particular importance to advancing the issue of core labour standards. The Commission adjusted its proposal in response to the concerns expressed by member states. Informal meetings of EU trade ministers guided the process. At such a meeting in March 1998 ministers agreed that the European Union should call for a comprehensive, wide-ranging Millennium Round. While the leadership ambition was clear, EU ministers were not ready to define the content of the negotiations in specific terms. The Financial Times (30 March 1998) reported that France rejected public mentioning of further liberalization of farm trade and was reluctant to embrace a detailed proposal on a ‘comprehensive’ WTO Round. The 1998 ministerial meeting in Geneva – the location of the WTO headquarters

Setting the global trade agenda

137

– set the scene for attempting to launch a new round at the next ministerial meeting, which was to be held in Seattle, USA, in December 1999. The European Union stepped up its efforts to craft a sufficiently specified common position. As demandeur for a new round, and in order to spur the preparatory process in Geneva, the European Union submitted more discussion papers – negotiated in the 133 Committee – on more subjects than any other WTO member state. The level of consensus among EU member states concerning the form and content of a new round increased, and was confirmed at an informal EU ministerial meeting in Berlin in May 1999. In July the Commission tabled its formal proposal in the form of a Communication to the Council and the European Parliament, The EU Approach to the Millennium Round (European Commission 1999a). In order to facilitate formal EU unanimity the Commission chose an additive method. All issues of interest to all EU member states were added to the draft agenda for the round, which thus was to be ‘comprehensive’ in nature. The Commission judged that it could not obtain consensus in Council for a mandate comprising of ‘market access issues’ only, i.e. negotiations on liberalizing trade in agriculture, industrial goods and services. Trade Commissioner Lamy stated that ‘a market access only Round’ could be ‘superficially tempting’. However, since it would foreclose updating of WTO rules and face resistance from parts of civil society wanting measures in support of sustainable development, the environment, health and safety, Lamy rejected the idea (Lamy 2001a). Rule making – in areas such as investment and competition and relating to the environment – had to compensate for the insufficiency of market access and accompany liberalization in agriculture in order to ensure political support for the final results. In addition, the Commission proposed that the WTO – outside the formal scope of the round – become more active in promoting core labour standards. The Commission stressed that the WTO had to be given a key role in global governance and ‘harnessing globalization’, while helping to increase trading opportunities through liberalization. ‘Specific reciprocity’ (Keohane 1986), which prevails in distributive, ‘mercantilist’ market access negotiations, would not suffice. An agreement based on ‘diffuse reciprocity’, which characterizes regulatory negotiations, had to be obtained. Results on rule making were, according to the Commission, not only indispensable to the European Union, but in everyone’s interest: ‘they are systemic, not mercantilist or zero sum issues’ (Carl 2001). As to be predicted in a three-level game, the Commission exploited the interplay between internal EU interests and external preconditions to gain acceptance for its proposal. Thereby it also rallied support for a free-trade-oriented approach, which otherwise would not necessarily have materialized. The Commission framed the EU interest in terms of the need for the European Union to exert leadership in the WTO, in the absence of US willingness to do so. This line of reasoning was particularly effective vis-à-vis France, which attached value to an EU line that could outweigh US policy. In addition, the Commission framed the proposed EU position in terms of support for developing countries. The need to meet the concerns of the increasingly influential developing countries vouched for a ‘comprehensive round’ in

138 Anders Ahnlid which developed and developing countries alike would be able to place their respective priority issues on the agenda. The Commission and ‘the Northern Liberals’ used the ‘leadership role’ – in particular vis-à-vis developing countries – to underpin the need for EU liberalization, viewed as ‘concessions’ by ‘Club Med’. The combination of the comprehensive approach, the need for leadership and the focus on developing country interests raised the political cost for any EU member state to veto a potential deal on the launch of a round. On the basis of the Commission communication, the 133 Committee and the Council proceeded towards the adoption of conclusions on the Preparation of the Third WTO Ministerial Conference on 26 October 1999 (Council of Ministers 1999a). The October conclusions constituted the formal EU position on the ‘comprehensive round’. The part of the conclusions dealing with agriculture was aligned to conclusions adopted by the Council of Ministers of Agriculture. It balanced between protection and reform in an acceptable way for the Commission and member states. ‘The Northern Liberals’ emphasized the reform language and counted on strong external pressure on the European Union to deliver. ‘Club Med’ stressed the need to safeguard ‘non-trade concerns’ and the ‘multifunctional nature’ of agriculture, which implied that trade-restrictive measures were more acceptable in this sector than in other sectors of the economy. ‘Club Med’ was given additional comfort through the proposal to defend geographical indications. In the negotiations on the conclusions a compromise was struck between ‘Club Med’ countries, led by Spain, which did not accept that the round should lead to a free-trade-oriented reform of the WTO anti-dumping agreement, and ‘the Northern Liberals’, in particular Sweden, which advocated such reform. France led the quest for conclusions that would safeguard the audio-visual sector in the service negotiations. Others were prepared to accept language that gave France sufficient security. Finally, the formulation of the EU objective concerning core labour standards warranted debate. Countries like Germany and Sweden sought an offensive EU attitude, while other members, such as Spain and the United Kingdom, held back. According to the compromise reached, the issue should not be put on the negotiating agenda but the relationship between trade and labour standards was to be analyzed by a ‘joint ILO/WTO Standing Working Forum on trade, globalization and labour issues’. The preparatory process led to a convergence of views among EU member states. The preferences of the member states in ‘Club Med’ had become less defensive. And where protectionist interests were still dominating, the conclusions contained language that provided sufficient reassurance against detrimental reform for the countries concerned. In addition, necessary ‘concessions’ would be reciprocated by equivalent contributions by third parties or, alternatively, balanced by gains in other areas of the agenda. For the offensive ‘Northern Liberals’, the primary objective was to get the round launched. In the internal negotiation these countries were prepared to ‘pay’ to reach that objective by accepting the ‘comprehensive approach’ and by accepting some defensive elements in the Conclusions. The offensive member states counted on the external negotiating dynamic to press the

Setting the global trade agenda

139

European Union towards liberal results in many areas, including agriculture, once the round got started. The breadth of the Commission’s ‘additive’ or ‘comprehensive’ approach and the strong emphasis on reciprocity helped forge consensus internally and make both ‘Northern Liberals’ and ‘Club Med’ agree to the EU position. But the proposal for a comprehensive round caused difficulties externally and added an extra challenge to the negotiating game at the international level, where it faced the real litmus test.

The EU position at the international level The European Union needed a broad round to achieve a balanced and hence workable agenda from its point of view. At the same time, the USA, where industry support was fragmented, continued to favour sectoral negotiations. Many developing countries were not interested in – or even outright hostile to – new negotiations of any sort. As expected, the Commission and the member states anticipated the reactions of other WTO members in the external negotiating arena and adjusted their positions accordingly. The external WTO negotiation on the agenda for a new round had begun well before the decision on the EU October conclusions. The adoption of the conclusions anchored the EU line politically. They strengthened the negotiating position of the European Union in the WTO by binding the hands of its negotiator, i.e. the Commission. Thus, the internal institutional setting empowered the Commission in the external arena. Still, to convince third countries about the virtue of a comprehensive new round EU-style – including far-reaching rule making on issues new to the WTO and engagement in politically sensitive areas such as environment and labour standards – was a formidable task. Dilemmas of leadership and credibility The European Union faced two related difficulties. First, the newly defined role of the Union as leader in the WTO posed a dilemma. While the notion of leadership helped rally EU member states behind the Commission stance, it challenged the USA and other third countries in the WTO. As recognized by Lamy (2001b), in ‘classical negotiator-speak, that sets us up to “pay” for the launch’. Second, the Commission had to overcome the strong hesitation and reluctance to a comprehensive approach held by others due to doubts about EU credibility and true motives. Lamy turned to Churchill’s view on democracy when defending the ‘comprehensive round’, i.e. that it ‘is probably the worst approach – except for all the alternatives. It is without doubt the only approach that stands any chance of securing support from the whole of the WTO membership’ (Lamy 2001a). Just as agriculture was the key determinant of the internal EU position on the need for a comprehensive round, it was the most important factor when third parties assessed the proposal by the European Union. The Union could gain acceptance for a broad round from sceptics and opponents only if it showed

140 Anders Ahnlid genuine willingness to open up its protected agricultural market. The European Union reputation as foot-dragger in the past led many participants to question if it advanced the comprehensive approach in good faith. Was the Commission asking for the impossible from others in order to shroud its own inability to deliver market opportunities in agriculture? Lamy recognized this problem as well, explaining that some participants: will always suspect the EU of tactical games in all this – that the built-in agenda will be held hostage until we have launched a broader Round; and that once we have launched the Round, we will then hold agriculture hostage to our ‘unreasonable’ demands on a host of other, esoteric areas of no interest to the rest of the world. (Lamy 2001a) The decision-making procedure of the European Union, in which member states possess veto power, added to the challenge to the Commission. On the one hand, the possibility of a veto in the sensitive area of agriculture served to accentuate uncertainties regarding the real willingness of the European Union to pay the price for the round. On the other hand, the Commission sought to use the threat of a veto as leverage for its negotiating position in relation to third countries in the WTO. The Commission sought to apply the additive method at the international level, just as it had successfully done at the EU level. WTO member countries had different priorities in a round, ‘but despite, or perhaps because of this, most agree that a comprehensive round, in which benefits can be gained by all, promises the best outcome for their particular priorities’ (European Commission 1999a: 25). The Commission also underlined that some agricultural exporting countries’ support for issues such as investment and competition depended on their hope of obtaining a successful outcome in agriculture negotiations. This led to the vital conclusion that the European Union: must be open to considering, as part of a comprehensive package, issues of interest to others, including further market access in sensitive areas or further rule making where we are not seeking change or where difficulties can be anticipated. These proposals will need to be looked at constructively for our advocacy of a balanced negotiation to be taken seriously. (European Commission 1999a: 25) This was a consequence that ‘the Northern Liberals’ among EU member states cherished, but ‘Club Med’ feared. In the preparations for the WTO ministerial meeting in December 1999, the European Union progressively managed to persuade an increasing number of trading partners to follow its lead. But support was weak and often came ‘in the face of reluctance and procrastination’ (European Commission 1999b). The

Setting the global trade agenda

141

OECD countries and several countries in Latin America and Asia now favoured a round. Suspiciously, the strongest EU support came from Japan, Switzerland and Norway, countries with even more protectionist agricultural policies than the European Union. The USA and other round supporters wanted a slimmer market access agenda. The position taken by most developing countries – including India and many African countries that rejected the launch of a round – was even further away from the EU proposal. For these countries the cost of rejection was low. They did not see a credible response to their difficulties in implementing the results of the Uruguay Round. Neither did they perceive a credible developed country stance on market openings for their products. The European Union and, in particular, the US position on core labour standards added to their opposition. And many small developing countries felt excluded from the informal negotiating process. Accompanied by demonstrations and riots in the streets of Seattle, the Commission did not manage to dispel deep-rooted mistrust and gain sufficient credibility. The EU stance on agriculture was an important reason for the failure to launch a new round in Seattle. In the informal negotiating process the Commission was forced to give up an explicit reference to the EU supported notion of ‘multifunctionality’. The Commission was also prepared to accept a negotiating mandate for agriculture that would lead to substantial reductions in export subsidies ‘in the direction of [their] progressive elimination’ (WTO 1999: 52). While ‘the Northern Liberals’ generally welcomed such a move, the debate in the 133 Committee and Council, meeting in Seattle, showed that France and Ireland were unable to accept such language (Inside US Trade 1999). Third countries, and particularly the Cairns group, saw this as confirmation of their suspicion that the European Union was unable to move meaningfully in agriculture. In his speech to the Conference, Commissioner Lamy stated that others should ‘count the European Union in when it comes to market access negotiations . . . and yes – in agriculture too’ (Lamy 1999), but in the negotiating game he failed to deliver. The European Union also failed to rally support for tackling ‘new issues’. Lamy spoke of the need to create a trading system for ‘controlled globalization’. The demonstrations on the streets of Seattle served to underline the need for control of this sort, Lamy argued. The draft text discussed in Seattle included negotiations on trade facilitation and transparency in government procurement, but postponed any decision on negotiations on investment and competition. Environmental issues were not mentioned in the draft. At its last meeting at Seattle, the EU Council ‘expressed its disappointment that an opportunity has been missed in Seattle’. Persistence of divergences concerning the level of ambition for a new round was seen as one of the main reasons for the lack of result, as were deficient conference procedures. Concerning the future EU stance the Council ‘confirmed that the elements contained in the EU comprehensive approach as defined in the Council Conclusions of 26 October 1999 should be pursued’ (Council of Ministers 1999b).

142 Anders Ahnlid Building confidence and forging consensus Two years of re-assessment followed the Seattle meeting. The Commission was mandated to engage in new ‘exploratory talks’ with WTO members on how to revive momentum for a round. Priority was given to the building of confidence vis-à-vis developing countries. The European Union sought to exercise leadership by turning to ‘carrots’ rather than ‘sticks’. This distinguished the EU approach from the US strategy before the launch of the Uruguay Round in 1986 – when the USA threatened to ‘go unilateral’ and, in practice, abandon the multilateral route if others did not follow suit and agree to a new round. While the European Union had already emphasized the importance of developing country concerns in the October conclusions, this emphasis was increased after Seattle. The October conclusions were not formally changed, but a Commission paper to the 133 Committee (European Commission 2000), informally endorsed by EU ministers in February 2001, de facto modified the conclusions and gave new impetus to the preparatory process in Geneva. The EU trade ministers, in February, stressed that the ‘EU must continue to play a leading role in promoting a New Round by mobilizing political involvement on all sides’ (European Union Presidency 2001). The importance of agriculture for developing countries was highlighted. The negotiations on the in-built agenda on agriculture and services began in 2000 without hope of any major achievements as long as the negotiations took place in isolation. The Commission explained that real progress would be possible only within the framework of a round. Anti-dumping, which could not be mentioned in the October conclusions, was now explicitly singled out as an area where the European Union had to take developing country concerns into account. In order to meet opposition against the EU proposals on investment and competition, a ‘plurilateral’ approach was suggested, implying that developing countries that so wished would be able to stay outside any future agreements in these areas. Thus the European Union was willing to let disadvantaged developing countries ‘free ride’ on deals clinched by others. In the internal EU debate the Commission used the positions taken by developing countries to strengthen its argument as to why EU member states would have to accept a new line. In this way, the Commission pushed for additional room of manoeuvre in the preparations for the fourth WTO ministerial meeting, to be held in the autumn of 2001 in Doha, Qatar, where the next opportunity to launch a round would be at hand. An additional ‘carrot’ was the decision by the European Union in February 2001 to abolish all tariffs and quotas on imports from the 49 Least Developed Countries (the so-called Everything But Arms initiative). It signalled that the European Union was both willing and able to make and implement decisions that involved a political cost – including in the area of agriculture – for many member states. The European Union also decided to lead efforts to secure developing country access to medicines for diseases such as HIV/AIDS under the WTO rules on intellectual property rights. Finally, the European Union backed the establishment of a special procedure for the handling of developing country difficulties to

Setting the global trade agenda

143

implement the Uruguay Round agreements and stressed the importance of traderelated technical assistance. In the lead up to the Doha meeting the Commission combined formal rigidity with pragmatism in the preparatory process. The Commission did not formally deviate from the October conclusions, but it used the flexibility embodied in the conclusions, and accepted in the internal preparatory process, to move the EU position closer to developing countries and, thus, in a more free-trade-oriented direction. At the same time the need for others to accept the EU agenda was emphasized. At a preparatory meeting in June 2001 an EU negotiator declared that the agenda of the new round: must be balanced. That means that it must include enough issues of interest to all. [But] the European Community accepts the principle of a ‘better balance’ for developing countries with respect to their key economic and other interests. (Carl 2001) More specifically, on the issue of market access in relation to other negotiating areas, the European Union recognized that: many [WTO] Members are probably expecting more from the EU than from many other participants . . . we are conscious that others are looking to the Community for a major contribution on market access, both on agriculture, services and industrial issues. We are ready to play our part if the overall balance is right. (Carl 2001) But the Commission also maintained that the agenda of the round had to go beyond market access and include areas such as investment, competition, health, consumer safety and environment. Lamy made it clear that ‘trade rules must not override the other concerns of society’ (Lamy 2001c). The comprehensive approach, now officially labelled ‘inclusive’, continued to meet resistance, in particular from developing countries. Thus, the Commission intensified its process of ‘persuasion’ and confidence building. The European Union managed to build up an increasing amount of support for an inclusive round step by step. The Everything But Arms initiative contributed, as did the EU line on access to medicines. At Doha, the European Union continued using carrots rather than sticks and worked hard to gain acceptance for a WTO waiver that gave 78 former colonies of EU states in Africa, the Caribbean and the Pacific continued preferential treatment in the EU market. The waiver served as a side payment to the ACP countries for their acceptance of the mandate for the negotiations that was hammered out in Doha. The terrorist attacks against the USA on 11 September 2001 contributed to a cooperative atmosphere in Doha, as did the renewed working relationship between the European Union and the USA. At the EU–US summit meeting in Gothenburg in June, ‘the two big elephants’ agreed to cooperate closely in the

144 Anders Ahnlid Doha preparations. And common ‘quad’ positions (including the European Union and the USA as well as Japan and Canada) were worked out on the ‘new’ issues – investment, competition, trade facilitation and transparency in government procurement – before Doha. In Doha the formulation of the mandate for the negotiations on agriculture was key to success. Again interest focused on the negotiating mandate for export subsidies. The USA, the Cairns group and many developing countries held that the coming round should put a definitive end to the use of such subsidies. The Commission worked on a formulation, according to which the negotiations should be conducted ‘with a view to phasing out’ export subsidies. France and Ireland threatened to veto such an outcome. After lengthy debate the Commission came back with a new text, which stated that ‘without prejudice to the final outcome’ negotiations should take place ‘with a view to phasing out’ export subsidies (WTO 2001). This was acceptable to all. Exporters of agricultural products got a text that – for the first time ever – included the notion of an end point to export subsidies. France, Ireland and the like could show that no decision had been taken that prejudged the final outcome. The Commission managed to ‘sell’ this concession dearly. In return, the European Union got acceptance for its part of the agricultural agenda relating to ‘nontrade concerns’. The agenda provided for some additional work on geographical indications. The European Union got a negotiation on ‘trade and environment’. Even if the mandate did not go as far as the European Union would have wished, placing the matter on the agenda for a WTO negotiation was a political breakthrough. The Doha Declaration also stated that negotiations on investment, competition, trade facilitation and transparency in government procurement were to start after the next WTO ministerial meeting in 2003, provided that a decision by ‘explicit consensus’ could be taken at that meeting on the modalities of the negotiations. This weakening of the deal was part of the price the European Union had to pay for being demandeur in order to satisfy India and some other developing countries, which would have preferred no negotiations at all on these issues. The only major departure from the EU October conclusion related to core labour standards. The Doha Ministerial Declaration did not contain the EU objective of a ‘joint ILO–WTO standing, working forum’. Nevertheless the text noted the cooperation between the WTO and the ILO, and the gains in other areas were sufficiently large for the Commission and the member states to declare victory in Doha. Even France – which conducted a traditional act of brinkmanship towards the end of the conference – hailed the outcome. The Doha Development Agenda – a negotiation on the basis of a single undertaking and thus ‘a round’ – was launched and set to conclude no later than 1 January 2005.

Agenda set – but is the European Union ready to lead all the way? The resemblance between the Doha Development Agenda and the EU proposal was sufficient to permit the conclusion that the European Union was able to

Setting the global trade agenda

145

provide effective leadership. The European Union managed to launch a new trade round, which would advance free trade – including in agriculture – despite internal protectionist pressures. In the three-level negotiating game, the Commission successfully played a key role as conduit between domestic preferences in EU member states and the positions of third countries. The Commission’s skilfully crafted strategies – which exploited the institutional setting of the European Union and the WTO – paved the way for the successful outcome. Thereby a balance was struck between conflicting internal EU interests in agriculture and new issues, on the one hand, and constraints in the form of resistance from third countries, on the other hand. The ability of the European Union to lead in the pre-Doha agenda-setting phase was negatively affected by CAP and the new conditions for trade policy making, but not detrimentally so. With a treaty-bound free-trade objective as the starting point, the Commission invented new methods to circumvent hurdles and largely managed to find ways and means to move the EU position in a free-trade-oriented direction. The internal strategy to frame the EU stance in terms of leadership and responsibility for developing countries was largely successful. A mandate on agriculture that EU members were about to turn down in Seattle was accepted in Doha. The internal institutional setting empowered the Commission in the external arena. In Doha, WTO members by and large accepted the political need in EU member states to start negotiations in the new areas, including ‘trade and environment’, albeit not exactly in the way suggested by the European Union. But the 1996–2001 campaign also points to important problems involved in EU leadership. The overall strategy pursued was still largely conditioned by agricultural concerns. It remains to be seen to what extent the Commission and ‘Northern Liberals’ among the member states manage to use the external pressure – legitimized through the Doha agenda – to promote CAP reform. The European Union moved on agriculture in the early phase of the round, but not sufficiently so, in the view of others. And the European Union still suffers from a lack of credibility, accumulated over the years. The additive or comprehensive approach – necessitated by the need for consensus decisions both in the European Union internally and in the WTO – also involves problems. Distributive gains stemming from market access were not deemed to produce a sufficiently attractive result for the European Union. Regulatory outcomes were also sought, partly to outweigh ‘losses’ in agriculture, partly in response to calls from civil society. Hence, the quest for rules in new areas to ‘harness globalization’. This will probably make it more difficult to strike a deal in the Doha negotiations, since specific reciprocity will not suffice and diffuse reciprocity is likely to render the negotiating game more difficult at all levels. This was illustrated by the unravelling of the compromise from Doha on the new issues at the 2003 ministerial meeting in Cancun, Mexico. No ‘explicit consensus’ on modalities for negotiations was reached, and the meeting failed to push the Doha negotiations forward. The difficulties to act as leader in a negotiation involving – presently – 148 WTO members, including a growing number of increasingly vocal developing countries, became evident. Under these circumstances, and given the

146 Anders Ahnlid uncertainties involved, ‘the single undertaking’ can be either a magic spell or a lethal curse for global trade negotiations. Yet, for the European Union, with its lingering defensive posture in agriculture, there is no alternative to its use. The fact that the European Union by and large passed the test of leadership in the agenda-setting phase pre-Doha is no guarantee that the same will hold in the following phase, i.e. in the real negotiations. It remains to be seen if the European Union will be able to lead WTO members to a successful conclusion of the Doha Development Agenda. Developments so far have clearly pointed to the difficulties involved in doing so.

Notes 1 ‘The European Union’ is used instead of the legally correct term ‘the European Community’, except for actions before 1 November 1993, when the Treaty on European Union entered into force. 2 According to Article 133 of the Treaty the objective of EU trade policy is ‘to contribute in the common interest, to the harmonious development of world trade, the progressive abolition of restrictions on international trade and the lowering of customs barriers’. This objective follows, in turn, from the overall economic, social and ecological objective found in Article 2.

References Ahnlid, A. (2000) ‘The Uruguay Round Services Negotiations, Economic Basis, Political Basis’, in V. Kremenyuk and G. Sjöstedt (eds) International Economic Negotiation: Models versus Reality, Laxenburg: International Institute for Applied Systems Analysis. Balladur, E. (1995) Deux ans à Matignon, Paris: Plon. Brittan, L. (1996a) ‘Commission Calls for Sharper EU Trade Policy’, in Monthly Bulletin of European Union Economic and Financial News, March. Brittan, L. (1996b) ‘Expanding World Trade: The WTO Road to Singapore’, speech before the Graduate Institute of International Studies, Geneva, 1 April. Carl, P. (2001) ‘The Agenda of Doha Must be Balanced With a “Better Balance” for Developing Countries’, speech before the Informal General Council of the World Trade Organization, Geneva, 25 June. Council of Ministers (1999a) Preparation of the Third WTO Ministerial Conference: Council Conclusions, the European Union, 26 October. Council of Ministers (1999b) Establishment of the Union Position on the Texts Submitted to the Conference for Approval, Conclusions, the European Union, 3 December. Croome, J. (1995) Reshaping the World Trading System, Geneva: World Trade Organization. European Commission (1999a) The EU Approach to the Millennium Round: Communication from the Commission to the Council and to the European Parliament, Brussels, 8 July. European Commission (1999b) Press release, 8 July. European Commission (2000) State of Play and Strategy for the New Round, paper prepared for the 133 Committee, 13 December. European Union Presidency (2001) Informal Discussions of Ministers Responsible for Trade Policy: Background Note from the Presidency and Chairman’s Summing Up, Press release, Brussels, 25 February. Financial Times, 6 August 1996 and 30 March 1998.

Setting the global trade agenda

147

Inside US Trade (1999) ‘Agriculture Ministers Move Towards Draft Ministerial Text’, Inside Washington Publication, 3 December. Keohane, R. (1986) ‘Reciprocity in International Relations’, International Organization, 40: 1–27. Lamy, P. (1999) Statement Before the WTO Ministerial Conference, Seattle 30 November– 3 December. Lamy, P. (2001a) ‘Special Treatment for Agriculture: The Way Ahead’, speech before the Oxford Farming Conference, 4 January. Lamy, P. (2001b) ‘US–EU: The Biggest Trading Elephants in the Jungle – But Will They Behave?’, speech before the Economic Strategy Institute, Washington, 7 June. Lamy, P. (2001c) ‘Towards a Common Goal: Making the WTO Work Better for all Members and all Citizens’, speech before the WTO NGO symposium, Geneva, 6 July. Patterson, L. A. (1997) ‘Agricultural Policy Reform in the European Community: A ThreeLevel Game Analysis’, International Organization, 51: 135–65. Putnam, R. (1988) ‘Diplomacy and Domestic Politics: The Logic of Two-Level Games’, International Organization, 42: 427–60. WTO (1999) Draft Ministerial Text, from the Ministerial Conference in Seattle, reprinted in WTO:s ministerkonferens, Seattle 30 November–3 December 1999, Stockholm: Ministry for Foreign Affairs, 2000. WTO (2002) ‘Ministerial Declaration’ from the Ministerial Conference in Doha, reprinted in WTO:s ministerkonferens, Doha 9–14 November 2001, Stockholm: Ministry for Foreign Affairs.

10 The European Union as actor in international relations The role of the external environment for EU institutional design Joakim Reiter

Introduction The European Union has commonly been depicted as a weak and reactive, or sometimes even non-visible, actor in international relations. Underpinning this weakness, it is often claimed, is the circumstance that the EU member states hold sway and hesitate to act as an entity. To paraphrase a former US Secretary of State: there is still no single phone number one can call to talk to Europe. The reality of the matter is that it depends. A single phone number would no doubt suffice if one wanted to talk to the European Union about international trade, agriculture or, at least since 1999, monetary policy. Even on global environmental issues and some aspects of foreign policy, as well as on a few domestic issues, a single phone number could be sufficient – at least if one would be satisfied with a somewhat fuzzy description of a common EU standpoint. The problem is that it would be impossible, in most cases, to bring up more than one topic at once. Instead of a phone book of 25 EU heads of state, one would end up with a list of perhaps five to ten top EU functionaries and probably still have to keep track of the EU heads of state as a backup. In short, what constitutes and who represents the European Union in global affairs remain elusive to any outside observers, just as it probably does to the European public at large. For the European Union in global affairs, diversity remains the rule – not the exception. Both within and across issue areas, there are large variations in the characteristics and functioning of the European Union as an actor in international relations. In theoretical terms, there are great differences in EU ‘actorness’. In one issue area, the European Union might have reached a high level of actorness – speaking with one voice or even a single mouth – while in other areas it is hard to view the European Union as an actor at all. The purpose of this chapter is to probe more closely into the puzzle of the issue specificity of EU actorness. It provides an attempt to trace the diversity in the structures of the Union’s internal policy coordination and EU actorness to the external environment and, more specifically, to the international cooperative arrangements in which the European Union and its member states participate.

The EU as actor in international relations 149 First, however, we will describe the conventional explanation of the diversity in EU actorness, which starts from the premise of the evolving internal integration of the Union – slowly stretching the community’s competence into a wider spectrum of issue areas – and we will demonstrate the inadequacies of this explanation.

