Europeanization and Transnational States
This book investigates what happens to an organized political unit when it be...
11 downloads
616 Views
2MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Europeanization and Transnational States
This book investigates what happens to an organized political unit when it becomes part of a larger entity and, in particular, how increased European integration will transform the members of the European Union towards transnational states. Europeanization and Transnational States provides an extensive comparative survey of the central governments in four Scandinavian countries and analyses the ways in which the European Union has influenced the day-to-day work of their state administrations. It includes coverage of Denmark, a long-standing member of the European Union; Finland and Sweden, countries that became members in 1995 and Norway, a non-member. The book utilizes various theoretical perspectives – such as adaptation to external pressure, strategic choice, path-dependencies and translation – to explain the changes related to increased European integration in central government agencies. It concludes that the consequences of Europeanization can be described as the growth of a transnational administration where identities as well as loyalties are created in processes that transcend the borders of states. Bengt Jacobsson is Professor of Management at So¨derto¨rn University College, Sweden. Per Lægreid is Professor of Administration and Organization Theory and Research Director at the Stein Rokkan Centre for Social Studies, the University of Bergen, Norway. Ove K. Pedersen is Professor of Comparative Politics at the University of Copenhagen, Denmark.
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 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 Widgre´n 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 democratization 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 Europe, Globalization and Sustainable Development Edited by Brian Baxter, John Barry and Richard Dunphy 13 European Union Council Presidencies A comparative perspective Edited by Ole Elgstro¨m 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
Europeanization and Transnational States Comparing Nordic central governments
Bengt Jacobsson, Per Lægreid and Ove K. Pedersen
First published 2004 by Routledge 11 New Fetter Lane, London EC4P 4EE Simultaneously published in the USA and Canada by Routledge 29 West 35th Street, New York, NY 10001 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.”
# 2004 Bengt Jacobsson, Per Lægreid and Ove K. Pedersen All rights reserved. No part of this book may be reprinted or reproduced or utilized 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 Jacobsson, Bengt, 1954– Europeanization and transnational states : comparing Nordic central governments / Bengt Jacobsson, Per Lægreid and Ove K. Pedersen. p. cm. Includes bibliographical references and index. 1. Scandinavia–Politics and government–1945- 2. European Union–Scandinavia. 3. Nationalism. I. Lægreid, Per, 1949II. Pedersen, Ove Kaj, 1948- III. Title JN7042.J33 2003 320.948–dc21 ISBN 0-203-63398-9 Master e-book ISBN
ISBN 0-203-63738-0 (Adobe eReader Format) ISBN 0–415–29978–0 (Print Edition)
2003008664
Contents
List of Illustrations Preface 1
Europeanization of Nordic states
ix xiii 1
The comparative approach 3 The evolution of a transnational administration 5 Theoretical explanations 11 The empirical data 21 The organization of the book 24 2
Transnational states
27
Introduction 27 What is transnationalization? 27 Levels of affectedness 28 Contact with and participation in EU organizations 34 Independent judgement 39 Interest organizations and private companies 42 Transnational states 46 3
Segmented states Introduction 49 Segmentation 49 Coordination 57 Multiplier effects 61 Transnationalization revisited 70
49
viii
4
Contents
Robust and flexible states
73
Introduction 73 A Nordic model with scope for variation 73 The transnationalization of administration 76 Robustness and flexibility 85 Robust and flexible administrations 101 5
Translating states
105
Pressure from the EU: form and length of affiliation 107 Historical legacy and administrative models 110 National strategies and the organization of EU work 111 Blending the theoretical perspectives 115 A translation perspective 119 Summary 124 6
Democratic challenges
127
Introduction 127 The democratic consequences – a method 130 The ideals of everyday integration 133 An evaluation – six challenges 138 Conclusion 144 7
Understanding transnational states
149
Introduction 149 Theories of transnational relations 149 Transnationalization and the transnational state 156 Summary 161 Notes Bibliography Name Index Subject Index
167 169 185 189
Illustrations
Tables 1.1 2.1
2.2 2.3 2.4 2.5 2.6 2.7
2.8
2.9
Survey responses divided by country, ministerial area, administrative level and size Have the unit’s activities been affected to a significant degree by the EU/EEA agreement in the following areas, according to country? Roughly how much of the unit’s total work time was devoted to EU-related work during the past year? What degree of impact, all in all, has the EU had for the unit’s area of activity? Is the unit more or less affected by the EU today than it was four years ago? Contact with EU organizations once a month or more often Participation in meetings with EU organizations once a month or more often If, during the past year, the unit’s employees have participated in working groups or committees within the Commission or European Council, what kind of instructions have they usually received? If, during the past year, the unit’s employees have participated in working groups or committees within the Commission or European Council, what kind of instructions have been given by political sources (government, cabinet) to ministries as distinct from central agencies? How often in the course of the past year have employees of the unit been in contact with interest
22
29 31 31 33 34 36
40
42
x
Illustrations
organizations and private companies in relation to EU matters? 2.10 If the unit has participated in public commissions or national work groups in relation to EU work, to what extent have interest organizations and private companies been represented? 3.1 EU administration and the core segment 3.2 Extent of the EU administration and the core segment in various policy areas 3.3 Proportion of EU administration and of core segment in ministries and agencies 3.4 In the course of the past year, it has become more usual to define standpoints of common national relevance concerning EU-related questions within the unit’s field of activity 3.5 In the course of the past year, EU work has resulted in increased coordination between the unit and other units working within the same sector 3.6 In the course of the past year, EU work has resulted in increased coordination between the unit and administrative bodies in other sectors 3.7 Units belonging to the state’s internal coordination segment 3.8 Units that, in the course of the past year, have had monthly or more frequent contact with EU bodies and with national interest organizations and private companies in connection with EU-related work 3.9 Proportion of units that, in the course of the past year, have had contact with the units in other Nordic countries once a month or more often in connection with EU-related activities 4.1 Special units/subsidiary units and/or posts with special responsibility for EU concerns 4.2 The significance of factors on changes in the unit’s organization and working practices over the past four years 4.3 In the course of the past year, how often have the unit’s employees had contacts with the government or the political leadership in relation to EU concerns?
43
44 51 53 54
58
59
59 62
64
66 88 89
95
Illustrations
4.4
4.5
In the course of the past year, how often have the unit’s employees had contacts with the government or the political leadership in relation to EU concerns, according to administrative level? The most common forms of contact between the units and the government/political leadership in relation to EU concerns
xi
95
96
Figure 6.1
The transnational administration assessed in relation to democratic ideals
145
Preface
This book presents a study of the ways in which the European Union has influenced the day-to-day work of the state administrations in the Nordic countries. It is a comparative study, based on an extensive survey of the central governments in four countries: one country which is a long-standing member of the European Union (Denmark), two countries which became members in 1995 (Finland and Sweden) and one non-member (Norway). The book reflects an interest in the changes in the central state governments related to the increase in European ambitions and rule making. We focus on the day-to-day work of state organizations, not intergovernmental conferences and summit meetings. We believe that the expansion of a transnational administration – where identities as well as loyalties are created in processes that transcend the borders of states – is both underestimated and undertheorized in the current discussion about Europe. The book is the result of a long-term, close cooperation of a Nordic research group now connected to the Department of Administration and Organization Theory and the Stein Rokkan Centre for Social Studies at Bergen University, the Department of Political Science at Copenhagen University and So¨derto¨rn University College in Stockholm. The cooperation has been intense concerning the design of the survey, the collection of data as well as the analysis of the data. It has been a truly comparative as well as a collaborative effort. Vital in this cooperation have also been our colleagues Anders Esmark, Gry Larsen and Go¨ran Sundstro¨m who participated in the writing of the first book that was a result of this research program: Europaveje. EU i de nordiske centralforvaltninger (Copenhagen: Jurist- og Økonomforbundets Forlag, 2001). We also would like to thank our Finnish colleagues: Lauri Karvonen, Siv Sandberg and
xiv
Preface
Krister Sta˚hlberg who helped us to make sense of the data from the Finnish administration. A note of thanks also goes to Danish colleagues Christian Frankel, Erik Højbjerg and Dorte Pedersen who participated in writing EU i forvaltningen. Broen fra Slotsholmen til Bruxelles (Copenhagen: Jurist- og Økonomforbundets Forlag, 2002) partly within the Nordic Project. Our project has depended on resources provided by the Nordic Council, the Swedish Research Council and the Norwegian Research Council. For excellent support during different phases of the project, we are also grateful to Maila Solheim for excellent technical assistance in supervising the preparation of the manuscripts and to Synnøve Serigstad who compiled the indexes. Stockholm, Bergen and Copenhagen, April 2003 Bengt Jacobsson, Per Lægreid and Ove K. Pedersen
1
Europeanization of Nordic states
The administration of European societies is in the process of changing. Here the European Union is playing an important role. The EU is particularly important because the scope of its ambition has developed and is now relatively extensive. Not only does it aim to create a single market but also to establish tighter links between the European states. It hardly seems possible to insulate any area of activity from the process of European integration. For many years the impact of the EU could be described as relatively limited and applying mainly in certain political fields, but the adoption of the Single European Act, the creation of the single market, the Maastricht Treaty on the European Union, the Treaty of Amsterdam and the endorsement by all member states of the Treaty of Nice have led to substantial increases in cooperation and integration within the EU. This work presents a study of the ways in which Europeanization has influenced the day-to-day work of the national administrations in the Nordic states. A comparison has been made of the impact of the EU on the central government agencies of the four Nordic states. We intend to describe and explain similarities and differences, and will also evaluate Europeanization from a democratic perspective. One main thesis in this book is that we are experiencing a transnationalization of the relationship between the Nordic states and the European organizations. We also perceive a clear segmentation of the national administrations. There is a greater ambition to coordinate activities within the state as a result of the EU. In addition we will demonstrate how the extensive exchanges with the EU that take place in some administrative areas can both augment existing patterns of contact but also impair others. However, while this enables us to discern changes in the everyday integration of the civil services, the national administrative models appear, nevertheless, to be relatively robust.
2
Europeanization of Nordic states
Our study concerns the day-to-day integration that Europeanization promotes. Earlier studies have shown that the distinctions between the national and the European are becoming blurred, and that it is only possible to understand the significance of the EU for states in the light of the high degree of intermixture that exists. In this study we aim to delineate both the formal, hierarchical relationships between those involved and also the informal ones that are based to a greater extent on networks. We would like to point out that we are not primarily interested in dramatic events and episodes, but the everyday exchanges between various actors. Our study does not focus on what takes place at Inter-Governmental Conferences and other summit meetings between leading politicians but on day-to-day contacts on EU issues between domestic agencies, the European organizations, other international organizations, interest organizations and companies (Mo¨rth 1996). The dependent variables in this book will be the changes in central government agencies related to increased European integration. We will examine the various dimensions of Europeanization in the central states of the different countries. We intend to study changes in the internal organizations and methods of collaboration, degrees of interaction and dependence in the various sectors. We shall study the changes that have taken place in external relationships and in networks, in patterns of contact and forms of participation and also what the EU has meant for identities and influence. We pay special attention to the relationships of the administrative agencies with both their political leaders and also EU organizations and other actors. We share our focus on states in the Europeanization process with many other scholars. There are, for instance, studies that emphasize the significance of the EU in strengthening territorial states (Milward 1992, Moravscik 1993), studies that focus on the issues related to problem solving and political legitimacy that EU integration creates for the states (e.g. Scharpf 1999), and observers who view the states as increasingly obsolete actors (Ohmae 1995). Studying the day-to-day integration of administrations makes it possible for us to create a broader and deeper image of what is happening to the states. No contours of a European community without states are visible to us. We will argue that states still constitute very significant actors, but that they are being transformed by the ongoing Europeanization process (cf. Kohler-Koch and Eising 1999). The states do not remain as they always have been. Almost half of all the units in the state administrations of the Nordic countries
Europeanization of Nordic states
3
consider that the EU has had a major impact on the area in which they operate. Both the structures of the states and their strategies have been remodelled by Europeanization. Not only are contacts with other agencies and with the EU organizations increasing but contacts with companies and other interest organizations are being altered. It could be said that Europeanization is penetrating both external and internal state boundaries. Nobody knows what this process will look like in years to come. Enlargement of the EU to include new states whose historical legacies vary, as do their political, social and economic conditions, makes it more important than ever to use comparative approaches in an attempt to come to terms with what Europeanization means for the transformation of the member states.
The comparative approach Much of the literature on Europeanization consists of studies which, albeit with varying points of departure, concentrate on the impact of European policies on national administrations (see, for example, Caporaso 1996, Me´ny, Muller and Quermonne 1996, Rometsch and Wessels 1996, Bulmer and Burch 1998, Hanf and Soetendorp 1998, Cowles, Caporaso, and Risse 2001, Knill 2001). This study has found inspiration in this tradition. When we refer to Europeanization of the administrations of the Nordic states we mean the changes in the national administrations resulting from the impact of the European Union (cf. Olsen 2002a). EU ambitions and regulations can be seen as one potential factor to explain what is taking place in the national administrations. But we are aware that there are also other factors that may influence developments in the EU-activities of the national administrations (Goetz and Hix 2001). The importance of the pressure exerted by the EU is, therefore, supplemented by national strategies, historical traditions and the interplay between such factors. In presenting our account of the way in which the state administrations have been influenced by the EU project, the comparative approach is central. The book is based on an extensive survey used to study the central governments of Sweden, Denmark, Norway and Finland. Our study concerns the administrations of these four relatively small states, integrated in a Nordic community, but where at the same time factors such as affiliation to the EU, EU strategies, general administrative systems and historical and political circumstances vary. This means that the Nordic states provide an
4
Europeanization of Nordic states
excellent laboratory in which to study the adaptation of organizations to an environment that is highly regulated and which makes demands concerning what should be done and what kind of behaviour is required. There is a great need of comparative studies of the impact of the EU. Comparative administrative research in Europe has, however, been comparable rather than comparative and has been criticized for being descriptive rather than analytical (Derlien 1995). Knill has pointed out that ‘. . . hardly any comparative studies exist which systematically investigate the conditions for administrative change from a crossnational perspective’ (Knill 2001: 15). This study is one attempt to remedy these circumstances. For this reason we have studied four classic nation-states in Northern Europe from which major changes have been required over a number of years. These demands have resulted largely from the ambitions aroused by European integration and the ensuing organizational needs. We will demonstrate that European ambitions have led to extensive changes in the administrations of the Nordic states but that these vary in different ways depending on the state or the sector described. We claim that the Nordic countries provide an exciting arena for studies whose specific concern is Europeanization. We are applying what is often referred to as a ‘most similar systems-approach’ (Przeworski and Teune 1970), in that the countries we are comparing are in many respects relatively similar. The problem with comparative studies of the Nordic countries has too often been their lack of variation, or in other words the explanatory factors have not varied to any great degree from one country to the other (Anckar 1993). Where Europeanization is concerned, however, there are distinct differences – one of the countries is a long-standing member of the EU (Denmark), two became members in 1995 (Sweden and Finland) and one has an ‘EU-light’ relationship through the European Economic Area (EEA) agreement (Norway). In terms of formal affiliation, therefore, as with EU policies overall, there are clear variations between the different countries. At the same time – and this is the point of the most-similar approach – there are fundamental similarities along many other dimensions. Norway, Sweden and Denmark as well as Finland are welfare states with extensive public sectors, and have enjoyed well established cooperation in many different fields for a long period. They display major similarities in their socio-economic and political circumstances,
Europeanization of Nordic states
5
economic development, educational systems etc. All of the Nordic countries are parliamentary democracies with well developed administrative systems. They are all unitary states and of roughly the same size. In each state there are firmly established interest organizations and a strong tradition of consensus. The variations that exist between the Nordic states, for instance with regard to EU strategies, administrative systems and the like, can, therefore, only be understood against a background of considerable homogeneity. This is an aspect that makes it particularly interesting to compare these states with each other. We have designed the study as one where processes of adaptation in the four countries act as dependent variables. One of the independent variables is made up of the pressure and the forces that are linked with Europe (others are, as we shall see, national systems that have evolved historically and the national strategies adopted for EU issues). The existence of such great similarities between these states provides ample opportunity to investigate the impact of the features that distinguish them, such as the different forms of affiliation and the varying periods for which they have been members (with Norway providing a special case because of its EU-light links), the alternative solutions to the organization of EU activities in the different states and the fact that the administrative systems of the four states differ. We claim that the comparative design we have adopted provides major benefits. Unlike many other cross-sectional studies (Bryman 2001), this survey is based on the long-term, close cooperation of a group of researchers on the design of the survey, collection of data, analysis and production of articles and books. There are very few studies of this kind that have been organized on a genuinely comparative basis, i.e. which possess a shared theoretical conceptualization and have resulted in such comprehensive and penetrating empirical studies. The comparative design, shared conceptual development, and coordinated implementation of the empirical studies have enabled practical realization of the comparative approach. Our ambition is that these studies will bear fruit for the understanding of states other than those that happen to be situated in the European North.
The evolution of a transnational administration For many years, a series of myths have dominated discussion of the consequences of participation in European cooperation for the states.
6
Europeanization of Nordic states
One of these concerns the dissolution of the states and their increasing obsolescence (Ohmae 1990, 1995). Even though it can be seen that the EU has taken over the legislative role of the states in many areas, expectations that a transfer of sovereignty to the EU would lead to the erosion of national political organizations find hardly any support in empirical studies (Camillieri and Falk 1992, Jessop 2001). Despite many years of creation of a new set of organizations and operations at a European level intended to regulate national agencies, activities and citizens, the result has hardly been the erosion of the national public organizations. However, to view states as uniform actors and to explain policy formation as a result mainly of state strategies and state actions have also become increasingly unsatisfactory. Admittedly, the actions of individual states may at times be significant, but these actions are to a great degree devised in a European context. For this reason there is rarely any strict division between the national and the European. National standpoints and strategies emerge from the exchange between administrative and political organizations of the states and their European counterparts. This means that the EU organizations play an important role in the processes that lead to the adoption of national decisions and standpoints, just as the national organizations play an important role in the processes that lead to the adoption of European decisions and standpoints. Not only the states and the European organizations are involved in these processes, but these can at times be relatively open to companies, interest organizations, lobbyists, Non Governmental Organisations (NGOs) and others. Transnationalization involves, therefore, not only a transnational exchange between states and European organizations, but also the participation of many other organizations in society. EU work in the states could involve a variety of different actors. It is also the case that in the EU greater cooperation is sought with other organizations. The Commission claims that ‘wide consultation with a variety of interested parties is an important means of ensuring that the Commission proposals are technically viable, practically workable and acceptable to stakeholders’ (White Paper on European Governance). It would be just as unwise to believe that nothing is happening to the role of the states in the course of European integration as to believe that they can merely be seen as relics of the past. In some processes, the states have become increasingly involved in their everyday routines with other states and European organizations. Most often this is fairly
Europeanization of Nordic states
7
remote from the summit meetings and Inter-Governmental Conferences. In order to understand how policies evolve in various areas, therefore, more is needed than mere study of national policies, national administrations and the European organizations so that the scope must be widened. At a European level as well, policies evolve to a large extent across boundaries, in processes involving large numbers of participants and where these participants can rarely be viewed as conducting only intergovernmental negotiations. In this work we shall argue that the consequences of Europeanization can be described as the growth of a transnational administration. In using this term we are referring primarily to an administration that is to a great extent involved in relationships to other actors, such as the administrations of other states, companies, interest organizations, EU bodies, international organizations and NGOs. Secondly, transnational administrations are characterized by the fact that those who act on behalf of the states are often doing so with no particularly strict or explicit mandates, or in other words they have wide discretionary scope for their own assessments. A third characteristic of transnational administrations is that they act to a great extent in accordance with regulations or beliefs that need not be linked to any national context but can well be European (or global). Transnational features in relationship to the EU may be extensive and then the national administrations have frequent contacts with other actors, contacts which permit a high degree of autonomous action. The actions of transnational administrations are influenced extensively by non-national regulations and beliefs. Or there may only be limited transnational impact, and if so civil servants have few transnational contacts or they act according to strict instructions or regulations that have been created in a national context. What our study reveals is that the states are increasingly becoming transnational, implying that large sections of the state administration have frequent contacts, a great deal of scope for their own independent action and are influenced by regulations that are not national. It is this transnationality, where European and national factors are intertwined, that we should like to emphasize. Transnational relations could be described as neither European nor national, but both. But the public and the private also become intertwined. Many other actors apart from public ones, such as companies, interest organizations, NGOs, are involved both in the creation of regulations and in the dayto-day exchanges. Transnational relations, therefore, are neither public
8
Europeanization of Nordic states
nor private, but both. And in our understanding of transnational relations we emphasize that administrations are often acting with no explicit, specifically national mandate. It can thus also be said that transnational relations are neither political nor administrative, but both. In studying day-to-day integration we shall be able to identify the extent to which the central administration is acting on its own initiative in European exchanges. We have presented above a preliminary definition of transnational administrations. Exchanges and contacts may be significant. But a high degree of transnationalization of states may be attained before we see any extensive participation and their administrations and political organizations need not be involved in extensive contacts with European organizations. There may be powerful organizations issuing regulations and influential regulatory systems that states comply with that have no contacts at all (integration without interaction). Analogously, we may also see extensive participation and comprehensive networks and meetings without this necessarily in practice implying that the exchange has any genuine impact (interaction without integration). States influence and are influenced by organizations at the European level. They are also implicitly bound up in exchanges with transnational companies, international organizations, interest organizations and NGOs that do not necessarily focus their interests on Europe, and they are influenced by the developments of regulations and discussions that are also taking place in a wider global community (Boli and Thomas 1999). It is, for instance, impossible to study the trade policies of the EU or any one state without considering the significance of relationships with the World Trade Organization (WTO), transnational companies and various NGOs. Environmental policies cannot be understood without reference to the ways in which states and the EU are involved in relationships with global organizations, epistemic communities, environmental organizations etc. Possibly the most widespread and accepted definition of transnational relationships is Risse’s ‘. . . regular interactions across national boundaries when at least one actor is a non-state agent or does not operate on behalf of a national government or an intergovernmental organization’ (Risse-Kappen 1995: 1). This definition resembles ours to a great degree but there are some aspects that we would like to particularly underline. There is, to begin with, a risk that Risse’s definition plays down the great variety of actors involved in the processes by lumping them all together as non-state. Secondly, we
Europeanization of Nordic states
9
wish to stress that states are not monoliths, but consist of many units and that these units may be relatively loosely linked to each other (segmentation). Thirdly, we emphasize how state organizations adapt over time to environments that in turn adapt to state activities (co-evolution). Fourthly, we stress that one should not look upon transnationalization as something that primarily exists ‘outside’ states, but instead that this process is embedded in everyday activities. Fifthly, we emphasize that influence on state activities can be exerted even without regular interactions. Interaction can take place without necessarily being significant, just as actions may be influenced even though there is no regular interaction. As has already been pointed out, this study focuses on the Europeanization of the national administrations, and is, therefore, based on the idea that influences go from the European to the national. Our belief presented above that the distinctions between what is European and what is national are blurred also means that European policies and ambitions are created through an exchange with various organizations in the respective member states. Just as national strategies and structures evolve through exchanges between many actors, the same applies to European strategies and structures. Corresponding to the Europeanization of the national administrative systems we can speak of a nationalization of the European administrative system. Variations in transnationalization In this book we will argue that there has been increasing transnationalization of the states. However, the degree of transnationalization varies. It varies between states. The Nordic countries, which all have firmly established and relatively strong states, display a pattern that differs from states which are less comprehensive and have less autonomy. But even between the Nordic states there are considerable variations. This can depend on different factors: the form of affiliation, the length of membership of the EU, strategies for EU activities, the public administration systems, their traditions and history. It should be noted that in strong states we will not necessarily observe limited transnationalization. It may be more rewarding (albeit more difficult) for companies and NGOs to act by forming coalitions with a strong state. Even if it may be easier to establish access to more fragmented states, such partnerships may prove to be less significant (Risse-Kappen 1995).
10
Europeanization of Nordic states
There are also variations between sectors. We will show that segmentation has mainly taken place in certain parts of the national administrations. It is above all in the areas where EU activities have developed furthest, such as agricultural policy and the single market, that the administrations have reached the highest degree of transnationalization. It may well be that the EU’s ambitions and organization creates transnationalization in this way. If certain elements of the administrations are intricately linked to Europeanization whereas others have little to do with EU activities, there is a risk that the EU involvement will result in the fragmentation of national administrations. If this happens we could have parts of the state that have closer contacts with functionally equivalent parts of other states and with European organizations than they have with other parts within their own state. We can also see variation at different administrative levels. In this study we have mainly considered the exchanges and opinions of the central administrators, and our studies reveal that on the whole it is the central administration and the political leadership that participate in networks with others. It may well be the case that the main participants in European exchanges are the central state agencies and that these are influenced by the regulations and ideas in circulation. If so, variations in transnationalization could also lead to fragmentation of the national community. We shall show that to some extent this is the case, meaning that transnationalization varies not only between sectors but also between various levels of the state administrations. At the same time we shall see that participation in a European network is also linked with extensive exchanges with other organizations, such as companies and interest organizations. Transnationalization may also vary during different periods. We often describe society today as being more economically, politically and culturally global than ever before. At the same time it is asserted that exchanges across frontiers were in some respects as extensive at the beginning of the last century before the First World War, and that transnationalization should not be regarded in any way as a process that only moves in one direction. When we discuss state administrations, it is rather the period from the mid-nineteenth to the mid-twentieth centuries that could be seen as parentheses (Knudsen 2003). If we consider the EU during the most recent decades, the process from the early 1980s can be seen as a gradual growth of the importance of the EU’s organization and the regulations it issues. The creation of the
Europeanization of Nordic states
11
single market has acted as the main dynamo of integration, and our study is situated in a period five years after the attainment of the single market. In this book we are going to present the ways in which the national state administrations in the Nordic countries have changed as a result of the EU. Our primary thesis is that during the last 10–15 years a development has taken place towards a transnationalization of the relationship between national and European organizations. We also claim that variation in this transnationalization exists in different elements of the states, or in other words that segmentation is developing. We perceive an increase in attempts to coordinate as a result of EU activities. We also discuss the circumstance that the EU gives rise to a multiplier effect, that it augments certain existing patterns of contact in administrations and at the same time weakens others. Europeanization reshapes administrations but at the same time the national administrative systems seem to enjoy a high degree of stability.
Theoretical explanations It is not our aim merely to describe the consequences of the EU but also to offer explanations. In this section we shall sketch in various theoretical perspectives on change that we will use: the first in which the states adapt to the ambitions and the dynamics of the EU; the second in which the states make strategic choices of what courses of action to take; the third emphasizing how patterns of behaviour that have evolved historically are stable and robust under the pressure to change from Europe. We also intend – partly on the basis of these three – to propose a fourth perspective that emphasizes translation. This underlines the driving force provided by the EU but points at the same time to the considerable scope that exists in the different states when it comes to translating the EU regulations. This translation is shaped to a great extent both by the traditions that have evolved over time and also in consequence of the method chosen by the different states to organize the way they work with the EU. We will also present an account of which more specific explanatory factors are linked to the different theoretical perspectives. If we believe that historical links are important, it becomes important to chart the activities of the states in relation to the administrative traditions that characterize them. If we believe that choices and strategies are vital, then importance is placed on what gains states seek from the EU and
12
Europeanization of Nordic states
the concrete methods adopted to organize EU activities. If we believe that it is the internal dynamics of the EU that matter and most of what happens reflects these dynamics, then the most important factor may be the form taken by the state’s affiliation and for how long it has been a member. We will analyse some of the potential explanatory factors and concentrate on the variations that exist between the states. Adaptation to the EU aspirations and regulations Seen from this perspective, developments within the EU play a decisive role for what happens in the national administrations. The development of the EU, its aspirations and its dynamics together with the ideas and regulations it engenders explain the actions of the national administrations. There are numerous studies that provide support for such a position. Adaptation to EU regulations in the Nordic administrations is ‘Lutheran’ to a great extent (Lampinen and Uusikyla¨ 1998, Sverdrup 1998, Tallberg 2000), and many examples can be provided of how the agendas of the member states are governed by what takes place in the Union. Initiatives and regulations that originate from outside the individual states provide the impetus for change. From a theoretical point of view this is a form of ‘external explanation’ in the sense that regulations and structures outside the individual states impel change. This kind of explanation leaves relatively little room for manoeuvre for the states themselves. Changes in the European environment are considered to determine to a great extent what happens nationally. Organizations are established, strategies created and directives obeyed, and the fundamental mechanism is for the national administrations to adapt. This adaptation can be brought about by coercive, but also through mimetic or normative processes (DiMaggio and Powell 1983). If we consider what has happened in the states discussing membership of the EU at the beginning of the twenty-first century, an explanation based on adaptation has a point. Systematic assessments are made of the degree to which the potential members can live up to the demands made of them, and the reference point is the existing system of regulation and organization in the EU. A state that cannot credibly demonstrate that it can comply with the membership criteria determined in different areas is ineligible. The regulations and criteria applying to membership of the European Monetary Union function in the same way. Compliance with the regulations is not a
Europeanization of Nordic states
13
straightforward process, and may contain no small amount of decoupling of presentation and practice, but there can be no doubt that the agenda is influenced by the EU. With an explanatory approach that emphasizes developments in the European Union and the adaptation of the member states, one central issue is what the formal system of affiliation looks like. How the pressure to change experienced can be expected to be different in member states or those wishing to become members to that in those that are not members or do not want to join. Here too there are differences between the Nordic states. Denmark has been a member of the EU since 1973, Finland and Sweden became members in 1995. Neither Denmark nor Sweden are participating completely in the integration process, for instance neither country is fully involved in the cooperation on monetary policies. Norway is not a member but is linked to the single market through the EEA agreement. This allows Norway access to the preliminary work and the implementation activities of the European Commission but not to the formal decision making process in the Council. The fact that Norway remains outside major aspects of the EU could be an important explanation for the way in which its work with the EU is organized. At the same time this position is the result of a deliberate national strategy. Norway has a tradition of being positive to free trade and intergovernmental political cooperation, and sceptical of political integration and supranational commitments, and it is this strategy that has created its formal exclusion (Olsen 2001). If the contact patterns of the Norwegian administration turn out to differ from those of the other states, it could, therefore, be interpreted to mean that the formal system of affiliation is decisive. Since this is a consequence of national traditions and strategies, this clearly shows that our different explanations cannot be considered in isolation from each other. Length of membership is another significant variable that can be seen as one explanation that emphasizes the impetus of the EU and national adaptation. The comparatively recent membership of Sweden and Finland could both mean that they have not yet had the time to establish the contacts that members with longer standing have acquired and also that their attention is particularly acute where EU issues are concerned. Denmark’s longer membership can both have the effect that EU activities have become an everyday concern for Danish civil servants, but could at the same time make it more difficult for them to know what is national and what has been initiated by the EU
14
Europeanization of Nordic states
(Siedentopf and Ziller 1988). There has been more time for the European and the national agenda to merge than in the other countries. National strategies Another perspective that is often encountered emphasizes choice and a country’s own national strategies. This point of view claims that the actions of national administrations are governed by strategic considerations, such as how states opt to organize their EU activities, what kind of coordinating bodies are created, and how the parliamentary bodies are incorporated into or isolated from the processes. One example of this perspective on European integration, which is often referred to as liberal intergovernmentalism, can be found in Moravcsik (1995, 1997). A lot of the ‘down-to-earth wisdom’ frequently found in EU contexts, implying that we should coordinate our actions and speak with one voice, are based to a great extent on this approach. With a perspective like this, much of what is observed can largely be explained in terms of national strategic choice. There is either less emphasis on the EU as an autonomous force or it is claimed that adaptation takes place because of deliberate actions. Even though a great deal of activity in the national administrations is governed by the EU agenda, its organization and regulations, the strategic perspective can still be ‘salvaged’ by claiming that the states have, after all, chosen to become members. What looks like a deterministic adaptation to amended European regulations is, therefore, actually merely the consequence of the original decision to become a member (or not to in the case of Norway). One important factor in this context is the dominant attitude to the EU in the various countries. In remaining aloof from major aspects of the integration process, Norway stands out as most negative. As has been pointed out, Norway has traditionally been somewhat sceptical of political integration and supranational commitments. After having spent some time in the same position as Norway, Sweden and Finland became full members. What is particularly obvious is the determination with which Finland has asserted its intention to work as a full member of the EU, whereas both Sweden and Denmark do not cooperate fully in the workings of the monetary union. Denmark has the longest experience of EU membership. There has been an intensive domestic debate on the EU in the Danish community, manifest in several referenda. Denmark has also
Europeanization of Nordic states
15
succeeded in including a number of derogations in the treaties. Citizens of the Nordic countries have been divided in their attitudes to the European project (Miles 1996). Hesitation about the EU can thus be found everywhere, but finds its strongest political expression in Norway, Denmark and Sweden. The European route has, partly as a result of its historical experiences, become particularly important for Finland, where national support has been aroused for a form of positive EU activism. The different countries have also organized their work with the EU in somewhat different ways (we will come back to these differences in more detail in later chapters). In Denmark as well as in Norway and Finland, there is a system of permanent coordination groups specially created for work with the EU. In Sweden it is largely up to each ministry to decide how coordination is to be arranged. No formalized system of special coordination groups has been established. The permanent coordination groups in Denmark allow private interests greater access to the coordination system than in the other countries. In Denmark the Foreign Ministry still plays a central role for coordination, while a shift has taken place in Finland and Sweden to bring coordination closer to the Prime Minister. As far as the parliaments are concerned, the Danish parliament (folketinget) seems to have a stronger position with regard to the government than the parliamentary committees in the other countries. Historical links A third perspective stresses stability and historical links and that established activities rarely undergo drastic reorganization but are relatively robust in the fact of external changes (Selznick 1957, March and Olsen 1989). This view is profoundly sceptical of those who argue for dramatic changes. Change – even when it is caused by the EU – is considered to be linked to history and highly dependent on the nature of prevailing traditions and the dynamism of the administration itself and the surrounding political climate. Often processes are analysed from this point of view as path dependent (Krasner 1984). Starting from a historical-institutional perspective, we cannot expect the norms and traditions, routines and established practices that have evolved nationally in a long historical process to be particularly susceptible in the short term to change (Pierson 1996). The processes of change are incremental.
16
Europeanization of Nordic states
One presumption is that the reforms and changes required that are not in conflict with the established administrative culture and that fit in with prevailing doctrines on administrative policies are those that have greatest chance of taking effect. The position adopted by an individual country to the transformational force of the EU will, therefore, depend on how well the national administrative traditions and systems fit the EU profile (Olsen 2002b). The expectation that the EU will play any decisive role in remodelling the administrations of the Nordic countries does not seem realistic from this perspective. We should rather expect established casts of thought and behaviour in the various countries to be capable of incorporating what is new. Old forms of regulation and organization do not disappear, but they are supplemented by the new ones. In this context it is particularly interesting to analyse the administrative models that exist in each country. It is customary to distinguish between eastern Nordic and a western Nordic administrative models – between those that are dualist and those that are monistic (Lægreid and Pedersen 1999). The eastern Nordic, dualistic, model finds clearest expression in the organizational structure of the Swedish national administration. Here a clear distinction is made between the government and its offices, on the one hand, and the central agencies on the other. The roots of this administrative model are usually traced back to the 1720s. The autonomous agencies are formally independent of the individual ministers but are subordinate to the government as a whole (except for a few well defined cases). At various times debate has arisen about this dualism but a system consisting of autonomous agencies and relatively small ministries is still normally said to characterize the Swedish administration (Premfors 1999). The western Nordic administrative model which finds expression in Denmark and Norway is characterized by ministerial rule, in other words what is usually referred to as monism (Lægreid and Pedersen 1999). Ministers are formally responsible for their subordinate agencies and this means that the autonomy of these agencies is restricted. A minister can be held accountable by parliament for all the decisions made by the administration. The roots of this administrative model can be found in the era in which absolute monarchy was introduced in 1660. Both the Danish and Norwegian administrative models are characterized as belonging to the ministry-directorate model. The autonomy of the directorates in relationship to the ministries may vary, but the ministers are always accountable for the actions of a directorate.
Europeanization of Nordic states
17
The differences between the east Nordic (dualistic) and the west Nordic (monistic) administrative models may be considered large (Petersson 1994, Lægreid and Pedersen 1999), but it can also be maintained that in practice they need not be particularly clear when, for instance, informal contacts are taken into account (Jacobsson 1984, Lindbom 1997). Traditionally Finland has been placed in the east Nordic model, but recent changes have given the Finnish model more monistic features. There is less collective decision making and increasing emphasis placed on the accountability of the various ministers. One possible consequence of the difference between the east Nordic and west Nordic models – both for processes of change in general and work with the EU in particular – is that greater sectorization is likely in Norway and Denmark. The ministerial system could be expected to lead to greater integration of politicians and officials in every sector, combined with weaker mechanisms for coordination between the sectors. This is perhaps most obvious in Norway, as the Danish Ministry of Finance has a more powerful coordinating role than the Ministry of Government Administration has in Norway (Lægreid and Pedersen 1999). In Sweden – as a result of the formal system – collective decision making in the government could lead to greater integration and coordination in the government offices, but weaker coordination between the government and autonomous central agencies. Towards a fourth perspective: translation Hitherto we have accounted for three different theoretical perspectives and discussed which more specific explanatory factors can be linked to the various perspectives. We shall later in the book (in chapter 5) return to these potential explanations and discuss their relative capacity to provide explanations. All three perspectives can help to increase understanding of the transformation of the national administrations. And we will discuss how these three perspectives can be combined and blended into a fourth one. But our fourth perspective can also be seen as a distinct theoretical perspective in its own right. The point of departure for this translation perspective is the dubiety of making clear distinctions between the national and the European. Europeanization means after all that factors that are normally considered national (strategies, policies, organizations) become
18
Europeanization of Nordic states
inextricably bound up with what is considered European. What is presented as a national strategy in any area has often been formulated through exchanges with European organizations and regulations. National strategies evolve in the EU, but these are at the same time the product of European exchanges (Jacobsson 1997a). Fundamentally the actions of state organizations are determined – as are their aims and how they want to be perceived – by their relationships with other actors. To understand the actions of state organizations we have to know about their networks. What state organizations do is determined also by the wider system of ideas and regulations that prevail in the areas in which they operate. There is a panoply of organizations – in addition to the EU there are various Intergovernmental Organizations (IGOs), NGOs, companies, professional and scientific organizations, and so on – that produce regulations that apply to states or parts of states. From a formal point of view, some of these regulations may be legally binding and others may take the form of recommendations or standards, thus constituting voluntary regulations (Brunsson and Jacobsson 2000). There are also a large number of comparisons, evaluations, rankings, audits, and so on that may be expected to affect the way in which states conceive of themselves and their activities. In this way states may be considered to be governed by their environments. Even if this environment does not consist only of the European Union, the EU has come to occupy a special position. The interests and strategies of states are created to a great extent by European exchanges. In certain areas and where certain activities are concerned states are allowed only little scope for manoeuvre (the new candidate countries have to adopt the acquis communautaire that already exists). So the image of states with exogenically given preferences, i.e. as actors that know what they want and are capable of forming national standpoints and strategies, has to be modified. It would seem that states are able to learn a great deal from the ongoing extensive European and transnational exchanges about both what they want and what their identities are. However, it should also be noted that the regulations that are often felt to be binding, the EU’s acquis communautaire, do not automatically lead to adaptation. It has taken 4–5 years to negotiate the current enlargement of the union with its new members. This process has involved negotiations about how the acquis is actually to be interpreted, how it is to be put into effect, what derogations will be
Europeanization of Nordic states
19
conceded and what timetable is to apply for the implementation of the acquis in the various candidate countries. So what may appear to be a relatively unproblematic compliance with a system of regulations allows considerable scope for interpretation and translation. We are, therefore, going to present an image of adaptation that in the first instance does not involve complying with regulations but rather an adaptation that results from participation in a very concrete community. Cross-border activities are significant. National, regional, European and global levels are often described as distinct spheres in which policy is created. Either the levels are opposed to each other (for instance in questions about whether the European level will prevail over the national) or the creation of policy is discussed in terms of events that take place simultaneously at different levels. Our perspective is based on the view that these ‘levels’ overlap and are inextricably interrelated in the sense that strategies can be both European and national at the same time. The formation of regulations and decision making in environmental policies for instance is both global, European and national. One way in which this is expressed is that the states both comply with regulations and create them. It is primarily the governments that create the regulations in the Council of Ministers that they will then comply with. The motive force for change comes, therefore, from ‘outside’ but the states are both inside and outside. They are – naturally to varying degrees – part of the rule creation system itself. It is also important to observe that compliance with these regulations usually leaves scope for national administrations to translate ideas and rules in ways that suit their own activities. In practice, regulations are almost never introduced in every detail but there is more or less latitude about how they are to be translated and edited (Czarniawska and Sevo`n 1996, Sahlin-Andersson 1996, Røvik 1998, Christensen and Lægreid 2001, Kjaer and Pedersen 2001). For instance, it seems more probable that a state will adopt ideas and regulations that match the ideas and regulations on which it already bases its activities. If environmental policy has been developed according to a specific pattern, it will be easier to adapt to a European environmental policy that is compatible with the national one. It is easier to introduce a model or a procedure in practice if the ideas on which it is based have already found institutional form in the way things are already done. If a state has been cooperating with the European Union for some length of time – as Sweden and Finland had
20
Europeanization of Nordic states
before they became members – the process of adaptation will not be as overwhelming as it would have been if they had belonged to some other community. Compliance with regulations can involve changes in practice to adapt with something coming from without, but it can also involve presenting existing practices in ways that agree with the regulations. In various contexts, the member states must demonstrate that their actions are in agreement with the EU regulations, and there are in principle two different ways of doing so apart from changing practices to accord with the regulation. One way is to devise a new label for activities that are already taking place to show that the regulations are being followed. The other is to de-couple the presentations of activities from the activities themselves (Weick 1976, Meyer and Rowan 1977, Brunsson and Jacobsson 2000). The possibility that Europeanization involves an increase in this kind of de-coupling cannot be excluded, one reason being that the state administrations are more often than previously subject to requirements that contradict each other. We should, therefore, like to leave some scope for the fact that national administrations have possibilities of translating European ideas and regulations in widely differing ways. Some of these translations may be regarded as strategic adaptations (Oliver 1991); others determined more by the situation or the process; while some translations may be seen as expressions of how robust and stable existing administrations are. Note that a central element in this concept of translation is that there is no ‘perfect’ rendering of, for example, an EU regulation in Danish practice. Translations of this kind are always complex and the process can always involve change for both the translator and what is being translated. A translation perspective seems to make it possible to explain the relatively incompatible results that have been presented in the literature on the consequences of the EU for national administrations. In some states EU policies and regulations make a major impact, in other states very little at all. In some sectors the effects of the EU may be significant, whereas in other aspects of the same state they may be restricted (Cowles, Casporaso and Risse 2001, He´ritier et al 2001, Knill 2001). It is obviously the case that the translation of the EU takes on very different appearances in different states and in this book we are going to specify what these translations look like in the four Nordic states.