The diversity in EU actorness and the role of community competence There are many variations of the EU policy process. In the academic literature, these variations range from intergovernmentalism, multi-level governance, to forms of supranational policy making (Wallace 2000: 28–35). In some areas, where EU integration is most progressed, decision making has been fully transferred from the national to the EU level. This entails thorough policy coordination among member states. In most cases, there has been a corresponding shift in powers from member states to the Commission and hence in the institutional design of the European Union. Trade policy is perhaps the best example. But there are others. For example, the Common Agricultural Policy (CAP) has virtually denationalized agriculture and replaced it with a European policy regime. In other policy areas – the single market, competition, as well as the social and environmental domains – the European Union has developed a regulatory model, which in a flexible manner combines transnational EU standards with different countryspecific standards. In yet other areas, such as employment policy, taxation and security policy, the EU level only provides a forum for more or less non-binding policy coordination. In such cases, the role of the Commission is more to stimulate debate and to promote policy convergence and/or, by benchmarking, to spread best practices. Similar variations exist in the EU policy process in international deliberations. In any given international context, the actions and behaviour of the Commission and the European Union member states fall into one of four broad categories – each representing a higher level of EU actorness: ●

●

●

●

The European Union can act in an uncoordinated way, i.e. allowing for each member state to pursue more or less its own agenda without meaningful (or binding) prior coordination with other member states. The European Union can attempt to convey a single message, through the adoption of common understandings, while allowing for each member state to act separately and have different priorities. The European Union can act as single entity, speaking with one voice, but with an ever-changing messenger, due to the fact that the rotating Presidency has the authority to represent the Union. The European Union can be a unitary actor, whereby the Commission speaks on behalf of the Community (hence having the role of a single messenger).

The conventional wisdom is that the variation in actorness can be best explained by the integration process of the European Union and the internal distribution

150 Joakim Reiter of competence in the union that follows from this integration process. The issue of competence concerns the assignment of policy powers between the European Union and its member states in any given issue area. EU competence, called community competence, stems from the treaties and their progressive amendments through intergovernmental conferences (IGCs). The competence of the European Union can also derive from the existence of Community measures. In sum, the European Union has exclusive competence in a given area by virtue of a treaty article or the adoption of internal measures (Macleod et al. 1996: 72–4). As a rule of thumb, the logic goes, the more extensive the community competence, the greater is the EU actorness in external relations. In practice, however, it is more complex. In some policy areas the competence can be mixed between member states and the European Union. Also, in many cases, there is a lack of clarity about competence, partly due to the evolving nature of the issues at stake. Trade policy is illustrative of this point (Woolcock 2002: 75–9). Trade policy is one area where the exclusivity of community competence appears, at first glance, to be both extensive and clear-cut. Through the establishment of a customs union, the Treaty of Rome granted the then European Economic Community exclusive competence for a common commercial policy. Since the treaty article (Article 113, now 133) referred to tariffs, anti-dumping and subsidies, the distribution of competence was relatively straightforward for these issues. But the treaty article did not provide an exhaustive definition of trade policy. Therefore, the gradual extension of the trade agenda in the international trading system has repeatedly forced the European Union to redefine the community competence. As a consequence, the question of community competence has been subject to lengthy discussions among the Commission and member states, not least in conjunction with treaty amendments at IGCs. But on many new trade-related issues community competence still does not apply exclusively. Both the Amsterdam Treaty of 1996 and the IGC at Nice in 2000 underscored that member states and the Community share competence on a number of issues, such as investment, environment, competition, services and intellectual property rights. The new draft treaty establishing a Constitution for Europe, that was agreed upon among EU heads of state in June 2004, does entail some expansion of the community competence into these new trade-related areas, but it involves only market access considerations. The new treaty would not challenge the basic premise that the community competence in many new trade-related issues, concerning everything other than market access, depends on the existence of internal measures.1 Hence, the legal uncertainties in the application of the principle of internal competence are abundant. As an explanation for EU actorness in external relations, the question of competence thus provides only a valuable starting point. It probably gives a more or less accurate overall picture and overview of the general differences in the way the European Union behaves in various issue areas. For example, the issue of competence could possibly explain at least in part why the European Union manages to speak with a single mouth in the area of trade, while it

The EU as actor in international relations 151 is, at the very most, capable of speaking with a single voice in the area of development cooperation, finance and security. The devil is in the details, however. The practical difficulties and the evolving nature of community competence raise serious questions about the validity of internal competence as the sole, or even key, determinant for EU actorness in external relations. Much remains unaccounted for. Even though it does explain differences in EU actorness between various policy areas, internal competence falls short of explaining variations in EU actorness between areas of mixed competence. Why, for example, does the European Union manage to speak with one voice and assert collective positions in international environmental negotiations while failing to do so in the area of finance? If competence was everything, surely the European Union should act similarly in both contexts. One could even expect a higher level of EU actorness in finance, not least since 1999. Internal competence also fails to provide an explanation of the evolution of community competence. Transfers of competence involve conscious decisions by member states. For example, why do member states in international trade negotiations – on mixed competence issues – approve the adoption of a pragmatic approach and give the Commission the authority to act as sole negotiator for the European Union, while member states resist any substantial role for the Commission on similar issues in international financial organizations? Perhaps more important, the picture becomes even more puzzling if one takes into account the discrepancies in the behaviour of the European Union within a specific issue area. For example, in trade policy the Commission acts as sole negotiatior and spokesman for the European Union in the World Trade Organization (WTO), whereas both the Commission and the EU member states speak when the same trade issues are being discussed in the Organization of Economic Cooperation and Development (OECD). And in UNCTAD – the United Nations Conference on Trade and Development – it is the EU Presidency that speaks on behalf of the European Union just as often as the Commission. The internal competence issue thus fails to further our understanding of why the European Union, as an actor, is represented in different ways in the WTO, the OECD and UNCTAD, despite the fact that all three organizations deal with trade issues. The same unexplained variation in EU actorness can be found in the area of international finance. Here, the depth and scope of internal EU policy coordination is much more extensive in the International Monetary Fund (IMF) than, for example, in the Financial Stability Forum (FSF) or the G–7, despite the fact that the internal EU competence could not be considered more progressed for the issues dealt with by the IMF than by the FSF. All this diversity in EU actorness remains unaccounted for by internal competence. The variation in EU actorness in external relations is much greater than would be predicted by internal competence alone. Substantial leeway seems to exist for adopting flexible approaches for the European Union – both between and within separate issue areas – regarding the way the Union behaves and acts in international deliberations.

152 Joakim Reiter

The external environment as determinant for EU actorness In order to find other potential explanatory variables for the European Union as actor in external relations, it is helpful to consider EU actorness as, by and large, a result of internal bargains over institutional design. In line with the broad tradition of rational-choice analysis, the issue-specific design of EU actorness should reflect the intentions of the EU member states to maximize the benefits of international cooperation.2 The integration process of the European Union is in itself a repetitive game over institutional design. Consequently, member states choose or do not choose to act collectively. Over time, design differences in EU actorness are not random. What is more, the development of the institutional design of EU actorness tends to entail hard bargaining between member states as well as between member states and the Commission. Institutional design affects outcomes and also the internal redistribution of benefits and costs. Naturally, the Commission has tended to fill the role of a strong proponent for the idea that the European Union should speak with one voice or even one mouth. But member states ultimately decide, since it is they who have to delegate authority in order for the European Union to be able to act collectively. In the process of developing institutional design, member states are still very much in the driving seat, and they tend to care about the steering. When making decisions on the institutional design of EU actorness, we would expect that at least three overarching considerations to influence the preferences and interests of EU member states.3 First, there must be a perceived need for reforming EU actorness. Member states must consider the existing institutional design of EU actorness in a specific issue area to be inadequate in order for the European Union – collectively and/or the individual member states – to reap the full benefits of their actions in external relations. Second, a new institutional design of EU actorness must be credible. Member states must believe that the new design, in terms of more collective decision making, will deliver superior results, i.e. that the reform will be effective. Third, the new institutional design must be nonthreatening. Member states must perceive the new design, by and large, as something of a positive-sum reform that will not provide an impediment to their attainment of national goals. Therefore, EU member states should be more prone to act in a concerted way, or even to speak with one voice, if this would realistically enhance their benefits from external actions without constituting any threat to each and every one of the member states. In evaluating the cost and benefits of increased EU actorness, member states cannot discount the specific international context in which the reform of the European Union’s institutional design is taking place. For example, since the EU member states are also members of a number of international regimes and organizations, the delegation of additional authority to the EU level has implications for the role and position of individual EU member states in those organizations. These implications are, in turn, dependent on the specific characteristics of the international organization. If all international organizations and contexts were alike, the design options for EU actorness would be the same in

The EU as actor in international relations 153 all organizations. But the functioning and characteristics of international organizations vary. Therefore, we would expect the EU member states’ evaluation of available forms of EU actorness to differ with each international setting and organization. By taking the external environment into account, it is thus natural for EU member states to opt for different designs of EU actorness. In sum, the external environment influences the way member states view the need for a reform in EU actorness as well as the attractiveness of stronger collective decision making. But the external environment in a systemic sense, e.g. the anarchical world order, appears too broad to account for the diversity in EU actorness. Instead, it is the issue-specific international settings – in terms of the various international regimes and cooperative arrangements that the European Union participates in – that provide constitutive structures for EU actorness. By viewing the EU policy process as a result of intentional institutional reform, the diversity of EU actorness in external relations must also be traced to the specific characteristics of each international forum or context in which the European Union engages.

Hypotheses on the relation between international cooperative arrangements and EU actorness Starting from the proposition that there is an interrelationship between the external environment and the design of the European Union in global affairs, the variation in EU actorness can be explained in terms of categories of systemic and organizational characteristics of the international regimes, i.e. cooperative arrangements, in which the EU member states participate. Three interrelated categories of explanatory variables can be identified, each of which produces a number of hypotheses about the diversity in EU actorness: ●

●

●

degree of hierarchy in a given issue area in terms of distribution of power among the parties and the decision-making procedures of the international cooperative arrangements; forms of cooperation, such as scope and purpose of the existing international cooperation arrangements; membership structure, i.e. the number of participating EU member states, in the international cooperative arrangements.4

The degree of hierarchy constitutes a first defining feature of the issue-specific international cooperative arrangements. In its broadest interpretation, it refers to the underlying distribution of power and the potential dominance of one or a few states in the cooperative arrangement. Some countries have managed to attain positions, formally or informally, in certain areas that give them disproportionate influence compared to their overall size and might. In international cooperation on finance and the control of arms trade, for example, otherwise small and less important countries are at the forefront of decision making. In its more narrow interpretation, hierarchy is equated with control and the decision-making procedures in

154 Joakim Reiter international organizations. For example, in some organizations all members have equal influence in decision making, while other organizations have explicit weighted-voting rules or allow vetoes for member states. The degree of hierarchy should influence, in a predictable manner, the prospect for furthering EU actorness. If only a few of the EU member states have retained a privileged position in the international cooperative arrangement, it is more unlikely that these member states will agree to develop EU actorness. If, on the other hand, all EU member states are on equal footing, the prospect for collective decision making in the European Union, including agreement to speak with one voice (if not one mouth), should be greater. This should hold true for both the distribution of power and decision-making procedures.

Proposition concerning hierarchy A higher level of EU actorness is more likely if the underlying distribution of power is relatively even and decision making is based on consensus in an issue-specific cooperative arrangement.

The forms of cooperation, i.e. scope and purpose, represent the second category of potential explanatory variables as to how the European Union chooses to design its own institutions in external relations. Scope refers to the depth and breadth of the issues that are subject to international cooperation and fall under the mandate of a specific cooperative arrangement. The purpose, or aim, of international cooperation can range from exchanging information, exchanging experiences (disseminating best practices) and providing a forum for discussions and negotiations on soft law to negotiating hard law in terms of binding and enforceable commitments.5 Both scope and purpose ought to influence EU actorness. Presumably it is easier for EU member states to act as a single entity when they are dealing with narrowly defined areas of international cooperation. A narrow agenda poses less of a potential threat to the future pursuit by EU member states of their own national goals. Everything else being equal, therefore, the greater the scope of international cooperation, the less prepared the EU member states will be to adopt extensive collective decision making in the European Union and to further EU actorness. As regards the purpose of international cooperation, EU member states are expected to favour greater EU actorness when the international cooperative arrangement is oriented towards negotiations leading to binding agreements. In such negotiations the European Union is probably more effective when the negotiating position is closely coordinated and, ultimately, when it speaks with one voice. However, if the focus of international cooperation is on information exchange or the dissemination of best practices, the interests of the European Union – as a group as well as individual member states – are more likely to be served by maximizing the flow of information and hence the number of ‘EU voices’ in the discussions.

The EU as actor in international relations 155 Proposition concerning the form of cooperation A higher level of EU actorness is more likely when the scope of the international cooperative arrangements is narrow, and the purpose of the international cooperation is to establish binding rules.

The membership structure of the international cooperative arrangements represents a third category that produces a hypothesis of its own. The membership structure can either be inclusive or restrictive, based on geographical origin, the level of development or any other relevant criteria. For the purpose of the argument presented here, however, the total number of participants in an international cooperative arrangement is of primary importance in relation to the representation of EU member states. This factor should also have obvious implications for the willingness of EU member states to reform EU actorness. If only a few EU member states actually participate as full members in international cooperation in a specific area, we would expect them to be hesitant to constrain their present room for maneuver by increased concerted actions by the European Union. The negative attitude towards a higher level of EU actorness should also be expected to increase in cases where the participating EU member states already constitute a dominant part of the total membership. This is because closer coordination among EU member states would heighten the risk for non-EU members to view the international cooperative arrangement as a purely ‘European affair’. If, on the other hand, all EU member states participate in international cooperation in a specific issue area, increased policy coordination among EU member states should be more readily attained, especially in cases where the EU member states represent only a minority of the total membership of the international cooperative arrangement.

Proposition concerning membership structure A higher level of EU actorness is more likely when all or a vast majority of the EU member states are part of the international cooperative arrangement and when the EU member states do not provide the bulk of the total membership.

These three propositions can be summarized in a highly simplified and stylized model (Figure 10.1). It is worth underlining that these three categories of explanatory variables, and the propositions that follow from them, are all interrelated. Taken as a whole, they provide a general, testable model of the level of EU actorness for each specific international cooperative arrangement. For the purpose of this chapter, the model will be tested on international cooperation in the areas of trade and finance, which both are areas of great importance to global economic governance and areas where the European Union now has extensive powers.

156 Joakim Reiter

Hierarchy

Forms of cooperation

Membership structure

More EU actorness

Less EU actorness

Distribution of power

Even

Unequal

Voting rules

Consensus

Weighted

Scope

Narrow

Broad

Purpose

Binding agreements (hard law)

From info-exchange to peer review and soft law

Participation

Inclusive, all EU member states

Restricted, only a few EU member states

Figure 10.1 Stylized model of propositions on the role of international cooperative arrangements for EU actorness.

The European Union as actor in trade and finance Trade and finance represent two important areas that are at the core of post-war cooperation and, hence, the contemporary international system. In these areas there exist a number of international cooperative arrangements that precede and have evolved in parallel with European integration. Trade and finance are, of course, separate areas of international cooperation. The history of European integration also reflects these important differences in trade and finance. By virtue of the establishment of a customs union, cooperation in trade policy has preceded that in finance. Naturally, therefore, the exclusivity of community competence in trade is more extensive. Financial policy is still subject to mixed or national competence. At the same time, however, these differences should not be exaggerated. As stated above, mixed competence is still the rule on a number of important trade issues. The Maastricht Treaty, establishing the Economic and Monetary Union (EMU), and not least the beginning of the third stage of EMU from January 1999,6 also mean that monetary policies have become a field of exclusive competence for the Eurosystem and that the competence over exchange rate policy is shared between the member states, the Eurosystem and the Commission. In addition, during the last decade, a number of internal measures have been adopted to promote financial integration in the European Union, including directives on financial regulation (Evans 2000: 25). In this section four different case studies will be conducted of cooperative arrangements in trade and finance – the WTO, the OECD, the IMF and the FSF. The case studies aim to test if it is possible to trace the differences in EU actorness, both between and within the two issue areas, to the cooperative arrangements in which the European Union participates. The WTO and the OECD illustrate the diversity in EU actorness on trade policy. The IMF and the FSF provide, in turn, two case studies in the area of finance.

The EU as actor in international relations 157 EU actorness in cooperation on international trade policy The Organization for Economic Cooperation and Development and the World Trade Organization are very different settings for international trade policy making. But there are some notable similarities as well. In neither cooperative arrangement is it possible for a single state, or a tiny fraction of the membership, to dominate the organization’s work. The cornerstone of decision making in the WTO and the OECD is that member states are treated equally. All member states are represented in the decision-making bodies of the organization and decisions are, in almost all cases, taken by unanimity. With the exception of hierarchy, however, the two organizations differ on the remaining categories of the defining characteristics of an international cooperative arrangement. The WTO is an umbrella organization covering the multilateral trade agreements in goods (GATT), services (GATS) and intellectual property (TRIPS). While the scope of trade policy as such has increased dramatically in the last decades, starting from the original GATT of 1947, the WTO still only covers trade or trade-related issues. This should be contrasted with the scope of the OECD. Much like GATT and later the WTO, the Convention establishing the OECD, which was signed in December 1960, stipulated that one of the key aims of the new organization was to promote policies designed ‘to contribute to the expansion of world trade on a multi-lateral, non-discriminatory basis in accordance with international obligations.’ To that end, the Trade Committee was charged with the responsibility to deal with various aspects of trade policy and primarily with nontariff issues (Timmins and Timmins 1985: 54–5). However, trade is just one of the areas that the OECD covers. In line with the statutes, the OECD deals with economic policy at large. The themes of the OECD are thus both numerous and quite disparate. Similarly, the purposes of the WTO and the OECD are different. The WTO provides a forum for multilateral trade negotiations with the ultimate aim of creating new binding and enforceable commitments for its member states (i.e. hard law). The OECD has also hosted a number of trade negotiations, such as sectoral negotiations in agriculture, shipbuilding and steel, as well as negotiations on export credits and a multilateral agreement on investments (Blair 1993: 109–256). However, in most instances, the work of the OECD tends to be more oriented towards discussion, the dissemination of best practices and the development of guidelines and non-binding commitments (soft law). A key objective of the Trade Committee is to support and facilitate trade negotiations in the WTO. The advantage of the OECD is thus that it provides more of a setting for reflection and discussion – something there is less scope for in the WTO process.7 Finally, the organizations are very different in terms of membership structures. While GATT had originally only 23 member states, GATT/WTO has over the decades evolved into an organization with almost universal reach. At present, the WTO has 148 members. Both the EU member states and the European communities (represented by the Commission) are full members of the organization. This could be contrasted with the membership of the OECD, which comprised 20 countries from the start and, despite gradual accession, has only 30

158 Joakim Reiter member countries at present. This also means that half of the total membership are also members of EU–15, or comprise two-thirds of EU–25. What is more, after EU enlargement, not all EU member states are any longer part of the OECD. Also, the European Communities (the Commission) is an observer but is taking part in the activities of the OECD, both at the Committee level and in the Council, the highest decision-making body of the organization. However, the Commission has no right to vote (Figure 10.2). In line with the hypotheses above, EU actorness should be much more evolved in the WTO than in the OECD. Based on the broad scope of the OECD, the organization’s focus on best practices and soft law, as well as the dominance of EU countries in the total membership and the fact that not all EU members are part of the organization, we would expect EU member states in the OECD to be reluctant to agree to speak with one voice. Instead, it would serve EU interests better if all EU member states participated in the OECD discussions, with prior coordination on a common EU line. In the negotiation-oriented WTO, on the other hand, the EU member states have a shared interest to agree to act as a single entity. The welldefined scope of the organization also means that intense coordination poses less of a threat to the EU member states. Reality seems to fit nicely with these predictions about the effect of the external environment on EU actorness. The European Union acts and behaves very differently when dealing with trade issues in the WTO and the OECD. In the WTO – in what appears to be a reflection of the external environment – the European Union speaks with one voice, and even one mouth. EU member states are represented in the various committees and in the General Council but refrain from speaking in those forums. The EU positions are instead agreed upon internally, through the establishment of a negotiating mandate and through continued consultations in the so-called 133 Committee. The Commission has then been given the responsibility to act as sole negotiator at WTO meetings. Even at the ministerial meetings, which

Hierarchy

Forms of cooperation

Membership

WTO

OECD

Distribution of power

No dominance by a single country

No dominance by a single country

Voting rules

One country, one vote Consensus

One country, one vote Consensus

Scope

Narrow

Broad

Purpose

Binding agreements (hard law)

From info-exchange to soft law

Participation structure

Inclusive, all EU member states are part

Restricted, all EU–15 are members but not all EU–25

Figure 10.2 The WTO and the OECD.

The EU as actor in international relations 159 take place every other year, an individual EU member state does not participate in the negotiations and drafting of the Ministerial Declaration but acts through the Commission. In most of the OECD deliberations, however, all EU member states still retain their right to engage in the discussions, despite the fact that the Trade Committee deals with many issues directly related to what is being negotiated in the WTO. Non-binding forms of policy coordination usually take place before Committee meetings. Also, the Commission often has a leading role for the European Union in the OECD discussions. But it is common for EU member states to present views of their own. Naturally, EU member states are expected not to depart, at least not too blatantly, from common EU positions – if such have been established. Beyond that, however, the EU member states are more or less free to engage in any way they like. At ministerial level, the EU line is, if anything, even looser. EU actorness in international financial cooperation Hierarchy in the area of finance is characterized by the dominance of a small group of countries and financial centers (Cohen 1998b: 113–30). This holds true, albeit for different reasons, for both the International Monetary Fund and the Financial Stability Forum. In the IMF, which from its creation in 1945 has been at the centre of international financial cooperation, the high degree of hierarchy is reflected in the weighted-voting rules of the organization and the distribution of seats in the Executive Board. Since the IMF is a financial organization, the votes of each member state are determined on the basis of that member’s contribution to the fund, the so-called quota. The highest decision-making bodies are the Board of Governors and the International Monetary and Finance Committee (IMFC), in which all member states are represented. In between the annual meetings of the Board of Governors and the biennial meetings of the IMFC, however, the Executive Board handles the IMF’s day-to-day business. The Board consists of 24 executive directors. Eight member states have directors of their own: the USA, the United Kingdom, France, Germany, Japan, Russia, China and Saudi Arabia. The remaining 176 countries are divided into groups. These groups share the other 16 posts as executive directors and pool their votes together accordingly. Nonetheless, it should be noted that, in reality, voting is rare – the deliberations among the IMF’s executive directors usually produce consensual decisions. The institutionalized hierarchy under the IMF could be contrasted with the principle of equality in decision making in the FSF. In the FSF all participants are of equal standing. However, less than a dozen states are full members. The FSF, which was established by the G–7 Finance Ministers and Central Bank Governors in 1999, includes – apart from financial regulators from the G–7 countries – representatives from Australia, Hong Kong, Singapore and the Netherlands, as well as from international financial and regulatory institutions and the European Central Bank (ECB).8 As such, the FSF constitutes an exclusive club for the main actors in financial regulation.

160 Joakim Reiter The scope and purpose of both organizations are relatively broad. The scope of the IMF’s main activities ranges from surveillance of its member states’ economic policies, promotion of monetary cooperation and exchange stability to financial assistance for balance of payment adjustments in terms of providing credits and loans, as well as technical assistance in the areas of fiscal and monetary policies. At a general level, one could claim that the IMF – through its various activities – is responsible for crisis management of the financial system. Almost by definition, therefore, the purpose of the organization is very broad. It ranges from information exchange, discussions on standards (e.g. in statistical reporting) and peer reviews to hard decisions on financial or technical assistance. The scope of the FSF is in many ways just as extensive as that of the IMF. The mandate of the FSF is to assess vulnerabilities affecting the international financial system; to identify and oversee action to address these; and to improve coordination and information exchange among various authorities responsible for financial stability (Evans 2000: 25). As such, the FSF constitutes the core body for crisis prevention in the financial system. The purpose of the FSF is to function as a network and discussion forum for the identification and formulation of broad standards, guidelines and best practices on pertinent issues. In this way, the FSF encourages, in a non-committing way, convergence towards common approaches. All standards and guidelines are non-binding. One outstanding feature of both organizations is how their membership structures relate to the EU–15 or EU–25. Particularly intriguing is the division of EU member states (with the exception of France, the United Kingdom and Germany) into various groups in the IMF’s Executive Board. These groups can contain both EU and non-EU states. For example, Ireland shares a group with Canada and some Caribbean states; Sweden and Denmark with Iceland and Norway; Spain participates in a group with eight Latin American countries. Both the Commission and the ECB also participate in IMF activities, through their observer status, but only the ECB has the right to speak. In the case of the FSF, five EU member states take a direct part in the work, as well as the ECB. The European Communities (i.e. the Commission) are not represented (Figure 10.3). Taking these features of the external environment into consideration, we would expect a relatively low level of EU actorness in both organizations. On all accounts – the tendency to hierarchy, the broad scope and the focus on soft law, and not least the membership structure – these two cooperative arrangements appear to predict reluctance among EU member states to agree to any deeper internal policy coordination. Also, in line with the hypotheses, EU actorness should be more evolved in the IMF, due to the organization’s inclusive membership structure and its hard decisions on loans and credits. But EU actorness in the IMF would not involve individual EU member states abstaining from speaking their minds and from making statements of their own. Conversely, in the FSF, despite the consensual way of making decisions, any form of EU actorness should be very hard to achieve. This is to be contrasted with the prediction of internal competence, whereby the existence of exclusive and mixed competence in this field would lead us to assume deep policy coordination among EU member states and, on some issues, a willingness to speak with one voice or even one mouth.

The EU as actor in international relations 161

Hierarchy

Forms of cooperation

Membership structure

IMF

FSF

Distribution of power

Dominance by a small group

No dominance but only a small group

Voting rules

Weighted

One country, one vote Consensus

Scope

Broad

Broad

Purpose

From information exchange to soft law and hard decisions on loans

Identification and convergence on best practices

Participation

Inclusive. All EU member states are part, but fragmented

Restricted. Not all EU member states are part

Figure 10.3 The IMF and the FSF.

In reality, concerted actions by EU member states remain the exception in the IMF and are non-existent in the FSF. In the FSF there have been no attempts to act as a single entity or to coordinate positions through common understandings. In the IMF, on the other hand, EU member states do meet on a regular basis to discuss the issues of the Executive Board. Also, there have been attempts at improving coordination in Brussels through the establishment of a Working Party on the IMF and related matters. However, coordination only aims to produce common understandings. At IMF meetings EU member states speak unconstrained and, in the Board, internal coordination is even less prevalent. If common positions have been agreed upon, the EU Presidency presents the EU line (the ECB in case of Eurosystem issues) but all other member states can participate in the discussion as well. In discussions on common understandings EU member states that constitute part of small groups and mixed groups (Ireland and Spain) of the Executive Board are often extremely reluctant to more collective decision making and deeper coordination. Among strong EU member states (Germany, France and Italy) loyalty has often proved to be stronger to G–7 positions than to EU positions. All in all, EU actorness in both the IMF and the FSF seems to fit with the propositions based on the external environment.

Conclusion It has been argued here that in order to understand the diversity of the European Union as actor in external relations internal factors are a necessary, but insufficient, explanation. Internal competence certainly provides a useful starting point and overview. However, it does not seem to be able to fully explain the large diversity – both within and across various issue areas – in the characteristics and

162 Joakim Reiter functioning of the European Union as an actor in international relations as well as the variety in the European Union’s internal set-up of institutions and networks for the coordination of policy making in international organizations. The key theoretical argument of this chapter is, instead, that the external environment – and more specifically the international organizations, of which the European Union is a member – also interrelates with, and directly influences, EU actorness. This also seems to stand the test of the case studies conducted in the area of trade and finance. The internal policy–coordination process, the role of the Commission and the member states, and the informal and formal negotiating networks among member states and with the Commission, reflect in important respects the constitutive elements and institutions of the external environment. The characteristics and functioning of EU networks in international cooperation with non-members are not exogenous variables, merely defined by the internal integration process of the European Union, but vary depending on the international organizational setting in which the cooperation take place. EU actorness is, fundamentally, an expression of semi-rational design with a view to making the European Union institutionally adaptive to its environment. Put in another way, the EU policy coordination process and its networks are both sensitive and respond in a predictable manner to the external environment in ways that fundamentally shape the capabilities and actorness of the European Union. In consequence, irrespective of EU integration, we would expect the diversity in EU actorness to remain. In addition, the ongoing enlargement of the European Union could very well serve as an impediment to deeper policy coordination in external relations on many issue areas where international cooperation is presently restricted to a limited number of EU member states. Based on the findings of this chapter, therefore, a single phone number to Europe is just as unforeseeable today as it was 30 years ago.

Notes 1 Regardless of this, however, it is worth underlining that the European Union has managed in most cases to handle the issue of competence in a pragmatic manner. The Commission has been given de facto negotiating authority across the board. 2 A similar argument is developed for international institutions in Koremenos et al. (2001: 762), where it is stated: ‘states use international institutions to further their own goals and they design institutions accordingly’. For the purpose of this chapter, this argument does not preclude the possibility that the institutional design of the European Union in international cooperation can fail to produce the intended results. Nor does it mean that there is a certain inertia in the internal bargains over institutional design in the European Union that favours tried solutions and old patterns of behaviour over novel changes. 3 These three factors have been developed on the basis of six elements, as identified by Cohen (1998a: 191–2): cognitive consensus; credibility; non-threatening; sponsor; ratification; and speed. 4 Koremenos et al. (2001) have developed a similar categorization of important features of international organizations. It covers five elements – membership, scope, centralization, control and flexibility – which largely overlap with, and have formed the basis for, the three categories presented here. 5 A third aspect of the category, called institutionalization, could have been used in the

The EU as actor in international relations 163 analysis. This refers to the role of the secretariat, mechanisms for monitoring compliance and the like. However, in this chapter it is very hard to treat this separately from the other two aspects and, not least, from purpose. The level of institutionalization tends to follow from the purpose of the cooperation arrangements. In organizations focusing on hard law, cooperation tends to be member driven. Conversely, when the purpose of cooperation is soft law and peer review, the secretariats of the organizations are usually larger and more empowered. 6 The third stage involves the establishment of a European Central Bank, the fixing of exchange rates and the introduction of a single currency. 7 These differences in purpose are also reflected in the institutionalization of the WTO and the OECD. The WTO is strictly member driven – the secretariat has around 550 employees – and has developed a quasi-automatic dispute settlement mechanism (with third party adjudication) to oversee compliance. The OECD, on the other hand, has a comprehensive secretariat (a staff of approximately 2,300 experts) with substantial inhouse statistical and analytical capacities. The OECD’s way of working consists of data collection, analysis, mutual examinations, mutual surveillance and peer pressure. 8 In total, 26 national authorities from 11 countries participate in the deliberations of the FSF. Additionally, there are 15 representatives from different international financial institutions and specialized standard-setting bodies, such as the IMF, World Bank, Bank for International Settlement, OECD, Basel Committee for Banking Supervision, International Accounting Standards Board, International Association of Insurance Supervisors, International Organization of Securities Commissions, Committee on Payment and Settlement System and Committee on the Global Financial System. The European Central Bank has one representative.

References Blair, D. J. (1993) Trade Negotiations in the OECD, London: Kegan Paul International. Cohen, B. J. (1998a) ‘When Giants Clash: The OECD Financial Support Fund and the IMF’, in V. K. Aggarwal (ed.) Institutional Designs for a Complex World, Ithaca: Cornell University Press. Cohen, B. J. (1998b) The Geography of Money, Ithaca: Cornell University Press. Evans, H. (2000) ‘Plumbers and Architects: A Supervisory Perspective on International Financial Architecture’, Financial Services Authority (FSA) Occasional Paper, January. Koremenos, B., Lipson, C. and Snidal, D. (2001) ‘The Rational Design of International Institutions’, International Organization, 55 (4): 761–99. Macleod, I., Hendry, I. D. and Hyett, S. (1996) The External Relations of the European Communities, Oxford: Clarendon Press. Timmins, D. B. and Timmins, W. M. (1985) The International Economic Policy Coordination Instrument: Some Considerations Drawn from the OECD Experience, London: University Press of America. Wallace, H. (2000) ‘The Institutional Setting’, in H. Wallace and W. Wallace (eds) PolicyMaking in the European Union, Oxford: Oxford University Press. Woolcock, S. (2002) ‘Utvecklingen av EU:s handelspolitik’, in J. Reiter and C. Jönsson (eds) Handelspolitik i förändring, Stockholm: SNS Förlag.

11 The European Union and the United States of America The politics of ‘bi-multilateral’ negotiations Michael Smith

Introduction This chapter starts from the evident significance but also the problematical nature of bilateral EU–US negotiations within the global political economy since the end of the Cold War. It is clear from a variety of recent studies that the intense interdependence and ‘competitive cooperation’ between the EU and the USA create numerous opportunities for bilateral negotiation and the handling of both disputes and areas of joint action (Levine 1996; Peterson 1996; Philippart and Winand 2001; Pollack and Shaffer 2001; Peterson and Pollack 2003; Smith 1998). What the same studies also make clear is that these bilateral negotiation processes exist alongside and in close relationship to broader multilateral frameworks, which in themselves generate plentiful occasions for competition and compromise. It is relatively unclear, in both analytical and policy terms, what the implications are of this close co-existence and the intermeshing of both bilateral and multilateral negotiation processes. From the perspective of negotiations conducted by or through the European Union, this phenomenon of ‘bi-multilateral’ negotiation is thus deserving of closer attention. The argument that follows is intended to provide both analytical clarification and an indication of the ways in which empirical investigation can make some sense of this increasingly pervasive form of negotiation. The chapter deals first with some of the issues raised by recent literature on EU–US relations and argues that existing approaches can be complemented and enriched by use of a negotiation perspective. Such a perspective can identify the multitude of negotiation processes that occur in and around EU–US relations, and can explore the ways in which coexisting processes of bargaining and problem solving find expression in different areas of policy. The chapter next attempts to clarify what is meant by bi-multilateral negotiations and to sharpen some of the questions this raises about the nature of negotiation processes and outcomes, in the light of the negotiation literature. It proposes a number of research questions dealing with issues in EU–US relations from the perspective of bi-multilateral negotiation. In particular, it points to the different ways in which bi-multilateral negotiations can occur: as part of the construction or reconstruction of international regimes, as part of a designed and institutionalized process and as a result

The EU and the United States of America 165 of ‘contagion’ through which bilateral issues call into question multilateral processes. The chapter goes on to address a range of recent cases from EU–US relations, with the aim of identifying analytical and policy issues. The conclusion discusses the further development of the framework and the selection of case studies for more detailed investigation.

EU–US relations in the post-Cold War era A very large amount of literature exists on the subject of EU–US relations since 1990, and this chapter will not refer to it specifically here (see the references cited above). Rather, the goals are to identify some key trends in the literature, to justify the addition to this literature of a negotiation perspective and to identify where that perspective might take us, in both analytical and policy terms. Three key areas of the existing literature can be identified for the sake of discussion here (see Pollack and Shaffer 2001: Introduction). First, some approaches are essentially intergovernmentalist in their foundations. These approaches conceive of EU–US relations after the Cold War as a reflection of the shifting power structure in the international arena, and as relations either between states or between statelike entities. A number of studies based on this approach tend to privilege the ways in which the changing international security arena has put pressure on the EU–US relationship, and sometimes (in mercantilist vein) to look at the ways in which this set of changes has or has not fed into other areas such as political economy (see, for example, Gompert and Larrabee 1997; Sloan 2002). From the point of view of the European Union, the key issues here have been seen as the continued (if reformulated) predominance of the USA, the reassertion of US priorities, and the tensions this creates between developments in the security realm and the realm of political economy. These tensions reflect the fact that whilst the European Union has continuing weakness in the political-security realm, it performs a number of key ‘state functions’ in the international political economy. Liberal intergovernmentalist approaches (see, for example, Nye and Keohane 1993) have tended to give this aspect of EU–US relations greater prominence than the ‘hard security’ issues favoured by realist or neo-realist commentators, and to emphasize the need for effective transatlantic bargains as the foundation for economic governance. Approaches from an intergovernmentalist perspective thus do not always neglect the political economy aspects, but they tend to view them against the background of the dominance of the USA in the realm of ‘hard security’ and related areas (Peterson and Ward 1995). EU–US relations are thus seen as a subset of the ways in which international structure, US power and US preferences have made themselves felt in the post-Cold War era; the European Union is judged by the extent to which it has or has not matched up to this ‘American challenge’ (Peterson and Pollack 2003: Conclusion). A second set of transgovernmentalist approaches is based on the growth of governance structures, both at the transatlantic level and more broadly in the global political economy. These approaches see transatlantic relations as a reflection of

166 Michael Smith the demand for the establishment of rules and institutions, as well as a reflection of the supply of those rules and institutions themselves. They draw attention to the massive growth in institutional contacts and institutional structures across the Atlantic, in particular through processes of transgovernmental cooperation, and they often link it with the growth of mechanisms of global governance. Some of the attention of this perspective is focused on the construction and operation of transatlantic networks, often with a public–private complexion. As such, these approaches partake of the broader interest in policy networks and policy communities, and they attempt to discern the impact of these networks and communities on the social learning and normative convergence of those involved. The studies in the collection edited by Pollack and Shaffer make a number of key distinctions here, and link these to elements of communicative rationality which are clearly important in sustaining the notion of transatlantic governance (Pollack and Shaffer 2001: Part III). Finally, a number of approaches centre on transnational relations. These approaches take as their baseline the fact that despite radical changes in the political-security domain, or the building of new institutional frameworks, these do not explain everything about EU–US relations. They point to the fact that EU–US interdependence and what can be termed integration have continued to intensify during the post-Cold War period, using indices drawing attention to commercial policies broadly defined and to private linkages created by economic agents such as multinational companies as much as to actions in the realm of ‘high politics’ (Cowles and Smith 1999; Smith 2001). They note that these structures of transnational relations have continued to deepen and to strengthen despite apparent threats of divorce or divergence in the transatlantic arena. Essentially, much of the argument from this perspective is liberal in nature, pointing to the ways in which private agents can maintain relationships and contribute to transatlantic integration, and arguing that the role of governmental authorities is to ensure broad stability within which private agents can operate to their mutual benefit. Whereas for intergovernmental approaches much of the attention is on a hierarchy of government functions (or what were referred to above as ‘state functions’), for transnational perspectives much of the attention is on networks, often displaying a mixture of governmental and private participants (Risse-Kappen 1995). Although empirically such processes overlap with transgovernmentalist approaches, the emphasis on private management as opposed to designed institutionalization provides a critical point of distinction. These existing conceptualizations contribute much to our understanding of the broad changes that have affected EU–US relations since the end of the Cold War. Each of them also takes a particular perspective on the nature of bargaining and problem solving at the transatlantic level. For intergovernmentalists, such bargaining is linked inextricably with the underlying global power structure and the resources available to government bodies in entering into political and economic bargains. For those involved in the development of transatlantic institutional frameworks through transgovernmental cooperation, there is an inevitable concentration on the politics of institutional construction and reconstruction, in a

The EU and the United States of America 167 broad sense, but also on the functioning of policy communities at the transatlantic level. For those engaged at the transnational level, the focus is necessarily on the functioning of market-based or other networks in which ‘government’ is provided through very different channels and types of communication. Thus, intergovernmentalists are better at demonstrating the influence of overall structural factors and either EU or US responses to them. At the same time, transgovernmentalist approaches can illustrate more effectively the middle-range implications of transatlantic regime building and ‘community building’ and can link this to broader issues of global governance. Transnationalist approaches by the same token are effective at uncovering the ‘hidden wiring’ of public and private transactions, but less powerful at taking into account the broader power structure or the mechanisms of institutionalization.

A negotiation perspective What is proposed here – as a complement to and a means of enriching existing approaches – is an approach based on negotiation theory, which at least in principle gives both a new perspective on EU–US relations and provides the basis for a distinctive set of research questions. In dealing with EU–US relations, we are confronted by a set of relations with a number of key properties, not all of which can be accessed or linked through established approaches. Among those properties are: ●

●

●

●

●

●

●

●

multi-level relationships with strong elements of public–private interaction as well as intergovernmental interaction; growing structures of institutionalization, for example through the New Transatlantic Agenda (NTA) and the Transatlantic Economic Partnership (TEP), but issues about the gaps between institutional potential and institutional performance; strong linkages between what goes on at the transatlantic level and what goes on at the global level in the context of institutions such as the World Trade Organization (WTO), the G–7 and a range of other multilateral bodies; intensifying integration especially in areas of the political economy, but unevenness of that integration across sectors; a proliferation of relevant networks, both transgovernmental and transnational, but lack of clarity about how those networks relate to each other and interact with each other; co-existing ‘languages’ of EU–US discourse, ranging from coercion to coalition building and collaboration, albeit often in an uneasy coexistence and reflecting elements of normative consensus but equally strong areas of normative confusion; issues of choice about forums for interaction, about strategies and about priorities which affect both EU and US policy makers; as a result, an ‘adversarial partnership’ or relations of ‘competitive cooperation’ which are arguably suboptimal for all concerned (but which may also serve the purposes of policy makers on both sides of the Atlantic from time to time).

168 Michael Smith These qualities make EU–US relations complex and often messy, with growing institutionalization ranged alongside continuing and even intensifying competition and the potential for crisis arising from the ‘hyperinteractivity’ of transatlantic dealings. One thing, though, seems clear: that what goes on in huge areas of EU– US relations can be characterized as negotiation. To be sure, this negotiation is not all of one type or within a settled framework of rules or assumptions. Bargaining and problem solving seem to coexist; mutual interests are accompanied by threats and coercion; protestations of commitment are uttered almost simultaneously with recriminations and statements of potential or actual defection (Pollack and Shaffer 2001: 5). But that is why a negotiation approach is arguably an important addition to the armoury of those studying EU–US relations. Negotiation perspectives ask a lot of the questions that inevitably arise when one looks in any detail at EU–US relations, and they promise to provide at least some of the answers in the form of detailed negotiation analysis. EU–US relations, conceived in terms of negotiation, also raise one of the crucial issues of negotiation in the post-Cold War period. How does the continuing bilateralism of many negotiation processes in the global arena coexist with and interact with the explosion of multilateral negotiations? Bilateral EU–US negotiations are in the vast majority of cases embedded in broader multilateral negotiations, and the relationships between these levels and arenas must form a key datum for the pursuit of negotiation – and the improvement of negotiating practice – among both Europeans and Americans. Both at the bilateral and at the multilateral level, EU and US negotiators are engaged in what has been termed the ‘management of complexity’ (Zartman 1994); to use another image, both EU and US negotiators are engaged in ‘negotiating the world economy’ (Odell 2000), and the ways in which they do so are on a priori grounds going to be significant both to the parties and to the world political economy. With this initial position in mind, this chapter aims to explore the nature of EU– US negotiations by using the device of bi-multilateral negotiation, to identify the issues this raises for negotiation strategy and negotiation outcomes, and to explore the implications of these processes for the broader global order. The argument implicitly focuses on the world political economy and on the political economy of EU–US relations; at this stage this is enough to take on, without looking more at the issues of ‘hard security’ which would generate a new if not completely different or separate range of avenues for enquiry (see for example Smith 2004a). By looking at issues in the political economy, this chapter also focuses on an area in which EU–US relations find the European Union on its strongest ground. The development of ‘state economic functions’ (Smith 2004b) and the accumulation of governance potential at the EU level create a situation in which the question ‘who speaks (or negotiates) for Europe?’ is subject to fewer qualifications and less contestation than elsewhere. This does not mean that EU negotiation capacity is a constant in such areas (Meunier 2000; Meunier and Nicolaidis 1999, 2000; Young 2000, 2002), and indeed such variation might well be a powerful analytical dimension for later detailed research, but it does mean that questions about competence, resources and agency are held within a narrower spectrum. For the purposes of

The EU and the United States of America 169 this chapter, such a limitation can be justified in the knowledge that future and fuller exploration of the issues will reveal more complexity and variability.

‘Bi-multilateral’ negotiations: a framework for analysis This part of the chapter attempts to clarify the concepts which will later be used to probe a range of case studies. The section proceeds by presenting the qualities of bilateral negotiations, multilateral negotiations and then bi-multilateral negotiations. In doing this, the exploration is based on six features of any negotiation process, reflecting broad aspects identified by much of the negotiation literature (for example Kremenyuk 1991; Zartman and Berman 1982; Jönsson 2002), and thus enabling the comparison to be made more effectively. The six features reviewed are: ●

●

●

●

●

●

occasions (why negotiate, and why negotiate on a given subject at a given time?) contexts (what are the constraints and opportunities built into the negotiation context, including institutional, resource, cultural and other features?) participants (who participates, what assumptions are made about agency and about roles?) agendas (what are the negotiations about, how are agendas set and how do they change?) strategies (how do participants make choices, how do they communicate, how do they conceive of the negotiation process, what goals do they set?) outcomes (what formal agreements result, what informal outcomes in the shape of learning or understanding emerge, how are any agreements or understandings implemented?)

Bilateral negotiations In the extensive literature on bilateral negotiation processes a number of key strands can be discerned (Kremenyuk 1991; Zartman and Berman 1982; Jönsson 2002). In bilateral negotiations, negotiating occasions occur (it can be argued) because of internal or external pressures on the participants, because of recognition of mutual interests or because of the attempt to promote the interests of one party or the other. The context is shaped by positional factors (the place of the participants in the broader structure of the world political economy), by domestic factors broadly defined and by the prior institutional and other commitments of the participants (which may of course mean that many negotiations are effectively programmed into the relations between the participants). The participants are two (or the agents of two entities), although this rather obvious feature is importantly modified by the internal make-up of the two participants and by the respective resource positions and preferences of the two parties. The agenda(s) for bilateral negotiations are in turn affected by positional factors, by the intensity and priority of commitments and by the extent of programming

170 Michael Smith available in the circumstances. Agendas may well be complex and cross-cutting, and one of the issues for the participants in bilateral negotiations is that of agenda management, both for their own benefit and for the sake of the negotiation process itself. Strategies for participants in bilateral negotiations can be analyzed from a number of perspectives, depending on whether one’s initial position is rationalist or reflectivist: perhaps the most plausible approach is to position oneself with the social constructivists to the extent that apparently rational sets of preferences and strategies are also intensely reflective of prior relations, expectations and mutual learning, especially where bilateral negotiations (as in the case of EU–US negotiations) have lengthy histories and many lessons have been (mis)learned. Finally, the outcomes of negotiations within a bilateral process can also be conceived in terms of social learning by the participants and of institutionalization; the consequences can be of three kinds – distributive, integrative or what can be described as ‘generative’ (the types of consequences that generate new areas of types of understanding and shared experience). Where bilateral negotiations are at issue, analysis can focus on the ways in which these outcomes are in theory or in practice to the benefit of either of the parties or of their mutual relations. Multilateral negotiations There is a significant literature on multilateral negotiations, prompted by the observable growth of these processes and of associated institutions in the 1980s and 1990s (Hampson 1995; Zartman 1994). Perhaps equally significantly, there is a literature on multilateralism itself (Ruggie 1993); which makes the important, not to say fundamental distinction between multilateralism as a ‘fact’ and multilateralism as a principle. In multilateral negotiations the occasions for negotiation have frequently been defined in terms of the creation, maintenance or modification of international regimes. These regimes may be formally encapsulated in institutions, or they may take a less formal but powerful normative form, shaping not only the process of multilateral negotiation but also the behaviour of individual actors within the relevant domain. This has implications for the assumed motivations of the participants, which can be many and more diverse than those expressed in bilateral negotiations. The context for multilateral negotiations is best described in terms of complexity and diversity, reflecting not only the positions of the many participants but also the impact of broader international change processes to which multilateral negotiations are not only a response but are also subject. Importantly, multilateral negotiations are almost by definition subject to cultural pluralism and to crosscutting cleavages among participants; this is however modified importantly by the scope of the negotiations, whether they are effectively sectoral and insulated or cross-sectoral and open to wider influences. The participants are thus many and varied, and the problems of synthesizing and understanding information (widely defined) about other participants are basic to participation; this process may be facilitated by agents such as secretariats or other international representatives. The number of participants itself can vary, often influenced by conventions or rules

The EU and the United States of America 171 about inclusion and exclusion, and there can be mechanisms for distinguishing between core participants and others. Agendas in multilateral negotiations as already noted can be (often are) complex and fluctuating, and the art of agenda construction is not merely a technical one, since it can have important political consequences. Different participants or groups of participants can come to the table with profoundly different understandings of what is desirable or possible. Such problems have implications for strategies, since the construction and management of coalitions is often seen as integral to multilateral negotiations, and the generation of leadership either via coalitions or through other channels is a key requirement in the eyes of many analysts. Where there is a requirement for leadership, there is automatically a question of legitimacy and efficacy: how do leaders gain and maintain a leadership position, especially where there are diverse and complex issues to be confronted? Finally, multilateral negotiations raise in distinctive forms the issue of outcomes; there can be a strong contrast between outcomes cast in terms of results (especially formal agreements) and outcomes cast in terms of process or understanding and the generation of a normative consensus. Formal agreements are often large scale, subject to package deals and often therefore suboptimal; informal process and understanding can often be difficult to assess and embed within the acquis of the negotiations. As in the case of bilateral negotiations, outcomes are often difficult to discern in the short term and may best be expressed in a willingness to go on negotiating in what is a virtually continuous process. Multilateral negotiations, more perhaps than bilateral negotiations, also have to grapple with problems of ratification, implementation and compliance; agreements formally made may be informally undermined, postponed or re-interpreted. Bi-multilateral negotiations Bi-multilateral negotiations are in a sense defined by what has been said above about both bilateral and multilateral processes, and from this follow important analytical and policy issues. The occurrence of bi-multilateral negotiations is defined by the coexistence of occasions both at the bilateral and at the multilateral level; this poses a challenge for negotiators in terms of ‘event management’, strategic choice and the planning of negotiations that may be related in a linear way or in forms of parallelism. Occasions also raise major issues of motivation: the coincidence or the close relationship between bilateral negotiations and multilateral events puts a new twist on the formation of preferences and priorities and may tax the attention of those engaged in the coexisting processes. In negotiations, timing is a vital variable, and in bi-multilateral negotiations it is likely to be at least a two-dimensional matter, affecting both adjustment to different ‘rhythms’ and schedules and the potential manipulation of timing as a matter of strategy (see below). Another key issue is the ways in which bi-multilateral negotiations arise. The analysis suggests that this can happen in at least three ways, each of which has implications for the process and outcomes of negotiations. First, bi-multilateral

172 Michael Smith negotiations can occur as part of the construction or reconstruction of international regimes. Second, they can occur as part of a designed and institutionalized process (either reflecting choices by participants at the bilateral or at the multilateral level, or at both). Third, they can occur as a result of ‘contagion’ through which bilateral issues call into question multilateral processes, and in which the multilateral processes in some way become hostage to the bilateral process. These three types of negotiating occurrence will have important shaping effects on the ways in which bimultilateral negotiations are defined, take place and produce outcomes. Context enters into bi-multilateral negotiations in a very specific way: the bilateral will form part of the context for the multilateral, and vice versa. Not only this, but institutional structures and positions might be different for different parties in the different negotiation arenas, with significant results for their resources or their ability to recognize contextual constraints or opportunities. Participation again enters into the bi-multilateral mode in a significant way: the bilateral participants as members of the larger group may well also have affiliations and institutional links that can be exploited or may prove problematic. This of course has a strong resonance, but with a different twist, of the well-known issues attending two-level or three-level games (Putnam 1988, 1993; Evans et al. 1993). But whereas in two- or three-level games it is assumed in effect that the chief of government (COG) participates at two levels or three levels with different constraints and resources, in bi-multilateral negotiations this cannot be assumed: the agents involved may or may not be the same, and the difficulties of coordinating across linked arenas may or may not be resolved. The other essential issue of the two- or three-level game, that of ratification, again has a different twist in a bi-multilateral context, since the existential linking of the two arenas may not overcome profoundly different rules of engagement or agreement. The question of what happens when the ratification of bilateral and linked multilateral agreements is at issue is clearly one that demands further exploration, but which lies outside the scope of the current chapter. In terms of agendas, the specific properties of bi-multilateral negotiations again create potential difficulties: how do the results of bilateral negotiations find expression in multilateral agendas, and vice versa? This is of course linked to issues of strategy, in which one obvious result of bi-multilateral negotiations is the formation, reinforcement or modification of coalition arrangements at the multilateral level. There is an ‘audience problem’ here, given that by definition the participants in bilateral negotiations have taken an explicit or tacit decision to exclude other potential coalition members, thus raising issues of transparency or legitimacy. To this extent, bi-multilateral negotiations raise in a potentially acute form the issue of ‘minilateralism’ as explored for example by Miles Kahler (1993). There is also in the broadest terms a twofold problem of modalities and externalities. In respect of modalities, the interface between bargaining and problemsolving approaches is a potential source of friction, misunderstanding and management problems in bi-multilateralism. This links in turn to the issue of externalities, whether these are designed and managed or not: the impact of multilateral strategies on bilateral strategies, and vice versa, is again one that demands further exploration.

The EU and the United States of America 173 Finally, for bi-multilateral negotiations there is a potentially severe problem of outcomes which is partly embodied in the above-mentioned issue of externalities: one of the central qualities of these negotiations is the mutual dependence of outcomes at the different levels and in the different arenas. How are unintended outcomes to be handled in such a context, and how are the issues of ratification and implementation to be grasped when they might impact upon negotiations at the other level in unpredictable ways? For example, how can the credibility of commitments made in different contexts and at different levels be maintained? It can be seen from this discussion of the analytical issues that one of the key problems in bi-multilateral negotiations is what might be termed ‘co-dependency’ – of occasions, contexts, participants, agendas, strategies and outcomes. This is not simply a contingency or an accident: it is an integral part of the negotiation process, although it can come about in a number of ways. Further, it is argued on the basis of the review in the earlier parts of this chapter that because of the intensity and range of EU–US relations, there is within them a high incidence of bimultilateral negotiations and thus of ‘co-dependency’, which raises issues for analysis and for the practice of negotiation itself.

Bi-multilateral negotiations in EU–US relations The argument so far has indicated the distinctiveness and some of the implications of bi-multilateral negotiations at the general and conceptual level. In the remainder of the chapter the focus is on the ways in which this analytical tool can be used to explore EU–US negotiations, both at the broad level and in the context of some case studies from recent experience. At the beginning of the chapter it was established that EU–US relations demonstrate a high intensity of interactions and negotiation processes, and that in many cases these processes demonstrate the mutual entanglement of the bilateral and the multilateral. The first task in applying the bi-multilateral framework is thus to identify some of the key research questions to which EU–US negotiations give rise. At the broad level, a number of research questions arise, among which are the following: ●

●

●

What influence do EU–US bilateral negotiations have on the occurrence of key multilateral negotiations, and on the willingness of the European Union and the USA to commit to those negotiations? Conversely, what influence do processes of multilateral negotiation have on the occurrence of EU–US bilateral negotiations, and on the willingness of the European Union and the USA to commit to those negotiations? What influence do EU–US bilateral negotiations have on their participation in multilateral negotiations and on their aims in those negotiations? Conversely, what influence do multilateral negotiations involving the European Union and the USA have on their participation in bilateral negotiations, and on their aims in those negotiations? Do EU–US bilateral negotiations increase agenda complexity in multilateral

174 Michael Smith

●

●

●

●

negotiations, or do they reduce that complexity? Conversely, does EU and US involvement in multilateral negotiations increase agenda complexity in their bilateral negotiations, or do they reduce it? Do EU–US bilateral negotiations facilitate coalition management in multilateral negotiations or act as an obstacle to it? Conversely, does EU and US involvement in multilateral negotiations involving coalition management increase or reduce their capacity to undertake bilateral negotiations? Do EU–US bilateral negotiations facilitate leadership and consensus building in multilateral negotiations, or do they make it more difficult? Conversely, does EU and US involvement in multilateral negotiations involving leadership and consensus building facilitate or make more difficult their capacity to engage in bilateral negotiations? Do EU–US bilateral negotiations facilitate outcomes for multilateral negotiations, or are they a barrier to agreement? Conversely, does EU and US involvement in multilateral negotiations facilitate outcomes in their bilateral negotiations, or is it a barrier to bilateral agreement? Can the EU and US involvement in bilateral negotiations act as a management device, a displacement device or an incentive to agreement in the multilateral arena? Conversely, does EU and US involvement in multilateral negotiations act as a management device, a displacement device or an incentive to agreement in bilateral negotiations?