Europeanization of Nordic states
21
The empirical data The results that will be presented in the following chapters are based on a survey conducted during 1998 of major elements of the central state administrations in Denmark, Finland, Norway and Sweden. All of the units in the ministries and agencies or directorates of a certain size have responded to a questionnaire on their EU activities. The units to which the questionnaires were sent were those which are state run, directly accountable to the government, operate throughout the country as a whole and are permanently staffed by administrators. Regional and local state authorities are excluded from the survey, as are those that supply only services (such as universities and higher education institutions, theatres, libraries, museums and courts of law) and state-owned companies. On the other hand; the study does include the units that have overall responsibility for and administer the allocation of resources and supervision of universities, courts of laws and museums for instance. The questionnaires were directed to units and not to individuals. One person has responded on behalf of a unit and these respondents have either been the heads of the units, some other senior official, an EU coordinator or some other member of the staff of the unit with extensive knowledge of the work it does in connection with the EU. This has enabled us to ensure that our responses came from individuals with comprehensive knowledge of the area covered by the questions and who are, therefore, well qualified to supply reliable information. The questionnaire was intended to reflect the great variations that exist in state administrations. For this reason we decided not to allow too large units with heterogeneous activities to answer only one questionnaire. Instead we focused on somewhat smaller units. The terms used to denote the units that responded to the questionnaire differ both from country to country and also within each country. In this work we have chosen to refer to all of them as units. The questionnaire has, firstly, been sent to all the units in the ministries. Secondly, it has been sent to all the units directly accountable to the top management in agencies with a staff exceeding 40. Thirdly, agencies with fewer than 40 employees have received one questionnaire, addressed to their top management. Those receiving our battery of questions constituted a modified total population with regard to the central administrations of the four Nordic countries. A total of 1323 units from the different countries
22
Europeanization of Nordic states
responded to the questionnaire: 311 in Norway, 373 in Denmark, 381 in Sweden and 258 in Finland. The response rate in the different countries was 86 per cent in Norway, 83 per cent in Sweden, 80 per cent in Denmark and 77 per cent in Finland. The consistently high response rate indicates that we are working with reliable material and that the respondents do represent those to whom the battery of questions was addressed (Jacobsson and Sundstro¨m 1999, Larsen, Lægreid and Wik 1999).1 Table 1.1 shows the distribution of the various units by country, ministerial area and administrative level. The size of the units is also
Table 1.1 Survey responses divided by country, ministerial area, administrative level and size (1998, per cent) Norway Ministerial area: Finance Agriculture and Fisheries Defence Justice Foreign Affairs/Prime Minister’s Office Health and Social Affairs Trade, Energy, Environment Communications Ecclesiastical Affairs, Education, Research, Culture Interior, Labour Market Other
Denmark
Sweden
Finland
Overall
8 9 2 7
10 12 1 3
12 7 6 8
16 11 5 3
11 9 4 5
4 16 18 10
10 10 23 6
5 11 20 9
7 12 15 11
6 12 19 8
10 16 0
10 16 0
11 12 0
7 10 5
9 14 1
Administrative level: Ministries Directorate, central agencies
28 72
31 69
21 79
22 78
25 75
Unit staffing size: Under 20 20–49 50–99 100 or more
27 45 18 10
48 34 8 10
38 37 12 13
33 34 18 15
37 37 14 12
260
309
345
244
1158
N
Europeanization of Nordic states
23
given. These figures do not include 13 per cent of the units that were totally unaffected by the EU. In Norway these amounted to 16 per cent, in Sweden 9 per cent, Denmark 17 per cent and 5 per cent in Finland. At ministry level 8 per cent were totally unaffected and 14 per cent of the agencies. The largest number of unaffected units can be found in the Defence sector (21 per cent) and the smallest in Communications (6 per cent) and Agriculture and Fisheries (7 per cent). The advantage of using a standardized battery of questions is that it enables the collection of systematically comparable data from all four countries. We have devoted a great deal of effort both to determining comparable populations and to formulating questions that make comparisons possible. The research group also discussed the wording of both the questions and the alternative responses very thoroughly. The aim was to avoid as far as possible any misunderstanding or distortion in the responses that could be attributed to the design of the test battery. The desire to attain a high degree of comparability in the data was also the reason for opting to provide predetermined alternative responses for most of the questions. Mention should be made of the methodological problems that result from addressing questionnaires to units and not to individuals. It may be difficult for an individual responding on behalf of a unit to gain any clear overview of all the unit’s activities that are relevant in this study. The size of the units, therefore, plays a major role in assessing the reliability of the responses. This is also the reason why we have chosen to send not merely one but several questionnaires to ministries and the larger agencies. This procedure requires extra caution in the use of attitude questions – it can be difficult for the respondent to state how everyone else in the unit feels and thinks ‘on average’ about various issues. Nevertheless, we have decided to include a few attitude questions. When we account for them we shall remind the reader of the difficulties that exist in interpreting the responses. The aim of the questionnaire survey was to provide knowledge of the extent to which various elements in the central administrations of the Nordic countries are affected by EU activities, what tasks they have in this area and what changes that have taken place in recent years. We try to ascertain how great a proportion of the administration is affected by the EU and also the extent of this impact on various elements in the administration. We are also interested in how EU activities are allocated in the different areas, expressed in terms of the EU’s various ‘pillars’, and also what kind of tasks EU activities mainly comprise.
24
Europeanization of Nordic states
Are we talking primarily of dealing with individual cases, providing information and advice, devising laws and regulations or something else? In addition we are trying to see how the internal organization and the staff of the ministries and agencies/directorates are affected by the EU. We ask, for instance, if the units have taken on new staff or provided staff training to deal with EU routines. Several questions also deal with the units’ contact patterns in their EU-linked tasks, both internally within the administration and externally with other actors, such as the various EU organizations, the administrations of other countries, companies and interest organizations. We also include questions about the degree of disagreement and possibilities of exerting influence in EU activities. The survey finishes with a number of attitude questions about the impact of EU activities. Most of the questions concerned activities during the previous year or changes that had taken place during the previous four years, i.e. after the EEA agreement came into force. In all, the questionnaire contained over 140 questions linked to more than 30 areas. In this book we will provide an account with our comments of how the entire population that is not totally unaffected by the EU has responded, and also how these responses vary from country to country, sector to sector and at different levels. Presumably the political and administrative situation in the different countries at the time of the survey may have had a significant effect on the responses. It should be noted that Finland was to assume the Presidency of the EU during the latter half of 1999, which may have influenced the pattern taken by the Finnish responses.
The organization of the book In this introductory chapter we have accounted for our theoretical premises and we have specified the content of the concepts of Europeanization and transnationalization. We have stressed that the study is based on a carefully implemented comparative design and that the Nordic states are particularly suitable for this kind of comparison. In addition to the reason for the comparative design we have accounted for the organization of the survey. This chapter has also contained a discussion of the main thrust of the study – the emergence of a transnational administration. Four different explanatory perspectives have been proposed: pressure from the EU, national strategies,
Europeanization of Nordic states
25
historical links and a translation perspective. We have also described how the various theories will be operationalized in the study. Chapter 2 illustrates how the national administrations in the four Nordic countries have changed as a result of the EU. One principal argument is that during the last 10–15 years there has been a development towards transnationalization of the relationship between the national and European organizations. The aim of this chapter is to describe this transnationalization. We will in the chapter look at the ways transnationalization manifests itself in the central administrations of the Nordic countries. The focus in chapter 3 is the segmentation of states. The conditions in various states are shaped by exchanges with both other actors in their own societies, such as companies and interest organizations, and with actors in other states, which leads to segmentation. We will reveal increasing attempts to coordinate as a result of EU activities. We also discuss the circumstance that the EU creates multiplier effects; the extensive exchanges with the EU to be found in some sections of the administrations may not only augment existing contact patterns but also weaken others. Chapter 4 contains a discussion of the underlying question of whether the changes that have taken place can be described as leading to fundamental changes in the administrative model. The empirical studies indicate that this can hardly be said to be the case. The established administrative models seem to be robust and also to possess a great degree of flexibility. At the same time we claim that the administrative models are being challenged by the processes that are taking place. One can ask what the stability of the models signifies. Does it mean that the administrative models can encompass ‘any kind of practice at all’? It is interesting to note that practices are changing while the formal structures are stable – and this pattern seems at least in part to conflict with one of the arguments advocated in the development of neo-institutional theory. There it is presupposed that the creation of legitimacy results principally in changes in formal structures (fac¸ades) while the practices endure unchanged. Here, too, we can admittedly see a de-coupling, but in reverse. Practices change radically, but the structures survive. In chapter 5, we shall explain both the transformation that has taken place and how it varies in the different countries. Our starting point is three distinct forms of explanation to enable understanding of Europeanization: adaptation to the EU system, national strategies and historical traditions. We also propose a fourth interpretative frame:
26
Europeanization of Nordic states
translation. Translation could first be seen as a result of a combination, or a blending, of the other perspectives, resulting in a kind of multifactorial understanding of Europeanization. But translation could also be seen as a theoretical perspective in its own right, illuminating how states and European organizations are embedded in each other. While chapters 2, 3 and 4 are based on description of results and of patterns of Europeanization, the aim of chapter 5 is to present an explanation of why things look like they do. In chapter 6 we shall then offer an evaluation of the consequences that the emergence of a transnational administration will have both for the power relationship of the actors and the parliamentary system as a whole. A method is developed for making this evaluation. The practices we have described and analysed are related to a number of democratic ideals and from this analysis we can derive the ways in which democracy is being challenged by Europeanization. Apart from transnationalization, we see tendencies towards bureaucratization, incrementalism, politicization, fragmentation, exclusion and captivity. Europeanization challenges in various ways our staunchly defended democratic ideals. Finally, in chapter 7 we shall discuss the integration dynamic that we have observed and introduce our own theoretical perspective for continued discussion of transnationalization and European integration. We argue that our theoretical perspective differs from the three most frequent approaches – intergovernmentalism, neofunctionalism as well as multilevel approaches – that are usually applied in order to understand Europeanization. It could admittedly be asserted that we have merely studied four states in the north of Europe that are not particularly representative. But we claim that out study is of general interest. The picture we describe is both more extensive and more coherent than in many of the studies previously made of Europeanization. The Nordic states constitute in this case an outstanding laboratory for comparative studies.
2
Transnational states
Introduction The aim of this chapter is to describe transnationalization. We will look at the ways transnationalization manifests itself in the central administrations of the Nordic countries. We start by discussing what we mean by the term before going on to describe the empirical findings on various dimensions: level of affectedness, contact and participation patterns with EU organizations, independent judgements, and relations to interest organizations and private companies. The chapter concludes by reviewing the principal aspects of transnationalization.
What is transnationalization? The main thesis of this book with regard to the description of how central administrative apparatuses adapt to increased European integration is summed up in the concept of transnationalization. Transnational relations are understood as regular interactions across national borders in which either the administration itself or the actors with whom the administration maintains contacts act without a specific and clear national mandate when participating in negotiations and decision-making processes. Transnationalization encompasses intergovernmental relations that develop directly between state administrations and the European Council and supranational bodies, for instance through interaction with the Commission and its subsidiary committees, but also through relations with actors outside the central administrations and EU institutions, such as interest organizations, private companies and standardization organizations. In addition there are patterns of internal interaction that have developed within the respective administrative apparatuses in response to increased
28
Transnational states
European integration (Esmark 2002a). A characteristic feature of transnationalization is that its effects tend to propagate so that they are felt not only by civil servants in senior positions and ministerial units, but also further down in the hierarchy, by agencies and relatively low level specialists. Transnational relations usually display a strong aspect of informality and a certain openness and flexibility in their patterns of engagement, although they also tend to need an element of regularity and routine. Thus we use the term transnationalization in a fairly broad sense. In order to find out whether the administrations are becoming increasingly transnationalized as a result of EU work, we will focus on the following dimensions. Firstly, to what extent does the EU affect the administration and how much of its work is concerned with EU affairs? Secondly, to what extent has the administration developed contact routines and patterns of engagement with foreign administrations and with civil servants in EU institutions such as the Commission, the European Council, the European Parliament and the EU Court of Justice? Thirdly, what degree of freedom do civil servants enjoy in using these channels of communication? Are they given clear, precise, written signals from their political leaders, or do they have to depend on their own judgement? Fourthly, to what extent does interaction take place between the administration and interest organizations and private companies in EU-related activities? Do other international organizations and private standardization organizations play an important part in the transnationalization of administration?
Levels of affectedness On the basis of our transnationalization thesis we would expect Europeanization to lead to sweeping changes in national administrative bodies (Jacobsson and Sundstro¨m 1999). Extensive efforts are currently being made to adjust to EU requirements, which leads to greater intermixture and this can be seen in high levels of EU influence, the considerable amounts of time spent on EU concerns, and the major impact of the EU on areas of central concern (Lægreid 2000). Those who work in the field do not share the idea that the EU exerts only a marginal influence on central administration. Civil servants who speak on behalf of the various units report that the EU has led to extensive and profound changes in many areas. Although certain activities and administrative fields are more affected than others, the
Transnational states
29
EU exerts a very broad influence on state administration. Only 13 per cent of units claim to be unaffected in any way by those aspects of the EU contained in its Three Pillars or by any related work of standardization.1 The feeling of being affected is greater in the new member states, which are in the midst of a period of adjustment and adaptation to the EU, than in Denmark, where it is likely that EU requirements have already been more closely integrated into administrative routines. We should, however, also note that Denmark has negotiated a fair number of exemptions from EU requirements. The EU has, therefore, had an impact on most central administrative units in the Nordic countries. With regard to the EU’s Pillar 1, which concerns the internal market, economic and social questions, more than half of their activities are affected (Table 2.1). This is hardly surprising, given that this is the pillar in which the EU’s exclusive authority is most clearly enshrined. The central administration of Sweden and Finland is affected to a greater extent by their membership of the EU than in Norway and Denmark. This is especially true with regard to judicial and police cooperation, which is no doubt due to the derogation that Denmark has negotiated in this area and to the fact that these concerns are not covered by the EEA agreement, but it is also true for the elements of Pillar 1. In all four countries the overwhelmingly dominant impact is felt in relation to the internal market concerns linked to Pillar 1. With regard to foreign and security policy, and judicial and police cooperation, only a few units in each country acknowledge any impact. In the years since 1998 there have, however, been considerable developments in the EU’s involvement in these fields, and one can expect the focus on Pillars 2 and 3 to increase in forthcoming years. Some of the units are considerably affected by standardization concerns, for instance, through their involvement with various European standardization organizations. These are private organizations financed Table 2.1 Have the unit’s activities been affected to a significant degree by the EU/EEA agreement in the following areas, according to country? (1998, per cent) Norway Internal market 48 Foreign policy, security policy 5 Judicial and police cooperation 9 Standardization 17
Denmark
Sweden
Finland
Overall
50 6 9 15
58 8 17 18
61 9 22 23
54 7 14 18
30
Transnational states
primarily by the companies responsible for developing standards, which also participate in their various committees, and by project grants and EU funds. The main task of these organizations is to determine the standards to lay down products and production processes. Since the late 1980s the EU has delegated much of this regulatory work to such organizations, and this has posed major challenges to national administrative structures in the Nordic countries (Vad 1998). Rather than develop a directive for a certain area, the EU often chooses to refer to standards, or it proposes that standards should be developed (Brunsson and Jacobsson 2000). Standardization results in sets of rules that are distinct from directives, insofar as they are voluntary, and from norms, insofar as they are explicit, and can be clearly attributed to certain authors. Standardization requires a considerable investment of effort and is of major relevance for many areas of policy, such as domestic environmental policy (Frankel, Højbjerg and Pedersen 1998). In the field of work environments and product safety, for example, the formulation of the machine directive involved no fewer than 40 technical committees and 200–300 workgroups (Middelthon 2000: 103). Another indication of how closely administrations are intertwined with the EU is the amount of time spent on EU-related work. More than a third of the central administrative units in Nordic countries devote a large proportion of their working hours to such concerns (Table 2.2). The percentage ranges from around 40 per cent in Finland and Sweden to 36 per cent in Denmark and 23 per cent in Norway. In interpreting these figures we must remember that in 1998, the year in which they were collected, Finland was preparing to assume the presidency of the EU in 1999, and that Norway was pursuing a cautious and hesitant policy towards the EU. In the case of Sweden the figure might indicate that that country’s new membership together with its great ambitions made considerable demands on administrative time, whereas Denmark had by now developed effective routines for handling EU concerns so that they consumed less time than in the new member countries. In general, the EU has had a considerable impact on central administration in the Nordic countries (Table 2.3). These effects and the amount of time required have significant consequences on how administrative units work. Almost half the units report very or moderately high levels of impact in their respective fields of activity. In Norway the proportion is less than one third, whereas 56 per cent of the units in Sweden and Finland report a substantial impact, a discrepancy
Transnational states
31
Table 2.2 Roughly how much of the unit’s total work time was devoted to EU-related work during the past year? (1998, per cent)
A very large amount A moderately large amount A modest amount Very little
Norway
Denmark
Sweden
Finland
Overall
2 21 46 31
8 28 35 29
15 26 38 22
14 29 33 25
10 26 38 27
Table 2.3 What degree of impact, all in all, has the EU had for the unit’s area of activity? (1998, per cent)
A very high level A moderately high level Neither high nor low A modest level A very low level
Norway
Denmark
Sweden
Finland
Overall
7 24 25 23 20
17 27 18 16 20
24 32 10 15 18
18 38 16 14 13
17 30 17 17 18
that illustrates the difference between EEA affiliation and full EU membership. As a country sceptical to EU integration in the areas covered by Pillars 1 and 2, and which is no longer in the same transitional phase as Sweden and Finland, Denmark falls in an intermediate position. Much of the work of adapting to the EU is concerned with the development of various kinds of regulations; this activity involves either harmonizing national regulations with EU law, or direct participation in the development of new EU legislation. This too has had an impact on the main functions of the various administrative units. For the central administrative bodies, the most important EU-related work consists of the formulation or revision of laws, regulations, agreements, conventions, directives and standards. One out of every three units refers to the development of regulations as its principal task with regard to the EU (Lægreid 2001a). In this respect there is little variation between the countries, which suggests that this activity has more to do with the implementation of EU legislation within the framework of national law than with the work of formulating regulation within the EU itself, in which Norway’s participation is more restricted. This interpretation is also supported by the fact that 37 per cent of the units in Norway, Sweden and Finland report that, as a result of the EEA agreement, considerable or very considerable changes have been made
32
Transnational states
to existing regulations through the introduction of new rules affecting the department’s area of activity. In addition to changes in regulations, the principal functions of many units include coordination and negotiation. One in four of the units stresses the importance of its coordinating role in its EU work. Such responsibilities are less widespread in the Norwegian central administration, due to the fact that Norway does not participate in EU decision-making processes within the Council system. Thus the core of EU-related work consists of classic political tasks such as regulation, legislation, coordination, negotiation and the spread of information. The view that EU work primarily involves dealing with single issues gains little support from this material. An important difference compared with the past is that a significant fraction of the legislative and coordination work concerns the adjustment of national arrangements to meet the requirements emanating from Brussels, and this means that factors beyond the direct control of the respective countries become increasingly decisive for their activities. EU membership and EEA affiliation imply dynamic relationships involving a continuous process of change. This means, for example, that account must be taken of how long a country has been a member or signatory and the point in time at which the study was carried out. The late 1990s was a period of very active EU cooperation characterized by the implementation of the internal market and increased integration following the Maastricht Treaty of 1993, which broadened the possibilities for transnationalization (Frankel 2002). One way to illustrate this ongoing adjustment is to investigate how legislation has been amended in response to the EU. Thirty-seven per cent of the central administrative units in the Nordic countries report considerable or very considerable levels of amendment to existing legislation or that new rules affecting the department’s area of activity have been introduced as a result of EEA affiliation or EU membership since the relevant agreements were made. This means that legislation has been subject to the same degree of amendment in the period following the change to the new form of affiliation as they were in the run-up to membership. Evidently the units see the EU not as something that calls for a single process of change, but as something that involves the ongoing adaptation of regulations relating to their various activities. National legislation is increasingly integrated with EU requirements, and in the Nordic countries it is becoming ever more Europeanized.