Table 11.1, based on the discussion earlier in the chapter and on the research questions raised here, is a preliminary indication of the ways in which the bimultilateral framework helps to focus attention on negotiation processes in the EU–US relationship. It also suggests that there is a substantial basis for proceeding with a ‘sampling’ process which will explore a range of recent EU–US negotiation episodes.

Sampling bi-multilateral negotiations in EU–US relations Having established some of the key general research questions relating to EU–US bi-multilateral negotiations, the next task is to examine a range of promising case studies, with the aim of establishing their potential for further detailed research. Such detailed research is outside the scope of this chapter, but the following ‘probes’ are designed to demonstrate the outlines of a longer-term research agenda in this field. The samples have been chosen to represent key issues in the recent encounters between the European Union and the USA in the global political economy. Sample 1: the Doha Development Round The first sample is the process by which the agenda for the Doha Development Round was established and pursued during the period 1998–2003. A number of forces contributed to the initiation of a process that led first to the Seattle

Multilateral Regime creation and regime maintenance Complex, linkages; normative as well as substantive agendas; information and uncertainty factors; cultural factors Multiple public/private; same individuals/groups as in BL? Issues of leadership? Use of minilateral devices? Linkages/diffusion; sectoral vs those concerned with order and institutions Results-oriented bargaining vs process-oriented strategies. Issues of leadership, management, coalitions/consensus Normative consensus; regime creation/maintenance/revision; collective goods?

Bilateral

Multiple bilateral; intense bilateralism and intense transgovernmentalism

EU–US power/resource balance; domestic forces; multilateral institutions and rules

‘Political’ vs ‘official’ vs ‘expert’; public and private; linkages between participants (designed or unintentional?)

Broad NTA/TEP; inter-regional cooperation; ‘integrative’ vs ‘competitive’ definition of issues

BL bargaining and problem solving; dependence on resources and skills

Competitive cooperation; learning and understanding at transatlantic level; institutionalization; generative as well as distributive functions

Components

Occasions

Contexts

Participants

Agendas

Strategies

Outcomes

Externalities from both BL and ML; costs and risks; credibility of BL and ML commitments; defection vs coalition building; internalization of ML norms?

Use of ML context/rules/norms as resources or stakes; integrative vs distributive bargaining across levels; choices, linkages (forum shopping?)

Complementarity/competition of BL and ML; role of ideas about ML institutions and their uses

Overlapping participants/roles; selection of participants? Third party issues?

Interaction of BL and ML; normative uncertainty? Cultural issues (transatlantic and ML?)

Sequential or simultaneous? Competitive or complementary? Designed or contingent?

Bi-multilateral

Table 11.1 Bilateral, multilateral and bi-multilateral negotiations in EU–US relations: components and issues

176 Michael Smith Ministerial Meeting of 1999 and later to the Doha Ministerial Meeting in 2001 and the Cancun Ministerial Meeting of 2003. In all of these stages, the participation of the European Union and the USA was seen as essential at the multilateral level; equally, in all of them there was a strong EU–US bilateral agenda in which the existence of ‘competitive cooperation’ was apparent (Baldwin et al. 2003; Allen and Smith 2002, 2003). Throughout the build-up to the various ministerial meetings, the European Union and the USA were active in establishing their own negotiating positions, but also in attempting to have these accepted as the focus for the multilateral process. In areas such as agriculture and the broad ‘development’ agenda, the European Union and the USA vied for influence and were also frequently engaged in intense bilateral exchanges. As the Cancun meeting approached, the European Union and the USA produced, on a bilateral basis, a key document dealing with agriculture, which aroused intense suspicion among other members of the WTO and which arguably contributed materially to the collapse of the Cancun meeting. But the Cancun fiasco also revealed other aspects of the negotiation process: whilst the European Union and the USA had been mutually absorbed in their bilateral context, the growth of the Group of 20 developing countries had led to the emergence of a powerful caucus dedicated to resisting any EU–US bilateral stitch-up. Construction of the agenda for the Doha Development Round thus involved structured negotiation occasions between the European Union and both the USA and other WTO members, but also a host of less formal and structured encounters. Arguably, the focus on EU–US bilateral concerns and their projection onto the multilateral arena meant that important signals about changes in the negotiation context were ignored, and the ‘Cancun surprise’ reflected this. Sample 2: bananas, beef, steel and taxes: EU–US bilateral disputes and their contagion The second sample brings into focus the way in which during the late 1990s and 2000 a range of EU–US bilateral disputes contributed to a souring of the atmosphere around the WTO and to a questioning of the WTOs dispute settlement procedures (Allen and Smith 2000, 2001, 2002, 2003). The European Union and the USA have generated a number of the most pressing bilateral disputes in the world political economy since the end of the Cold War, and these have been focused to an increasing degree by the ways in which WTO procedures effectively build in a process of bi-multilateral negotiation. EU–US disputes about bananas, steel, Foreign Sales Corporations, hormone-treated beef and a wide range of other issues have thus come to characterize the contemporary international political economy. Without going into detail about the individual disputes, it is important to note that the changes in the WTO regime during the mid-1990s have institutionalized an extended form of bi-multilateral negotiations, in which the parties have the opportunity not only to pose as the defenders or opponents of the procedures themselves but also to engage in successive rounds of bilateral problem solving which can lead to forms of social learning and adaptation. Where these disputes go sour, the effects can be felt across the whole of the WTO system, as

The EU and the United States of America 177 was the case with the banana dispute of the late 1990s. It is also the case that other parties can be drawn willingly or unwillingly into successive phases of the negotiations: for example, the central American countries in the case of bananas, or the central and east European countries, China and Korea in the case of the steel dispute. It is important in these types of cases to note the capacity of both the European Union and the USA to recruit other parties in their support or their defence, but it is also important to judge the extent to which in such cases the bilateral negotiations remain distinct and maintain the commitment of both the European Union and the USA. Sample 3: Helms–Burton and the ILSA – US unilateralism, EU responses and mutual coercion The third sample consists of negotiations in which initial US unilateralism and extra-territorial activism led to a strong EU response in which the WTO was used as part of the defensive or deterrent strategy. Cases in point are the long-running disputes over the Helms–Burton Act and the Iran–Libya Solidarity Act (ILSA) in the mid-to-late 1990s (Allen and Smith 1999, 2000, 2001). The initial legislation in the US Congress promised punitive action against those consorting with ‘pariah states’ or trafficking in the assets these states might have acquired from US companies. Thus, the Helms–Burton Act applied these types of sanctions to dealing with Cuba, an area in which EU member states and companies have strong interests; the later ILSA extended similar measures to Iran and Libya. Bilateral negotiations between EU and US government representatives were complicated by the fact that congressional oversight and EU member state suspicion were involved. The European Union adopted a position in defence of the WTO regime, arguing that the unilateral and extra-territorial nature of the measures gave grounds for a complaint against the USA and that WTO judgements should be enforced against Washington. Over a period of five years or more, a process of bilateral crisis management, accompanied by appeals to WTO principles, had the effect of making the WTO regime a stake in the EU–US struggle – a struggle that was eventually defused but which had the potential to re-erupt at short notice when further rogue regimes were identified. Sample 4: commitment and defection: the Kyoto Process The fourth sample is provided by the EU–US tensions that emerged in the late 1990s and persisted into the new millennium, centred on the ratification and implementation of the Kyoto Protocol (Allen and Smith 2000, 2001, 2002; Bodansky 2003). This measure dealt with the emission of greenhouse gases and the consequent phenomenon of ‘global warming’, and it established a framework for dealing with successive reductions in emission of the damaging gases. The USA was initially supportive of the process, but at the end of the Clinton Administration it reversed its position and refused to ratify the Protocol – a major problem for its implementation, since the USA accounts for a large proportion of

178 Michael Smith global emissions of greenhouse gases. Through the early years of the new millennium, there was a succession of EU–US bilateral efforts to rescue at least part of the deal, with notably more commitment on the EU side than on that of the new Bush Administration after 2000. At the same time, the European Union set out to establish what can only be described as an anti-US coalition in the context of continuing efforts to achieve multilateral ratification of the Protocol – an effort which attracted the lasting hostility of the Bush Administration, but which also brought to prominence the role of other major players such as Russia (whose ratification was vital to implementation of the Protocol). For the European Union the position was complicated further by the need to negotiate emission-trading procedures within the Union itself, whilst at the same time trying to deal both with the USA on a bilateral level and with the rather unruly remainder of the Kyoto coalition in the multilateral arena. This problem of positioning had clear roots in the bi-multilateral nature of the negotiation process, and meant that, particularly for the European Union, the problem of creating and maintaining a credible posture was especially difficult. Sample 5: the race to agree: the EU, the USA and China’s admission to the WTO The fifth sample is provided by the negotiation of China’s admission to the WTO, which was formally achieved at the beginning of 2002 (Allen and Smith 2001, 2002). For our purposes, the key feature of this process is that both the European Union and the USA had a major stake in achieving a good deal with the Chinese, and that in accordance with the rules of the WTO, the Chinese had to negotiate deals with all of the existing WTO members before they could enter. Both the European Union and the USA were thus negotiating bilaterally with the Chinese, whilst the result would be a major change in the scope and membership of the multilateral trading regime. What emerged during the latter part of 2000 was a kind of ‘negotiation race’ in which the respective offers made by the Chinese to the European Union and the USA generated a competitive dynamic in areas such as financial services, fed by the demands in both the European Union and the USA of major interest groups. In the terms proposed by this chapter, here there was a process of regime modification under way, but because of the rules and conventions built into the regime, there was an incentive for major stakeholders to conclude advantageous bilateral deals – and to renegotiate them, indeed, if the opposition seemed to have stolen a march at any given stage. All of the parties at various stages could take advantage of linkage strategies, with uncertain consequences for the WTO regime in general. Sample 6: GE–Honeywell and beyond – competition policy and EU–US negotiations Throughout the late 1990s and 2000s there was an increasing recognition of EU– US contacts in a wide range of competition policy and anti-trust issues, both at the

The EU and the United States of America 179 level of specific disputes and at the broader level of international regime building (Allen and Smith 2000, 2001, 2002; Devuyst 2001). Such contacts were part of a more extended process through which ‘new agenda’ items in commercial policy were introduced alongside the more traditional trade policy items (another is the trade policy consequences of environmental change). This had been recognized at the transatlantic level since the early 1990s and had formed a significant item in both the development of the Transatlantic Declaration from 1990 onwards and the framing of the New Transatlantic Agenda in 1995. The changing nature and institutional foundations of competition and anti-trust policies in both the European Union and the USA during the 1990s, together with the intensification of globalization processes, created fertile conditions for disputes and also for negotiations. Central to the disputes that emerged was the increasingly active role of the Competition Directorate General in the European Commission, and of its Commissioner, ‘super’ Mario Monti. In a series of major rulings during the late 1990s and early 2000s Monti demonstrated his eagerness to take on some of the largest US corporations. In particular, his actions in the proposed merger between General Electric (GE) and Honeywell created a series of recriminations across the Atlantic. Such tensions were significant not only on a bilateral level but also on the level of the multilateral regime, given the attempts to extend WTO involvement into areas of competition policy. They were given added dimensions by the changing nature and culture of anti-trust policies in the USA itself under the Bush Administration, which made it more difficult to achieve the kind of informal problem solving that had characterized much of the 1990s. Of the cases surveyed here, this is perhaps the one with the least immediate multilateral contagion, given that the types of mergers and acquisitions involved might be found only in relations between relatively few countries or regions, but it could be argued that with the advance of globalization it is one that is likely to be increasingly influential in the future. It is also important because, along with some other areas, it relates at least as much to culture and implicit norms as it does to explicit procedures.

Conclusions What does the preliminary review of samples conducted here show about the analytical leverage to be gained through a focus on bi-multilateral negotiations in EU–US relations? There are really two sets of conclusions to be outlined here: first, conclusions about the potential of the bi-multilateral framework in general and, second, conclusions about the sample of cases surveyed. With respect to the more general issue, the discussion of the sample seems to provide support for the utility of a bi-multilateral (as opposed to a two-level or other) framework. The cases outlined cover a wide range but common to them all is the difficulty of separating the bilateral from the multilateral dimension, and the ways in which this affects the occurrence, progress and outcomes of the negotiations that took place. It is clear that in terms used earlier, the construction or reconstruction of international regimes provides a fertile soil for the occurrence of bi-multilateral processes. It is also clear that the extent to which bi-multilateralism

180 Michael Smith is designed into the process (either by the participants themselves or as the result of earlier institutional choices) is an important aspect of a number of the cases. Finally, it is also clear that the effects of ‘contagion’, through which bilateral issues call into question multilateral processes or in which multilateral processes, can become hostage to bilateral processes can be illustrated from the sample. In all of the cases reviewed, it is clear that, with significant variations, the element of ‘co-dependency’ is present, and that it is a systematic or organic part of the negotiation process. With respect to the sample of cases examined above, it is important initially to observe that it is a highly selective and arguably biased one. By focusing particularly on issues with a WTO context (in all but one of the cases) it can clearly be argued that this gives EU–US relations a particular salience, and that it introduces a bias in terms of the institutional context. The WTO in a sense institutionalizes bimultilateralism in a particular form, and this means that the sample is bound to display it. This is of course an argument for extending the range of the sample in future research, not for abandoning the idea of bi-multilateralism. The Kyoto case seems to demonstrate that in the creation of new regimes there is a good deal to be said for a bi-multilateral approach, and this in turn reflects the salience of the European Union and the USA in the fields covered by many new or extended international regimes. On the other hand, where there are significant shifts in coalitions within the multilateral context, as with the Doha Development Round, it can be argued that these affect the salience of the European Union and the USA and that solutions based on an exclusive EU–US bi-multilateralism are likely to prove ill-founded. It is apparent from the Helms–Burton/ILSA case that there are circumstances in which the multilateral system can be held hostage or used by parties to bilateral disputes, and that this risk has not been reduced by the changes in the WTO system. In something of the same fashion, the Chinese membership issue demonstrates the ways in which institutional rules can generate bimultilateral processes and produce incentives to ‘competitive negotiation’ or ‘negotiation races’ linked to the granting or withholding of benefits. As noted at the beginning of the chapter, the argument made here is a very preliminary one, but it appears to have potential as a way of gaining leverage on EU–US negotiation processes in the post-Cold War era. There is preliminary validation of the bi-multilateral focus, and of some of its key elements in relation to processes of regime construction and reconstruction; likewise, there is confirmation that this can be a profitable way of approaching EU–US negotiation processes. Although the criteria for selection of cases have not been fully exposed, it appears that in addition to those proposed earlier in the chapter, the following are important potential principles for selection: first, the extent of EU–US interests and stakes in a given area; second, the density and shape of the bilateral and multilateral institutional processes (that is to say, choosing differently structured areas); third, the prominence of negotiations for world order concerns and use of multilateralism as a norm, as well as multilateral institutional contexts. These and other criteria should enable cases to be selected that give a more systematic insight into the processes of EU–US negotiation, and they should provide the basis for more extended research into this area.

The EU and the United States of America 181

References Allen, D. and Smith, M. (1999) ‘External Policy Development’, in G. Edwards and G. Wiessala (eds) The European Union: Annual Review of Activities, 1998, Oxford: Blackwell. Allen, D. and Smith, M. (2000) ‘External Policy Developments’, in G. Edwards and G. Wiessala (eds) The European Union: Annual Review of Activities, 1999, Oxford: Blackwell. Allen, D. and Smith, M. (2001) ‘External Policy Developments’, in G. Edwards and G. Wiessala (eds) The European Union: Annual Review of Activities, 2000, Oxford: Blackwell. Allen, D. and Smith, M. (2002) ‘External Policy Developments’, in G. Edwards and G. Wiessala (eds) The European Union: Annual Review of Activities, 2001/2, Oxford: Blackwell. Allen, D. and Smith, M. (2003) ‘External Policy Developments’, in L. Miles (ed.) The European Union: Annual Review of Activities, 2002/3, Oxford: Blackwell. Baldwin, M., Peterson, J. and Stokes, B. (2003) ‘Trade and Economic Relations’, in J. Peterson and M. Pollack (eds) Europe, America, Bush: Transatlantic Relations in the TwentyFirst Century, London: Routledge. Bodansky, D. (2003) ‘Transatlantic Environmental Relations’, in J. Peterson and M. Pollack (eds) Europe, America, Bush: Transatlantic Relations in the Twenty-First Century, London: Routledge. Cowles, M. G. and Smith, M. (1999) ‘Public Goals and Private Strategies in the Transatlantic Economic Partnership’, paper presented to the Sixth International Biennial Conference of the European Community Studies Association, Pittsburgh, June. Devuyst, Y. (2001) ‘Transatlantic Competition Relations’, in M. Pollack and G. Shaffer (eds) Transatlantic Governance in the Global Economy, Lanham, MD: Rowman and Littlefield. Evans, P., Jacobson, H. and Putnam, R. (eds) (1993) Double-Edged Diplomacy: International Bargaining and Domestic Politics, Berkeley, CA: University of California Press. Gompert, D. and Larrabee, S. (eds) (1997) America and Europe: A Partnership for a New Era, Cambridge: Cambridge University Press. Hampson, F. O. (1995) Multilateral Negotiations: Lessons from Arms Control, Trade, and the Environment, Baltimore, MD: Johns Hopkins University Press. Jönsson, C. (2002) ‘Diplomacy, Bargaining and Negotiation’, in W. Carlsnaes, T. Risse and B. A. Simmons (eds) Handbook of International Relations, London: Sage. Kahler, M. (1993) ‘Multilateralism with Small and Large Numbers’, in Ruggie, J. (ed.) Multilateralism Matters: The Theory and Praxis of an Organisational Form, New York: Columbia University Press. Kremenyuk, V. (ed.) (1991) International Negotiation: Analysis, Approaches, Issues, San Francisco, CA: Jossey-Bass. Levine, N. (ed.) (1996) The United States and the European Union: Economic Relations in a World of Transition, Lanham, MD: University Press of America. Meunier, S. (2000) ‘What Single Voice? European Institutions and EU–US Trade Negotiations’, International Organization 54 (1): 103–35. Meunier, S. and Nicolaidis, K. (1999) ‘Who Speaks for Europe? The Delegation of Trade Authority in the EU’, Journal of Common Market Studies 37 (3): 477–501. Meunier, S. and Nicolaidis, K. (2000) ‘EU Trade Policy: The Exclusive Versus Shared Competence Debate’, in M. G. Cowles and M. Smith (eds) The State of the European Union, Volume 5: Risks, Reform, Resistance and Revival, Oxford: Oxford University Press.

182 Michael Smith Nye, J. and Keohane, R. (1993) ‘The United States and International Institutions in Europe after the Cold War’, in R. Keohane, J. Nye and S. Hoffmann (eds) After the Cold War: International Institutions and State Strategies in Europe, 1989–1991, Cambridge, MA: Harvard University Press. Odell, J. (2000) Negotiating the World Economy, Ithaca, NY: Cornell University Press. Peterson, J. (1996) Europe and America: The Prospects for Partnership, second edition, London: Routledge. Peterson, J. and Ward, H. (1995) ‘Coalitional Instability and the New Multidimensional Politics of Security: A Rational Choice Argument for US–EU Collaboration’, European Journal of International Relations 1 (2): 131–56. Peterson, J. and Pollack, M. (eds) (2003) Europe, America, Bush: Transatlantic Relations in the Twenty-First Century, London: Routledge. Philippart, E. and Winand, P. (eds) (2001) Ever Closer Partnership: Policy-Making in US–EU Relations, Brussels: PIE–Peter Lang. Pollack, M. and Shaffer, G. (eds) (2001) Transatlantic Governance in the Global Economy, Lanham, MD: Rowman and Littlefield. Putnam, R. (1988) ‘Diplomacy and Domestic Politics: The Logic of Two-Level Games’, International Organization 42 (3): 427–60. Putnam, R. (1993) ‘Two-Level Games: The Impact of Domestic Politics on Transatlantic Bargaining’, in H. Haftendoorn and C. Tuschhoff (eds) America and Europe in an Era of Change, Boulder, CO: Westview. Risse-Kappen, T. (ed.) (1995) Bringing Transnational Relations Back In: Non-State Actors, Domestic Structures and International Institutions, Cambridge: Cambridge University Press. Ruggie, J. (ed.) (1993) Multilateralism Matters: The Theory and Praxis of an Organisational Form, New York: Columbia University Press. Sloan, S. (2002) NATO, the European Union and the Atlantic Community: The Transatlantic Bargain Reconsidered, Lanham, MD: Rowman and Littlefield. Smith, M. (1998) ‘Competitive Cooperation and EU–US Relations: Can the EU be a Strategic Partner for the US in the World Political Economy?, Journal of European Public Policy 5 (4): 561–77. Smith, M. (2001) ‘Regionalisation, Globalisation and EU–US Relations: The Political Economy of Uneven Integration’, paper presented to the British International Studies Association Annual Conference, Edinburgh, December. Smith, M. (2004a) ‘Between Two Worlds? The European Union, the United States and World Order’, International Politics 41 (1): 95–117. Smith, M. (2004b) ‘Foreign Economic Policy’, in Carlsnaes, W., Sjursen, H. and White, B. (eds) Contemporary European Foreign Policy, London: Sage. Young, A. (2000) ‘The Adaptation of European Foreign Economic Policy: From Rome to Seattle’, Journal of Common Market Studies 38 (1): 93–116. Young, A. (2002) Extending European Cooperation: The European Union and the ‘New’ International Trade Agenda, Manchester: Manchester University Press. Zartman, I. W. (ed.) (1994) International Multilateral Negotiation: Approaches to the Management of Complexity, San Francisco, CA: Jossey-Bass. Zartman, I. and Berman, M. (1982) The Practical Negotiator, New Haven, CT: Yale University Press.

12 The Cotonou agreement Asymmetric negotiations and the impact of norms Ole Elgström

Introduction Since 1975 the so-called Lomé conventions (the first agreement was signed in Lomé, the capital of Togo) have constituted the most significant part of the European Union’s relations with the Third World. They have been heralded by EU officials as showpieces of development assistance. These aid and trade agreements are of major importance for the development efforts of a large number of countries – 46 in 1975, 71 today – in Africa, the Caribbean and the Pacific (ACP). In the 1990s proposals for radical change were introduced and these were seen as threatening the traditional foundations of the Lomé agreements. After long, tortuous negotiations a new agreement was agreed upon in February 2000. The new convention, signed by the parties in Cotonou later the same year, bears witness to novel ideas but is nevertheless a continuation of the traditional EU– ACP ‘partnership’. The aim of this chapter is to analyse the post-Lomé agreement, and the processes leading up to it, using a negotiation perspective. It will explain the contents of the agreement, and the changes that have taken place since the previous agreements, by highlighting major elements of the bargaining situation facing the parties and how this situation has developed over time. The argument contained herein includes two basic contentions. The first is that, in general, the Lomé negotiations reflect the impact of norms and identities and not only the effects of traditional bargaining power. Norms play an important role in forming EU negotiation positions and EU responses to ACP initiatives. It is argued that a gradual transformation has occurred in EU–ACP relations. The moral norms and powerful self-images that created a European identity in relation to the former colonies – and emphasized concepts of partnership and obligations – have been weakened, while other norms, stressing liberalism and democratization, have become more influential in the negotiations. The normative basis for EU policy towards the Third World has been thoroughly reshaped. The result is the introduction of World Trade Organization (WTO) compatible free-trade rules and stricter conditionality. The partner identity, institutionalized into the EU aidproviding machinery, has however acted as a bulwark against total disruption of the tradition-bound Lomé relationship.

184 Ole Elgström Second, it is claimed that internal negotiations within the European Union itself are more influential to the outcome of the Lomé negotiations than the actual negotiations between strong and weak. When faced with supplicant, weaker negotiation partners, the European Union first has to find a common position through a cumbersome, often adversarial, internal bargaining game before it presents a strong, united front to the opponent. The effect is a ‘take-it-or-leave-it’ approach that is difficult for the weaker party to change substantially. The principles of the new agreement mirrors the EU negotiation mandate that was the outcome of internal negotiations of the member states and institutional actors. The ACP countries have been able only marginally to affect the post-Lomé process. In comparison with existing accounts of the Lomé regime, the present analysis is different in its explicit focus on the bargaining situation, including its insistence on the prominence of norms as explanatory factors. An analysis solely based on material power resources could not, it is argued, explain either the shifting negotiation outcomes over time or the specific contents of the Cotonou agreement. After a brief description of the general characteristics of the Lomé conventions, this chapter will outline the principal features of the Lomé negotiations. They are described as an asymmetric bargaining game in which the internal deliberations of the stronger side is a major predictor of outcomes. Norms and identities are treated as major determinants of EU positions. Thereafter, the analytical framework is used to analyse the recently negotiated convention. A discussion of how the explanatory factors have interacted to produce a new pattern of EU–ACP relations will be undertaken. The chapter ends with a number of concluding remarks.

The Lomé conventions 1975–95 The Lomé agreements are aid and trade pacts between the EU members and a large number of their former colonies, with a heavy focus on African countries which are among the world’s poorest. Trade has been used as an instrument to promote growth and development, inter alia by abandoning the sacred General Agreement on Tariffs and Trade (GATT) principle of reciprocity. Most products (some exceptions exist within the agricultural sector) from the ACP states can enter the EU market without tariffs. The ACP group thus enjoys non-reciprocal preferential access to the EU market. As a complement, the STABEX and the SYSMIN programmes aiming at the stabilization of export revenue earnings were created during Lomé I (1975) and II (1980). These insurance schemes, which compensated ACP countries that suffered short-term declines in their export earnings, were greeted as unprecedented attempts to solve the problems of rawmaterial producing Least Developed Countries (LDCs). Fundamentally, the aid component of the Lomé conventions aims at improving living conditions in the recipient states. The size of the budget is determined by intergovernmental negotiations between the EU members before each new agreement. Lomé is thus financed outside the regular EU budget. The budget (the European Development Fund, EDF) covered ECU 13.1 billion for 1995–2000. This means that the European Union is one of the major multilateral donors. The

The Cotonou agreement 185 Lomé assistance programme is run by the Development Directorate (formerly DG VIII), while – oddly enough – aid programmes that are within the ordinary EU budget are administered by the External Relations Directorate (DG I). EU assistance is seen as complementary to the bilateral aid programmes of the members. The multilateral share of total EU aid has remained about 15 per cent since the mid-1970s. Efforts to increase the weight of joint EU policy, or even to coordinate more closely all EU assistance, have met with little success, despite the Maastrich Treaty’s call for greater policy coordination with the member states. Traditionally, the Lomé agreements covered five years. Lomé II and III were largely similar to the first convention. Changes in bargaining power did not translate into any major changes in aid priorities. EDF allocations continued to increase. It could be argued that, increasingly, the major goal of the ACP countries was to defend and conserve the Lomé regime. Lomé IV, which covered ten years (1990–2000), introduced structural adjustment lending and human rights conditionalities. These trends, reflecting a growing scepticism over the effectiveness of foreign assistance, were strengthened in the Mid-Term Review of 1995. It was against this background that negotiations that were to decide the fate of EU–ACP partnership were initiated in the second half of the 1990s.

Principal features of the Lomé negotiations An asymmetrical bargaining game The Lomé negotiations have from the beginning been similar to other multilateral North–South negotiations in their basic power asymmetry. Different authors have used different terms to symbolize this kind of bargaining, which is founded on inequality; John Ravenhill writes about ‘collective clientelism’ (Ravenhill 1985), while Christopher Stevens calls the same phenomenon a ‘patron–client relationship’ (Stevens 1986). The terms refer to a situation where a weak party negotiates with a strong party in order to obtain favours through an exchange of noncomparable resources. From this perspective, the ACP countries are seen as demandeurs or supplicants. Western Europe is by far the most important economic partner for the ACP countries. Almost 50 per cent of the total African exports were purchased by Europe in the early 1990s. Western Europe is also the biggest supplier of Africa’s imports, and the dominant donor of development assistance. More than half of the total aid to Sub-Saharan Africa comes from Europe (Rye Olsen 1997: 1). Although about 85 per cent of this was bilateral aid from EU members, and only 15 per cent EU aid as such (Grilli 1993: 57), the figure still underlines the relative importance of the EU countries for African ACP members. At the same time, the economic position of many ACP members is deeply worrying: 32 of them belonged to the UN list of 41 LDCs, and half of them had structural adjustment agreements with the International Monetary Fund (IMF) in the late 1980s (Ravenhill 1992: 42). By all standards, the bargaining strength of the ACP countries is thus very small. Their relative bargaining power is extremely weak, regardless of whether

186 Ole Elgström we look at their aggregate or their issue-specific power (Habeeb 1988). Their need for aid resources is desperate, their alternatives are few in a world where total aid flows are stagnating. Their dependence on the European Union is far-reaching. The miniscule secretariat at their disposal seems almost ridiculous in comparison with the well-oiled administrative machinery of the European Union. According to traditional realist-inclined bargaining theory, asymmetric dependence implies asymmetric bargaining power, which, in its turn, produces asymmetric outcomes, reflecting ‘the power of the purse’ (Elgström 1992: 22–3). We would therefore – if we only paid attention to traditional power resources – expect the European Union to totally dominate negotiation processes and the outcome to reflect the interests of the Union. In any conflict, the European Union would prevail. In principle, the power situation was the same in 1975 as it is today. We would thus, from a power perspective, expect few changes over time. If anything, the asymmetry has become more poignant over the last 20 years. Most of the ACP members experienced severe economic decline in the 1980s. Contextual developments have served to further undermine the relative power of the ACP countries. New members of the European Union have diluted the former strong regional emphasis on Africa. For example, the accession of Spain and Portugal certainly advanced the attention paid to Latin America within the European Union. This development coincided with an increased general interest in developing countries with a more promising economic performance than the ACP members (Babarinde 1995). Even more devastating for ACP interests was the fall of the Berlin Wall and the resulting ‘need’ for EU aid to central and eastern Europe; these events constituted, according to many observers, a direct threat to traditional foreign aid as they competed about the same scarce resources (Lister 1997: 138–43; Babarinde 1995: 477). Finally, the trend towards ever lower global tariffs that were negotiated in the GATT process, further undermined the relative advantage of EU preferences within the Lomé framework. The combined effect of these factors has been an increase in the perceived power asymmetry, resulting in a decline in the significance of the Lomé process. Lomé’s continued pride of place in EU development cooperation in the early 1990s was, as Enzo Grilli put it, more for historical and humanitarian reasons than for any economic self-interest in Africa’s development (Grilli 1993: 346). In the cognitive worlds of EU decision makers, the Lomé conventions have lost in value. The major EU development interests are perceived to lie elsewhere, in east and central Europe, in the Mediterranean, in Latin America and even in Asia. The importance of internal EU bargaining Several scholars have underlined the major importance of internal EU deliberation in order to understand the Lomé talks. ‘The most important negotiations continued to occur within the European Community, in particular between the member states at the Council of Ministers level’, remarks John Ravenhill (1992: 58). In practice, the member states first have to conclude their negotiations

The Cotonou agreement 187 and agree upon a mandate before actual talks between the European Union and the ACP states can begin (Nötzold and van der Ropp 1990: 190; Rye Olsen 1997: 9). These observations seem valid not only to EU–Third World negotiations but to all asymmetrical negotiations in which the European Union is involved. Thus, the same phenomenon has been noticed in accession negotiations with central and east European countries. The fact that such EU negotiations are ‘two level games’, involving distinct phases of internal and external bargaining, brings with it certain problems (Friis 1997: 91; Putnam 1988). Internal EU negotiations are often very complex and time-consuming. After a deal is concluded, it becomes extremely difficult to unravel it: if one article is changed, other aspects of the internal deal will then have to be re-negotiated. The result is that the outsiders, in this case the ACP states, are faced with a ‘take-it-or-leave-it’ situation. As the hands of the EU negotiator are tied by the internal agreement, the power of the Union increases. Protracted negotiations become legion. Lykke Friis also points to what she calls a ‘club logic’ within the European Union: even ‘friends of the LDCs’ among the members prioritize interests of other members before the interests of any external actor (Friis 1997: 95). In brief, the European Union is a difficult negotiation partner. The internal cohesiveness – once an internal deal is made – and the problems of unravelling intra-EU agreements make the European Union a tough actor to deal with. In combination with the fundamental asymmetry of power, the prediction must be that the expected outcome in any EU negotiation with weaker countries will be very closely linked to the initial position of the European Union. The abovementioned traits also contribute to what Christopher Hill has called the ‘capability– expectation gap’: the EU system can seldom deliver what the actors they negotiate with expect to get from a deal (Hill 1993). This might lead to frustration and to threats to the European Union’s external bargaining reputation. The Lomé negotiations reveal the complex, multi-actor character of internal EU negotiations. These are not straightforward intergovernmental negotiations despite the intergovernmental nature of EDF funding. The Commission has regularly acted as an independent player (or as several players, as there have been splits within the Commission regarding development policy), and has often tried to pursue its own interests – sometimes by acting as champion of the interests of developing countries (Nötzold and van der Ropp 1990: 191; Stevens 1986). The European Parliament has increasingly developed into an influential actor in its own right (Ravenhill 1992: 44). It already has manifold ways to influence the outcome of the Lomé negotiations. The EU–ACP Joint Assembly (institutionalized meetings between members of the European Parliament and of the parliaments of ACP countries) is an appropriate forum for sending signals to the Commission. The constitutional necessity to have the approval of the European Parliament on any agreement gives the Council a strong incentive to listen to Parliament advice. The necessity to analyse EU–ACP negotiations as a two-level game is only emphasized by the inclusion of EU institutional actors.

188 Ole Elgström Norms, identity and interests According to the two theoretical approaches described above – highlighting asymmetrical power and the European Union’s status as a ‘difficult negotiating actor’ – EU–ACP negotiations ought to be characterized by hegemony and unilateral exertion of power. This picture is correct – but only to a certain extent. It is equally true that the poor countries have not exclusively played the role of pawns in the bargaining games. Several observers have remarked upon the relative influence of the less favoured nations in the Lomé negotiations. The ACP countries are claimed to have reached substantial success, especially during the first rounds of negotiations, and to have been able to preserve and even enlarge these advantages (Hewitt 1989). We are thus faced with a puzzle: how is it that the Lomé negotiations, though primarily guided by the interests of the northern countries, have continued to demonstrate significant signs of weak state power? It is the view of the author that the answer to this puzzle lies primarily in the realm of norms, morals and identity and in the ability of the ACP states to capture and utilize such sentiments within the Union. It is proposed that the European Union’s development cooperation policy has been heavily influenced by a European aid identity, which was created in the 1960s and 1970s and which has continued to be reflected in negotiations to date. This identity was mirrored in normative ideas about what constituted a legitimate international political order. In the 1990s, however, powerful forces have challenged this identity and consequently EU–ACP relations are in a state of flux. This chapter argues – consistent with constructivist ideas in social science (Adler 1997; Finnemore 1996) – that state (and EU) interests are constructed through social interaction. The prevalence of strong norms (defined as shared standards of appropriate behaviour held by a community of actors) plays an important part in these processes. State (and EU) preferences are, therefore, partly constituted by social norms and culturally determined identities. Interests are not inherent qualities of states and cannot be readily deduced from objective conditions and material characteristics. Material facts – ‘objective power relations’ – are not selfevident. They acquire meaning through human cognition and social interaction (Finnemore 1996: 6). National interests are not given, but in constant development and partly depending on predominant ideas and norms. The activities of moral entrepreneurs – be it non-governmental organizations (NGOs), inter-governmental organizations (IGOs) or state representatives – can hasten norm spread. At the time of Lomé I in 1975, a European partner identity had developed (Grilli 1993: 93; Ravenhill 1992). It consisted of a number of interrelated beliefs: in special ties between the EC and Africa, in a special responsibility for the EC, in economic complementarity and the possibility of mutual benefits, and in interdependence between rich and poor. The partner identity was translated into a specific conception of a strong EU interest in lasting and equal relations with the ACP countries. The result was the rhetorical emphasis on Lomé being a contractual relationship between equal partners – a partnership – but also a practical devotion to joint planning and joint institutions (Greenidge 1998: 59).

The Cotonou agreement 189 The legacy of colonialism was probably the most important factor in creating the partner identity (Grilli 1993: 330–2). Not least in France, the ideas of continued responsibility and unique ties were strong and widespread (Hewitt 1989; Grilli 1993). The term ‘EurAfrica’ was coined to symbolize this special relationship. A ‘sense of obligation’ or ‘richesse oblige’ existed, expressed in a feeling of responsibility for the former colonies (Zartman 1992: 6). To this was added the prevalent belief in international interdependence; in this case that the West European states needed African raw materials, while the European Union could provide finance and knowledge for African development (Lister 1997: 172–3). An unstated, but still important factor for France and other actors in the EC was the possibility to interpret Lomé as a sign of continued great power status. Africa could, in this perspective, be seen as an EU sphere of interest. The proclivity of EU spokesmen to portray the agreements as a showpiece to the world, and as ‘better’ than aid programmes of other states reinforces this interpretation (Lister 1997: 130; Hewitt 1989). The practical effect of the partnership principle was to enhance the power of the ACP group. It became very difficult for the European Union to diminish its aid (at least in nominal terms) or to introduce any measure which could be interpreted as a deterioration of existing agreements. The preeminence given to the Lomé convention in EU official rhetoric as ‘the main pillar of the Community’s development policy’ implied that ‘there would have been considerable embarrassment if the convention had not been successfully renegotiated’ (Ravenhill 1992: 56). John Ravenhill talks about the ‘foot in the door’ effect: ‘each time the convention is renegotiated, the ACP states can use the existing benefits they enjoy as a lever with which to pry open the door a little further’ (Ravenhill 1992: 57). This, it could be argued, held true in all negotiations before the 1990s (including Lomé IV, signed in 1990). Today, this picture has changed. The partnership rhetoric is still there, as are elements of altruism and moral commitment (Lister 1998: 22). The underlying tone is, however, different. The Lomé principles are no longer sacred. The European Union as a donor is much more demanding than it was ever before. The reason is partly the shift in power noted above: the EU countries have today very limited material interests in the ACP states. However, more importantly, the change is also due to a shift in ideational priorities. The spread of new, powerful transnational norms has eroded traditional ideals and thoroughly changed the ideological basis for the European Union’s negotiation approach. In Lomé IV, aid for structural adjustment was introduced as a major new feature. The underlying idea was that a sound macroeconomic environment is a precondition for economic development. In doing so, the European Union chose ‘to accept the World Bank and IMF formulae’ (Lister 1997: 116). This is a clear example of ideational diffusion from IGOs to other actors (cf. Finnemore 1996). A new international norm priority, introduced by the IMF/World Bank, was adopted by most donors and was soon an almost unchallenged dogma. Similarly, the once heralded notion of EU assistance as being ‘non-political’ is no longer valid (Grilli 1993: 101–7). A linkage between giving aid and human

190 Ole Elgström rights was introduced in Lomé IV, and this was considerably strengthened in the so-called Mid-Term Review of 1995. In this document, demands for democratization and the rule of law appeared for the first time, and a unilateral suspension mechanism in cases of flagrant misbehaviour was included (Crawford 1998: 132–7). These steps constituted an adaptation to prevailing international norms and reflected a ‘virtual international consensus in favour of aid conditionality’ (Lister 1998: 31). The introduction of political conditionality is, therefore, another example of ideational diffusion, which has transformed EU policy. In general, the spread of international norms condoning humanitarian intervention has tended to legitimize an extended use of political conditionality. The ACP tried to oppose the new conditionality, referring to it as an ‘affront to the partnership spirit’, but to no avail (Babarinde 1995: 469). In the mid-1990s a combination of shifts as regards material power and predominant ideas had created a new aid climate. The partner identity had been eroded – even if fragments remained and were still important enough to make any total dissolution of the Lomé system a difficult decision. It had been replaced by an attitude of benign neglect, signifying a lack of a clear EU identity with regard to the ACP group. EU interests – constructed in an interplay between norm adaptation and changing perceptions of power – had been transformed: the overall interest in Africa (let alone the Pacific or Caribbean) had diminished, whilst ideological concerns about human rights, democracy and good governance had gained a more prominent position.

The post-Lomé negotiations This part of the chapter will analyse the reshaping of EU–ACP relations that have taken place in recent years. To this end, the analytical instruments discussed above are used. After a brief overview of the events leading up to the Cotonou convention, this section will focus first on the internal negotiations that resulted in the EU negotiating mandate. Following this, a description and analysis of the post-Lomé negotiation process and the Cotonou agreement will be undertaken. Throughout, it will be emphasized how norm evolution has transformed EU–ACP relations since 1995. The event that set the ball rolling was the publication, in November 1996, of the Commission Green Paper on the future of the Lomé Convention. The Green Paper was officially described as simply a think piece, not necessarily reflecting Commission standpoints, but it was widely perceived as a trial balloon to see how prepared EU members were for fundamental changes in EU foreign assistance. Whatever its motives, the Paper succeeded in usurping the development agenda and became the focus of debates on aid and trade in the years to come. This signified an increased role for the Commission in development policy planning. The Paper set forth wide-ranging changes. It was permeated with a desire for a ‘revived partnership’ with a much stronger emphasis on a strengthened political dialogue including unambiguous demands on human rights and good governance. It introduced the revolutionary idea that future relations should perhaps be based

The Cotonou agreement 191 on reciprocity, and the almost heretical notion that the ACP group was perhaps not an ideal partner to cooperate with in corpore; perhaps it should be broken up and new agreements signed with smaller groups of LDCs. In brief, the GP called for what the French call a banalisation of the EU–ACP special relationship, excluding the special properties of the Lomé agreements that the ACP states had praised and changing the terms to conform to WTO norms (cf. Whiteman 1998: 33). The Commission was acutely aware of the sensitive nature of these ideas and expected a lively debate. Indeed, it facilitated such a debate by arranging a number of seminars on the Paper in each member state and in ACP settings. This debate can be understood as internal pre-negotiations that allowed a broader than usual range of actors to participate. Most member state governments also seem to have produced official responses to the Green Paper in the form of government positions (in the autumn of 1997). With all these reactions as inputs, the Commission, in January 1998, produced a recommendation regarding the Union’s future partnership with the ACP countries (Agence Europe 5/2 1998), leading to member state and European Parliament reactions and to negotiations in the Council. The British Presidency succeeded in producing a joint negotiation mandate in June. The ACP countries had a slower start, but finally reached consensus on a general position in time for the ACP–EU formal negotiations to commence as scheduled in September 1998. For more than a year, negotiations moved ‘at a snail’s pace’ and fundamental differences seemed to block any agreement (European Voice 1999). However, following a marathon negotiation session in early December 1999, ACP–EU ministers reached an agreement on most of the contentious issues that had plagued the negotiations. A ministerial meeting in February 2000 put the finishing touches to the new trade and aid deal. Internal EU negotiations on Lomé: new actors – conflicting norms As argued in the first part of this chapter, the outcome of the internal EU negotiations is probably the most effective predictor of the final ACP–EU negotiated agreement. This makes a two-level game analysis of the Lomé negotiations highly germane. General norm consensus is the phenomenon to be expected within the EU circle. The Green Paper had expressed this Western consensus in an incisive form, and in many instances the proposed standpoints became less radical in the EU negotiating mandate. Nevertheless, the core of the norms expressed in the Paper was still evident. Poverty eradication and the integration of the ACP states into world economy had become constitutional treaty obligations when they were included as goals for EU development cooperation in the Maastricht Treaty. A sound economic and political environment is another familiar element in the EU proposals. New to the post-Lomé situation, however, is the priority placed on free trade and WTO compatibility, and the strong commitment to a strengthened political dialogue. Integration into the world economy – itself a phrase of the 1990s – was now intimately linked to the creation of Free Trade Agreements with regional ACP

192 Ole Elgström sub-groupings. The importance placed on the ‘credibility’ to the multilateral trade system of any Lomé agreement signals that conformity with WTO rules has become a primary objective and an EU norm. Regarding the political dialogue, the expressed need for a ‘substantial improvement’ indicates that earlier efforts to influence the ACP by means of a dialogue have not been too successful: the French call the political dialogue the parent pauvre of Lomé, lacking ‘de contenu et de visibilité ’ (in their position on the Green Paper). Another new area of shared values is ‘differentiation’. The Green Paper’s suggestion that the European Union’s relations with the ACP countries should be based on geographical regions, and that agreements should take account of each region’s conditions, has been accepted as an EU norm, but complemented with the more conventional view that each country’s level of development should also determine what concessions it receives. However, within this framework of general consensus, the EU negotiating mandate was a product of compromise. It reflects the priorities of member states with differing development policies and diverse norms. In the internal policy process, three major cleavages can be discerned: ●

●

●

the chasm between traditionalists (who want to preserve ‘the spirit of Lomé’) and revisionists (who prefer an all encompassing revision), a division that partly corresponds to the division between regionalists and globalists; the divide between growth-oriented free-market proponents and social development-oriented aid supporters; the cleavage between free-trade enthusiasts and trade sceptics.

When studying the reports from the national debates on the Green Paper, it is quite striking to notice how many actors, not least among the NGOs, actually advocate the status quo and refer to the positive aspects of the Lomé agreements and to the potential negative effects of a free-trade approach (British Report 1997: 13; Agence Europe 23/4 and 20–21/4 1998). In the internal negotiations, these traditionalists joined forces with actors with strong emotional ties to the old agreements. The French position paper emphasizes the necessity to preserve the ‘originalité et integrité de convention de Lomé ’, while at the same time adapting it to new circumstances. In the presentations made by the Commission it is stressed that the fundamental challenge has been to introduce necessary fundamental rethinking ‘without calling into question the essential aspects of the “pioneer Lomé model”’ (Agence Europe 29–30/6 1998). The negotiating mandate bears witness to this ambiguity: the ACP umbrella is retained as are the references to ‘partnership, forseeability and contractuality’, but combined with geographical differentiation and radical change in the trade regime (Agence Europe 29–30/6 1998). The Green Paper was permeated with a market- and growth-oriented view of development. Social aspects of development, while not totally neglected, were not a primary concern. In the eyes of one critical observer, the Paper ‘is a document that is the product of the Commission’s irritation and frustration with recalcitrant client states which are unwilling to impose liberal economic policies’ (Grynberg

The Cotonou agreement 193 1997: 2). The Austrian position on the Paper saw it as a watershed: ‘Should market economy or development policy considerations . . . ultimately take precedence in the Convention?’ In the process leading to the negotiating mandate, a good deal of pressure was put on the Commission to include much more explicit language on social development. A united front of NGOs made this one of their first priorities (Eurostep 1998) and a major lobbying focus, and they were strongly backed by several member states (Council General Secretariat 1997, 187, 194, 210, 211) and by the European Parliament. A decisive turning-point came when the new British Labour government decided to make a ‘strengthening of the poverty focus’ and ‘securing commitment to international development targets’ priorities for its Presidency (see Survey of Current Affairs 1998). The result was a UK-Commission agreement that the May 1998 Development Council should ‘take forward the poverty focus and gender equality aspects of the Community’s development assistance’ (Survey of Current Affairs 1998). The mandate shows evident signs of these changed priorities. Social issues, gender – previously ‘woefully lacking’ (British Report 1997) – and sustainable development are given much more prominent positions. At the same time, the space given to the private market, to structural adjustment and to growth shows that the GP’s initial focus was still very much evident. The mandate is a good example of a compromise – at the expense of clear priorities – allowing both sides to claim victory. The sharp focus on WTO conformity in the negotiating mandate – turning to the third major cleavage – was itself a reflection of general Western norms on the blessings of free trade and is thus another example of transnational norm spread. In adopting the free-trade norm wholeheartedly, the Union’s proposals came to differ considerably from the Lomé tradition as far as trade was concerned. The objective to gradually integrate the ACP states into the world economy is shared by member states and EU institutions. Behind this facade of consensus there are, however, marked differences as regards how much emphasis free trade should receive. In a sense, it is a matter of trade versus aid. In another sense, it is a question of whether or not one believes in the thesis that free trade always carries with it advantages for everyone. The primacy of free trade, celebrated in the Green Paper, seems to have emanated from DG I. There, ardent believers in the blessings of reciprocity succeeded in having their standpoint becoming the guiding star of EU policy (Holland 1998). The more sceptical view of free trade is best represented by many NGOs, which labelled the Commission’s proposals ‘fundamentally unbalanced’ and maintained that it will hurt the economic prospects of many LDCs (Eurostep 1998). Several member states also voiced concern that unfettered free trade could have negative consequences for the poorest ACP states, and that it therefore had to be ‘cushioned’ with special treatment for the Least Developed Countries and other vulnerable economies (Council General Secretariat 1997, 187, 194, 221). These concerns were integrated in the final document – again after prompting from the British Presidency – in the form of special treatment for vulnerable ACP members and of intensified attention to the fight against poverty.

194 Ole Elgström Norm conflict or norm convergence? ACP–EU negotiations at the turn of the century When the two parties presented their negotiating mandates in September 1998, the official reaction was to emphasize the areas of agreement. As a result of the internal negotiation process, the European Union had, as described above, softened some of its standpoints and approached more traditional Lomé principles, although the core radical proposals in the Green Paper still remained untouched. Similarly, the ACP states had also moved away from their initial reluctance to consider any major change at all in comparison with earlier conventions (ECDPM 1998a; The Courier 1999). The parties confirmed their basic agreement on general aims, stressing the need for a ‘strengthened partnership’, and on principles like equality, ownership and differentiation according to ‘the various political and economic realities’ of the ACP countries (One World Europe 1999). Everyone today underwrites the need for a political dialogue in which questions of human rights are openly discussed – which proves that consensual norms in this area have moved a long way since the early 1980s. They agreed – and that had not been predicted by all observers at the time of the Green Paper – on the desirability of retaining the ‘special relationship’ between the groups and keeping aid and trade matters together in one comprehensive agreement. The apparent consensus hid, however, a number of areas where basic norms were diametrically opposed. Norm development in the European Union had created a situation where EU negotiation demands were no longer in tune with the ‘conservative’ positions of the ACP. The European Union was committed to abandoning the principle of non-reciprocity, even if only in a long-term perspective, and to create free-trade agreements with regional partners on the basis of reciprocity. The ACP states were fiercely against a rapid opening up of their economies. ‘Political dialogue’ still carries profoundly different meanings and connotations for the two parties. While development is the primary objective for the ACP, ‘seen as a basic human right that should not be subordinated to political goals’ (The Courier 1999), a sound political and economic environment is seen as a prerequisite for development by the Union. LDCs generally tend to view development assistance as an unalienable right, whilst developed states want to make sure that their aid has a chance to be ‘effective’ before extending any assistance. This gap was reflected in the tough negotiations on the political dialogue. The ACP countries feared that future discussions would, in practice, focus on their shortcomings and turn out to be traditional conditionality under another name. As a consequence, although much rhetoric was spent on the need for an ‘open and real’ dialogue based on mutual respect and equality, in EU policy papers (Council General Secretariat 1997, 210), it was never made clear how this would function in practice, or what would happen if the parties did not agree. The ACP countries, furthermore, demanded that EU policy must also be on the agenda of a ‘balanced and unconditional’ dialogue (ACP Group Negotiation Mandate 1998).

The Cotonou agreement 195 One of the most heated items of discussion in the whole post-Lomé process was the EU demand that ‘good governance’ was to be included as an ‘essential element’, potentially warranting aid suspension, in the convention. The ACP side feared that this ‘nebulous concept’ would serve as a catch-all to allow interference in their domestic affairs. In the final agreement, the European Union had its way. The compromise solution that made good governance a ‘fundamental’ instead of an ‘essential’ element, and that allowed only ‘serious corruption’ to be linked to the suspension of aid, cannot hide the fact that the European Union succeeded in introducing a new principle that the ACP states had initially ‘been firmly opposed to’ (European Voice 1999; Eurostep PAF 1999b). The role of NGOs in the development process was also contested. ACP governments were cautious towards opening dialogue to these new actors, who were often perceived as lacking legitimacy and potentially undermining state authority. The difficulty was resolved once the formula ‘partnership actors’ (presupposing NGO involvement in all stages of a project) was adopted, rather than the formula ‘cooperation actors’ (placing emphasis only on their role in implementation). Still, the participation of NGOs concerns only ‘information and consultation’ on development strategies – the most sensitive area for ACP governments – compared to their ‘involvement’ during the implementation stage (Agence Europe 19/12 1999; Commission 1999). Fields of contention also existed in the area of trade cooperation. One concerned the proposed ‘regionalization’ of the ACP–EU relationship. Many ACP representatives saw this as an attempt to divide and rule, to split the ACP into competing subgroupings with each having a weaker bargaining position than a united ACP group (ECDPM 1998b). Despite the obvious dissimilarities between the Lomé countries, and despite the fact that they do not appear as a united group in any other forum than in their relations with the EU, the ACP countries strongly underlined their common identity and their wish for a joint agreement (ECDPM 1998a, c): ‘The Group commits itself to preserving and reinforcing within the current geographical framework its political solidarity and unity’ (ACP Group Negotiating Mandate 1998). Regionalization was closely connected to the creation of Free Trade Areas (FTAs), as the European Union wished to set up ‘regional economic partnership agreements’ (REPAs), based on free trade, with the various ACP regions. The ACP group expressed serious doubts about the consequences such arrangements would have for them, fearing that integration into the world economy would imply marginalization. Whilst acknowledging the value of free trade in principle, the ACP countries in general advocated a continuation of the Lomé regime, and indeed demanded that trade preferences be maintained or even increased. Nonreciprocity was proclaimed ‘a pivotal part’ of the Lomé system (ECDPM 1998c; Agence Europe 3/10 1998). Moreover, in the area of trade, the general approach set out in the EU negotiating directive was preserved in the final agreement. For the first time the ACP explicitly commits itself to introduce a reciprocal trade regime – albeit in a distant future. WTO compatibility remains the key determinant. While the word

196 Ole Elgström ‘regional’ is dropped, future trade arrangements will still be based on regional integration initiatives (Eurostep PAF 1999a). The deep reluctance of the ACP to open up their economies was, in the bargaining process, translated into lengthy adjustment periods (up to 20 years – the ACP had wanted 30), exceptions and a joint agreement to seek an eight-year waiver from WTO demands on reciprocity. The European Union thus persevered with regard to the principles it wanted and conceded with regard to time-frames and in allowing some flexibility: commitment to REPAs are not immediately required. In brief, it seems more apt to characterize ACP–EU post-Lomé negotiations as expressions of increasing norm conflict than of norm convergence. Much of the consensus on objectives and principles is, of course, real and should not be underestimated. However, agreement is often only on the surface; on probing deeper, one finds competing norms and disagreement on both goals and means. The Cotonou agreement was the result of tough bargaining, notably in the areas of good governance and trade. To understand the outcome, a focus on norms and internal EU negotiations is required. Understanding the Cotonou convention The outcome of the post-Lomé negotiation process reflects the main principles of the EU negotiation mandate. The theoretically based prediction that the mandate, being the result of extended internal negotiations, would be exceedingly difficult to change, in all but details, is largely confirmed. In the words of an NGO commentator: ‘the EU’s movement on the issue [the future trade regime] is cosmetic rather than substantial . . . [T]he EU has, to a large extent, achieved its goal of pushing a lot of its proposals that it had set out before the negotiations’ (Eurostep PAF 1999a, b). The EU positions were based on new norms that had been gradually acquired in the 1990s. The Union wholeheartedly adopted liberal principles of free trade and democratic governance. The concrete consequences, in terms of the postLomé negotiations, were the European Union’s insistence on the abandonment of non-reciprocity as the guiding star of the agreement, the introduction of a trade regime based on liberal principles and on good governance as a key concept in the new text. On all these points, the Union saw its general approach win through. A novel problem for the ACP was the increasing influence of the WTO and the evolving EU norm to strictly follow WTO policy. This further constrained the ability of the ACP group to achieve its goals. Compromises were made, however. The most radical proposals of the Commission Green Paper, questioning the very ideas of having a separate partnership agreement, had already been discarded in the internal EU negotiation process. As has been demonstrated, a coalition of actors with competing norms succeeded in retaining elements of ‘the spirit of Lomé’ and in introducing a social development perspective into European Union’s aid policy. Further changes were made in the process of ACP–EU bargaining. Using arguments that related to both the traditional partnership ideology and to the moral consciousness of the rich countries,

The Cotonou agreement 197 ACP leaders gained concessions in many areas. The ACP did not immediately have to commit itself to free-trade agreements. Regionalization was downplayed. Good governance did not become another ‘essential element’. Promises of future possible improvements in ACP access to the European Union market were given. Finally, generous implementation frames were granted to postpone any deleterious effects of future free-trade agreements. Still, these concessions do not change the overall judgement that the essential principles of the novel EU norms – shaped by internal bargaining within the Union – are firmly embedded in the new agreement. This picture of EU dominance is strengthened by the unilateral character of the funding process. The EDF budget is still determined solely by the European Union. This time, the outcome of the internal budgetary wrangle was 13.5 billion euro – substantially below the 14.3 billion proposed by the Commission, a sum which would have maintained the value of the former EDF in real terms. The budget decision did little to assuage ACP apprehensions that the Lomé partnership is no longer a supreme priority in EU relations to the Third World. ‘[W]ith the dismantlement of the export revenue stabilising instruments, the Commodity protocols . . . and the imminent end to non-reciprocal trade preferences, the ACP countries no longer have the unique place they occupied among EU trade and development partners’ (Eurostep PAF 2000).