Transnational states
33
If we look beyond legislative change and ask how much greater the total impact of the EU was in 1998 compared with four years earlier, the impression of the dynamics of EU integration is strengthened still further. In 1998 over 70 per cent of the central administrative units in the Nordic countries were more affected by the changes than they had been four years previously. The trend is the same in all the countries, although the intensity of the relevant processes varies between them. This is even clearer in terms of this particular indicator. Whereas over half the central administrative units in Sweden and Finland were much more affected in 1998 than four years previously, the corresponding percentage in Norway and Denmark is under 20 per cent (Table 2.4). Here we see a clear reflection of the form of affiliation and of how long the affiliation has been in effect. Central administrations experienced a higher degree of EU impact as a result of becoming new members than following affiliation to the EEA, whereas the administrative apparatuses of countries that had not changed affiliation experienced less change over the preceding four years. We find corresponding differences with regard to the proportion of working hours devoted to EU-related tasks. This clearly suggests that a shift from affiliation to full membership had considerable consequences for the ways central administrations function compared to the impact of EEA affiliation. A further indication that adapting to the EU resulted in a dynamic period for the Nordic countries in the 1990s is that travel to foreign countries in connection with EU affairs increased for 31 per cent of the units, especially during 1998. Here, as well, the intensity differs from one country to another. Finland tops the list with 41 per cent, followed by Sweden (37 per cent), Denmark (24 per cent) and Norway (19 per cent).
Table 2.4 Is the unit more or less affected by the EU today than it was four years ago? (1998, per cent)
Much more affected Slightly more affected Affected to the same degree Less affected Don’t know
Norway
Denmark
Sweden
Finland
Overall
18 43 33 4 3
14 42 37 2 6
52 38 9 1 1
51 34 12 0 2
34 39 23 2 3
34
Transnational states
Contact with and participation in EU organizations In this study the term contact is used to cover both formal and informal, oral and written communication and interaction. When we look at the intensity of administrative contacts with EU organizations, a clear picture emerges. The crucial relationship for all the countries is with the Commission and the expert preparatory committees. Both the extent and the intensity of contacts with this aspect of the EU system have changed significantly. Among the affected units between half and three quarters were in contact with the Commission’s Directorates General at least a few times in the course of 1998. In Sweden, Finland and Denmark, 40 per cent were in contact with the Commission on a monthly basis or more often, whereas the corresponding figure for Norway is 21 per cent (Table 2.5). This is followed by contact with the Commission’s preparatory or expert committees. Contacts with the European Council and its subsidiary offices and committees come third, and are significantly less extensive than those with elements of the Commission. There is also a certain amount of contact with the comitology system, although less than with the European Council system. Table 2.5 Contact with EU organizations once a month or more often (1998, per cent) Norway EU Commission/Directorate General 21 Preparatory or expert committees of the EU Commission 14 Comitology or implementation committees 5 European Council/Coreper, with subsidiary working groups or committees 1 EU Parliament or subsidiary working groups 1 EU Court of Justice 1 Other EU organs (e.g. the Economic and Social Committee, Committee of the Regions, EEA, EMI) 1 EEA and EFTA organizations 18
Denmark
Sweden
Finland
Overall
41
43
40
37
22
31
36
26
8
14
15
11
22
24
18
17
2 2
2 3
7 2
3 2
4 1
9 2
4 1
7 6
Transnational states
35
There are also notable national variations in patterns of contact. Although Denmark’s level of contact with several EU institutions is comparable to that of Sweden and Finland, on the whole the former has both broader and more intensive contacts than either of the latter. Norway stands out as an ‘outsider’ in these figures; clearly it seeks contact via the channels that are open to it and maintains a relatively intense level of contact with the Commission system and the comitology institution, but not to anything like the extent found in member states. Even so, the national variations do not mask the obvious fact that all the countries concerned have fairly intense contacts, primarily with the Commission system. For all four administrative apparatuses, among the EU organs contact with the EU administration is the preferred choice. This contact exists not only with the committee and the comitology system, where the various national representatives are summoned by the Commission, but to an equal extent with the Commission’s Directorates General. In all four countries there is a significant amount of activity between the administrative apparatuses and the Commission as the EU body responsible for initiatives, decision making and implementation. Cooperation with the Commission can take several forms, from attempts to promote large-scale strategic national interests down to routine clarificatory questions and requests for information (Esmark 2002b). A characteristic feature of this contact is that it is not as formalized as participation in committees and the comitology system. Participatory activity is highest in the preparatory bodies. Again, Sweden and Finland turn out to be the most active in this regard. Sixtytwo per cent of the affected units in these two administrations participated in at least one meeting of a preparatory committee in 1998. The corresponding figures for Denmark and Norway are 52 per cent and 41 per cent respectively. Monthly or more frequent participation in preparatory committees is not so common, although the variation between the countries is the same (Table 2.6). Participation in comitology is significantly lower. Only 6 per cent of the units in the Nordic countries participated on a monthly basis or more often. This shows that the national administrations are especially active during the phase when initiatives are being formulated and up until the Commission drafts a proposal. It is difficult to obtain clear figures for the number of committees, but there exist between 60 and 100 permanent expert committees and about the same number of ad hoc expert committees. On top of this there are some 300–400 comitology committees (Nedergaard 2000, Esmark 2001).
36
Transnational states
Table 2.6 Participation in meetings with EU organizations once a month or more often (1998, per cent)
Participated in a preparatory/ expert committee of the European Commission Participated in a comitology committee Participated in Coreper or in working groups of the European Council/Coreper Participated in other EU organizations
Norway
Denmark
Sweden
Finland
Overall
7
15
21
26
17
2
3
11
7
6
0
14
17
15
12
3
10
13
11
10
Meetings of the European Council take place according to a fixed timetable and the uniform pattern of low participation by the member states and Norway is, therefore, what one would expect. Thus we see that the working methods of the EU bodies allow considerable variation, and here again it is the Swedish and Finnish administrations that are most active. Participation is most extensive in the Commission system, the comitology committees and the Council system, although there are clear differences from country to country. Norwegian participation is clearly lowest and Finland stands out with a particularly high level of participation in committees linked to the Commission. Despite the national variations, contacts and participation are compact, in the sense that it tends to be the same administrative units that maintain relations with the various EU organs. The national administrations also participate in EU activities through the secondment of national experts. In 1998 there were more than 700 national experts working in various EU organizations. Unlike the civil servants of the EU itself, these representatives are still employed by the respective national administrations and are seconded to the EU agencies because of their expertise in certain fields. The loan of national experts is one of the few areas where national variations are minimal. Just under a third of all units in the four national administrations seconded 1–2 national experts to EU institutions over a four-year period. The high levels of participation in various committees and expert groups, observer programmes and informal contact networks require
Transnational states
37
considerable travel back and forth between the respective countries. In the course of 1998 almost 60 per cent of the central administrative units had 10 or more days of foreign travel in connection with EUrelated work. This travel activity is least obvious in Norway and most common in Sweden and Finland (Lægreid 2000). The principal impression is, therefore, that contact with and participation in EU organizations is on the whole broad and intensive. Not only did the Swedish and Finnish administrations experience the most intensive periods of adjustment and the highest levels of impact in the years 1994–98, it was also these countries that had the most intense contacts with EU agencies in the course of 1998. They are followed by the Danish and Norwegian administrations. Norway sought contact and participation where the EEA agreement permits, although its involvement via the various channels that are open to all the countries was also less than that of the other administrations. Despite national differences the figures still show that a significant proportion of the administrative units in Sweden, Finland and Denmark, and slightly fewer in Norway, have stable contacts with and frequently participate in EU organizations such as the Commission system, comitology committees and the European Council. To varying degrees they also have contacts with organizations such as the EU Court of Justice and the European Parliament. Given the breadth of our definition of transnational relationships, we can characterize the relations with the Commission system and the comitology agency as transnational. But there is also a whole range of informal contacts with the European Council system, many of which also show transnational characteristics. Through their involvement in the preparatory expert committees the administrations contribute to the generation of specialized knowledge, participate in negotiations concerning the Commission’s proposals and thereby influence the content of those proposals (Egeberg and Trondal 1997, Højbjerg 2001, Esmark 2002b). Through their participation in administrative and regulatory committees they contribute to the content and implementation of legislation that follows from Council decisions. It is reasonable to assume that in dealing with these tasks, each administration will seek to safeguard its respective national interests, despite the fact that its operations are not supported by a national mandate in the narrow sense. Also where Council work is concerned, which is in fact generally governed by national mandates, various transnational relations develop as a result of participation in working groups and in Coreper.
38
Transnational states
Some national administrations try to persuade the Commission to take up central issues, and in this respect their capacity to influence the Commission and to determine its agenda is crucial in upholding national interests (Pedersen 2001). These efforts are aimed primarily at the Directorates General but also involve the Commission itself. But contacts between the four administrations and the Commission system are hardly restricted to such strategic attempts to influence the agenda and initiatives of the Commission; they also encompass straightforward, routine inquiries for clarification. The results of the current study do not allow us to distinguish more specifically between these two aims, but we can conclude that transnational relations with the Commission system are without parallel the most intensive of all those maintained by the administrations in their implementation of EU-related activities. Patterns of contact and participation also reflect varying degrees of proactive or reactive activity. On average one in four of the central administrative units affirmed that they had taken a proactive stance in the course of the foregoing year by taking the initiative within the EU to promote suggestions for new resolutions or rules (Esmark 2001). This indicates that the units are generally more reactive than proactive in their transnational relations with the EU. The Swedish and Danish units stand out as the most active in promoting their own initiatives within the EU. It can be seen that highly intensive contacts and participation need not necessarily result in greater influence and impact in European cooperation. As far as the Swedes and the Finns were concerned, European cooperation in 1998 was still undergoing a period of trial and adjustment, whereas the Danish administrative units had sufficient experience of membership to have more or less mastered the rules of the game. This means that despite less intensive contacts and participation, it was Denmark that reported greater success rates in promoting its interests within the EU. With regard to the promotion of national interests, these units claimed the highest success rates, on their own assessment, in gaining acceptance in the EU bodies for their views and wishes. For example, half the units report that they were successful or very successful in dealing with the Commission, the Directorates General or expert committees, whereas the average for all the units was 34 per cent. All in all, the Danish units are both proactive and relatively influential. The Swedish units are also fairly proactive, although they report less success in achieving acceptance of their views. The Finnish
Transnational states
39
units are more reactive, but are more successful than the Swedes. Norway is generally reactive and has also had little influence (Esmark 2001: 127).
Independent judgement One way to investigate whether or not civil servants act on the basis of their individual judgement in transnational relations is to ask what kind of instructions they receive from their superordinate authorities and the political level when participating in various groups and committees within the Commission or the European Council. It is common for representatives to manage without any clear instructions from higher administrative levels when participating in EU committees or working groups. Almost three out of every ten units consider this to be the case. Even so, they rarely travel to Brussels without any kind of guideline at all; in nine cases out of ten they receive some kind of guidance from their respective units (Sundstro¨m 2001). General and oral directives are more common than precise and written ones, which suggest that EU work involves a significant element of informal activity. There are only small differences from one country to another with regard to guidelines. It is rarer for civil servants in Norway and Denmark to receive precise instructions from a higher administrative level than their Finnish counterparts. The main finding in Table 2.7 is that civil servants who take part in workgroups and committees within the Commission and the Council only rarely have the advantage of clear and precise guidelines from the political level, meaning that the transnationalization requirement that agents act without clear national mandates seems largely to be satisfied. The Norwegian administration seems to receive relatively little attention from political leaders. This might indicate that Norwegian politicians view EU activities as something they have little chance of influencing. The Norwegian administration participates first and foremost in matters relating to the internal market but is unable to contribute to decisions made by the European Council. Another explanation might be that, aware of the broad scepticism towards the EU among the population, Norwegian politicians are only too happy to leave contact with the EU to the administration. At the time when the survey was conducted, Norway had a government with a No-to-the-EU stance, which may have contributed to this apparent passivity.
40
Transnational states
Table 2.7 If, during the past year, the unit’s employees have participated in working groups or committees within the Commission or European Council, what kind of instructions have they usually received? (1998, per cent) Norway
Denmark
Sweden
Finland
Overall
Instructions from higher administrative level: None General Precise Written Oral
33 56 15 20 36
29 57 19 29 31
28 57 22 23 49
21 54 30 27 40
27 56 23 25 40
Instructions from political (government, cabinet) level: None General Precise Written Oral
48 45 10 17 11
42 50 17 26 8
29 60 28 23 38
61 27 11 15 14
42 48 19 21 21
The Finnish administration is even more independent of political involvement than the Norwegian. For them it is simply exceptional to receive instructions or guidance from politicians. Six out of ten Finnish units receive no directives from the political level, and also in other respects, the impression is of an administration left to its own devices insofar as work with the EU is concerned. Perhaps there is a reluctance to associate the existing contacts with the political levels with anything that might be regarded as ‘direction’. One interpretation might be that politicians are generally prepared to let the administration do what it thinks best while being willing to intervene on specific issues, although only very selectively. It is also possible that the Finnish administration is adept at anticipating the views of its politicians. On the other hand, there are ways of coordinating policy that do not require direct signals from politicians. For one thing, the appointment of civil servants with a particular party line is common in Finland. For another, there exist various informal mechanisms for coordinating policy, such as the so-called ‘Evening school’, where bureaucrats and government members meet to discuss common concerns (Selovouri 1999). But despite the possibilities for political control, it is still possible to describe the Finnish administration as relatively independent
Transnational states
41
of its politicians with regard to EU concerns. Finland stands out as a country in which few units refer to a high degree of guidance from the government or cabinet in relation to their activities with the EU. Also in Denmark it is fairly common to act without the involvement of politicians, even if less so than in Finland and Norway. The Danish administration reports that it frequently fails to receive general instructions from its politicians. Written guidance is, however, clearly more common in Denmark than the other countries, which might be linked with the existing formal system for coordination. By comparison, oral instructions alone from politicians are rare in Denmark. The formal coordination system reduces the need for direct control of the administration. One outcome is that most Danish units say they find it easy to know what politicians want in relation to EU matters, even though their contacts are not especially close-knit. Sweden is notable as the Nordic country where EU work is most highly permeated by politics. It is uncommon for the units to have no instructions from political sources, and there are a greater proportion of oral guidelines than elsewhere, which is linked to the high level of informal contact between the Swedish administration and the country’s politicians. Even so, these active contacts do not imply that the political assessments of the government and the cabinet are accorded exceptional weight compared with the other Nordic countries (Jacobsson 2001: 235). This can be taken as an indication that much of the political control in Sweden is of a fairly ritual nature. The differences between the countries also become apparent when we distinguish between various levels of administration (Table 2.8). In Sweden the political leadership emerges as particularly active in guiding EU activities, and this applies also to the administration’s contacts with the EU. In no other country do the central agencies receive so much informal policy guidance. The distance between politicians and ministerial employees in Sweden is indeed fairly small. The passivity of the Finnish political leadership can be noted in relation to both ministerial units and central agencies. In Denmark, the system for the written transmission of policy guidelines to ministerial units is far more extensive than in the other countries. The relation between the administrations and politicians is also complicated by the fact that the time scales used by the EU do not always correspond to those used in the various countries (Ekengren 1998). Especially in Sweden, where there are high levels of contact between the administration and political leadership, tight schedules
42
Transnational states
Table 2.8 If, during the past year, the unit’s employees have participated in working groups or committees within the Commission or European Council, what kind of instructions have been given by political sources (government, cabinet) to ministries as distinct from central agencies? (1998, per cent) Norway
Precise Written Oral
Denmark
Sweden
Finland
Min.
C.A.
Min.
C.A.
Min.
C.A.
Min.
C.A.
15 28 23
5 10 3
29 42 19
13 19 3
46 20 61
21 25 29
17 22 13
10 14 14
Key: Min = Ministries; C.A. = Central agencies
complicate the work of preparing matters for presentation to the government and the political leadership.
Interest organizations and private companies Transnational relations also encompass private actors, and we will now examine patterns of contact and participation in EU-related activities involving private companies and interest organizations. First, we will focus on contacts between the administrations and private interests within the various countries. In Denmark the most intensive contacts are with interest organizations, whereas in Finland the contacts tend to be more with the industries themselves (Table 2.9). In Sweden contacts are more evenly balanced between companies and associations. The Norwegian units have the least intensive contacts with companies and interest organizations. This pattern of variations recurs when we distinguish between forms of contact. In Finland and Sweden most contacts are informal. In Finland only 12 per cent claim to have contacts of a formal nature. The figure for Sweden is 32 per cent. By comparison, 64 per cent of the Danish units and 52 per cent of those in Norway reported formal contacts in terms of written communications, meetings and hearings. In other words, contacts with private interests are most formal in Denmark, closely followed by Norway, and most informal in Finland, followed by Sweden. Contact with broader European interest organizations is far less extensive than with their national equivalents, yet still significant. This
Transnational states
43
Table 2.9 How often in the course of the past year have employees of the unit been in contact with interest organizations and private companies in relation to EU matters? (1998, per cent) Norway
Denmark
Sweden
Finland
Overall
Once a week or more interest organizations private companies
5 11
13 11
12 14
5 18
9 13
On a monthly basis interest organizations private companies
16 11
22 12
17 14
18 15
18 13
Just a few times interest organizations private companies
43 38
29 33
43 31
39 35
36 34
Never interest organizations private companies
36 40
36 44
36 41
38 32
37 40
is especially true in the case of Sweden, where approximately half the units had had at least some contact with interest organizations elsewhere in Europe during the preceding year (Pedersen 2001). The relations with private interests indicate that the national administrations do not work alone in formulating and approving the standpoints to be taken in European decision-making processes. The central administrations in all four countries have extensive contacts with the private sector in connection with EU activities, and it is, therefore, given some influence in this context. This is especially true of the member states, while slightly less so for Norway. Regular contacts are not the only means by which the private sector is drawn into EU procedures. It may also be the result of representation and participation in public commissions of various kinds. These might take the form of special committees set up to coordinate and improve ways of safeguarding national interests within the EU, as is the case in Denmark, or they might be public commissions, work groups and coordinating committees. There are significant differences in how the various countries have organized their coordinating systems, with regard to the private sector, and in the extent to which the respective
44
Transnational states
administrations make use of public commissions and ad hoc groups in their handling of EU affairs. Sweden and Finland have to a greater extent than Denmark and Norway established work groups and commissions to deal with EU issues. Significantly more units in the two former countries have taken part in such groups and commissions than in Denmark and Norway. In Denmark only 14 per cent of the units participate in commissions and groups of this kind on a monthly basis, whereas in Finland the figure is 66 per cent and in Sweden 47 per cent. On the other hand, a greater number of the Danish units have contacts with representatives from interest organizations (Table 2.10). This suggests that many groups and commissions in Sweden and Finland are used to achieve coordination between state authorities, whereas in Denmark and Norway they are more mixed and include representatives from both the agencies and the private sector. This would also explain why the units in Sweden and especially Finland regard the relevant forms of contact as more informal, while at the same time claiming that work groups and commissions play a significant role in handling EU-related matters.
Table 2.10 If the unit has participated in public commissions or national work groups in relation to EU work, to what extent have interest organizations and private companies been represented? (1998, per cent) Norway
Denmark
Sweden
Finland
Overall
To a high degree interest organizations private companies
14 10
25 3
18 9
6 6
16 7
To a moderate degree interest organizations private companies
30 30
29 16
31 27
22 18
28 22
To a small degree interest organizations private companies
16 22
9 17
19 30
32 51
19 31
Not represented interest organizations private companies
39 39
37 63
33 34
38 23
36 40
Transnational states
45
Sweden and Norway allow both interest organizations and companies a greater degree of representation than do Denmark and Finland. In Denmark there is a clear tendency to favour the former, whereas in Norway and Sweden there tends to be more representation of the latter. In Finland neither interest organizations nor companies are engaged to anything like the extent they are in the other countries. Sweden has the most open administration in the sense that the total involvement of the private sector is highest there. This Swedish openness is based on a combination of informal contacts and participation. If we also include independent experts as a third category of external interests, further differences then become apparent between the countries. Expert representation in EU-related workgroups and ad hoc groups is most common in Norway, and least so in Denmark. In Norway, 40 per cent of the units claim to have met some or many independent experts in the various organs working with EU affairs. The corresponding figure for Denmark is only 20 per cent, while for Sweden it is 36 per cent and for Finland 25 per cent. These figures show that interest organizations are generally better represented than independent experts, but that experts are better represented than companies. Experts and companies are now challenging the traditional representation of the interest organizations in collegial organs. Altogether, 28 per cent of the administrative units in the Nordic countries participated in ad hoc groups in which interest organizations were represented to a large or moderate degree, whereas the corresponding figures for experts and companies are 22 per cent and 18 per cent respectively. Only 36 per cent had experienced situations in which the interest organizations were not represented, whereas 40 per cent report the same for experts and 51 per cent for companies. In Norway, independent experts and the larger interest organizations are given preference over private companies. In Denmark the organizations are clearly better represented than the companies, and the latter better than experts. Finland is relatively open to private enterprise. Sweden is the country with the broadest involvement of external actors such as interest organizations, private companies and independent experts. On the basis of these figures we can conclude that none of the national administrations takes sole responsibility for the formulation and approval of the standpoints to be taken in European decisionmaking processes. All in all there can be little doubt that the four
46
Transnational states
countries have chosen to give the private sector a voice in their handling of EU-related matters, both through representation in public commissions and in terms of routine contacts.