Conclusion This chapter has analysed the Cotonou agreement, and the Lomé regime in general, from a negotiation perspective, with an emphasis on how shared and competing norms influence negotiation processes. It has been argued that the foreign aid identity of the European states in the 1970s, together with the perceived increasing power of the resource-rich LDCs, created an environment where the ACP countries succeeded in getting a relatively favourable deal. The partnership ideal of the 1970s has continued to exert influence on later negotiations, including the post-Lomé process, but its importance has waned as other norms pertaining to democratic governance, a sound macroeconomic environment and the blessings of free trade have gained in leverage. In the new agreement, the principles of ‘good governance’, ‘WTO compatibility’ and ‘integration into the world economy’ have penetrated and transformed the traditional Lomé partnership. The norm-based power of the weak has been replaced by a situation of total power asymmetry, where the normative consensus of the European Union leaves little room for concessions. A second main conclusion is that the results of the internal negotiations within the European Union, before actual talks with the ACP begin, continue to be the single best predictor of what an ACP–EU agreement will eventually look like. The normative consensus of the European Union is a negotiated consensus. In EU prenegotiations, sensitive compromises are reached between groups with different interests and values; several examples of this have been given as regards the postLomé negotiations. This creates a frustrating situation for the ACP countries, who

198 Ole Elgström cannot unravel the cornerstones of the EU consensus and have to seek concessions only on the margin.

References ACP Group Negotiating Mandate (1998) ACP/28/028/98Rev.2, 30/9 1998. http:// www.oneworld.org/acpsec/gb/lome/future/negman_e.htm Adler, I. (1997) ‘Seizing the Middle Ground: Constructivism in World Politics’, European Journal of International Relations, 3: 319–63. Agence Europe, 5/2 1998, 20–21/4 1998, 23/4 1998, 29–30/6 1998, 26/9 1998, 3/10 1998, 19/12 1999. Babarinde, O. (1995) ‘The Lomé Convention: An Aging Dinosaur in the European Union’s Foreign Policy Enterprise?’, in C. Rhodes and S. Mazey (eds) Building a European Polity? The State of the European Union, vol. 3, London: Lynne Rienner and Longman. British Report on the Green Paper (1997). Commission (1999) Information memo 9 on EU–ACP negotiation 21/12, http://europa. eu.int/comm/development/event/ Council General Secretariat (1997) Note for the Members of the Development Cooperation and ACP Working Parties. Subject: Future ACP–EU Relations. Doc No. 199/97 (contribution by Austria), No. 210/97 (by Sweden), No. 187/97 (by Finland), No. 194/97 (by Denmark), No. 211/97 (by Ireland), No. 221/97 (by Belgium). Crawford, G. (1998) ‘Human Rights and Democracy in European Development Cooperation’, in M. Lister (ed.) European Union Development Policy, London: Macmillan. ECDPM (European Centre for Development Policy Management) (1998a) ‘Where Does the Post-Lomé IV Debate Stand?’, Lomé 2000, No. 7. ECDPM (1998b) ‘Formulating the ACP Negotiating Mandate’, Lomé 2000, No. 8. ECDPM (1998c) ‘ACP and EU Positions Diverge’, Lomé 2000, No. 9. Elgström, O. (1992) Foreign Aid Negotiations, Aldershot: Avebury. European Voice (1999) Vol. 5, No. 29. Eurostep (1998) Summary Response to the European Commission’s Proposed Negotiating Mandate for Lomé V (i) http://www.oneworld.org/eurostep/lom8012.htm Eurostep PAF (1999a) No. 153, 29/10, http://www.oneworld.org/eurostep/paf153.htm Eurostep PAF (1999b) No. 159, 10/12, http://www.oneworld.org/eurostep/paf159.htm Eurostep PAF (2000) No. 167, 11/2, http://www.oneworld.org/eurostep/paf167.htm Finnemore, M. (1996) National Interests in International Society, Ithaca and London: Cornell University Press. Friis, L. (1997) When Europe Negotiates: From Europe Agreements to Eastern Enlargement?, Copenhagen: Institute of Political Studies, University of Copenhagen. Greenidge, C. B. (1998) ‘The African, Carribean and Pacific Group of States’ Experience of Partnership with the European Union’, in M. Lister (ed.) European Union Development Policy, London: Macmillan. Grilli, E. (1993) The European Community and the Developing Countries, Cambridge: Cambridge University Press. Grynberg, R. (1997) Towards a North–South Monologue: A Pacific Response to the Green Paper on Relations Between the EU and ACP Countries, Maastricht: ECDPM Working Paper no. 25. Habeeb, W. M. (1988) Power and Tactics in International Negotiations: How Weak Nations Bargain with Strong Nations, Baltimore and London: Johns Hopkins University Press. Hewitt, A. (1989) ‘Development Assistance Policy and the ACP’, in J. Lodge (ed.) The European Community and the Challenge of the Future, London: Pinter.

The Cotonou agreement 199 Hill, C. (1993) ‘The Capability–Expectations Gap or Conceptualizing Europe’s International Role’, Journal of Common Market Studies, 32: 305–28. Holland, M. (1998) ‘Vices and Virtues: Europe’s Foreign Policy and South Africa 1977– 1997’, European Foreign Affairs Review, 3: 215–32. Lister, M. (1997) The European Union and the South, London and New York: Routledge. Lister, M. (1998) ‘Europe’s New Development Policy’, in M. Lister (ed.) European Union Development Policy, London: Macmillan. Nötzold, J. and von der Ropp, K. Frhr. (1990) ‘Lomé IV: A Change for Black Africa’s Return to the World Economy?’, Aussenpolitik, 2: 181–92. OneWorld Europe (1999) Guides: EU–ACP relations: A new momentum? http:// www.oneworld.org/europe/en/guides/lome/ 1999–12–21. Putnam, R. D. (1988) ‘Diplomacy and Domestic Politics: The Logic of Two-Level Games’, International Organization, 42: 427–60. Ravenhill, J. (1985) Collective Clientelism. The Lomé Conventions and North–South Relations, New York: Columbia University Press. Ravenhill, J. (1992) ‘When Weakness Is Strength’, in I. W. Zartman (ed.), Europe and Africa: The New Phase, Boulder, CO: Lynne Rienner. Rye Olsen, G. (1997) ‘Western Europe’s Relations to Africa in the Post Cold War Era’, The Journal of Modern African Studies, 34: 1. Stevens, C. (1986) Negotiating Lomé III: An Example of North–South Bargaining, IDS Sussex Discussion Paper. Survey of Current Affairs (1998) Vol. 28, No. 1, UK Foreign and Commonwealth Office. The Courier (1999) No. 173. Whiteman, K. (1998) ‘Africa, the ACP and Europe: The Lessons of 25 Years’, Development Policy Review, 16: 29–37. Zartman, I. W. (1992) ‘Introduction’, in I. W. Zartman (ed.) Europe and Africa: The New Phase, Boulder, CO: Lynne Rienner.

13 Breaking the waves The European Union’s enlargement negotiations with EFTA and central and eastern Europe Lykke Friis

Introduction: the more the messier? Enlargement of the European Union, with the exception of Greece in 1981, has always taken place in waves. Several countries negotiate in parallel and join at the same time. Indeed, in late 2002 the European Union concluded accession negotiations with a record high of ten countries. Despite the recurrent tendency to enlarge in waves, empirical studies and theoretical conceptualizations are almost silent on how the various negotiations affect each other. Often accession negotiations are looked upon as a bilateral exercise between, for instance, Sweden and the European Union. This coincides with the EU’s official rhetoric, which is repeated again and again in European Council conclusions: applicants are negotiating on their own; the vanguard of the group will not have to wait for the rearguard. Just from the mantra of international negotiation theory – ‘the more the messier’ and ‘two’s company and more’s a crowd’ – one would imagine that size matters (Midgaard and Underdal 1977; Winham 1977; Zartman 1994). Although multilateral negotiation theory is still rather undeveloped, scholars share the view that the overarching characteristic of multilateral negotiations is complexity. Consequently, scholars focus on how negotiators can limit the complexity to make it manageable for negotiations (Zartmann 1994: 3). As a preliminary hypothesis we will therefore expect that the European Union will have a clear interest in linking the various accession negotiation tables together in order to ease the negotiations and hence influence the final outcome. To be sure, EU accession negotiations are not strictly multilateral. Nor are they classical bilateral negotiations, but rather a mixture of the two. On the one hand, accession negotiations have a clear bilateral dimension, in the sense that the negotiations do not take place around a big table with all candidates present. All applicants travel to Brussels individually and negotiate with the European Union without any other applicant present. On the other hand, negotiations are also endowed with a multilateral dimension. After all, the European Union has launched a negotiation process with several countries with the aim of reaching one single goal: EU enlargement. Moreover, the negotiations are conducted in parallel.

The EU’s enlargement negotiations 201 This chapter sets out to investigate the dynamics of these special kinds of negotiations. How does the fact that the European Union is always negotiating with several applicants at the same time affect the overall negotiation outcome? Can a candidate ‘break the wave’ and negotiate independently, or is it rather swept away by the wave? The chapter approaches these questions on the basis of comparison. More specifically, we look at the European Union’s two latest accession rounds, EFTA (the European Free Trade Association) and eastern enlargement.1 The comparative method has the advantage of avoiding the single case-study trap, which characterizes the field of EU enlargement studies. Very often scholars focus on one enlargement round at a time (for an exception, which examines the method of enlargement, see Preston 1997). Although these studies provide important insights into each specific round or case, they have difficulties in producing generalizations about the enlargement phenomenon (cf. Schimmelfennig 2000). The core conclusion of the chapter is that one single accession negotiation cannot be explained by focusing exclusively on the specific negotiation table. The very fact that negotiations are conducted in parallel leads to a nested-game situation, where both the European Union and the applicants have a clear interest in linking the various tables together. In order to understand, for instance, Poland’s accession negotiation, one will therefore have to keep a close eye on all the parallel tables.

The EFTA enlargement wave: and then they were 15 The decision to launch accession negotiations with the EFTA countries was taken at the Lisbon European Council on 26–7 June 1992. Here, member states also emphasized the classical principle that negotiations would be ‘conducted in parallel, while dealing with each candidature on its own merits’.2 However, an analysis of, especially, the endgame of the accession negotiations underlines that the negotiations were clearly linked and that they should therefore be seen as a whole rather than as four independent negotiations. The negotiations with Austria, Finland and Sweden were officially opened on 1 February 1993, and with Norway a few weeks later on 5 April 1993. In line with the general tendency of postponing the heavyweight issues till the endgame of negotiations, the true action did not start until December 1993. At that time, member states, together with the applicants, managed to clear away two important issues. The first topic to be handled was the common foreign and security policy (CFSP). As neutral countries, Austria, Finland and Sweden all harboured concern that the CFSP would undermine their policy of neutrality. Although the background for each applicant was different – indeed, Norway, was a member of the North Atlantic Treaty Organization – the problem was solved in a ‘one-sizefits-all’ solution: the various applicants signed an identical joint declaration in which they accepted all aspects of the CFSP (without reservations), without mentioning the word neutrality. The overall purpose of agreeing to one common

202 Lykke Friis text was to avoid ‘singularizing’ Austria publicly (Avery 1995: 5). Such a ‘singularization’ could possibly have triggered debate in Austria, seeing that it was the only country that had made a reservation concerning its neutrality in its original application for membership. The other issue that was cleared away was connected to the Single Market. In several areas, such as environment, health and safety, the newcomers operated with stricter standards than the European Union. In these areas the applicants asked for a transition period in which they could maintain their national rules. As a starting point, the European Union offered Austria a transition phase of three years, which was accepted. However, Sweden, for its part, refused to accept the three-year offer and instead secured itself a four-year transition period. In the immediate period afterwards, Finland and Norway obtained the same, as did Austria, which had immediately knocked on the EU door demanding a similar deal. This example shows how the different negotiations very quickly become intertwined. Apparently applicants can influence the outcome of their accession negotiations by linking them to the action at the other tables: ‘If you can offer a four-year transition period to Sweden, why should we accept a three-year deal?’ The actual endgame of the EFTA enlargement emphasizes how the European Union can also influence the overall outcome by linking the various negotiation tables. Especially in the run up to the endgame in March 1994, the Commission reached the conclusion that Sweden and Finland would have fewer difficulties to resolve than Austria and Norway (Avery 1995: 7). Finland only had two major problems left. First of all, it wanted a breathing space on agriculture, so that its farmers would not have to adopt the European Union’s lower prices immediately. Alternatively, it should be possible to offer farmers compensatory payments – either from the European Union or the national budget. Second, Finland argued that a new objective 6 for aid to sparsely populated areas should be introduced in the EU’s regional policy. Since Sweden had calculated that – unlike Finland – it would become a net contributor to the EU budget, it also demanded a so-called ‘phasing in’ of its budget contribution.3 Austria and Norway, however, in addition to agriculture, budget and regional policy, had a number of requests left on their plates. For Austria, the requests centred on two issues. The first issue was the transit question: For how long a period could Austria limit the passage of heavy lorries through its territory? The second question was linked to the right of EU citizens to buy secondary residencies (i.e. vacation homes) in Austria. Inspired by Denmark’s permanent derogation on this issue, Austria at least wanted a long transition period. For Norway, the additional problems centred around one issue: fisheries. The very fact that Sweden and Finland had fewer problems ‘determined that the Union’s priority’, to quote a Commission official who was heavily engaged in the EFTA negotiations, ‘would be to conclude, if possible, with Finland and Sweden first’ (Avery 1995: 7). By closing negotiations with the two least ‘obstinate’ applicants first, the European Union could pressure Austria into signing a similar deal – playing on the fear that prolonged negotiations could leave it outside the immediate accession wave: ‘Do you really want to keep on negotiating? We can

The EU’s enlargement negotiations 203 proceed with Sweden and Finland. You would then have to wait till the next enlargement round, and who knows when that will take place?’ The Commission’s ‘divide-and-rule’ strategy, which was supported by the member states, had a considerable effect on the negotiations. After complex negotiations, the European Union was indeed able to close negotiations in the afternoon of 1 March 1994 with Sweden, immediately followed by Finland. After the European Union came up with the proposal that it would be able to co-finance transitional aid for the incoming farmers in a so-called ‘agro-budgetary package’, Sweden withdrew its demand for a budget rebate, whereas Finland accepted an immediate alignment of agricultural prices. With these two deals under their belt, EU negotiators turned to Austria. Now that the others had given in on two issues, Austria suddenly looked like a very difficult and obstinate negotiation partner if it insisted on all its various demands. As a clear example of how the European Union can ‘move’ the negotiation game by linking the various negotiating tables, Austria relinquished its demands on the budgetary and agricultural issues. After all, once the European Union had struck a deal on these two issues with Sweden and Finland, it was unrealistic that Austria would obtain more. Following the example of the Single Market, the European Union would thus risk that Finland and Sweden would reopen their deals, demanding to be treated in the same way as Austria. Instead of insisting on a special deal on agriculture and the budget, Austria accepted a deal similar to those agreed with Sweden and Finland and gave absolute priority to the transit question:4 for how many years could Austria keep its ‘ecopoint’ system limiting the passage of heavy lorries through its territory? Rather quickly, the two parties decided upon a three-stage transitional period of nine years (three times three years). However, the overall procedure for passage from one stage to the other (majority voting or unanimity) caused substantial problems. The longer the negotiations lasted on 1 March, the greater the pressure on Austria: would the country that had applied for membership first really not be able to close negotiations on the same day as Finland and Sweden? And could one even imagine that it would not be part of the first wave? After all, with two closed negotiations the European Union could indeed embark upon enlargement. In light of this pressure, it was perhaps not too surprising that Austria showed an increased willingness to compromise and closed negotiations at 9 p.m. on 1 March. The very fact that Sweden, Finland and Austria had finalized their negotiations increased the pressure on Norway. Since Norway had started the negotiations at a later stage, the time was not yet ripe for a Norwegian endgame. The Norwegian case highlights a slightly different dynamic than the Austrian one. In its own right, Austria clearly responded to a group effect; once Sweden and Finland had signed, most of the ground was cut from under the feet of its negotiators. Conversely, in the Norwegian case, the group dynamic was more related to the European Union; after having finalized negotiations with the first three EFTA states, the European Union had a clear interest in closing negotiations with Norway in order to finalize the EFTA enlargement round. If Norway closed negotiations at a much later stage, the European Union would have to embark

204 Lykke Friis upon a separate ratification procedure, but also upon another round of institutional and budgetary adjustments, before it could accept Norway. In a clear economy-of-scale logic, it was therefore in the European Union’s interest to meet Norway halfway on the all-important fisheries question. In reality, the group effect accelerated Norway’s accession negotiations. Indeed, Norway and the European Union were able to clear the fisheries question two weeks after Austria, Finland and Sweden had concluded negotiations.5 To be sure, had Norway’s Achilles heel been connected to issues already settled with Austria, Finland and Sweden, it is very unlikely that a similar acceleration could have taken place. The European Union would simply have been afraid that its generosity towards Norway would have triggered demands for Swedish, Finnish and Austrian re-negotiation.

The eastern enlargement game: and then they were 27? Just like the EFTA negotiation round, the negotiations with central and eastern Europe were kicked off with the classical mantra that each applicant would negotiate independently: ‘Negotiations with the different applicant countries will be conducted on the basis of the same principles and criteria, but separately and according to the individual merits of each applicant country. Their progress and conclusion are not required to take place in parallel’ (Presidency Report 2000). However, as can be seen below, the European Union had difficulties in translating these words into action. Negotiations with Cyprus, the Czech Republic, Estonia, Hungary, Poland and Slovenia started in March 1998. In February 2000 negotiations were also opened with Bulgaria, Latvia, Lithuania, Malta, Romania and Slovakia. For many months the negotiations developed at a snail’s pace. From March 1998 to April 2001 the member states and the applicants were still engaged in the opening rounds (cultural policy, common foreign and security policy, statistics and so on). In this period, apart from preparing the negotiations, member states mainly focused on their internal budget negotiations and the Nice Intergovernmental Conference. In reality, this meant that there was neither the time nor the political will to start the real negotiations, in which both parties would be granted transitional arrangements. These negotiations had to wait until the Treaty of Nice was finalized. Whereas negotiations before March 2001 progressed rather slowly, the negotiations during the Swedish and Belgian Presidencies in 2001 were characterized by an impressive speed, and, even more importantly, by the closure of a number of heavyweight chapters (free movement of capital and workers, environment, etc.). If one studies the negotiations in detail, it is remarkable how successful the European Union was in linking the various negotiation tables. Just as in the EFTA round, the European Union’s strategy was to close negotiations with one country and then present this as a fait accompli to the others. Since the European Union used different countries to strike the initial deal with, applicants were not able to tie the various heavyweight chapters together and demand a package deal: ‘We will only accept a transition period on free movement of labour if we get a long

The EU’s enlargement negotiations 205 transitional period on environment in return’. Every time the initial country had struck the first deal, a number of other countries followed suit – even though they had originally asked for longer transitional arrangements.6 To be sure, the tendency to follow in the wake of the first country to strike a deal also reveals a novel feature in EU enlargement history: for the first time, negotiations developed into a race, and the actual closing of chapters was vital. In earlier enlargement rounds, such as the EFTA round, there was not the same focus on how many chapters the individual countries had closed. Focus was solely on the applicants’ progress in the negotiations and their adaptation to EU legislation. The reason why countries suddenly counted chapters – and were eager to catch the number one spot in the chapter statistics – was exactly due to size. All were afraid to drown in the large group and therefore tried to secure their place in the first enlargement round by closing chapters as quickly as possible. Also because of this, countries tended to take to their heels once one of their ‘competitors’ had signed. Perhaps not too surprisingly, the European Union immediately decided to play on this in the negotiations. To quote a representative from the Swedish Presidency: We knew we could play on the applicants’ competitive instincts. So we made sure to run a very transparent negotiation process, so each applicant could measure itself against the others. The result was that the applicants pushed each other ahead. Lithuania and Latvia were star pupils here. Poland, conversely, started to lag behind and hence we could increase the pressure on the Poles. (Author’s interview, February 2002) The first heavyweight chapter to be closed during the Swedish Presidency was environment. Although many analysts had predicted that this chapter would have to wait until the endgame, Slovenia closed the chapter in March 2001. Based on the European Union’s common position, Slovenia obtained a number of transition phases between five and ten years. The ten years were linked to EU legislation on clean ground water. In the beginning of June 2001 Estonia, the Czech Republic and Hungary followed suit with similar arrangements. After Slovenia had signed, the countries realized that they had very few chances of improving the deal. In late June 2001 Lithuania and Cyprus also closed the chapter, followed by Poland in October. By the end of 2001 all applicants, except Romania, Bulgaria and Malta, had closed the chapter on more or less the same conditions as Slovenia. The next heavyweight chapter to be dealt with was free movement of capital. The core question here was the length of the transitional period the applicants could obtain on EU citizens’ right to buy, for instance, farmland in central and eastern Europe.7 The first country to strike a deal was the Czech Republic. Although its initial request for a transitional period had been ten years, its negotiators accepted the European Union’s proposal for a seven-year transitional period regarding agricultural land and five years on vacation homes. Once again, a group effect materialized, Hungary – and then later Slovakia and Bulgaria – accepted

206 Lykke Friis more or less the same arrangement. By the end of the Belgian Presidency, it was only Poland and Romania that had not closed the chapter. Once again, this underlines the fact that countries realized they had better accept the deal since it is very difficult to obtain more at a later stage. After all, if the European Union gives more to one applicant, it risks being confronted by an avalanche of re-negotiation demands. An EU negotiator, precisely in connection with Poland’s request of a 12–year transition period, has put it as follows: It is very difficult for us to offer more to Poland. It would definitely have consequences for our negotiations with the other countries. Either they would immediately ask for a similar deal or they would use it as a hostage: ‘we got less on free movement of capital than Poland, so now you had better give us more on other issues’.8 (Author’s interview, January 2002) The final heavyweight chapter to be closed under the Swedish and Belgian Presidencies – free movement of persons – was probably the most controversial one. Many member states feared that immediate access to central and eastern European workers would put their job markets under pressure. All central and eastern Europeans, conversely, took the view that free movement of persons was an integral part of the Single Market. If the European Union would not open its job market, applicants would be left with second-class membership. In spring 2001, following the ‘one-size-fits-all’ approach, the member states agreed on a negotiation mandate offering all applicants the same seven-year transition phase.9 This proposal immediately caused a lot of protest. For instance, tiny Slovenia, with its approximately two million citizens, was infuriated by the European Union’s ‘offer’, since it was clearly motivated by fear of the Polish and not the Slovenian labour force. Today, far more EU citizens work in Slovenia than vice versa. The Czech Republic’s chief negotiator, Pavel Telicka, took a similar stand on the matter: I think the EU has the best possible opportunity to prove how able it is to differentiate between those candidates which are more problematic and those candidates, including the Czech Republic, which are less problematic. I wonder, whether the EU would be able to differentiate in practice. . . . Everyone has to agree that there are differences between the candidates as regards their potential threat for the labour markets in the EU. It is clear that the Czech Republic is not a problem.10 (Uniting Europe, 26 March 2001) Clearly, in line with the example in the EFTA case (where the European Union was afraid to ‘singularize’ Austria on the neutrality policy) the European Union was indeed afraid to point fingers at Poland. To quote an official from the Belgian Presidency:

The EU’s enlargement negotiations 207 The Slovenians are so angry about free movement of workers, but we cannot avoid giving them a transition period, because then we would completely isolate Poland, and that would be politically unacceptable. (Author’s interview) Despite its harsh criticism of the proposal, Hungary was the first country to accept the deal. Immediately afterwards, both the Commission and the Swedish Presidency claimed that the Hungarian agreement was a ‘model’ for others to follow.11 Only a few weeks later, Latvia and Slovakia accepted similar agreements, which, according to the Swedish Presidency, were ‘exactly the same as Hungary’s’.12 Poland’s chief negotiator, however, refused to be pressured into accepting the offer: We have always said that each country has to be treated individually. There was never a common position among the candidate countries. This is not a race. We are talking about serious issues . . . At this stage, the Hungarian proposal is not acceptable for Poland. (Uniting Europe, 2 July 2001) In autumn 2001 the Czechs followed suit. After tough negotiations, the Czechs succeeded in obtaining some minor modifications to the Hungarian model. Immediately – and confirming the ‘one-size-fits-all’ trend – the European Union decided to offer the Czech solution to all countries that had already closed the chapter.13 The Czech chief negotiator, however, was anything but pleased with the deal, which he only accepted on the basis of ‘political realism . . . taking into account that a number of other candidates have already concluded, and the European Union was not ready to move further’ (Uniting Europe, 31 October 2001). In the final days of the Belgian Presidency (and after the Polish elections), Poland also agreed to the ‘Czech model’. Apparently, the new government reached the conclusion that it was unrealistic to obtain more now that so many other countries had signed. In January 2002 the European Commission published its first information note on the all-important chapters that could be solved only in the endgame: agricultural policy, regional policy, and the European Union’s overall budget.14 Although there were once again clear differences among the applicants, especially regarding agriculture, the Commission came forward with one proposal for all. Almost predictably, this triggered a number of protests from the applicants, for instance from the Czech minister of foreign affairs, Jan Kavan: As far as the ability of the candidate countries’ agriculture to absorb restructuring reforms is concerned, the Commission’s evaluation seems too sweeping to us, overlooking the principle of differentiation. In the Czech Republic, 3.2 per cent of the workforce works in agriculture, which is the lowest figure among all the candidate countries.15

208 Lykke Friis Unlike when some of the other heavyweight chapters were up for discussion, some of the applicants (Hungary, Poland, Slovakia and the Czech Republic) decided to coordinate their response. In a common declaration, the countries rejected the Commission’s initial proposal.16 Until then the central and eastern European applicants had followed in the footsteps of the EFTA countries and refrained from coordinating their positions.17 Hence, the European Union was able to embark on a ‘divide-and-rule’ practice, where it used Slovenia to strike the first deal on environment, the Czech Republic on free movement of capital, and Hungary on the free movement of persons. Instead of joining forces, all countries accepted the deal that the first country had struck, hence precluding any kind of package deal: ‘We applicants will accept only the “2+3+2 deal” on the free movement of persons if we get our ten-year transition phase on free movement of capital’. Precisely by not coordinating, the applicants were thus cutting the ground from under the feet of each other. The core reason for the reluctance to coordinate and agree upon common positions is rather ironic: applicants fear that – if they join forces – they will be treated as a group, where one applicant cannot get a better deal than the others. As this chapter has shown, with or without coordination the applicants are largely subjected to one-size-fits-all solutions, whereby it is very difficult for an applicant to ‘break the wave’. To be sure, applicants can get special deals when they are the only ones asking for it. Sweden, for instance, was able to obtain a permanent derogation on the tobacco product moist snuff (snus). However, in all those areas where the majority of applicants are asking for transition phases, it is more difficult to point to clear examples where applicants have obtained a special deal. As a ground rule, countries have argued in vain that they are indeed in a ‘special situation’ and therefore deserve something special.

Conclusion: from bilateral negotiations to nested games The core conclusion of this comparative study is that accession negotiations should not be looked upon as a series of bilateral exercises between the European Union and the various candidate countries. If one takes the European Union’s mantra of individual, bilateral negotiations at face value, one ends up with a distorted picture of accession negotiations. Indeed, if one tries to explain, for instance, Austria’s accession negotiation by looking only at the Austrian table, one misses a lot of the important action. Often the central events that influenced Austria’s accession took place at the Swedish and Finnish negotiation tables. Overall, the parallel accession negotiations should therefore not be conceptualized as individual negotiations, but as a complex system of nested games. Actors – both the applicants and especially the European Union – have a clear interest in linking the various negotiations together. As could be seen from the two case studies, these linkages largely worked in favour of the European Union. From a purely economy-of-scale logic the European Union first closed negotiations with one country and then offered more or less the

The EU’s enlargement negotiations 209 same to the other candidates in the hope that they would all follow suit. The purpose of this strategy was not only to cut through the complexity of the negotiation but also to play the candidates off against each other (‘divide and rule’). In both cases this strategy was highly successful, since applicants realized that it would be difficult to obtain a better deal once the European Union had settled the issue with one applicant. Since the European Union is apparently unwilling to offer different deals to various countries, such a differentiation would have required it to re-open the deal with the first country. Precisely because this was judged unrealistic, countries that had vehemently rejected the first offer later decided to accept it. This strategy of linking negotiations was particularly successful in the endgame of the EFTA enlargement round. At that time, the European Union could also play on the fear of being left behind: ‘We have just closed negotiations with Sweden and Finland, so if you want to join in this round – and not wait for many years – we suggest that you compromise’. These poorly concealed threats were anything but weakened by the fact that the applicants refrained from coordinating their positions. Austria could only watch in despair how its room for manoeuvre was basically undermined in the negotiations, given that Sweden and Finland had both made a number of compromises. All in all, no applicants in either enlargement round were thus truly able to ‘break the wave’. On the contrary, many negotiators found their negotiation position swept away by the other applicants. It should be said, however, that the nested-game character did not work exclusively in favour of the European Union. As can be seen from the Norwegian case, applicants can obtain some concessions in the final hours, simply because the European Union has an institutional interest in avoiding single-round accessions. After all, every enlargement demands that member states embark upon a tiresome ratification process and adjust the European Union’s institutions and budget to the newcomers. Again from an economy of scale logic the European Union therefore has an interest in avoiding piecemeal enlargements. With these conclusions in mind, this chapter points out that more research on the group effect is worth undertaking both from a theoretical and an empirical point of view. How can the various interlinkages, for instance, be incorporated in a theoretical understanding of EU enlargement? In that sense, this chapter only confirms that our theories, both integration and international negotiation theory, have difficulties in explaining the enlargement phenomenon. As pointed out by many scholars, the enlargement phenomenon has indeed been largely ignored by the various theorists (Wallace 2000: 150; Schimmelfennig 1999: 1; Schmitter 1996: 14). In any case, the endgame of eastern enlargement has already delivered a new fascinating case study of possible interlinkages. To what extent did the negotiations at the European Council in Copenhagen confirm our conclusion that it is not possible for a candidate to ‘break the wave’?

Notes 1 The chapter does not cover the entire complex negotiation, but concentrates on the period from 1998 to early 2002.

210 Lykke Friis 2 Lisbon European Council, June 1992. 3 Both countries also argued that special economic aid should be given to ‘Arctic agriculture’. On agriculture, Sweden’s situation was less complex since it had already lowered its prices considerably. 4 Already before the endgame Austria had made an important compromise on the question of foreigners acquiring secondary residences. Instead of a permanent exception, it accepted a transitional arrangement. On 1 March, it was agreed that the arrangement would run for five years. 5 Due to internal disagreement over the institutional question in the European Union (number of votes contributing to a blocking minority) it was not until 27 March 1994 that the negotiations with all applicants were actually signed. 6 It should be said, however, that the EU countries themselves created a linkage between two important chapters – if only by treating them simultaneously. First, EU member states indicated that they preferred to have a breathing space before they opened their own job markets to central and eastern Europeans. Then, member states specified that they could accept that central and eastern Europe barred EU citizens from buying, for example, vacation homes and agricultural land in central and eastern Europe for some years. 7 A number of countries (Estonia, Lithuania, Latvia and Slovenia) had already closed the chapter without transition phases, seeing that the necessary EU legislation had been introduced in the wake of the Europe Agreement and the economic reform process. 8 In March 2002, during the Spanish Presidency, Poland finally closed the chapter on capital. At first sight it actually managed to obtain a deal which was substantially better than the other countries – a 12-year transition period on the acquisition of farmland. However, on closer scrutiny the difference is diminished. Farmers from EU countries who have leased land can thus buy the land after a period of respectively seven or three years lease – depending on whether the land is located in the western or eastern part of Poland (Uniting Europe, 27 March 2002). The very fact that Poland managed to get a different, and to some extent better, deal seems to indicate that power matters. After all, Poland is by far the biggest applicant and is also the only candidate where two highly EU-sceptical agricultural parties have made it into the Parliament (League of Polish Families and Self Defence). Since a third agricultural party (PSL) was part of the coalition government in Poland, it was indeed vital for Poland to obtain a somewhat better deal. After Poland’s deal several countries toyed with the idea of re-opening the chapter, but refrained. 9 The offer was very complex and was coined the ‘2+3+2 proposal’, since it made room for several review periods when member states would analyse whether the transition period should still be maintained. 10 The Hungarian chief negotiator Juhasz argued along similar lines: ‘We are asking for an individual approach. . . . This means that the EU should examine whether any threat is coming specifically from Hungary or not. To answer this question, the EU needs to do individual examination. This is in line with EU rules’ (Uniting Europe, 12 February 2001). 11 Uniting Europe, 18 June 2001. Cyprus and Malta were not ‘offered’ transitional arrangements, given that EU members were not afraid of an influx from these countries. After all, the wage level in these countries is equivalent to that of the European Union. 12 Uniting Europe, 2 July 2001. 13 The Czechs obtained the possibility to limit the inflow of labour from other new member states during the period when old member states use a transitional period towards the Czechs. The Czech motive was that, for instance, Polish labour would consider moving to the Czech Republic as long as the German labour market was closed. 14 European Commission, ‘Commission Offers a Fair and Solid Approach for Financing EU Enlargement’, Press Release, IP/02/170, 30 January, 2002. 15 Speech at the European Policy Centre, Brussels, 22 February 2002.

The EU’s enlargement negotiations 211 16 Statement by the Prime Ministers of the Visegrad countries, published by the Presidency in Office, 15 February 2002. 17 In the EFTA wave, countries preferred to embark only on a very loose coordination for fear that their accession might be delayed by difficulties in the negotiations with others. To quote Avery (1995: 12): ‘“Nordic solidarity” was seldom much in evidence.’

References Avery, G. (1995) ‘The Commission’s Perspective on the EFTA Accession Negotiations’, SEI Working Paper (12), Sussex: European Institute, Sussex University. Midgaard, K. and Underdal, A. (1977) ‘Multiparty Conferences’, in D. Druckman (ed.) Negotiations: Social-Psychological Perspectives, London: Sage. Presidency Report to the European Council on Enlargement of the Union and Agenda 2000, Brussels, Annex, The General Negotiating Framework. Available online at http://www.europarl. eu.int/enlargement/cu/agreements/101297c_en.htm (accessed December 1997). Preston, C. (1997) Enlargement and Integration in the European Union, London: Routledge. Schimmelfennig, F. (1999) ‘The Double Puzzle of EU Enlargement: Liberal Norms, Rhetorical Action, and the Decision to Expand to the East’, ARENA Working Papers, WP 99/15. Schimmelfennig, F. (2000) ‘The Enlargement of European Regional Organizations: Questions, Theories, Hypotheses, and the State of Research’, paper presented at the Workshop Governance by Enlargement, Darmstadt, June 2000. Schmitter, P. (1996) ‘Examining the Present Euro-Polity with the Help of Past Theories’, in G. Marks et al. (eds) Governance in the European Union, London: Sage. Uniting Europe, 12 February 2001, Agence Europe. Uniting Europe, 26 March 2001, Agence Europe. Uniting Europe, 18 June 2001, Agence Europe. Uniting Europe, 2 July 2001, Agence Europe. Uniting Europe, 31 October 2001, Agence Europe. Uniting Europe, 27 March 2002, Agence Europe. Wallace, H. (2000) ‘EU Enlargement: A Neglected Subject’, in M. G. Cowles and M. Smith (eds) The State of the European Union: Risks, Reform, Resistance, and Revival, Oxford: Oxford University Press. Winham, G. R. (1977) ‘Negotiation as a Management Process’, World Politics, 30 (1): 87– 114. Zartmann, I. William (ed.) (1994) International Multilateral Negotiation: Approaches to the Management of Complexity, San Francisco: Jossey-Bass.

14 Conclusion Ole Elgström and Christer Jönsson

Characteristics of EU negotiations The EU negotiation system is extremely complex, involving a large number and varieties of issues, actors and roles. Despite the multi-faceted nature of these layered and heterogeneous negotiations, it is still possible to distinguish some common features and prominent patterns. In a sense, the EU negotiation system is sui generis: while some of the traits presented below can be found in other international or national settings, the overall aggregation of shared features seems to be unique, giving EU negotiations a particular flavour (cf. Elgström and Smith 2000). From another perspective, the EU system of negotiation may be claimed to represent a model for what future international negotiations might look like, in a world chararacterized by increasing interdependence and democratization (cf. Giddens 1998). The European Union constitutes, according to this view, a laboratory where experiments with multi-level governance negotiation processes are taking place, and the traits described below represent a structured attempt to pinpoint the peculiarities of such a system. First, EU negotiations are permanent, linked and continuous. It is difficult to define when a particular negotiation starts or ends. Different negotiations are interlinked or overlap. The outcome of one negotiation process creates a new bargaining situation and becomes the starting point for new negotiations, often involving the same actors. Phenomena like path-dependence, lock-in and institutionalization become pervasive (Laffan 2000). The embeddedness of EU decision making also limits the freedom of maneouvre of participating actors. The option to leave a negotiation, to use the exit alternative, is often not available, in effect transforming the ‘three-fold choice’ (Iklé 1964) of negotiators (to accept an offer, to try to improve the outcome or to abandon the negotiations) to a ‘two-fold’ variety. This has important effects on strategy and tactics. From another perspective, the possibility of close international cooperation increases with the length of the shadow of the future. If actors know that their future interaction will last for a very long time, the potential long-term gains from cooperation increases and becomes much more certain, making reciprocity stable (Axelrod 1984: 173). This increases the chances for diffuse reciprocity, meaning that concessions in a bargaining exchange do not have to be made at the same time, and do not necessarily have to be absolutely equivalent (Keohane 1986).

Conclusion

213

Expectations of this kind enormously facilitate the settlement of complex negotiations and tend to encourage integrative negotiation behaviour (Elgström and Jönsson 2000). Several chapters in this volume stress this fundamental set of qualities and point to a dominating problem-solving mode of negotiation in the European Union. A defining characteristic of the EU negotiation milieu, also speaking in favour of an integrative, problem-solving mode of negotiations, is the existing community of values. Compared to other international negotiations, the EU member states are quite homogeneous in terms of shared norms and world-views. This framework of collectively held general values does not preclude the simultaneous existence of different national and, not least, administrative cultures, which affect negotiating styles and approaches to procedure. Nevertheless, the shared values create a basis for a European identity, however fragile, and helps to create a positive environment for successful negotiations. At the same time, the value-based nature of the European Union provides an impetus for efforts to export their common ideals, thus encouraging it to become a norm entrepreneur (Elgström, Chapter 12, this volume). A second distinguishing feature of EU negotiations is their high degree of institutionalization. They are embedded in a rich framework of formal and informal norms and rules. Stable patterns of practices have developed over time and resulted in a high degree of predictability. Response patterns become almost automatic and taken-for-granted. Norms that govern procedures and even substance have been created. A negotiation framework that specifies when, how and where negotiations are to take place has emerged through a process of meta-negotiations (negotiations concerning the rules of the game). In this volume, the momentous effects of decision-making rules and procedures have been highlighted, as have the informal norms that guide actor behaviour (Bjurulf and Elgström, Chapter 4, this volume). We have also underlined the importance of the organizational manifestations of EU institutionalization. Supranational actors have been shown to play substantial roles in all stages of the decision-making process (Bjurulf and Elgström; Tallberg and Jönsson; Sannerstedt; Chapters 4, 6, 7, this volume) and this gives EU negotiations a distinctive quality, with member states much less in control of the process than in ordinary international negotiations. Elsewhere, the power advantages of institutional actors have been elaborated (Tallberg 2003; Beach 2002): they can – it is claimed – use their unique access to information and their ability to shape the agenda in order to influence negotiated outcomes. A third characteristic factor is the curious blend of informal and formal processes that constitute EU negotiations. We have in this volume emphasized the major importance of informal networks. EU networks typically include both state and non-state actors, and supranational institutions carry out important, boundaryspanning functions in these multi-layered structures (Jönsson and Strömvik, Chapter 2, this volume). Our attention has focused on the contacts, ties and connections between actors, apart from the often hierachical formal organization. We also like to note the central role played by personal relationships in networks that evolve through frequent interaction. In networks, which often develop around formal organizational units, interpersonal links tend to become extremely

214 Ole Elgström and Christer Jönsson important. This goes for networks on all levels; from those centred around expert committees up to the European Council. A club atmosphere seems to emerge where network members share the same world-view and the same problems, and frame policy issues in the same way. One may even speak of the emergence of a particular EU negotiation culture, where participants resort to ‘Eurospeak’, a form of communication that marks their identity as members of the right networks. New members are socialized into this system, while the problems of outsiders to gain access to, and to understand, the system is accentuated. High-ranking politicians testify to the relief and immense gratification they experience when they meet with colleagues and find they all share the same domestic problems and the same pressures. That such a negotiation climate has an impact on processes and outcomes seems inevitable. Obviously, informal networks coexist with formal negotiating bodies. There is a constantly fluctuating boundary between formal and informal negotiations. The EU negotiation system is therefore beset by coordination and synchronization problems, and need ‘negotiations about negotiations’ more than traditional forms of international bargaining (Elgström and Smith 2000). There is also a tendency for informal networks to be transformed, often on the initiative of the Commission, into formalized bodies. One aspect of EU negotiations that informs both formal and informal processes is their highly legalistic nature. The fact that EU decision making occurs in the ‘shadow of the court’ means that all decisions have to be grounded in EU treaties and refer to appropriate legal paragraphs. All law-creating negotiations tend to be formalistic and focused on formulations and legal precedents. EU negotiations therefore tend to be text negotiations where privilege is given to legal specialists, often to be found in the supranational institutions, and to their expertise in EU law (cf. Cede 2004). Fourth, EU negotiations are – in contrast to other international negotiations but like national decision making – shaped by the existence of an elaborate implementation machinery. The Commission has a duty to oversee the implementation of EU law, while national authorities still carry most of the actual burden of transforming law into practice. All these processes take place in the shadow of the European Court. Though implementation is in no way always guaranteed – and supranational actors often have to engage in compliance bargaining with recalcitrant member states (see Tallberg and Jönsson, Chapter 6, this volume) – there is a high probability that decisions will actually become reality. This fact has an impact on EU negotiations. On the one hand, it has been argued that actors that expect to be tightly bound for a long time by a new agreement are more prone to be careful and see to it that the settlement is not in conflict with important interests. Therefore, they fight vigorously for their positions (Fearon 1998). On the other hand, actors also know that if other actors do make concessions, these are likely to become effectual. Agreements are likely to be kept. This would highly increase an actor’s willingness to reciprocate, as the level of uncertainty would diminish, thus producing an integrative negotiation climate. We believe the effects of the second mechanism to dominate in the European Union context, and that the promise of faithful implementation therefore encourages effective negotiations.

Conclusion

215

Varieties in EU internal negotiations Despite the many common traits that distinguish negotiations in the European Union from international multilateral negotiations in general, an equally prominent feature of EU negotiations is their variability. As noted in the introduction to this volume, all EU negotiation processes do not look alike. They differ, inter alia, in the extent to which informality and networks dominate the process, in the role of institutional actors and in what is the dominant mode of negotiation. Some negotiations are controlled by member states and are permeated by competitive bargaining. Others are carried out in informal networks, including state officials, institutional actors and NGO representatives, and more resemble problem-solving exercises. The contributions to this volume have highlighted several factors that may explain this confusing variation. In this conclusion, we consider a few of these: variation over time, stage of decision making, type of issue area and level of politicization. Variation over time The European Union is a dynamic entity that is constantly reconstructed, both by formal treaty change and by informal evolution in practices and norms. Institutional transformations are a result of the ever ongoing negotiations that characterize the EU system, while at the same time constituting a framework for, and a constraint on, EU negotiations (cf. Bjurulf and Elgström, Chapter 4, this volume). Another way of expressing this is to emphasize learning and institutionalization as crucial processes that form and change the European Union over time. Actors learn from their experiences. For example, procedures that result in time-consuming quarrels and in non-decision are abandoned in favour of more efficient decision-making modes. Contextual developments may strengthen the need for procedural change. At the same time, modes of negotiation, once adopted, tend to be institutionalized and become taken-for-granted parts of institutional life. A dominant policy-making style therefore tends to be continually reinforced and may even expand in its decision-making environment. In a previous publication, we noted that an integrative, problem-solving mode of negotiation dominates EU negotiations (Elgström and Jönsson 2000). According to Hayes-Renshaw and Wallace: Council negotiations have over the history of the EC been subject to a push/ pull between more integrative and more confrontational modes. In the mid1980s the more integrative mode gained ascendancy. It is hard otherwise to account for the speed and the ease with which both the principles and the details of the SEA [Single European Act] were agreed. (Hayes-Renshaw and Wallace 1997: 259) We argue that the gradual turn to a more consensus-oriented approach, which seems to have continued to spread within the EU machinery, was a result of

216 Ole Elgström and Christer Jönsson learning, institutionalization and environmental pressures. Traditional bargaining led to protracted negotiations and to a lack of policy innovation. The normative consensus around the common market and the Single European Act made the shift in negotiation mode possible. The success of the SEA process consolidated and institutionalized problem solving as an element in the ‘community method’ of decision making. Subsequent expansions of the scope of the First Pillar procedures have incresed the use of integrative negotiations. In these ways, the European Union is changing itself into a problem-solving apparatus (while still retaining important elements of tough bargaining; see below). Likewise, the functions of EU institutions, as well as of non-governmental actors, have evolved and been clarified over time. For example, the roles of the Council Presidency and the Commission as partly competing agenda setters and mediators in EU negotiations have been chiselled out over time by changes in functional needs, by external developments and by institutionalization processes (Bjurulf and Elgström, Chapter 4, this volume; Elgström 2003; Tallberg 2004). Strong networks have appeared in many policy areas: many of these involve supranational bodies in linking-pin roles; some have been formalized and transformed into Commission-led committees ( Jönsson and Strömvik, Chapter 2, this volume). Institutionalization processes and path-dependency, giving advantages to pre-existing networks and to a certain way of conceptualizing and problematizing an issue area, tend, argue Rosén and Jerneck (Chapter 5, this volume), to stifle efforts to reform the Union. To break up embedded habits, dramatic crises or major external pressure may be needed. In times of crisis and uncertainty, it is easier for competing advocacy coalitions to be heard, giving them the chance to pressure for novel ways to frame the issue at hand, and paving the way for alternative networks to enter the scene. Stage of decision making In the literature on EU decision making, it is sometimes argued that different theoretical approaches are more or less appropriate when analyzing different stages of the process. According to Jeremy Richardson, an epistemic community approach is best suited for understanding agenda setting, a network approach for analyzing policy formulation, an institutional approach for decision making, and an interorganiztional approach for implementation (Richardson 1996: 5). We have in this volume observed both consistency and variety in the phases of the policy-making process. Networks are ubiquitous; we find them in the preparatory stages, often linked to existing expert committees and working parties, but also during decison making and implementation. However, networks seem to be most important in the earliest stages of the process, especially in the agenda-setting phase, where advocacy coalitions compete to advance their particular framing of the problem at hand. Later stages of decision making tend to become increasingly formalized and intergovernmental. Institutional actors perform important functions during all stages of EU policy processes. For example, the agenda-setting role of the Commission in early phases

Conclusion

217

of EU decision making has been highlighted in existing research (Pollack 1997; Schmidt 2000), as has the management and brokerage roles of the Presidency (Elgström 2003) and the Council secretariat (Beach 2002) in the decision-making stage of the process. In this volume, the Commission’s position during the implementation phase as a third-party prosecutor within the formal framework of the treaty’s sanctioning procedures has been brought to the fore. Compliance bargaining, Jönsson and Tallberg propose, is in essence a bilateral process between the Commission and member states, taking place in the shadow of the law. While a common negotiating culture, characterized by cooperative problemsolving mode and loose coalitions, can be claimed to dominate EU decision making, variation does exist (Sannerstedt, Chapter 7, this volume; cf. Elgström and Jönsson 2000: 697–9). Initial deliberation in expert groups is heavily imbued with a spirit of expertise. The aim is to find substantive solutions that are sustainable, and integrative agreements are often found. In brief, the technical nature of negotiations is very strong at this early stage. Council working groups – in the decisionmaking phase – are much more politicized. National interests are strongly and vigorously defended, which does not preclude the existence of a strong spirit of cooperation. At this stage, problem solving is closely connected to efforts to find solutions that do not go against the strong interests of others. Finally, implementation committees (the comitology) are also characterized by a mode of negotiation where finding ‘the best solution’ is put at a premium. At this stage, negotiations are often heavily influenced by the Commission and controversies between competing national interets are rare. The problem-solving, consensual and management-oriented negotiation style of most EU negotiations has led some observers to describe EU negotiations as instances of ‘communicative action’, and as examples of deliberative, democratic processes (Joerges and Neyer 1997; Puetter 2001; Niemann 2002), where the goal is to reach a ‘reasoned understanding’ and where preferences may change by means of persuasion and good arguments. The EU decision-making milieu is claimed to demonstrate many parallels to the Habermasian ideals of shared values, recognition of equality and equal access to deliberative arenas: the concept of communicative action may generally add to our understanding of EU negotiations, given the fact that arguably the most important precondition for communicative action, the existence of a shared lifeworld, is particularly well developed in the EU, given the dense patterns of interaction, institutionalisation and socialisation in the Union. (Niemann 2002: 39–40) In our view, it has to be borne in mind that all real-world negotiations include aspects of distributive as well as integrative behaviour (Hopman 1996; Sebenius 1992). Even when actors try to find a jointly beneficial solution, they at the same time try to maximize their own benefits. Bargaining and problem solving represent ends of a continuum between which actual negotiations occur, with their place largely determined by contextual factors. EU negotiation may indeed often bear

218 Ole Elgström and Christer Jönsson more similarities to deliberative processes than most other international settings, but in some circumstances – as we shall see below – competitive bargaining is still the rule. Furthermore, all actors are not represented in EU networks, information is not evenly spread and power resources (and not only good arguments) are regulary used to influence co-negotiators. Future research is well-advised to focus more on the intricate interplay between problem solving and bargaining and on the institutional determinants of negotiation behaviour. Type of issue area and level of politicization Policy types, levels of politicization and modes of negotiation are sometimes brought together in studies of the European Union. John Peterson (1995: 70) recommends a research strategy that ‘encourages us to look for identifiable patterns of decision making at different levels of analysis’, and introduces a distinction between ‘history-making’, ‘policy-setting’ and ‘policy-shaping’ decisions, linked to, respectively, a ‘distinctly political’, a ‘political and technocratic’ and a ‘technocratic and consensual’ rationality (Peterson and Bomberg 1999: 9). From another perspective, different types of policies have been supposed to generate distinctive styles of negotiation (Pollack 1994: 96). For example, Lowi’s classification of policy types into constituent (in which the basic rules of the system are under consideration), redistributive (in which a transfer of financial resources from some actors to others are involved), distributive (in which Community funds are allocated within sectors) and regulatory (in which member states agree to adopt common regulations) policies, can be used to compare modes of negotiation over policy areas. Previous research has thus hypothesized that non-politicized, regulative and distributive issues should be characterized by integrative, problem-solving negotiations, while redistributive and constituent issue areas, where high levels of politicization make protection of national self-interests a major goal, should lead us to expect distributive, tough bargaining behaviour (cf. Elgström and Jönsson 2000: 701). The contributions to this volume tend to support these hypotheses. Agricultural reform negotiations, which may be expected to have highly redistributive effects, turn out to be predominantly competitive. Those aspects of transparency negotiations that involved constitutional consequences were also among the most acrimonious. Negotiations in expert groups and comitology committees that mostly deal with technical, non-politicized matters are characterized by joint problem solving, while working groups under the Council, that more frequently handle politicized issues, are more prone to experience defense of national interests. One interesting and rather surprising finding in this volume is that networks flourish and exert influence not only in low politics issues, but also in matters of security, a vital national concern. The role of informal networks in the ‘least likely’ case of the highly sensitive decision to send an EU military mission to Congo indicates that networks are omnipresent in the EU system ( Jönsson and Strömvik, Chapter 2, this volume).

Conclusion

219

Variation in EU external negotiation behaviour The chapters in the second part of this volume illuminate important varities in the approaches and roles of the European Union as an actor in international negotiations. Not only does the European Union differ in what constitutes and who represents the Union in different international arenas (Reiter, Chapter 10, this volume), but there is also a remarkable diversity in European Union attitudes and strategy. In some negotiation contexts, the European Union is seen as an introspective, inflexible actor with traditional military and commercial great power aspirations. In others, the Union is perceived as ‘new kind of great power’, as a civil power with normative ambitions. It is sometimes described as a foot-dragger, at other times as a progressive leader (Elgström and Strömvik, Chapter 8, this volume). Four different types of explanations to this rich variation can be found in the contributions to this book: 1) the peculiar characteristics of the European Union’s internal institutional set-up; 2) contextual factors in terms of asymmetries in power or orientation to change; 3) the characteristics of the organizational arena in which the negotiations take place; 4) the history and identities associated with a particular negotiation. In the literature on the European Union as an external actor it is often argued that variations in the actor capacity of the European Union explain the different degrees of impact that the Union has in different policy areas. The argument is that the peculiar legal-institutional structure of the Union often leads to non-optimal negotiation behaviour. In arenas where the European Union, because of historically determined structural features linked to the competition between intergovernmentalism and supranationality, demonstrates a diffuse divison of labour, unclear competencies and a lack of distinct and homogeneous objectives, the European Union is claimed to be less influential than in arenas where it exhibits clear goals, concerted ambitions and distinct role assignments (cf. Bretherton and Vogler 1999; Ginsberg 2001; Whitman 1998; Zielonka 1998). Reiter (Chapter 10, this volume) argues that the internal distribution of competencies between member states and the Commission provides a necessary, although insufficient, explanation of variations in EU actorness. Elgström and Strömvik (Chapter 8, this volume) maintain that the institutional complexity of the European Union creates problems of synchronization and coordination, and that it often lends a conservative bias to EU positions. What they all have in common is the attribution of explanatory value to the internal structures of the European Union. The second explanation refers to broad contextual factors. Elgström and Strömvik argue that the European Union is most likely to be a ‘difficult’, inflexible actor when it is defending the status quo and simultaneously has an advantage of power. In situations of power symmetry and/or when the European Union is change oriented, the Union is hypothesized to be more likely to demonstrate leadership and flexibility. Sophie Meunier (2000) has linked negotiation context to institutional design, by proposing that EU-specific, institutional factors (like restricted delegation and unanimity voting) play a bigger role when the European Union acts as a defender of present conditions.