Transnational states In summary, transnationalization is apparent in the significant level of the impact of the EU on national administrations and in most of their areas of activity, while at the same time highly stable relations have been developed in terms of regular contact patterns, rights to participation and actual involvement in a range of EU bodies. The Nordic central administrations are not closed to influence from Europe, but are involved in extensive interactions. It would be wrong to claim that stability and the status quo determine administrative development. The pattern that emerges does not support the idea of a uniquely Nordic administrative regime based on a single model of administration in which the various units enjoy considerable autonomy in relation to the EU. Transnational relations clearly reach beyond the forms of intergovernmental communication entrenched in the practices of the various foreign ministries; they extend outwards into the various ministries and downwards into central administrations, including directorates and agencies. There can be no doubt that the trends revealed by this study are evidence of significant transnationalization of central administration in the Nordic countries. In 1998, the vast majority of the central administrative units in the four countries were more or less affected by the EU within their respective fields of activity. Within the administrative apparatus a certain amount of organizational and staff restructuring has taken place, and there has been considerable activity in terms of further training, while the distribution of responsibilities has also been altered (Lægreid 2001b). All four countries have established special collegial groups to oversee coordination. The administrations use a significant proportion of their time on EU-related concerns and are materially affected by the EU within their field of activity. This is especially true of issues relating to the internal market, which means that the EU has considerable implications for the activities of the majority of departments. EU activities tend to concentrate largely on the development and modification of rules and regulations, but are also manifest in extensive coordinating measures and through increased emphasis on control and supervision (Lægreid 2001b).
Transnational states
47
Among the various European agencies and bodies, the most important relationships are with the European Commission, its subsidiary expert and workgroups, the comitology committees, and with the European Council and its subsidiary workgroups. Also significant, though not equally so, are relations with the EU Court of Justice, the EU Parliament, the Economic and Social Committee, the Committee of the Regions, and the European Free Trade Association (EFTA). In addition there are other international organizations and standardization committees. Extensive contacts with and participation in various Brussels agencies are developing, especially with the Commission and its subsidiary organs. Most of the contacts are formal, although some are clearly informal. Many units have also seconded employees as national experts to the EU bodies, and there is extensive travel backwards and forwards to Brussels. Much of this activity is, however, reactive and many of those involved in it are uncertain about the ability of their respective units to exert an influence on EU organizations (Esmark 2001). Furthermore, many of the European and international relations that the administrative units are involved in are characterized by weak, ambiguous, and unclear political mandates and directives from their political leadership. Those who work in the transnational arena often have broad powers and are free to exercise considerable professional discretion. The guidelines they follow usually embody the tradition of their respective administrative institutions and are built on professional norms. This heightens the risk of the emergence of a network of bureaucrats and experts who shape the content of public policy largely without being subject to the influence and control of elected representatives. The likelihood of this happening is greatest in areas that require considerable technical knowledge and professional expertise. One mechanism that would reduce the danger of such a development would be a well-functioning coordination system within the central state apparatus. In relation to political leadership, the administrations are generally characterized by broad freedom of action in the way they handle EU affairs. Contacts between politicians and the administrations are of relatively low intensity, although there are exceptions. The contacts tend to be informal and many units receive little if any policy guidance when participating in various international EU bodies. Many find that tight time schedules make it difficult to consult their political leaders before taking decisions. All in all, the overall picture that emerges
48
Transnational states
suggests that, with regard to EU affairs, political leadership is rather passive. Most administrative units succeed in getting their views accepted by the government and political leadership. The majority also find it relatively easy to recognize what stance the political leadership wishes to adopt on various EU matters. The administrative units report that EU affairs very seldom provide grounds for conflict, which suggests either that these issues are largely entrusted to specialists and experts, or that politicians have developed other arenas in which to work with EU questions. With regard to interest organizations, there is a trend to continue the Nordic practice of according them broad rights to participate. These cooperative relations are well developed in terms of both contacts and participation in various collegial organs of a more or less formal nature. In addition professional organizations and private companies often bypass the national administrative channels and establish direct relations with the institutions in Brussels.
3
Segmented states
Introduction In this chapter we will describe how the transnationalization varies in the administration. We do this by examining three supplementary empirical theses concerning segmentation, coordination and multiplier effects. We argue that transnationalization is differentiated by being more comprehensive in some policy areas than in others; by posing new challenges to coordination processes; and by reinforcing established relations and contact patterns. Each of the main sections will begin with a discussion of how relevant terms are to be understood before going on to describe the empirical findings on the various dimensions. The chapter will end with a summary of how transnationalization can be specified.
Segmentation What is segmentation? Transnationalization is not equally prevalent in all branches of central administration, and we, therefore, need to ask whether there is any evidence of segmentation of the transnational relations. Are aspects of transnationalization such as involvement in European cooperation or frequent contacts with private companies and interest organizations only found in certain branches of national administration? Is the transnationalization of national administration most common in those areas of EU activity that are best developed, for instance in the fields of agricultural policy and the internal market? In this chapter we will investigate the degree to which such involvement is greater in certain segments of administration than in others. If it turns out that some aspects of administration are more deeply involved in European affairs
50
Segmented states
than others, then we must ask whether there is a risk of EU work leading to the fragmentation of state administrations. This risk involves the possibility that certain elements within the state might develop contacts with elements of other states and with European organizations that are closer than those they have with other sections of their own administration. EU administration and core segments It is by no means so that all administrative units have the same level of contact with or participate to the same extent in EU organs. This is hardly surprising, given that European cooperation does not extend into all areas of national policy. Some areas have been involved in European cooperation since as early as 1957, whereas many others remained unaffected until the 1990s. There are, therefore, significant differences between those elements of the administration affected by the EU and those that are not. It is apparent that the proportion of units that feel the impact of European integration declines rapidly when we shift from the higherlevel evaluations of involvement and impact and look instead at indicators such as time expenditure, frequency of contacts, and participatory activity. In other words, we can distinguish between degrees of transnationalization, and hence between a broader EU administration and a core segment contained within it. The EU administration is that part of the administration subject to a great deal of impact from the EU with regard to one or more of its Pillars or in terms of standardization work, and which is in contact with EU organizations once a month or more often. Within the EU administration we can identify a core segment of units that not only handle EU-related issues and are frequently in touch with EU organizations, but which also participate on a routine basis in the activities of those EU bodies as such. Thus 41 per cent of the Nordic administrative units qualify for the EU administration category (Table 3.1). This is largest in the Swedish administrative system, with Finland taking second place. EU administration is smallest in Norway. All units within the EU administration handle EU-related activities, albeit to varying extents. The proportion of units that use a large or very large proportion of their time on EU affairs is, however, much smaller than the EU administration as a whole, and this implies the existence of a core segment within the EU administration.
Segmented states
51
Table 3.1 EU administration and the core segment (1998, per cent) Norway
Denmark Sweden
Finland
Overall
EU administration: Affected by Pillars 1–3 or by standardization to a high degree and in contact with at least one EU body at least once a month 25
40
52
46
41
The core segment: Affected by Pillars 1–3 or by standardization to a high degree and participating at least once a month in meetings with at least one EU body, and using a large or a very large proportion of working hours to EU issues
17
27
23
18
7
As already stated, the core segment can be identified not just in terms of how much EU-related work it has to deal with. What we mean by the core segment is a group of units which, in addition to being more or less affected by cooperation, also take part on a monthly basis or more often in meetings at one or more of the EU organs. In total the core segment encompasses 18 per cent of the units in the Nordic countries. Relatively speaking, the segment is largest in Sweden and Finland and only very small in Norway. The core segment is distinct from the EU administration in terms of two crucial parameters. Firstly, the units that it encompasses devote a large proportion of their working hours to dealing with EU matters, and secondly, they participate actively in the work of the EU institutions themselves. The difference between the EU administration and the core segment is, therefore, a matter not only of the amount of EU-related work, but also of participation. Segmentation involves, on the one hand, a division of responsibilities between administrative units that are unaffected by Europeanization and units that are to some degree affected, and on the other, a division of labour between the units that belong exclusively to the EU administration and those contained in the core segment. The EU administration is characterized by having transnational relations, but not necessarily very frequently or in ways that require direct participation. The core segment devotes a large part of its time to EU-related matters and has close and intense relations with EU organizations.
52
Segmented states
Which units belong to which segments? Not surprisingly specific policy areas dominate the various segments. The highest proportion of units belonging to the EU administration and the core segment can be found in the agricultural and fisheries administration. In Sweden, Denmark and Norway, these are the two most Europeanized areas of policy. Respectively, 88 per cent, 62 per cent and 45 per cent of the units in fishery and agricultural administration in these countries belong to the EU administration. A relatively high proportion also belongs to the core segment. The EU administration and the core segment are also strongly represented in foreign affairs, trade and industry and the energy sector. The field of transport and communication is also significantly affected by Europeanization, although in this sector relatively few units have regular contacts with, or participate in EU organs. The weakest representation in the EU administration and the core segment is in the fields of justice and defence. The pervasiveness of transnationalization in any particular policy area has to be viewed in terms of how significant that area is to European cooperation. The field of agriculture is the oldest of the dominant core areas of the community. In addition the EU definition of the agriculture and fisheries sector corresponds well with the way this sector is defined at the national level. It was not until the 1980s that the internal market became the dominant project within the EU. The internal market affects many policy areas, the greatest degree of involvement being apparent in the various ministries of trade and industry. Since the joint act of 1986 the environment has become one of the highest priority fields of EU policy, although here as well there are links to the implementation of the internal market, not least due to the development of environmental standards based on new methods (Højbjerg 2001). In general, Table 3.2 seems to reflect the historical importance of the various policy areas within European cooperation. Of course, this does not apply to the various foreign ministries, which do not have the same relation to the European policy areas as other ministries. For each of the four countries, the position of the respective foreign affairs ministry does, however, illustrate that policy towards Europe is the concern of both the minister for foreign affairs and of other specialized ministers. But the relatively low place occupied by the Norwegian Ministry of Foreign Affairs also suggests that, in Norway, European
Segmented states
53
Table 3.2 Extent of the EU administration and the core segment in various policy areas (1998, per cent) Norway
Denmark
Sweden
Finland
EU adm.
core seg.
EU adm.
core seg.
EU adm.
core seg.
EU adm.
core seg.
Employment Trade and energy Finance Defence
12 28 24 17
4 7 10 0
33 54 41 39
8 21 21 11
50 63 55 15
19 33 28 0
57 56 50 8
43 44 24 8
Home affairs Justice Church, education, culture Agriculture/fisheries Environment Health and welfare
18 5
0 0
12 25
4 13
36 58
12 31
53 29
29 14
33 45 24 19
7 14 12 0
30 62 51 16
7 32 18 13
34 88 71 46
16 71 42 16
44 58 33 54
0 33 17 29
Transport and communication
23
4
39
11
32
19
41
22
Foreign affairs
25
25
55
34
81
44
56
44
Key: EU adm. = EU administrations; core seg. = core segment
politics is not considered a matter for foreign affairs to the same extent as in other countries. Finance, the labour market, communications, and welfare are the sectors most used to acting without the direct guidance of politicians. In core state areas such as defence, justice, and foreign affairs, policy tends to be more integrated. The most precise guidelines are received in the agricultural and environmental sectors. National variations between ministries and agencies are limited. In all four administrations Europeanization tends to be felt more strongly in the ministries than in agencies. But Europeanization is apparent on both levels of central administration and variations are nevertheless greater between the policy areas than they are between the administrative levels (Lægreid 2001b). In relative terms, the EU administration and the core segment are more strongly represented in ministries than in agencies in all four administrations (Table 3.3). This is most clearly seen in the cases of Sweden and Norway. Even so, the agencies are not exempt from EU-related concerns, especially when one considers that the absolute
54
Segmented states
Table 3.3 Proportion of EU administration and of core segment in ministries and agencies (1998, per cent)
Ministries Agencies
Norway
Denmark
Sweden
Finland
Overall
EU core adm. seg.
EU core adm. seg.
EU core adm. seg.
EU core adm. seg.
EU core adm. seg.
43 19
48 36
72 47
49 45
51 37
15 4
26 13
49 21
32 20
28 15
Key: EU adm. = EU administrations; core seg. = core segment
figures for the agencies are considerably higher than those for the ministries. Although centralization is not a fundamental dynamic of EU activities, the level of ministerial involvement rises in proportion to the amount of EU-related duties incurred by the respective units and their participation in the work of the EU institutions. To sum up, we can observe a division of labour in all four administrations. Beside a number of unaffected units, we find both an EU administration with some transnational relations and a core segment of units with close and intensive transnational relations. Not only are the Finnish and Swedish administrations affected most with regard to EU administration and the core segment, they also have the closest and most intensive contacts with EU organizations. These are followed by Denmark and Norway. There are also considerable variations in terms of different policy areas, so that those relating to trade and the internal market are the most transnationalized. Another indicator of segmentation is that less than a third of the units have routine contacts with private interests in connection with EU affairs, whereas the remainder have such contacts only rarely if ever (Pedersen 2001). This reinforces the impression that domestic administration is segmented as a result of EU integration. There is only one segment among these units that has relations with private interests in connection with EU tasks, just as only one segment among them is involved in EU-related activities. Segmented states The segmentation thesis also draws support from this material. Although the administrations are generally involved in transnational relations to a significant degree, there are marked variations between
Segmented states
55
various areas of the administrative system with regard to the strength and breadth of the transnational connections. Whereas some units are more or less untouched by the EU, others are deeply involved in it so that it affects their daily work. Where variations between the sectors are concerned, a dividing line runs between, on the one hand, areas of ministerial concern linked to the older aspects of EU integration contained in Pillar 1, the internal market and the four freedoms, where transnational relations are most comprehensive, and on the other, newer areas relating to the welfare state, foreign affairs and defence policy, and judicial and police cooperation. This means that the most affected ministries are those dealing with trade, agriculture and fisheries, and also transport and communication. Over 60 per cent of the units in the agriculture and fisheries administration can be characterized as EU administration, meaning that they are considerably affected by the EU within their field of activity, while at the same time having contact with EU organs once a month or more often. The agriculture and fisheries sector stands out as the field in which the most far-reaching organizational and staffing changes have been made in the process of adapting to the EU. No ministerial sector devotes more time to EU matters and none reports higher levels of EU impact than this one. This is also the area with the best-developed patterns of contact and participation, in relation both to EU institutions and to interest organizations. This sector is engaged in extensive coordination activities and receives numerous and precise policy signals from the political leadership, which often becomes involved in its own right in the relevant EU concerns. The level of impact is particularly apparent in the changes being made to rules and regulations, an activity that has also increased in recent years. In contrast to agricultural administration, the welfare sectors operate largely without specific policy signals from politicians. The probable explanation for this is that the welfare sector has well-developed contacts with standardization organizations that generally work with technical issues in which politicians are only marginally involved. The effects of the EU on the environmental, energy and trade sectors can be seen particularly in the extensive contacts these policy areas have with interest organizations and private companies, but also to some extent with EU bodies. These sectors are also heavily involved in coordination work and have high levels of contact with the political leadership. Foreign affairs administration stands out as having made especially large adjustments in order to cope with EU work. The days are past
56
Segmented states
when international relations were the preserve of the foreign affairs administration, with the Ministry for Foreign Affairs functioning as a superior coordinating organ through which all contacts and business were channelled (Trondal and Veggeland 1999, Trondal 2000b). Where EU affairs are concerned, trade and industry ministries are, within their field of activity, as much ‘foreign affairs’ ministries as are the various foreign ministries themselves. Matters relating to Europe are largely dealt with by the various specialized ministries, and the role of the foreign affairs ministries has been reduced, even though these are still strongly represented in the core segment and enjoy closer contacts with the political leadership than the units in many other ministerial fields. In the defence and justice sectors, adjustment to the EU has entailed only minor changes in organizational structures, skills, staffing patterns or division of responsibilities. Defence is the policy area that devotes least time to EU-related issues, feels the fewest effects from the EU in its own sphere of activity, has only low levels of contact with and participation in EU organs and only weakly developed coordination activities. The ecclesiastical, education, research and cultural affairs sector is similar in being only minimally affected. Also only marginal in the EU context is the judicial sector, a state of affairs apparent, for example, from the low degree of coordinating activities in this field, the lack of EU significance on the changes that have taken place in ministerial departments and the paucity of changes in the distribution of responsibilities in response to EU affairs. The trend towards segmentation can also be seen in the close interplay between the various factors indicative of administrative adaptation. Where one finds that organizational adjustments have been made in connection with EU work, it is highly likely that one will find adjustments of other kinds as well. If a unit is involved on one dimension, then most probably it has made changes on others, and if a unit has made major organizational adjustments, then it is probably deeply involved in the EU. This means that whereas some sectors develop their own specialized EU units, establish organs of collegial coordination, undertake changes in the distribution of responsibilities, engage new staff, allocate considerable resources to EU-related training, improve necessary skills, are more affected by the EU in their daily activities, devote more time to EU affairs, and experience a greater impact of the EU in their particular field, other units are hardly affected at all on the same dimensions. The consequence is that the
Segmented states
57
various dimensions of adaptation to the EU tend to reinforce one another, and this has resulted in the development of a division within administrative systems between a highly competent high flying EU administration that devotes considerable time to EU affairs on the one hand, and a rather more national administration with only distant links with the EU on the other. The new divisions of labour in central administration resulting from EU involvement can be seen in the crystallization of an EU administration which involves four out of every ten central administrative units. What characterizes these units is the considerable impact of the EU in their particular fields and that they have contacts with EU institutions at least once a month. Within this administration we can also distinguish a core segment that encompasses one unit in every seven. In addition to experiencing a considerable degree of EU impact, these last frequently take part in EU meetings and spend large amounts of time dealing with EU affairs. In a situation in which certain parts of a central administration are deeply involved with Europe-wide concerns, while others have little to do with Europe, the former might end up having more intensive contacts with organizations in the EU than with other parts of their own national administration. This gives rise to special challenges in coordination within the administrative system, and it is these challenges that we shall discuss in the next section.
Coordination What is coordination? Both transnationalization and segmentation constitute major challenges to administrations. For certain parts of an administration the conditions shaping their activities develop transnationally, such that they evolve into circumscribed segments that politicians have little opportunity to influence. A natural response to this kind of development is to increase efforts to strengthen coordination, for example, by enabling various national actors to act in similar ways in a range of arenas, or by comparing and evaluating different areas of policy so as to achieve uniform standards (Kassim, Peters and Wright 2000). In general, coordination can be understood as an activity whereby the behaviour of various actors is consciously related to one another within a common and comprehensive framework (Larsson 1986, Sundstro¨m 1999). We will examine both the extent and the organization of coordination.
58
Segmented states
Coordination can be pursued in various ways. Firstly, it can occur vertically within a particular policy area. Secondly, it can occur horizontally within a state across different policy areas. Thirdly, it can occur at a national level, when private actors also become involved. In relation to the discussion of transnationalization, this effectively implies the development of a pattern of coordination in which public servants from different countries and various European organizations meet, exchange views and experiences, and jointly discuss problems and solutions. While a transnational administration of this kind might enjoy fairly good contacts with private companies and interest organizations, coordination occurs primarily within the framework of the various sectors. Vertical, horizontal and national coordination Many units in the Nordic central administrations report that a considerable amount of their EU work concerns coordination. As an indicator of how relations within the state are changed in practice by closer proximity to the EU, we will examine the responses of the units to a number of general propositions. We asked the units to say whether or not they agreed that it had become more usual for them to define courses of action of common national relevance within their respective fields of activity. This question concerns coordination within the country as a whole and not just within the administration. Forty-five per cent of all the units agree with the statement, whereas a quarter express reservations (Table 3.4). In Norway there are far fewer who believe this kind of national coordination has become more common, whereas in Finland the majority report that it has. Table 3.4 In the course of the past year, it has become more usual to define standpoints of common national relevance concerning EU-related questions within the unit’s field of activity (1998, per cent)
Agree entirely Agree in part Disagree in part Disagree entirely Don’t know
Norway
Denmark
Sweden
Finland
Overall
2 24 20 20 34
6 38 21 1 34
11 35 12 14 28
20 45 8 3 24
10 35 15 10 30
Segmented states
59
Among other things, national coordination involves taking the views of interest organizations into consideration. Thirty-five per cent of the administrative units in the four countries report that coordination between the units and private interests has increased in connection with EU affairs. This indicates a clear choice in the Nordic countries to give private interests a say in these matters rather than to entrust the formulation of national standpoints and decisions entirely to the national administrative units. Two further general propositions were more directly concerned with how EU activities have affected the state’s internal coordination. One of these was that EU work had resulted in increased coordination between that unit and others within the same sector, while the other was that EU work had resulted in increased coordination between the unit and administrative bodies dealing with other policy areas. These figures allow us to conclude that, apart from Finland, the units are more prepared to agree that EU activities have increased vertical internal coordination within their respective sectors than horizontal, cross-sectoral coordination. Even so, many agree that horizontal Table 3.5 In the course of the past year, EU work has resulted in increased coordination between the unit and other units working within the same sector (1998, per cent)
Agree entirely Agree in part Disagree in part Disagree entirely Don’t know
Norway
Denmark
Sweden
Finland
Overall
5 42 18 13 22
6 55 14 1 24
16 42 7 12 22
14 53 15 3 15
10 48 13 8 21
Table 3.6 In the course of the past year, EU work has resulted in increased coordination between the unit and administrative bodies in other sectors (1998, per cent)
Agree entirely Agree in part Disagree in part Disagree entirely Don’t know
Norway
Denmark
Sweden
Finland
Overall
4 29 19 14 34
4 46 17 1 33
6 32 11 23 28
13 57 11 4 16
6 40 14 11 28
60
Segmented states
coordination has also been reinforced. Once again, Finland has the highest rate of agreement, and Norway the lowest. The difference is particularly noticeable with regard to cross-sectoral coordination. In Finland 70 per cent of the units agree that EU work has led to increased coordination across various sectors, whereas the corresponding figure for Norway is 33 per cent. Agreement with the latter statement is also relatively low in Sweden. The responses to these three statements reveal that the largest increase in state coordination in response to closer proximity with the EU has occurred in Finland and the smallest increase in Norway. Denmark and Sweden fall somewhere in between, with the former slightly more affected than the latter. Sweden is notable for greater growth in sectoral than in cross-sectoral coordination. Coordination and transnationalization Increased involvement in the EU has led to increased demand for coordination and for an effective increase in efforts to improve coordination in central administrations (Kassim, Peters and Wright 2000). In organizational terms this becomes apparent in the establishment of various forms of collegial coordinative bodies of such familiar types as special commissions, coordination committees, EU sections and EU advisory groups. There is at present a considerable trend towards coordination in relation to EU affairs. In particular involvement with the EU has led to increased vertical coordination within certain sectors, but there has also been horizontal coordination across sectors, and national coordination, whereby interest organizations and other relevant groups in the community become involved. There is much to suggest, however, that it is primarily the officials in the units who promote this coordination first and foremost, and that the political leadership is only involved to a minor degree. One exception is Sweden, where there are close contacts between political leadership and various government administrative offices. It has been claimed that increased integration in EU affairs leads to increasing demands for national coordination or for national interests to be shaped so that small countries can still defend their interests within the EU – the need to ‘rally around the Swedish stance’ (Jacobsson 1993). Our data indicates that this work of bringing together administrative units, politicians and interest organizations as part of the process of national coordination is most formalized in
Segmented states
61
Denmark, where the goal has perhaps been easier to achieve in a routine fashion (Pedersen 2002). In Finland, where coordination is pursued primarily by the administration, it is more difficult to characterize the result as national coordination. The same applies in the case of Sweden, where coordination is first and foremost sectoral. In Norway as well, coordination is essentially a concern for the administration, as seen for instance in the framework guideline arrangement (Sætereng 2001). Within central administrations it is possible to delineate a coordination segment that involves one unit in every seven (Sundstro¨m 2001). What these units have in common is, firstly, their agreement that EU activities have resulted in increased coordination within or between sectors, and secondly that they participate once a month or more often in collegial coordination organs and have weekly contacts with other state authorities concerning EU affairs. An interesting finding is that a significant number of the units in this coordination segment do not belong to the core segment, which indicates that many of the core segment units involved in transnational EU work do not play a crucial role in national coordination activities. This raises the question of the extent to which various channels overlap and complement one another or are mutual alternatives and rivals. This question forms the theme of the next section.
Multiplier effects What are multiplier effects? The term ‘multiplier effect’ refers to the way in which transnationalization enables an administration to use more channels of participation and contact than in the past. Transnational relations supplement and reinforce established networks and channels of contact. Strong multiplier effects occur where transnationalization provides administrative units with new channels and gives rise to external relations in terms of contacts and participation. Weak multiplier effects arise where transnationalization causes those administrative units that are most involved in European affairs to become increasingly independent of national linkages so that they begin to live lives of their own. The multiplier thesis implies that increased contacts with European organizations can be reconciled with more frequent contacts with the domestic political leadership, with other national administrative
62
Segmented states
units, with units in other Nordic countries and with interest organizations and private companies. If the multiplier thesis turns out to be correct, then it should entail that those administrative units with more extensive networks of contacts with the administrations of other countries and with EU organizations will have greater scope for action. The transnational administration functions as a link not just between national, Nordic and European concerns, but also between the state and its surrounding society. This means that multiplier effects are greater where units with the most intensive contacts with the EU also have the most intensive contacts with other actors such as governmental units, politicians, Nordic bodies, private companies and interest organizations. Multiplier effects and internal coordination One way to measure multiplier effects is to investigate how many units from the core segment also belong to the coordination segment, i.e. among the units that are most central to the state’s internal coordination. On average, 17 per cent of the units belong to the state’s internal coordination segment (Table 3.7). Only 8 per cent of the Norwegian units belong to this segment, as against 28 per cent of the Finnish. This suggests that, with regard to EU concerns, it is the Finnish central administration that is most coordinated internally. When we compare this coordination segment with the core segment, we see that 45 per cent of the units in the core segment fall outside the internal coordination segment within the state. Thus, on this dimension we find only a weak expression of the multiplier effect. It would seem that a certain segment of the administration is deeply involved in activities to enable coordination in relation to EU affairs, but that this segment is largely distinct from the sections of the administrations that are most integrated in the EU. Half the coordination segments consist of units working in the sectors that are often considered as the core of the EU – trade,
Table 3.7 Units belonging to the state’s internal coordination segment (1998, per cent) Norway
Denmark
Sweden
Finland
Overall
8
12
20
28
17
Segmented states
63
agriculture and foreign affairs. This indicates that the high degree of integration within the EU does not hinder units from coordinating on the national level. There are, however, also policy areas which are deeply affected by the EU, but which are nevertheless hardly represented in the coordination segment, such as environmental policy. We also see that the coordination segment includes a number of units that are not closely involved in the EU, but which have traditionally played a central role in the government’s internal coordination, such as the financial sector. This means that coordination is still possible between units that are closely involved with the EU, but which do not also belong to the coordination segment, although where this is so the coordination involves rather specialized issues. Multiplier effects and relations with political authorities We can identify a political segment consisting of units that, on the one hand, stress the importance of political priorities, and, on the other, have contacts with political leadership at least once a month (Jacobsson 2001). Thus the political segment encompasses the units on whom politicians have the greatest influence. For the Nordic countries as a whole, the segment encompasses 16 per cent of all the units. In Sweden it accounts for 20 per cent, whereas the percentage for the other countries is 15 per cent. Two thirds of the units in the political segment belong to various ministries, which means that 44 per cent of the ministerial units belong to the political segment as against only 7 per cent of the agency units. When we look at which units are involved, we see that it is the foreign affairs sector that dominates, where as many as 44 per cent of the units belong to the political segment. There is a clear pattern revealing extensive involvement by politicians in EU affairs that affects the respective foreign affairs ministries. By comparison, in the core segment, i.e. among the units that are most affected by and which have the highest levels of contact with and participation in EU organs, the foreign affairs administration was far less dominant. The core segment contained as many units from the agricultural sector as from the foreign affairs sector, and also other trade sectors showed closer relations with the EU. There might be several reasons for these variations in sector representation within the core segment and for the dominance of the foreign affairs sector in the political segment. One fairly obvious interpretation would be that politicians still tend to regard the EU as
64
Segmented states
primarily a foreign affairs concern, whereas the actual administrative dynamic suggests a rather different situation. The indications are that an administrative segment is in the process of developing that is closely associated with EU organizations, but not so closely linked to national politics. Approximately half of the units belonging to the political segment also belong to the core segment, i.e. half the units that have close links with politicians are also closely associated with the EU (Jacobsson 2001). A large section of the administration has close contacts with the EU without being so closely associated with politicians. This is the same pattern as was found in the coordination segment. Only 40 per cent of the units in the coordination segment also belonged to the political segment. This means that it is not the units with the closest relations to the EU that deal with coordination of EU affairs or that enjoy contact with politicians on EU questions, despite there being a positive connection between these activities. Less than a third of the units in the core segment can also be found in both the coordination and political segments. This implies that there are certain multiplier effects on these dimensions, but that they are relatively weak. Multiplier effects and relations with interest organizations In order to examine whether coordination between units and interest organizations plays any role in national coordination concerning the EU, we shall compare how many units have contact with both the EU and private interests in relation to EU affairs. The objective is to clarify whether in these cases relations between the units and private interests reflect a deliberate attempt at coordination to facilitate EU work. In all four countries there is a group of administrative units that have contacts with both private interests and EU organizations in relation to EU affairs (Table 3.8). In relative terms, this group is of Table 3.8 Units that, in the course of the past year, have had monthly or more frequent contact with EU bodies and with national interest organizations and private companies in connection with EU-related work (1998, per cent)
Administration in general EU administration Core segment
Norway
Denmark
Sweden
Finland
Overall
28 59 79
35 67 75
36 53 62
38 55 56
34 59 68
Segmented states
65
roughly the same size in all four countries. In addition there is a significant correlation between the units that have contacts with private interests and those with contacts to the EU organizations. The EU administration and the core segment generally have more intensive contacts with private organizations than does the administration as a whole. The principal impression is that multiplier effects occur where units in the core segment also have frequent contacts with interest organizations. Regardless of whether a country is a full member of the EU, European integration results in intensive relations between private interests and public authorities. This effect is particularly apparent in the context of the internal market project. In addition, the relationships arising from integration tend to be more mixed, some of them being informal, others formal. And finally, integration results in the establishment of new and more numerous channels of communication between private organizations and public authorities on both the national and EU levels. Private companies and interest organizations are able to use these channels in two ways. Firstly, they can bypass their national administrations and establish direct contacts with European organizations, or secondly, they can go through their national administrations, entrusting to them the defence of their interests in European decision-making processes. Much the same applies to the state units themselves. These can choose either to ignore individual private interests and concentrate instead on the protection of the community interests by means of political-administrative processes (‘within’), or they can grant the various private interests a voice by encouraging contact and representation (‘go through’) (Pedersen 2002). These results are supported by other studies that have examined the development of relations between administration and interest organizations in connection with the protection of interests in EU-related matters (Sidenius 1998, 1999). They also derive backing from more general descriptions of the development of relationships between authorities and interest organizations (Pedersen 1995, Christiansen 1998, Christiansen and Sidenius 1999), but they conflict with some Swedish studies indicating that, starting in the 1980s, there has been a general de-corporatization trend (Lewin 1992, Rothstein 1992, Hermansson 1993, Micheletti 1994, Hermansson et al. 1999, Rothstein and Bergstro¨m 1999). On their own these studies do not allow us to confirm the general claim of a de-corporatization trend. The latter is,
66
Segmented states
however, also difficult to refute. What we can say is that EU integration does not seem to have weakened relations between authorities and interest organizations, even in Sweden. On the contrary, it seems to have encouraged the development of more channels of communication between these actors; it has resulted in contacts of greater intensity, and a broader diversity of the means whereby the interests in question are cared for. Where interest organizations are concerned, the multiplier effects are apparent, on the one hand, in the relations established between those bodies and their respective administrations, and on the other, in the contacts which those bodies develop with the European organizations without going via the national administrations. At the same time, the various units are also engaged in a great deal of coordination activity that does not directly involve the interest organizations. In all four countries private and public interests have been affected by a certain amount of coordination. Even so, there are significant differences between the countries as to which interests are involved and to what extent. Multiplier effects and Nordic relations Looking at the levels of inter-Nordic contact patterns, we notice that three units in every ten have contacts with units in one of the other three countries once a month or more often, while one in ten has such contacts on a weekly or daily basis (Larsen 2001). Contacts between these units are, however, twice as extensive in those parts of the administrations most involved in EU-related activity compared with the administration in general. The difference between the administration in general and the units most involved in the EU is most pronounced in the Norwegian case (Table 3.9). This can be viewed, on the one hand, as a consequence of Table 3.9 Proportion of units that, in the course of the past year, have had contact with the units in other Nordic countries once a month or more often in connection with EU-related activities (1998, per cent)
Administration in general EU administration Core segment
Norway
Denmark
Sweden
Finland
Overall
27 53 74
29 53 60
37 56 64
38 55 64
33 55 64
Segmented states
67
Norway’s form of affiliation and of the use of contacts with the other Nordic countries as a channel to the EU. On the other hand, it can be seen as a consequence of the generally broader cooperation between the Nordic EU members, which also helps to increase the extent of the contacts for those units that are not as intensively involved in EU work in these countries. Taken as a whole, the countries clearly show a positive correlation between the extent of contacts and how affected the units are by EU-related activities. This is true both for direct contacts with units and for contacts with bodies serving Nordic intergovernmental cooperation, such as the Nordic Council. The difference between the units most involved in the EU and the administration in general is greatest in terms of contacts with authorities in EU countries outside the Nordic area. The data shows clearly that the Nordic countries still serve as models for one another more frequently than other countries (Larsen 2001). A full 64 per cent of units in the various Nordic administrations seek ideas or inspiration for the running of their activities from other Nordic countries. Fifty per cent claim to find their inspiration in other countries, and 31 per cent in EU organizations. These differences remain the same when we also focus exclusively on the EU administration. Despite having now been a EU member for many years, Denmark is still less inclined to draw its inspiration from EU organs than are the other Nordic countries. Both for the Nordic member states and for the Norwegian administration we can say that, the more involved the units are in EU-related work, the greater the contacts with both the authorities and the Nordic intergovernmental cooperation bodies. We can, therefore, conclude that, for the Nordic countries, EU membership has not led to a lower priority being set for inter-Nordic governmental cooperation. Nordic cooperation is seen as complementary to the direct contacts between units in EU-related affairs. Contact between the authorities is now twice as extensive as in the past. In the Nordic EU countries, the extent of contacts with other international organizations is greatest among the most EU-integrated units. In the case of Norway, on the other hand, we do not find the same tendency among the units most involved in EU affairs towards higher levels of contact with other international organizations, with EU organizations and the Nordic cooperation bodies, than in the administration in general. In other words, as far as contacts with
68
Segmented states
other international organizations are concerned, multiplier effects are not as powerful in the Norwegian administration as they are in the Nordic EU countries. To sum up, we can say that a multiplier effect is clearly manifested in the positive link between the extent of EU-related cooperation and the involvement of units in Nordic and international relations. Increased European integration constitutes a challenge for older forms of Nordic cooperation, which have been transformed, though by no means discontinued, in response. Even if Nordic administrative units still look to each other for ideas and suggestions for change within their respective administrative sectors, Nordic connections now take place less through inter-Nordic bodies and more in terms of direct bilateral relations between administrative units. Multiplier effects and transnationalization Whereas segmentation and, to some extent, coordination primarily affect relations within the state and between the government and national social interest, the multiplier thesis is concerned first and foremost with the interplay between these and international interactions, and between various international connections with a special focus on relations to the EU (Pedersen 2001). The greater the overlap and exchange between the various external relationships, the greater the support for the multiplier thesis, and conversely, the more isolated and shielded an EU channel happens to be, the less the support for the thesis. In this study, the multiplier thesis has been corroborated in a number of contexts. The most striking pattern to emerge is that the units with the closest relations to EU organizations, in terms of frequent contacts, considerable travel activity, and broad participation in EU bodies, are also affected to a considerable extent by other international organizations, and have well developed contacts both with the other Nordic countries and with national interest organizations. These units are, however, somewhat less integrated in terms of national coordination activities and relations with political leadership. A significant proportion of the units have close contacts with EU organizations without playing a central role in governmental coordination work and without having close relations with politicians. The support for the multiplier thesis is weakened by the observation that the core segment, the coordination segment and the political
Segmented states
69
segment do not always consist of the same units. In particular it is the various externally oriented relationships that are closely integrated, whereas linkage with intragovernmental coordination processes, and connections with political leadership are not so close. Nevertheless, the units that have close contacts with political leadership and are considerably involved in coordination activities are more closely integrated in European relations than those with little contact with political leadership and low involvement in coordination. One interpretation of this is that the EU issues involve a certain division of labour for the central administrations, so that some units are more concerned with outwardly oriented functions and with looking after ‘upstream activities’ that have to do with EU decision-making processes, whereas others are more inward looking and are primarily preoccupied with ‘downstream activities’ related to implementation and management of EU regulations vis-a`-vis national legislation and administration. The main picture is that the various channels tend to complement one another rather than compete as alternatives, as can be seen in the fact that interest organizations tend to adopt a ‘bypass’ strategy in addition to, rather than instead of, a ‘go-through’ strategy in their EU dealings, and that relations with the EU function as a supplement rather than an alternative to existing connections between the Nordic countries. Transnational administration is not independent of the national, but works as a link between the state and the surrounding society and between the national, the Nordic and the European. Nevertheless, tensions can arise between this kind of transnational administration and a more intergovernmental conception of relationships between nation-states and the EU, and this is due, in part, to the fact that the growth in sector-specific relations makes it difficult to operate with conceptions of a unified and integrated administrative apparatus that presents an integrated face to the outside world. In contacts with the EU a core segment has developed that is becoming ever more transnational and that derives its picture of the world from processes that are largely independent of politics, whereas there is also a political segment that works on the assumption that politics is an intergovernmental business in which the various foreign ministries play an important role. These two segments are only partially interconnected.
70
Segmented states
Transnationalization revisited The administrative apparatuses in the Nordic countries have undergone a process of transnationalization insofar as their daily activities have come to be increasingly affected by EU affairs, not just at the most central levels, but also lower down in the system, in various specialized units, boards, directorates and individual authorities (Trondal 2000a). Adaptation to the EU is not determined exclusively by the units most closely linked to the political leadership; it is also a very real concern for external and more independent professional units. Concerning matters connected to the internal market, the domain where the central administration is most involved with the EU, the agencies are just as affected as the ministries. The same applies in relation to legislative change. Connections are also extensive for both the ministries and the agencies in terms of participation in and contacts with the various organs in Brussels, although in this respect there is a clear tendency for relations to be better developed among the ministries. This reinforces the transnationalization thesis, which predicts well developed relations with the EU not only among units at the top of the administrative hierarchy, but also among those that are more peripheral. Transnationalization also becomes manifest through the development of extensive patterns of contact with and participation in EU affairs, and through changes in relations with interest organizations and with other countries. In addition, public servants and experts have considerable freedom and numerous opportunities to use their own professional judgement within the transnational networks that develop. Transnationalization, however, needs modification and qualification. Firstly, because it leads to segmentation, so that certain parts of the central administrations acquire pronounced transnational characteristics, whereas others are only transnationalized to a more limited extent. We can distinguish between a core segment within the EU administration of the Nordic countries, in which the features of transnationalization are dominant, an EU administration in which transnational networks are more in balance with traditional administrative tasks, and the remainder of the administration, where transnational networks are only poorly developed. Secondly, the processes of coordination face challenges vertically, i.e. within single ministerial domains, horizontally, i.e. across different sectors, and nationally, in terms of increased demands for clearly defined national
Segmented states
71
standpoints that also involve interest organizations. Thirdly, the occurrence of multiplier effects means that the growth of EU networks and contacts helps to supplement and reinforce certain existing external relations, such as those with other Nordic countries and international organizations, while at the same time weakening relations upwards towards political leadership and inwards towards national and governmental coordination processes.
4
Robust and flexible states
Introduction Roughly speaking we can distinguish between two types of study of the Nordic countries. One tradition emphasizes their inherent similarities and seeks to identify a significant Nordic model. The other tradition dwells on the differences between the countries. In this chapter we will examine both these traditions and ask whether adaptation to the EU takes place within established administration regimes, or whether it represents a fundamental break with the past, involving profound changes to the administrative models of the Nordic countries. Our conclusion is that the administrations adapt within robust and stable organizational forms. The fundamental formal administrative models remain firm while at the same time significant changes take place in administrative work (Lægreid 2001b, Christensen 2002, Olsen 2002b, Marcussen and Ronit 2003). We will first clarify our starting point by sketching some important features shared by the Nordic administrative regimes and their differences, as they were before European integration started in earnest. We will then sum up the similarities and differences between these countries in relation to their increased integration in Europe, which has been described in chapters 2 and 3. Finally, we will formulate a robustness and flexibility thesis that emphasizes how alterations and transformations are made within robust administrative forms that allow considerable variation in patterns of behaviour.