220 Ole Elgström and Christer Jönsson Third, Reiter (Chapter 10, this volume) also gives the external environment a prominent place in his explanation of EU actorness. In his opinion, the roles of the Commission and member states and internal policy-coordinating processes primarily reflect the constitutive elements of the issue-specific international organization in which the negotiations take place. According to Reiter, EU actorness is fundamentally an expression of a semi-rational design where the member states adapt to its environment in order to enhance their benefits from external action. The fourth and final type of explanation is offered by Elgström (Chapter 12, this volume). In his account of the Cotonou negotiations, he puts a premium on norms and actor identities, arguing that a historically based EU partnership identity in relation to the ex-colonies of its member states heavily influenced the European Union’s mode of negotiation in the post-Lomé process. Extending his argument, it might be proposed that negotiation behaviour may vary with EU identity in different settings. According to this explanatory approach, role conceptions and role expectations may account for variations in EU role behaviour. The variations in EU external behaviour described in the introduction to this section indicate that the European Union is plagued by role ambiguity. In the words of Laffan et al. (2000: 180), ‘there is an historic ambiguity of the EU’s international role’. This might of course reflect a genuine vacillation between different role conceptions, and an uncertainty as regards what role it wants to – or can – play in a world shaped by constant change. It might also, however, mirror the differences pointed to and developed above: EU roles may vary according to internal structures, external contexts or history-bound identities.

Consequences for democratic governance The pattern of EU member states constantly involved in negotiations with a variety of actors in informal networks raises thorny questions about democratic values. Networks link elites; presence or absence in networks is therefore a critical source of power and influence. One might argue that the large number of access points in informal EU networks makes for more opportunities for enterprising organizations and individuals to participate in the policy process than in many democratic states. And, as we have seen, the consensual and problem-solving nature of EU negotiations has led some Habermasian theorists to regard the European Union as an exemplar of deliberative democracy. Yet the European Union’s lack of popular legitimacy, its ‘democratic deficit’, has been a perennial stumbling-block. From a democratic viewpoint, the nature of the EU policy process – a variety of institutionalized negotiations in networks, as described in this volume – is problematic in several respects. Accountability is the touchstone of representative democracy. The government is accountable for its actions before the people in elections at regular intervals. For citizens to be able to make informed judgments of accountability two fundamental requirements must be fulfilled: transparency, in the sense that the voter should be fully informed as to what the government has in fact done during its term, and responsibility, in the sense that it should be clear to the voter who is responsible

Conclusion

221

for what action. The picture of EU negotiations, drawn above, is problematic on both accounts. Negotiations in EU networks rarely take place in the open. The accepted wisdom is that effective negotiations require confidentiality and a measure of secrecy, and one of the advantages of informal network structures is precisely their limited visibility. The lack of transparency, associated with negotiations in networks, is therefore a major problem in terms of government accountability before the voters; the heated debate about transparency in the European Union testifies to that (cf. Bjurulf and Elgström, Chapter 4, this volume). Added to the general problem of transparency is the difficulty of ascribing unequivocal responsibility to specific actors for decisions reached through negotiations in networks. Negotiated outcomes imply shared responsibility among a number of actors who have all had to modify their initial positions. Unpopular aspects of the outcome can always be blamed on other actors – ultimately on the ‘bureaucracy in Brussels’. In short, as agency and power become more anonymous, the difficulty of pinning responsibility to individual actors is magnified. The traditional image of individual governments being held responsible for discrete policy measures, affecting only citizens of its own state, is eroded by multi-level governance through negotiations in networks. ‘If the efficacy of the system of representative democracy is being strained and eroded in the face of regional and global interconnectedness, what mechanisms could ensure accountability in the new international order?’ (Held 1995: 138). Or, as the question is often framed in the European Union: How can multi-level governance assume the character of multi-level democracy? These fundamental questions remain unanswered. Another problematic concept is representation. It can be understood in terms of ‘acting for’ and ‘standing for’ others (Pitkin 1972). Representatives, in other words, are agents of principals (the people in representative democracy) as well as symbols of a larger community. As far as EU networks are concerned, it is questionable whether participants in policy making can be considered representative. In fact, informal networks often emerge as a way to avoid thorny questions of representation. Formal structures for coordination typically limit the number of organizations represented, while informal coordination facilitates the participation of a flexible number of organizations (cf. Chisholm 1989). Representativeness, of course, cannot be determined objectively, but is a matter of perception. Thus: The problem is not that the Brussels arena is a ‘closed shop’, but rather that the new political space in which national and European institutions and groups compete is composed of complex and differentiated networks. The fundamental issue is then how far these actors, including intermediary groups, are perceived as real ‘representatives’ by, of, and for the citizens. (Wessels 1997: 38) Yet another principal democratic problem concerns the delimitation of a demos, the people entitled to suffrage and other democratic rights and responsibilities. Nationalism laid a foundation for modern democracy, in this regard, by defining

222 Ole Elgström and Christer Jönsson the demos in terms of the nation which, in turn, was to coincide with the state. The demos of modern democratic practice thus became territorially bound. In the light of negotiations and power sharing between territorial states and non-territorial entities in the European Union, this seemingly firm fundament begins to crumble. The traditional understanding of democracy presupposes perfect coincidence between the demos with voting rights and the people affected by the resulting policy measures. This forges the link between voting and accountability judgments. In the era of globalization, however, ‘territorial systems of accountability no longer necessarily coincide with the spatial reach of sites of power’ (McGrew 1997: 12– 13). This applies, a fortiori, to multi-level governance in the European Union, which opens up new ‘possibilities of governance independent of the existence of a central authority and beyond the territorial congruence of those who govern with those who are subject to governance’ ( Jachtenfuchs and Kohler-Koch 1995: 5). In short, the peculiar EU policy process, with its reliance on negotiations and informal networks, raises the fundamental question of who is accountable to whom. Those who wield power and carry responsibility tend to be anonymous and not necessarily representative, and the policy process remains opaque, at the same time as a truly European demos, able and willing to hold the makers of European policies accountable, is still missing. Another relevant question concerns whether efficiency is bought at the expense of democratic values in the European Union. Does the fact that negotiations in network produce substantive results that most intergovernmental organizations cannot match justify the ‘democratic deficit’? Ultimately, it is a question of legitimacy. Do citizens in member states consider the outcome of the EU process valuable enough to legitimize the Union, or does the lack of participation, transparency and representativeness imply lack of legitimacy in their eyes?

Implications of enlargement Will negotiations in networks, the way we know them today, survive enlargement? One aspect of the question has to do with sheer numbers. Even though networks are flexible in terms of entries and exits, the connectivity or density is dependent on a limited number of participants. Connectivity or density refers to the extent to which participants (nodes) are connected with each other. Complete connectivity – when each node is connected to all other nodes – can exist only in smaller networks. It is doubtful that it exists even in the existing, comparatively dense EU networks. However, with increasing numbers of participants, it is probable that a few nodes will become more central in terms of being reachable from, and able to reach, most other organizations in the network. In other words, a greater element of hierarchy (more influence for the large member states at the expense of the smaller ones) is to be expected as a result of enlargement. The informality and atmosphere of trust that characterize negotiation in EU networks rest on a shared political culture. Experience from former waves of enlargement clearly indicates that it takes quite a while for newcomers to adjust to the peculiar EU culture. Moreover, it could be argued that the history of the

Conclusion

223

candidate states from eastern and central Europe has not prepared them for a political culture based on informal processes of continuous negotiation and networking. Nor has issue-based expertise been entrusted with the task of negotiating international agreements in these countries. The NGO community, in addition, is comparatively underdeveloped in these formerly socialist, centralized states. One can wonder whether the relative homogeneity amidst diversity that has hitherto characterized the informal negotiation culture in the European Union will survive the addition of a large number of new members with diverse backgrounds and political cultures. At any rate, we can expect a lengthy adjustment process. In sum, institutionalized, network-based negotiations in the European Union, as we have seen, have been far from unproblematic in the past and face new challenges in the years ahead. The future development of policy-making processes in the European Union requires close attention by scholars and practitioners alike.

References Axelrod, R. (1984) The Evolution of Cooperation, New York: Basic Books. Beach, D. (2002) Bringing Negotiations Back into the Study of European Integration, University of Southern Denmark: Department of Political Science, dissertation. Bretherton, C. and Vogler, J. (1999) The European Union as a Global Actor, London: Routledge. Cede, F. (2004) ‘The European Union as Epitome of a Negotiation Process’, in P. Meerts and F. Cede (eds) Negotiating European Union, unpublished manuscript. Chisholm, D. (1989) Coordination Without Hierarchy: Informal Structures in Multiorganizational Systems, Berkeley: University of California Press. Elgström, O. (ed.) (2003) Comparing European Union Council Presidencies, London: Routledge. Elgström, O. and Jönsson, C. (2000) ‘Negotiation in the European Union: Bargaining or Problem-Solving?’, Journal of European Public Policy, 7 (5): 684–704. Elgström, O. and Smith, M. (2000) ‘Introduction: Negotiation and Policy-Making in the European Union – Processes, System and Order’, Journal of European Public Policy, 7 (5): 673–83. Fearon, J. D. (1998) ‘Bargaining, Enforcement, and International Cooperation’, International Organization, 52 (2): 269–305. Giddens, A. (1998) Interview with Anthony Giddens, ‘Third Way’s the Charm’, Newsweek, 28 September 1998: 50. Ginsberg, R. (2001) The European Union in International Politics: Baptism by Fire, Lanham: Rowman & Littlefield. Hayes-Renshaw, F. and Wallace, H. (1997) The Council of Ministers, New York: St. Martin’s Press. Held, D. (1995) Democracy and the Global Order, Stanford: Stanford University Press. Hopmann, P. T. (1996) The Negotiation Process and the Resolution of International Conflicts, Columbia: University of South Carolina Press. Iklé, F. C. (1964) How Nations Negotiate, New York: Praeger. Jachtenfuchs, M. and Kohler-Koch, B. (1995) ‘The Transformation of Governance in the European Union’, paper prepared for Fourth Biennial Conference of the European Community Studies Association, Charleston, SC, 11–14 May.

224 Ole Elgström and Christer Jönsson Joerges, C. and Neyer, J. (1997) ‘Transforming Strategic Interaction into Deliberative Problem-Solving: European Comitology in the Foodstuffs Sector’, European Law Review Journal, 3: 273–99. Keohane, R. (1986) ‘Reciprocity in International Relations’, International Organization, 40: 1–27. Laffan, B. (2000) ‘The Big Budgetary Bargains: From Negotiation to Authority’, Journal of European Public Policy, 7 (5): 725–43. Laffan, B., O’Donnell, R. and Smith, M. (2000) Europe’s Experimental Union: Rethinking Integration, London: Routledge. McGrew, A. (1997) ‘Globalization and Territorial Democracy: An Introduction’, in A. McGrew (ed.) The Transformation of Democracy?, Cambridge: Polity Press. Meunier, S. (2000) ‘What Single Voice? European Institutions and EU–US Trade Negotiations’, International Organization, 54 (1): 103–35. Niemann, A. (2002) ‘Communicative Action, the Article 113 Committee and the WTO Agreement on Basic Telecommunication Services’, Dresdner Arbeitspapiere Internationale Beziehungen Nr. 6. Peterson, J. (1995) ‘Decision-Making in the European Union: Towards a Framework for Analysis’, Journal of European Public Policy, 2 (1): 69–93. Peterson, J. and Bomberg, E. (1999) Decision-Making in the European Union, Houndmills: Macmillan. Pitkin, H. F. (1972) The Concept of Representation, Berkeley: University of California Press. Pollack, M. (1994) ‘Creeping Competence: The Expanding Agenda of the European Community’, Journal of Public Policy, 14 (2): 95–145. Pollack, M. (1997) ‘Delegation, Agency, and Agenda Setting in the European Community’, International Organization, 51 (1): 99–134. Puetter, U. (2001) ‘An Institutional Framework for Arguing and Deliberation: The Eurogroup as a Generator of Informal Resources’, paper presented at the 4th PanEuropean International Relations Conference, University of Kent, 6–10 September. Richardson, J. (ed.) (1996) European Union: Power and Policy-Making, London and New York: Routledge. Schmidt, S. (2000) ‘Only an Agenda-Setter? The European Commission’s Power over the Council of Ministers’, European Union Politics, 1 (1): 37–61. Sebenius, J. K. (1992) ‘Challenging Conventional Explanations of International Cooperation: Negotiation Analysis and the Case of Epistemic Communities’, International Organization, 46 (1): 323–65. Tallberg, J. (2003) ‘The Power of the Presidency: Brokerage, Efficiency, and Distribution in EU Negotiations’, unpublished paper. Tallberg, J. (2004) ‘The Power of the Chair: Leadership and Negotiation in the European Union’, unpublished book manuscript. Wessels, W. (1997) ‘The Growth and Differentiation of Multi-Level Networks: A Corporatist Mega-Bureaucracy or an Open City?’, in H. Wallace and A. R. Young (eds) Participation and Policy-Making in the European Union, Oxford: Clarendon Press. Whitman, R. (1998) From Civilian Power to Superpower? The International Identity of the Eurpean Union, Houndmills: Macmillan. Zielonka, J. (1998) Explaining Euro-Paralysis: Why Europe is Unable to Act in International Politics, Houndmills: Macmillan.

Index

ACP (African, Caribbean and Pacific countries) 118, 121, 122, 143, 183–97; relations with EU 183–4, 188, 190, 199 Agenda 2000 63, 65–6, 68–76, 204 agenda-setting 50, 125, 127, 145–6, 216 Ajello, A. 25 Amsterdam Treaty 30, 39, 49, 57–8, 150 Annan, K. 24 Aspinwall, M. 47 Audretsch, H. A. H. 83, 90 Australia 159 Austria 67, 69, 107, 134, 193, 201–4, 206, 208–10 bargaining: competitive 39, 99, 102, 215, 218; distributive 99, 175; integrative 99; vs negotiation 2 bargaining power 7, 45, 60, 80, 83–90, 91, 94, 183, 185–6 bargaining situation 7, 81, 183–4, 212; definition of 2 Belgium 49, 53, 56, 59, 92–3 Best Alternative to a Negotiated Agreement (BATNA) 4, 103 Beyers, J. 105 Big 12 Round Table 23 boundary-role occupants 15 Brittan, L. 22, 133, 135–6 Bulgaria 204–5 Bulmer, S. 45 Bush Administration 177, 179 Cairns group 132, 141, 144 Canada 144, 160 capability–expectation gap 120, 187 China 159, 177–8 Chirac, J. 70 Chisholm, D. 16 Christiansen, T. 98 Churchill, W. 139 Claes, W. 16 coalition 2–5, 8, 17, 19, 23, 32, 35, 39, 50,

53, 57, 65–8, 70–2, 75–6, 99, 106–8, 110, 113, 167, 171–2, 174–5, 178, 180, 196, 210, 216–17; building 2, 19, 65, 99, 106–8, 167 co-decision 50, 52, 54–5, 58, 87 comitology 6, 37–8, 97–9, 101–12, 217–18 Commission 3–4, 7, 16–19, 21–4, 26, 35–7, 39–40, 46, 48–58, 60, 63, 65–9, 72–3, 83–94, 99, 105, 108–11, 118, 120, 122, 125–6, 130, 132–45, 149–52, 156–60, 162–3, 179, 187, 190–3, 195–7, 202–3, 207–8, 214, 216–17, 219–20; Development General Directorate 37, 185; DG V 22; and enforcement 80, 82–96; External Relations Directorate 185; networking strategy 17, 22–3 Committee of Permanent Representatives (COREPER) 76, 97, 103 Common Agricultural Policy (CAP) 6–7, 63–78, 130, 149 common foreign and security policy (CFSP) 21, 23–4, 120, 201, 204 Community Actions for the Elderly 21 competencies 120; exclusive 118, 119; mixed 118, 119 compliance bargaining 6, 21, 79–96, 214, 217 compromise 6, 19, 30, 42–3, 49–50, 52, 54–5, 57–9, 64–5, 70–2, 74–6, 84–5, 88, 90–1, 103–5, 112, 120, 124, 138, 145, 164, 192–3, 195–7, 203, 209–10; solutions 52, 85, 88; supranational 6, 64, 71–2, 74–6 conciliation 55–6, 59–60 conditionality 123, 183, 190, 194 Congo, Democratic Republic of: EU military mission 21, 23–6 consensus norm 52, 59 Cotonou agreement 7–8, 118, 183–99, 220 Council of Ministers 4, 45, 138, 186; High Representative of 54, 59, 120; Political and Security Committee (PSC) 24–5; Secretariat 23–6, 217; working groups

226

Index

6, 97–103, 106, 108, 112, 217; see also Presidency of the Council credibility 89, 139, 141, 145, 162, 173, 175, 192 Cyprus 204–5 Czech Republic 204–8 Daugbjerg, C. 71, 74 Davignon, E. 22–3 deadlock see stalemate Dehousse, R. 97, 104, 109 Delors, J. 108 democratic deficit 29, 220–2 Denmark 35, 36, 39, 42, 49, 54, 67, 92–3, 134, 136, 160, 202 Dierickx, G. 105 Doha Development Agenda see World Trade Organization (WTO) Economic and Monetary Union (EMU) 156 enlargement 63, 65–6, 74, 128, 134, 158, 162, 201–5, 207, 209–10, 222; negotiations 8, 118, 121–2, 200 epistemic communities 4, 18 Estonia 204–5 Eurolink Age 22 European Central Bank (ECB) 159–61 European Council 3, 23, 59, 65–6, 69–70, 200–1, 209, 214 European Council of Telecommunications Users Associations (ECTUA) 23 European Court of Justice (ECJ) 2, 53, 59, 80, 84–6, 88–9, 92–4 European Development Fund (EDF) 38, 184–5, 187, 197 European Free Trade Association (EFTA) 8, 200–6, 208–9 European Parliament 18–19, 21, 26, 35, 40, 46, 48–52, 55–9, 60, 83, 87, 110, 137, 187, 191, 193, 210 European Round Table of Industrialists 18 European Union (EU) ‘actorness’ 7, 148–63 European Union (EU) Military Committee 25 expert groups 6, 36–7, 97–106, 108–10, 112, 125, 217–18 Financial Stability Forum (FSF) 151, 156, 159–61, 163 Finland 39, 49, 54, 134, 201–4, 208–9 Finnemore, M. 30–2 Fischler, F. 66–7, 70 France 24, 37, 40, 49, 51–2, 54, 67–73, 78, 92–3, 107, 134, 136–8, 141, 144, 159–61, 189 Friis, L. 123, 187

game theory 2 gender equality 30, 32, 34–8, 41–2, 193 General Agreement on Tariffs and Trade (GATT) 157, 184, 186; Uruguay Round 66, 71, 75, 118, 131–2, 135, 141–3 General Agreement on Trade in Services (GATS) 157 Germany 24, 40, 49, 53, 67–73, 92, 94, 107, 134, 136, 138, 159–61 globalization 124, 133–4, 137–8, 141, 145, 179, 222, 224 governance: global 123–4, 137, 166–7; good 123, 190, 195–7; multi-level 13–14, 16, 19, 149, 212, 221–2 Greece 15, 92–3, 134, 200 Grilli, E. 186 Group of 20 176 G–7 151, 159, 161, 167 Habermas, J. 217, 220 Hall, P. A. 71 Hayes-Renshaw, F. 215 Helms–Burton Act 177, 180 Hill, C. 187 HIV/AIDS 142 Hjelm-Wallén, L. 125 Hong Kong 159 Hungary 204–5, 207–8 Iceland 160 impartiality norm 59, 60 impasse see stalemate India 26, 141, 144 institution 1, 4–7, 13–14, 42, 45–62, 72, 79, 80–2, 84, 87–9, 91, 118, 120, 133, 154, 159, 162–3, 166–7, 170, 175, 181, 188, 209, 213, 221; definition of 4 institutionalism 45–7, 61 institutionalization 1, 5–6, 34, 37–8, 76, 98, 124, 162–3, 166–8, 170, 175, 212–13, 215–16 intergovernmental conferences (IGCs) 91, 94, 150; Nice 2000 150 intergovernmentalism 21, 42, 45, 149, 165–7, 219 Intergroup on Ageing 21–2 International Atomic Energy Agency (IAEA) 82 International Labor Organization (ILO) 56, 138, 144 International Monetary Fund (IMF) 151, 156, 159–61, 163, 185, 189 International Telecommunications Users Group (INTUG) 23 Iran–Libya Solidarity Act (ILSA) 177 Iraqi war 24

Index Ireland 39, 49, 68, 72, 92, 94, 107, 141, 144, 160–1, 198 issue aggregation 2–3 issue disaggregation 2–3 issue networks 4, 27, 73–4 IT Task Force 22 Italy 68, 70, 92–3, 134, 161 Japan 22, 121, 126, 141, 144, 159 Kavan, J. 207 Keck, M. E. 32, 35 Kirchner, E. 98 Kissinger, H. 120 Kohl, H. 68 Kyoto Protocol 177, 187, 180 Laffan, B. 220 Lamy, P. 137, 139, 140–1, 143 Latvia 204–5, 207 leadership 3, 7, 23, 52–3, 65, 72–3, 75–6, 99, 108–10, 118, 121, 123, 125, 130, 132, 135–9, 142, 144–6, 171, 174–5, 219 Legro, J. 34 linkages 6, 8, 64–5, 71–2, 131, 135, 166–7, 175, 200–4, 208–9, 212 linking-pin 5, 18, 22–4, 26, 124, 216; organizations 18, 124; roles 216 Lithuania 204–5 logic of appropriateness 37, 47 Lomé convention 7, 182–6, 189–92, 196–9 lowest common denominator 60, 117, 119 Lowi, T. 64, 218 Lund, G. 57 Luxembourg 49, 92, 94 Maastricht Treaty 30, 38, 49, 89, 130, 156, 191 MacSharry reform 64, 66, 75 Malta 204–5 Mansholt, G. 63 March, J. G. 76 Maurer, A. 105 mediation 57, 65, 72, 75–6, 108–9, 124 mediator 3, 19, 52, 54–5, 57, 59, 65, 73, 108–9, 110, 216 Meunier, S. 121, 219 Mittag, A. 105 Monti, M. 179 Moravcsik, A. 42 multilateralism 170, 172, 179, 180 Murphy, A. 123 negotiation: asymmetric 121, 123, 183, 185; bilateral 65, 69, 87, 131, 169–74, 177, 200, 208; bi-multilateral 7, 164, 169; competitive 31, 40; definition of

227

1–2; distributive 30, 217–18; enlargement 8, 118, 121–2, 200–11; implementation see compliance bargaining; institutionalized 4, 86, 94, 220; interpretative 31, 34, 36, 39; multilateral 2–3, 5, 7, 19, 20, 91, 97, 99, 106, 108, 121, 164, 168, 169–76, 179, 200, 215; polylateral 3; problem-solving 2, 120, 218; redistributive 30; reform 63–78; stages 79; text 30, 33, 36–7, 39, 40–1, 214; vs bargaining 2 negotiation culture 6, 98, 111, 214, 223 negotiation style 97, 99, 101, 213 nested games 208 Netherlands 35–6, 39, 42, 49, 54, 69, 92–3, 107, 134, 136, 159 network 1, 3–5, 7, 13–28, 35, 45, 65, 73–4, 83, 107, 110, 123–7, 160, 162, 166–7, 213–16, 218, 220–3; types of 3–4 non-governmental organizations (NGOs) 18, 20–2, 35, 42, 51, 53, 124, 147, 188, 192–3, 195–6, 215, 223 norm entrepreneur 6, 29, 31, 34–6, 38–9, 40–2, 123 norm negotiations 1, 5–6, 29, 30, 33, 41–2 norm spread 29, 33–4, 42, 188, 193 norms 4–7, 29–43, 45–7, 54, 59–60, 76, 122–3, 127, 175, 179, 183–4, 188–94, 196–7, 211, 213, 215, 220 North American Free Trade Agreement (NAFTA) 81 North Atlantic Treaty Organization (NATO) 24–5, 49, 53–4, 117, 201 Norway 141, 160, 201–4, 209 Olsen, J. P. 76 openness see transparency Organization for Economic Cooperation and Development (OECD) 133, 141, 151, 156–9, 163 package deal 3, 6, 19, 30, 42, 57, 64–5, 72, 140, 171, 204, 208 package meetings 86, 94 partnership 86, 167, 181, 185, 188–9, 190–2, 194–7, 220 path dependency 75–6, 212, 216 Patterson, L. A. 130, 133 Persson, G. 43 Peterson, J. 218 Poland 201, 204–8 policy community 6, 23, 74–5 policy networks 3, 14, 20, 23, 73–4, 166 political culture 64, 222–3 politicization 6, 75–6, 215, 218 Pollack, M. 89 Portugal 92, 134, 186

228

Index

Presidency of the Council 39–40, 43, 47, 51–60, 65, 69, 70, 73, 99, 108–10, 113, 118, 126, 216–17; Belgian 204, 206–7; French 46, 51–4, 58; Greek 25; Swedish 40, 46, 54–5, 57–9, 204–5, 207 principle of non-reciprocity 194, 196 problem-solving 2, 7, 79, 105, 113, 120, 127, 213, 215–18, 220, 223–4 Putnam, R. 130, 133 qualified majority voting 50, 88, 121 Realism 42, 45, 165, 186, 207 reciprocity 136, 139, 184, 191, 193, 196; diffuse 20, 64, 137, 145, 212; specific 20, 137, 145 reform negotiations 6, 63–4, 72, 76, 218 representation 3, 16, 18, 32, 111, 155, 221 Richardson, J. 216 Risse, T. 32 role differentiation 2 Romania 204–6 Rome Treaty see Treaty of Rome Russia 159, 178 Saudi Arabia 159 Schmidt, S. 51 Schneider, G. 47 Schroeder, G. 70 shadow of the future 20, 212 shadow of the law 80, 87–8, 217 shadow of the veto 88, 140 shadow of the vote 2, 58, 80, 88 Sikkink, K. 30–2, 35 Singapore 132–3, 159 Single European Act 215–16 single negotiating texts (SNT) 51, 58 single undertaking 131, 135–6, 144–5 Slovakia 204–8 Slovenia 204–8 Smith, M. 7, 120, 164, 166, 168, 170, 172, 174, 176, 178, 180 soft power 123, 127 Solana, J. 24–5, 54 Spain 56, 59, 92–3, 134, 138, 160–1, 186 stalemate 1, 7, 19, 69, 71, 75, 99, 103–4, 108, 125 Stenelo, L.-G. 64 Sweden 15, 33, 35–6, 39–40, 42, 45, 48–9, 53–9, 67, 69, 98, 100, 107, 134, 136, 138, 160, 200–4, 208–10 Switzerland 141

telecommunications 21–3, 26, 132; deregulation 22 Telicka, P. 206 threats 89, 91, 102, 105, 108, 166, 168, 187, 209 three-level interactive game 130, 133, 137, 145 Trade Related Aspects of Intellectual Property Rights (TRIPS) 157 transgovernmental relations 3 transgovernmentalism 165–7 transnational relations 3, 166 transparency 6, 26, 29, 30, 33–4, 38–43, 45–7, 49–60, 118, 133, 141, 144, 172, 218, 220–2 Treaty of Rome 17, 21, 63, 150 Treaty on European Union see Maastricht Treaty trialogue 55–60 trust 3, 16, 19, 20, 49, 50, 56, 64, 102, 105, 111, 222 two-level game 130, 133, 172, 187, 191 unilateralism 177 United Kingdom 24, 36, 42, 49, 67–8, 72, 92–3, 107, 134, 136, 138, 159, 160 United Nations (UN) 24–6, 31, 117, 185 United Nations Conference on Trade and Development (UNCTAD) 151 United States of America (USA) 24, 75, 108, 121, 126, 130–2, 139, 141–4, 164–81; negotiations with EU 7, 164, 168, 170, 173, 178; relations with EU 164–8, 173–4, 178, 180 voting rules 4, 46, 58, 104, 112, 118, 121, 154, 156, 158–9, 161 Wallace, H. 215 Watson, G. 56–7 Webber, M. 71 Wessels, W. 105 win–win solutions 64–5 working groups 6, 20, 97–103, 105–6, 108–12, 125, 217–18 World Bank 163, 189 World Trade Organization (WTO) 7, 63, 66, 81, 121, 130, 134–6, 151, 157, 167, 183; Cancun meeting 2003 145; Doha Development Agenda 7, 130, 144, 146, 174, 180; Geneva meeting 1998 136; Seattle meeting 1999 133, 137, 141–2, 145