A Nordic model with scope for variation Among families of countries, the Nordic countries are generally regarded as having a very high degree of internal similarity, and in studying them
74
Robust and flexible states
from the outside it is common to ascribe them to a specifically Scandinavian or Nordic model (Deutsch 1957, Etzioni 1965, EspingAndersson 1990, Kuhnle 1991, Petersson 1994, Greve 1996, Lægreid and Pedersen 1999). With the exception of Finland they form a linguistic community, and are all constitutional monarchies. They are independent, small, parliamentary unitary states, with protestant state churches, and multi-party systems. In each country the largest party has traditionally been social democratic, although differences between parties are relatively small. These countries share the same well-developed allembracing welfare state with a large public sector. They have open economies that depend on export, and high standards of living. They are culturally homogeneous, with an egalitarian socio-economic structure. Their recent history has been characterized by an orientation towards consensus, peaceful coexistence, well-developed corporative arrangements and a strong tradition of significant local government (Bruun et al. 1990, Olsen 1997). The political control over public administration has been general and passive, allowing the executive a lot of leeway. This seems to reflect some major features of the political administrative system: a high level of mutual trust and shared attitudes and norms among political and administrative leaders and within the public sector. These countries have developed forms of cooperation characterized by their being strong nation-states that place great emphasis on the preservation of national sovereignty (Olsen and Sverdrup 1998). There are, however, also significant differences between the Nordic countries. Whereas Finnish government has over the past decades experienced a series of broad majority coalitions, the other countries in the region have had minority governments. In Denmark, and to some extent in Norway, these have been coalition governments, whereas in Sweden the Social Democratic Party has over long periods formed a minority government on its own. Finland is also distinguished from the other Nordic countries in terms of the close links that it had with the Soviet Union up until the collapse of the Eastern Bloc in 1989, a situation that limited the country’s opportunities for political and economic participation in Western Europe. Norway and Sweden, for example, were involved in EFTA from its inception in 1960, whereas for a long time Finland was linked to EFTA only in terms of an associative agreement and did not become a full member until 1986. It has been traditional to distinguish between an east Nordic administrative model, which involves independent central agencies that report to the cabinet as a collective body rather than to individual
Robust and flexible states
75
ministers, and a west Nordic model under ministerial control, where each minister is responsible for everything that happens in his subordinate authorities. A distinction is drawn between a dualistic Swedish-Finnish model with strong cabinets and strong, relatively autonomous central agencies, and a monistic Danish-Norwegian model, with strong specialized ministers with integrated ministries, directorates and agencies responsible for specific policy areas, and weaker cabinets (Lægreid and Pedersen 1999). The east Nordic model builds on a twofold organizational division between the respective ministries and central agencies, which implies that the position of the central agencies is formally independent of the political bodies. The characterization of the Swedish system as dualistic is intended to underline the twofold division between the ministries and the rest of the civil service, whereas the model based on ministerial control has been called monistic in order to emphasize the unitary principal in the relationship between central and peripheral authorities. Finland has traditionally been seen as belonging to the east Nordic model, but changes in recent years have given the Finnish model more monistic features. In formal terms, the differences between the east and west Nordic models are considerable, although in practice they are probably not so extreme, once the various informal contacts are also taken into consideration (Jacobsson 1984, Lindbom 1997). In the course of time, the Nordic countries have seen the evolution of more polycentric political-administrative systems, whereby the administrative apparatuses have become less hierarchical, unitary and tightly connected. Due to expansion, differentiation and the promotion of autonomy, the state has steadily become more complex (Premfors 1998), with the result that top civil servants have been represented as ‘key actors on different teams’ (Lægreid and Olsen 1984). Studies of political-administrative development in the Nordic countries also reveal clearly national characteristics, even where it is possible to identify shared features (Lægreid and Pedersen 1994, 1996, 1999). Compared with the Anglo-American countries, the Nordic countries appear less aggressive and radical in implementing administrative reforms associated with the New Public Management (NPM) movement (Pollitt et al. 1997, Christensen and Lægreid 2001). In contrast to the ‘rolling back the state’ strategy of the ‘Westminster model’, the Nordic countries have adopted a less radical course of administrative reform aimed at modernization, maintenance and increased efficiency (Pollitt and Boekaert 2000, Lægreid 2001c).
76
Robust and flexible states
Up until the mid 1990s, Sweden seemed to be the Nordic country that had gone furthest along the NPM road, while developments in Norway were more hesitant and reluctant. But even if Sweden has moved towards greater devolution, and focussed more closely on efficiency and marketization, traditional democratic values have continued to influence public sector reforms to a greater extent than in many Anglo-American countries (Ferlie et al. 1996). The Nordic countries can thus be seen to have retained a model of social responsibility in contrast to one based on extreme market efficiency (Fudge and Gustavsson 1989, Lægreid 2001c). Similarly, when we look at the development of the welfare state in these countries, we find it characterized by institutional stability, but also by signs of divergence in the 1990s (Eitrheim and Kuhnle 2000). New regional dynamics in Europe might weaken the exclusive ties between the Nordic countries within their region and contribute to involvement in different regional projects (Baldersheim et al. 2001). In the following we shall see how transnationalization has contributed to both the similarities and the differences among the central administrative apparatuses of the Nordic countries.
The transnationalization of administration Drawing on the information in the foregoing chapters, we will first summarize the most important similarities between the Nordic central administrations resulting from transnationalization. Subsequently, we will look at the differences between the countries. Similarities in the Nordic countries The general profile of the effects of Europeanization on the structure and activities of the central administrations can be summed up in four main points. 1
In the late 1990s, the central administrative apparatuses in the Nordic countries underwent significant changes in terms of the dimensions studied. The view that Europeanization does not lead to major changes in the structure and mode of operation of national administrations (Page and Wouters 1995) finds little support in this study. The central administrative bodies play a key role in dealing with EU-related issues and this has also given rise
Robust and flexible states
77
to a new breed of national Euro civil servants (Bergman and Damgaard 2000). Of all the units in the Nordic central administrations in 1998, only 13 per cent were wholly unaffected by the EU within their respective fields of activity. All the countries have established collegial coordination bodies. The administrations devote a considerable proportion of their time to EU affairs and are significantly affected by the EU within their various fields of activity, especially in matters relating to the internal market, which implies that the EU has a major impact on most policy areas. EU activities focus to a large extent on the development and modification of rules and regulations, but there is also a great deal of coordination and increased emphasis on monitoring and supervision (Lægreid 2001a). Extensive contacts with and involvement in the organizations in Brussels have been developed, especially with the Commission and its subsidiary bodies. Although these contacts are primarily formal, there are also many informal relationships. Many units have also seconded staff to EU organizations as national experts, and there is considerable travel backwards and forwards to Brussels. Much of this interaction is, however, reactive and many units are sceptical about the influence of their respective units among the EU organizations. At the same time, established forms of cooperation within the Nordic countries are undergoing changes. Although the region’s administrations still turn to one another when seeking ideas and inspiration on how to modify their administrative bodies, interNordic connections are being redirected, away from common Nordic institutions in favour of stronger direct bilateral relations between the various administrative bodies. The Nordic practice of encouraging the involvement of interest organizations continues. Corporative relationships of a more or less formal nature are well developed in terms of both contacts with and participation in various collegial bodies. In addition, these organizations and private companies make use of a ‘bypass’ channel to establish contacts directly with institutions in Brussels without going via the national administrations. Where relationships with political leadership are concerned, the administrations are generally characterized by a high degree of autonomy in EU affairs. Contacts between the political and administrative levels are not particularly intensive, albeit with
78
2
3
Robust and flexible states
exceptions. These contacts are largely informal and many units have to make do either without guidelines or with only general indications when taking part internationally in various EU organs. For many, tight time schedules make it difficult to refer questions to political leadership. The overall impression is that politicians are relatively passive in relation to EU affairs, and most units have little difficulty in getting their views accepted by their government and in political circles. The majority find it relatively easy to predict what stance politicians will adopt on various EU questions. The administrative units report only very low levels of conflict in relation to EU matters, which indicates that most of these issues are entrusted to specialists and experts, or that politicians have developed their own arenas for handling such concerns. In summary the picture that emerges provides support for the transnationalization thesis. There are variations in the intensity of adaptation depending on the extent of the change involved. There have been fewer changes in internal and formal organizational design than there have been in patterns of external contact and participation. Relations with the EU in terms of contact and participation are more intensive than they are with other Nordic countries. Channels of contact with political leadership concerning EU affairs are only relatively poorly developed. Coordination within the administrations is increasing, although civil servants dominate the process. Despite the broad contacts with and intensive participation in organs in Brussels, levels of conflict are low and influence is limited. This is especially true in relation to the European Council, but even where the EU Commission is concerned, only one unit in three claims to be clearly successful in getting its views and wishes accepted. Most of the affected administrative units are reactive in their stance to the EU. We have described an administration that is transnational but reactive in its dealings with the EU and a political leadership that is passive in relation to the administrations. The administrations respond to EU initiatives and follow them up without significant involvement from politicians. Our data allows us to conclude that the EU-related modifications in the central administrations have occurred not only in immediate response to changes in forms of affiliation, but have continued with great momentum throughout the ensuing years. One piece of evidence for this is that a majority of those polled estimated
Robust and flexible states
4
79
EU-prompted changes in rules and regulations to be more considerable during the four years following the change in form of affiliation than they were as a consequence of first becoming an EU member, or of first signing the EEA agreement. In 1998 the vast majority claimed to be more affected by the EU than they were four years previously. One indication that this is the case is that travel to foreign destinations in connection with EU work increased more in the preceding year than it did in the four years prior to it. The result is increased transnationalization. The adjustments have shaken up the established administrative models in the various Nordic countries. In its own discussion of this issue, Sweden claims to be adhering to its dualistic administrative model in its handling of EU work, but this study shows that Europeanization constitutes a challenge to this model. The same applies with regard to the Danish-Norwegian principle of ministerial rule, whereby it is difficult, especially in Norway, to make cabinet ministers accountable for affairs that have been entrusted to the bureaucracy, and central decisions are made in Brussels without any explicit method to allow the process to be influenced by the political leadership in Norwegian ministries. Formal corporative arrangements also remain more or less intact, although they too face challenges and are being expanded. Much the same can be said about Nordic cooperation, where, in the process of readjustment, the various administrative units continue to draw their inspiration largely from each other. Our material did not allow us to establish any de-corporatization trend, nor any reparliamentarization of politics in EU-related matters, such as has been claimed for developments in other policy areas in the Nordic countries (Petersson 1998). Within this general pattern we have revealed significant variations between different levels of administration and policy areas, and especially between different countries.
Variations between the Nordic countries In all four countries, adaptation to the EU has progressed along largely similar lines. There is increased involvement in Europe in terms of the expansion of contacts and participation and the standardization of systems of regulations, but the tempo and scope of adaptation vary from one country to another. There are clear variations between the
80
Robust and flexible states
administrative adjustments to the EU being made in these countries. One pattern that recurs throughout is that almost all indicators show Norway to have undergone the smallest EU-related changes in its central administrations, whereas Finland and Sweden are the countries with the largest changes. Denmark occupies an intermediate position. Norway The Norwegian administration has been considerably affected by the EU via the EEA agreement, especially in areas relating to the internal market. Implementation of the EEA rules was fairly rapid within the Norwegian administration. Even so, adaptation to the EU is clearly less extensive here than in the new EU members Sweden and Finland. Although the Norwegian central administration is not excluded from Europe, its involvement in European cooperation is still relatively limited. Less time is devoted to EU-related concerns in Norway than in the new member countries. Moreover, the networks of participation and contacts with EU bodies are only weakly developed in Norway. This is true with regard to the Commission system, and especially the Council system. Norwegian bureaucrats travel to Brussels less frequently than those from the other Nordic countries. The idea that Norway compensates for its lack of membership in terms of increased informal contact and adaptive measures gains little support from this study. For the Nordic countries generally, change in the form of affiliation to the EU is linked with changes in patterns of mutual cooperation in favour of stronger bilateral contacts, especially between the new EU members, and this trend has weakened Norway’s role within the Nordic cooperation. Relationships with interest organizations are also less well developed in Norway. The use of public commissions and national reference groups is less widespread in that country, and EU/EEA concerns have not stimulated as much coordination between administration and private interests in Norway as in the new EU member states. Furthermore, the degree of direct involvement of Norwegian politicians in EU work is less than in the other countries. In relation to the other Nordic countries, coordination in EU affairs in the Norwegian central administration is less well developed. The relative weakness of Norway’s transnational relations with the EU compared with the other countries is illustrated by the fact that only 7 per cent of the Norwegian central administration units fall within the core
Robust and flexible states
81
segment, which is defined as those units affected by the EU to a considerable degree, which devote a large part of their time to EUrelated matters and which participate at least once a month in meetings within EU/EEA bodies. The corresponding percentages in Sweden and Finland are 27 per cent and 23 per cent respectively, whereas the Danish core segment is more concentrated, accounting for 17 per cent of that country’s central administrative units. Neither does Norway have more international contacts outside the EU than the other Nordic countries. There has been less development in the relevant staffing levels and skills in the Norwegian central administration than in Finland and Sweden. In particular, Norway has created few new posts to handle EU-related concerns compared with the new member countries. There have also been fewer changes in the distribution of responsibilities in Norway, on both the horizontal and the vertical dimensions. Neither have new forms of monitoring and supervision been developed to the same extent in Norway, and the scale of monitoring activity has increased to a smaller degree. Contacts and coordination between national authorities in EU affairs are also less extensive in Norway than in Finland and Sweden. Compared with the member states, Norway rarely acts proactively towards the EU, and its sense of exerting influence among EU institutions is also weaker than in the other countries. Cultural differences between the Nordic countries are apparent in the greater emphasis on the political assessments of the government in Norway than in the member states, and in the relative lack of attention paid to consumer groups, private interests and other affected parties in Norway compared to Sweden and Finland. The fact that EU-relevant guidelines from the political leadership are accorded considerable weight in Norway, even when only vague and rarely in written form, might indicate that autonomous adaptation and the anticipation of reactions are widespread forms of interaction between political leadership and the bureaucracy in Norway. The most considerable cultural conflicts between national administrative traditions and the norms, rules of engagement and working practices of the EU are experienced by the new member states, whereas Norway, which has weaker transnational relations with the EU, is less affected by such cultural collisions. The situation in Norway could be summed up by saying that since the country is a non-member the high level of administrative involvement in the EU is more surprising than the fact
82
Robust and flexible states
that it scores lower than Sweden and Finland on most indicators of change. Denmark There are significant differences between the new member states and Denmark. These differences can be explained partly in terms of how long the respective countries have been members and by their various efforts to adapt, but it can also indicate that the common form of affiliation does not necessarily obliterate the more individual features of their separate administrations (Trondal 1999). In other words, the differences between Denmark and the other member states can be explained in terms of length of membership together with the robustness of differences that already existed between the administrative models of Denmark on the one hand and Sweden and Finland on the other. Sweden and Finland generally score higher than Denmark on many indicators of change. One interpretation of this pattern is that the central administration in Denmark is now in a different situation from that of Sweden and Finland, insofar as the former has already undergone 25 years of adaptation to the EU and is more willing to perceive its relations to the EU as an integral and natural part of its ordinary activities, so that it becomes difficult to distinguish between EU-related work and other tasks (Siedentopf and Ziller 1988). The greatest formal and structural changes in Danish administration were made around the time when it became a member, and later adjustments have been incremental (von Dosenrode 1998). Nevertheless, Denmark now seems hesitant and sceptical about further integration, as can be seen in its rejection of the Maastricht Treaty by referendum in 1992 and its subsequent reservations concerning union citizenship, the economic and monetary union, defence policy and judicial and internal conditions (Miles 1996). In contrast to Finland, but not to the other countries, there has been a lengthy and extensive public debate in Denmark about the country’s relationship to the EU, and this has influenced the impact of Europeanization on the central administrative apparatus. The Danish position is indeed full of contrasts. Despite the country’s Euro-sceptical attitude, in its day-to-day activities it is a well-adapted EU member, and it stands out as one of the most adept countries at implementing EU regulations (Pedersen 2000).
Robust and flexible states
83
Sweden and Finland Sweden and Finland are still becoming established as EU members and have demonstrated energy and eagerness to adapt. The contrasts with the period prior to EU membership are more dramatic than in Denmark’s case, where EU work has already become routine. Whereas changes in Finland and Sweden reflect the effort to adjust to new EU membership, recent changes in Denmark, as an established EU member, have more to do with finding a balance in relation to EU work. This is apparent not least in the fact that the Danish administration has developed more formal coordinative mechanisms and now has fairly settled procedures both within the administration itself and in relation to professional organizations. In addition to having closer informal relations with the EU, Denmark also behaves more proactively than Norway and Finland and reports greater success in gaining recognition for its standpoints in Brussels. Another feature that distinguishes Denmark from the other countries is the relatively central role that its Ministry of Foreign Affairs and parliament play in EU work. Although the Finnish administration can be seen as somewhat more advanced than its Swedish counterpart in its adjustments to the EU, our material does not reveal any dramatic differences in how the Swedish and Finnish administrations have mastered EU integration, and this finding tends to conflict with the results of other studies (von Sydow 1999, Raunio and Wiberg 2000b). The fact that, in many respects, Sweden and Finland stand out from the other Nordic countries might well suggest that the east Nordic administrative model is being recast into a new mould. The Swedish and Finnish administrations are becoming increasingly involved in an ever more tightly integrated Europe; whereas Norway’s administrative bodies are less integrated in European cooperation. Sweden and Finland are experiencing major collisions between their national administrative cultures and EU norms, rules of engagement and working practices. There are, however, also a number of interesting differences between Sweden and Finland. The Finnish administration tends to have only weak links to the political level in relation to EU work and to be more dominated by bureaucracy. In Finland more than in the other countries EU concerns are handled by an administrative apparatus which has considerable autonomy and a pragmatic, closed and technocratic culture, whereas the Swedish administration is characterized by greater openness and greater involvement of the government and political
84
Robust and flexible states
leadership. In this context it is worth noting that the Swedish debate about administrative policy in relation to the EU has placed great emphasis on the need for democratic control. Whereas Finland has given greater focus to the practical and pragmatic dimension, the Swedish administration has been more concerned with the ideals of openness and public transparency. Denmark seems to have reconciled itself to the more restricted openness that characterizes the decisionmaking culture within the EU. But although it has paid relatively little attention to this issue, Denmark still manages to take public transparency into account in EU matters. The way in which Denmark implements the openness principle has gradually been transformed to fit in with the political style of the EU (Grønbeck-Jensen 1998). Another interesting difference between Sweden and Finland is that, whereas Nordicism remains a highly influential idea in Sweden, the consideration of inter-Nordic relations is less important for the Finnish administration. This can be seen in the fact that the Finnish units only rarely attempt to promote their views within the EU in conjunction with other Nordic countries. In general, their orientation towards the other Nordic countries is more restricted. They tend to draw their inspiration in administrative matters from beyond the Nordic area, in particular from the EU and other international organizations, in addition to their Nordic orientation (Larsen 2001). Compared with Sweden, the Finnish administrative units operate largely independently of politicians. At the same time Finland stands out as the country where coordination of EU work has been most extensive; cross-sectoral coordination in particular has increased much more in Finland than in Sweden. The fact that the Finnish administration tends to show higher levels of adaptation to the EU can also be explained, on the one hand, in terms of the extra pressure to adjust that the country felt in the run up to the period beginning in the summer of 1999 when it held the EU presidency, and on the other in terms of the fact that, in contrast to Sweden and Denmark, Finland has had an EU policy that is more favourable towards adjustment and integration. Evidence for the latter is that Finland is the only Nordic country to have become a member of EMU. Summary The transnationalization of the administrative apparatuses in the Nordic countries can be seen in the increased influence of EU affairs
Robust and flexible states
85
on the daily routines of the various administrations, not just at the most central levels, but also down to the various specialized units, the committees, directorates and authorities. Extensive patterns of contacts with and participation in the EU have been developed, and relationships with interest organizations and other countries have been affected and modified. In addition, civil servants and experts have considerable freedom and numerous opportunities to use their professional discretion within the transnational networks that have developed. This transnationalization nevertheless calls for qualitative differentiation and distinctions. Firstly, there are significant differences from one country to the next. The administrative units in Sweden and Finland stand out as having developed the strongest transnational relations and those in Norway as having the weakest, while Denmark occupies an intermediate position. Secondly, the differences between countries are weaker when administrative level and policy sector are taken into account, and this indicates that the significance of national borders is modified by boundaries between sectors and levels. Thirdly, transnationalization occurs within robust administrative regimes. Although it poses challenges to the administrative distribution of tasks, to organizational models, forms of coordination and existing external corporative relations, it does not result in the dismantling of the respective structures. This will form the theme of the next section.
Robustness and flexibility Introduction How can one evaluate the changes in the administrative apparatuses as a whole? Has there been a fundamental transformation in the administrative models and regimes that characterize the Nordic countries, or are the transformations taking place within relatively robust and basic forms of organization that permit broad variation in patterns of action? What does Europeanization really mean for the Nordic administrative regimes? Are we dealing here with administrative systems that are subject to dramatic changes in the terms on which they operate or are willingly and rapidly adapting to new directives and rules? Or can we view the relevant changes as occurring within the framework of reasonably robust and stable administrations that have had time to evolve and which adapt only slowly and in accordance with well-established routines, procedures and cultures?
86
Robust and flexible states
Since the EU lacks an administrative policy of its own, it is unable to lay down clear and forceful guidelines as to how the various countries should organize their national administrative apparatuses (Page and Wouters 1995). For this reason it is unlikely that extensive changes will occur in the national forms of organization. Some claim that adaptation is essentially incremental and gradual and balances change and continuity (Sverdrup 2000), whereas others maintain that radical and comprehensive changes are taking place that constitute a trend towards administrative convergence (Majone 1996, Me´ny et al. 1996). What is meant by robustness and flexibility? The presentation given in this chapter generally supports an intermediate position that regards organizational forms as lacking in stability, yet resistant to radical transformations. This view implies a certain form of robustness, insofar as it acknowledges the possibility of relatively limited changes to higher order features of the system, such as those that have occurred, for example, in Sweden’s well-established corporative arrangements, in the small ministries of its integrated government offices, and in its larger independent agencies, or in Norway’s system of ministries and directorates in its sectorized ministerial administration, or the Danish arrangement consisting of ministries, directorates and agencies. But at the same time these higher order organizational forms constitute only fairly basic categories that allow considerable variation in internal organization, working practices and division of responsibilities within the established arrangements. In other words, they lead us to expect an ambiguous and loose coupling between the changes that occur in formal organizational structures and the availability of resources, actual working practices and the distribution of tasks. Even if the individual countries have considerable freedom to organize their EU administrations as they see fit, the EU imposes stricter requirements in some areas than in others (Hix and Goetz 2000). Taken together, these considerations imply a dualism with regard to organizational changes. On a higher level, the organizational design of the central administrations might appear fairly robust, but at the same time, major changes are possible within the loosely defined organizational categories that characterize the administrative models of the various countries. In other words, we find both robustness and
Robust and flexible states
87
flexibility, meaning that new organizational solutions for handling challenges from the EU tend to supplement existing arrangements without becoming alternatives to them or replacing them. In sum this implies an increase in the complexity of the civil service. Existing arrangements persist while at the same time new arrangements are added. Integrated forms of internal organization With regard to the organization of day-to-day work within the various administrative bodies, the Nordic countries have adopted solutions based on established organizational forms. The Nordic administrations have generally chosen an integrated approach to the organization of EU affairs, which are thus treated as an extension of the respective field of domestic policy in keeping with a concept of the EU as a fourth level of government. Not only do the concerns that arise have local, regional and national dimensions, they also have an EU dimension (Egeberg and Trondal 1997). In Sweden, for instance, all the ministries have integrated their handling of EU affairs into their systems for dealing with other issues relevant to the department or unit (Statskontoret 2000). The choice of an integrated approach rather than one that concentrates EU concerns to specialized units is in keeping, on the one hand, with the solutions adopted for the handling of international work in general (Egeberg and Trondal 1997), and on the other, with methods of handling EU affairs in other EU countries (Bulmer and Burch 1998, Hanf and Soetendorp 1998, Harmsen 1999). Only 7 per cent of the departments in the Nordic central administrations have their own EU units (Table 4.1). It has, however, become fairly common to have a special EU coordinator. Around a third of the units now have individuals with such coordinative responsibilities, who are either employed within their departments or closely linked to them. In terms of the common emphasis on integrated organizational solutions there is little variation between the countries. The formal organizational designs of the established systems do, however, permit flexible specialization, as can be seen in their engagement of staff with special responsibility for coordinating EU work, often as a supplementary task. Within the Swedish ministries, many units have their own officials with responsibility for EU matters in the respective domain (Statskontoret 2000). There are also a few examples of specialized EU units. In Denmark, for instance, the
88
Robust and flexible states
Table 4.1 Special units/subsidiary units and/or posts with special responsibility for EU concerns (1998, per cent) Norway Denmark Sweden EU unit within the department 6 EU unit linked to the department 7 Special posts, EU coordinator within the department 34 Special EU coordinator linked to the department 7
Finland Overall
6 11
9 6
7 4
7 7
25
28
22
27
9
14
8
10
Ministry of Economic and Business Affairs has an international secretariat working mainly with EU issues, and the Ministry of Food, Agriculture and Fisheries has an office for EU coordination. In addition, up until 2000 there was a special EU directorate with responsibility for agricultural transfers from the EU. In Sweden there is a special EU unit at the Ministry of Foreign Affairs and a coordinative unit in the government. In Finland also there was for a time a special EU secretariat within the Ministry of Foreign Affairs. In 1999 this was transferred to the Prime Minister’s office. Although adaptation to the EU has not led to comprehensive changes in the formal organizational designs of the various central administrations, the central state apparatuses of the Nordic countries have been subjected to considerable reorganization in the course of the past few decades (Lægreid and Pedersen 1999). Forty-three per cent report that they underwent major reorganization in the preceding four years, while 35 per cent claimed to have undergone moderate reorganization. This implies that by and large the reorganization that has taken place cannot be seen as an unambiguous reflection of Europeanization. The administrations have been changed not only from the outside in response to demands and influence from the EU, but also from above, as a result of the government’s administrative policy, and from below, in terms of various local, agency-specific initiatives. Only 35 per cent of the units claimed that changes in connection with EU membership amounted to moderate or major transformations in the unit’s organization and working practices during the past four years (Table 4.2). The changes that occur are not always equally attributable to active administrative policies within the respective governments’ reform and renewal programmes, or to more internally driven processes within the
Robust and flexible states
89
Table 4.2 The significance of factors on changes in the unit’s organization and working practices over the past four years. Proportion reporting that the respective factor has been of moderate or major significance (1998, per cent)
Adjustments in connection with EU membership The government’s administrative policy Local initiatives of the agency itself in response to internal or sector-specific conditions
Norway Denmark Sweden
Finland Overall
23
35
42
37
35
22
41
39
32
31
38
56
37
21
38
various agencies. The significance of EU adaptation for adjustments and changes in organization and working practices is lowest in Norway. Also the government’s overriding administrative policy is of relatively little significance as a source of organizational change in Norway, a fact that must be viewed in relation to Norway being a reluctant reformer (Olsen 1996). The significance of locally driven processes is greatest in Denmark and least in Finland, which indicates an important distinction between a decentralized administrative tradition in the former, and one that is more centralized in the latter. These driving forces are, however, not mutual alternatives and independent, but supplement one another in complex interactions (Lægreid 2001b). Collegial arrangements as supplement to basic organizational structures How the Nordic states organize the coordination of the EU work, might be seen as an independent variable that affects the emergence of the transnational state, as described in chapter 1. But the coordinating arrangements can also be seen as a dependent variable indicating flexible adaptation within robust administrative models. This is the approach we adopt here. In recent years, the formal organizational design that determines the ordinary hierarchy and patterns of specialization of the Nordic central administrations has been supplemented by various forms of collegial bodies (Christensen and Egeberg 1997). The same pattern is also apparent in the EU administration. Many of these collegial bodies are formalized and permanent, cutting
90
Robust and flexible states
across the pattern of divisions into specialized central administrative tasks. These collegial arrangements differ, however, from country to country. One important difference is the degree to which their representatives are drawn either exclusively from units in the administration or also from interest organizations in society. Another difference is whether these collegial bodies are permanent and involve voting rights for their participants, or more temporary and hence lack formal voting rights. In Denmark a system was established in 1973 with 18 interministerial special committees placed under the various ministries. The special committees were composed of members from various agencies who could call in representatives for different interests as the need arose (Damgaard and Nørgaard 2000). A special framework guideline procedure was established intended to marshal the government’s views on Danish attitudes to regulatory processes within the EU. In the 1980s the role of the special committees, of which there are now 32, was strengthened. These committees play a crucial role in Danish EU coordination (Statskontoret 2000), and almost half of the Danish units had participated in meetings of such special committees in the year prior to the study. The foreign ministry is represented in all of these groups and has the overall responsibility for coordination, and a mandating procedure has been established, whereby the market commission (now the Europe Commission) of the Danish Folketinget took on a clearer coordinative and administrative function in relation to the government in managing EU affairs (Højbjerg and Pedersen 2002). In Denmark there is also a higher-level coordination committee, the EU Committee, run by the foreign ministry and composed of civil servants from the ministries most affected by the EU. Important issues are also discussed in the government’s foreign policy committee. This body consists of ministers and is chaired by the foreign minister. In parliament, EU issues are dealt with in the European Affairs Committee. This is where all draft EU legislation is discussed and the position adopted by the government depends on whether there is majority support for a proposal in this body. The Norwegian authorities copied the Danish coordination system in 1994, with its special committees, its coordination committee, the government’s EU-committee and the framework guideline procedure, but without the mandating procedures. Today there are 22 special committees in Norway. These consist only of representatives from the administration. In both countries these committees are dominated by
Robust and flexible states
91
experts from the relevant ministries and directorates under the leadership of the most affected ministry. The Ministry of Foreign Affairs retains the right to participate in all special committees. The special committees are subject to a superior collegial coordinative body in the central administration under the leadership of the Ministry of Foreign Affairs. In Denmark this committee meets once a week and functions as a coordinating link between the special committees and the government. In Norway, which does not participate in the political activities of the EU associated with the Council system, the Coordinative Committee plays a less prominent role. The government submits information to the parliament through the EEA committee, a consultative body. Sweden and Finland have chosen another arrangement with fewer permanent organs and greater reliance on the integrated handling of EU affairs within the ordinary coordinative system that already existed within the central administrations. Finland, however, also has a system of cross-sectoral working groups (EU sections), which deal with EU questions in a range of fields, and a higher-level coordination committee (the committee for EU affairs). The Prime Minister’s Office and the Ministry of Finance are represented in every section. They also contain representatives from relevant interest organizations. In 1999 there were 38 EU sections in the Finnish central administration (Raunio and Wiberg 2000a). As in Denmark and Norway, these were led by cross-sectoral working groups drawn from the responsible ministries, with members from the Ministry of Foreign Affairs, other ministries, and the central civil service. Representatives from various interest groups within society also take part in these groups, which have more of an advisory function than their counterparts in Denmark and Norway. In Finland the coordination model was originally dominated by the Foreign Ministry, which was also entrusted with the EU secretariat. But this responsibility has shifted gradually to the Prime Minister’s Office. The EU secretariat was phased out and instead a Government Secretariat for the EU was created in the Prime Minister’s Office. Otherwise responsibility for drafting and formulating Finland’s standpoints on EU matters lies with the ministry primarily concerned. At the political level Finnish EU activities are led by a Cabinet Committee for EU Affairs, chaired by the Prime Minister. Here we find features that are shared by the Danish/Norwegian system. The Finnish parliament has a Grand Committee which acts as a consultative body.
92
Robust and flexible states
Sweden has not established any similar system of cross-sectoral special committees subordinate to a coordinative commission. No formalized system of special groups to coordinate governmental activities involving the state and society has been established in Sweden like those in Denmark and Norway. Sweden compensates for this deficiency with an extensive network of more informal advisory national project and reference groups, where representatives from various units can meet representatives from other authorities, interest organizations, private companies and independent experts. In Sweden there is a range of more informal inter-ministerial working groups to handle the preparation of EU matters. A number of reference groups have also been established, which provide for consultation and the exchange of information with interest organizations (Karlsson 1999). But these collegial bodies are less formal and permanent than in the other countries, and their numbers and composition are continually changing. In Sweden EU issues are dealt with in the same way as all other matters that concern the administration. Each minister is responsible for dealing with EU issues that affect her or his area of competence, and must consult and come to an agreement with other ministries. There is, however, an EU council (EU samra˚d) in which EU coordinators from the various ministries meet, but their consultations are mainly of an administrative nature. The Foreign Ministry was initially given the overall responsibility for coordinating EU issues, but as in Finland this responsibility has been transferred to the Prime Minister’s Office, where there is now a State Secretary for EU Affairs. Parliament occupies a somewhat subordinate role in the Swedish coordinating system, despite the special committee for coordination between the government and the Riksdag, the Advisory Committee on European Union Affairs. Although this committee was modelled to some extent on its Danish counterpart, it does not have the same importance as in Denmark, one reason being that the minority governments in Denmark have been less powerful than in Sweden (Sidenius et al. 1997, Hegeland 1999). The principal pattern that emerges is that a new European dimension has been added to the national central apparatuses, partly in terms of new responsibilities being integrated into the formal organizational designs, whereby the fundamental hierarchical and specialized structures of the central administrations remain essentially unchanged, and partly in terms of a process of adaptation to the EU through the establishment of collegial bodies that cut across the
Robust and flexible states
93
established divisions of responsibilities within the administration. This means that, in relation to EU work, the civil servants have a primary allegiance to their own units and sectors while at the same time participating in collegial bodies that deal with cross-sectoral coordinative functions. Challenges to the administrative models In the Nordic countries we can distinguish two levels of central administration. The inner central administration consists of units with close links to the political leadership that are directed by the minister. The outer central administration is the central bureaucracy, consisting of directorates (Norway, Denmark) or central agencies (Sweden, Finland). There is an important formal difference between, on the one hand, the dualistic Swedish system with its organizational distinction between ministries and independent central agencies that answer to the government collectively, and not to the separate ministers, and on the other hand, the monistic Danish/Norwegian system of ministerial rule, where the outer central administration answers to the ministers, who can be made responsible for all decisions taken by subordinate bodies. Finland occupies an intermediate position between these models. Our findings show, however, that many of the differences between the administrative levels cut across national borders, which indicates that the important formal distinction between the administrative levels in Sweden and other Nordic countries is less apparent in practice. There are clear variations between administrative levels, although they are not necessarily what one would expect on the basis of the differences between the east and west Nordic administrative models. Participation in collegial coordinative bodies is most widespread in the inner central administration, but this does not imply that the central agencies are entirely excluded from coordinative work. Coordination is, however, more widespread at the ministerial level, whereas singular resolutions are more common among the directorates and agencies. Although in relative terms the ministerial units have benefited more from training and the creation of new EU-related posts, the greatest number of new positions has been added in the outer central administration, since this level encompasses many more units. In Sweden especially, EU membership has had considerable significance at the ministerial level. This might be explained by the fact that government offices are sceptical about entrusting matters entirely to the
94
Robust and flexible states
central agencies, and to some extent this scepticism is a consequence of the Swedish administrative model. Contacts with the administrative units and the involvement of interest organizations in collegial bodies are more widespread at the ministerial than the agency level, whereas the degree of involvement of interest organizations and private companies is similar on both levels. The volume of contacts with parliament and the political leadership is much larger in the ministries than in the agencies. The Swedish ministries have particularly efficient channels of contact with political leadership. On the other hand, the Swedish agencies do not have fewer contacts with political leadership than their counterpart in Denmark and Norway, and the contacts that are maintained in the former are more informal than those in Denmark and Norway, a finding that is surprising when we consider the distinctions between the east and west Nordic models. Policy guidance from political leadership also tends to be more precise and more frequently in written form for the ministerial units than for the agencies and directorates, and conversely, the quantity of informal guidance higher for the agencies than for the ministries. In the west Nordic model based on ministerial rule, the formal mechanisms are often complemented by broadly accepted channels of informal contacts. In the east Nordic model, by contrast, informal contacts are sometimes considered problematic, since they can be interpreted as an expression of ministerial control, and hence in conflict with the essential independence of the Civil Service. In particular, the increased need for guidance and coordination in relation to EU work has sharpened the discussion about how effectively the east Nordic model serves its objectives. Do increased demands for national standpoints call for more informal contacts than the administrative model can provide, or are these problems exaggerated? Studies of the Swedish administration have indicated that there are extensive contacts between ministries and agencies (Jacobsson 1984). Table 4.3 shows the patterns of contact that exist between politicians and civil servants in EU-related questions. In Sweden and Finland, contacts between political leadership and administration are more frequent than in Denmark and Norway. This is slightly surprising when we know that the Swedish and Finnish administrative models embody certain restrictions on this kind of contact. Contacts are closest in Sweden, where only one in four units has no contacts whatsoever with political leadership. Almost half the units in the Norwegian administration have no contacts with political
Robust and flexible states
95
Table 4.3 In the course of the past year, how often have the unit’s employees had contacts with the government or the political leadership in relation to EU concerns? (1998, per cent)
Once a week or more often Once a month A few times Never
Norway
Denmark
Sweden
Finland
Overall
10 10 33 47
12 15 31 42
22 15 37 25
22 10 29 39
17 13 33 37
leadership in relation to EU affairs, and only 20 per cent have contacts once a month or more often. This might well be explained by the fact that Norway is not represented in the European Council. The Danish administration also has relatively few contacts with its political leadership, which suggests that there is something in the west Nordic minister-controlled model that leads to reduced contact. There are also differences in contacts between various levels of the administration, and the ministries display a different pattern from the agencies. As we see from Table 4.4, the Danish and Norwegian agencies are particularly limited in their contacts with political leadership in relation to EU concerns. In Sweden the ministerial units have particularly close contacts with politicians. Civil servants and the political leadership in Swedish government offices evidently devote considerable time and energy to discussing EU affairs. As many as 61 per cent of the units in the ministries are in touch with the political leadership concerning such matters at least once a week, and this is far higher than the corresponding figures for the other countries. It also appears that the Swedish and Finnish agencies have more contacts with political leadership about EU affairs than their Danish and Norwegian colleagues. The independence traditionally associated with the east Nordic model of administration, therefore, seems not to result in fewer contacts. Table 4.4 In the course of the past year, how often have the unit’s employees had contacts with the government or the political leadership in relation to EU concerns, according to administrative level? (1998, per cent)
Ministries Agencies
Norway
Denmark
Sweden
Finland
Overall
30 8
30 5
61 12
36 18
40 9
96
Robust and flexible states
Table 4.5 The most common forms of contact between the units and the government/ political leadership in relation to EU concerns (1998, per cent)
Formal, written Formal meetings Informal
Norway
Denmark
Sweden
Finland
Overall
35 35 49
50 31 33
10 31 76
13 20 61
26 30 57
What is special for Sweden is that the contacts with the political leadership seem largely informal (Table 4.5). Seventy-five per cent of all the Swedish units maintain that informal contacts are the most common type. Formal written communication is relatively uncommon both in Sweden and Finland. In Denmark greater emphasis is placed on formal written contacts, and informal contacts are less common. The structure of the Danish coordination system, with its broadly informal design, seems successful when judged by its patterns of contacts. At the same time, there are considerable differences between Denmark and Norway. The latter is significantly less formal and places less emphasis on written communications. Levels of informal contact are similar for the agencies and the ministries. In Finland and Sweden informal contacts between agencies and political leadership are even more common than they are between ministerial units and political leadership. The dualistic model that applies in these countries evidently does not lead to fewer informal contacts. Both in Sweden and Finland, ministerial units spend more time on formal meetings with the political leadership than the agencies. When we compare the east and west Nordic administrative traditions, we discover interesting and somewhat surprising patterns. In the dualistic systems, which strongly emphasize the boundary and division between politics and administration, contacts and informal relations are in practice extensive. In the monistic system, where the formal rules are built around ministerial control, we find in practice fewer contacts and more formal relations, relatively speaking. One explanation for this might be that the formal design of the ministerial control model is more effective in terms of political direction than the independent civil service model. Control mechanisms are more embedded in the structure and informal contacts are consequently less necessary.
Robust and flexible states
97
Another explanation might be that the development of patterns of contact is influenced by more than just the administrative model. The Danish coordinative system is built up as a hierarchical and formal system corresponding to an idealized model of parliamentary governance. The latter was created in the early 1970s, when Denmark became a member of the community. At that time, a system of organizations was developed based on the interplay between administration and politics, and this was later implemented (Pedersen 2002). Although the various special committees are free to make their own decisions, the system also assumes that professional bodies will be consulted. On receiving initiatives from the Commission, the necessary coordination between units functions automatically. The initiative is passed on within the administration and from the administration to relevant actors within the public sphere. Many actors are included in the formal written system. In Sweden it was declared at an early stage that ‘the Swedish administrative model is permanent’ and that the existing system of management should apply also for EU matters. No attempt was made to construct a special system for handling EU affairs similar to the one in Denmark. The system involving integrated participation of interest organizations and private companies that had been developed in Denmark reflected corporatist ideals that had become unfashionable by the 1990s (Rothstein and Bergstro¨m 1999). One interpretation of the pattern arrived at might be that the formal Swedish system of governance fails to cope with the EU workload, and that the units are, therefore, obliged to engage in a high level of informal activity alongside the formal arrangements. Whereas the Danish administration takes care of most issues via the formal system and only rarely acts informally, the situation in Sweden is the opposite. There the majority of EU concerns are dealt with informally. At the time when Sweden became a member of the EU the corporative pattern was hardly fashionable and many influential actors had played a part in building up a general system of management and communication between administration and politics based on management by objectives and results. It was assumed that this system would also be capable of handling EU routines. The high degree of informal activity that we have registered, especially in relationships between administration and politicians, can be seen as an indication that this was not the case. The pattern can be interpreted as reflecting significant mistrust in the formal management and communication system. The degree of informal relationships is also high in Finland
98
Robust and flexible states
and fairly high in Norway. Regardless of how the various countries have organized their internal management and coordination systems, there are factors that make informal contacts necessary. Adaptation within robust organizational structures Increased integration in the EU has neither undermined nor resulted in the dismantling of national administrative apparatuses. While the established forms of organization persist, the challenges posed by the EU result in fairly pragmatic adaptations in terms of staffing, reorganization and distribution of responsibilities. These changes supplement existing arrangements rather than replace them, and, therefore, tend to increase the complexity of the central administration. Although the formal organizational designs are relatively stable, EU involvement has led to changes in the distribution of administrative responsibilities both horizontally and vertically (Lægreid 2001b). The national administrative apparatuses seem to have considerable inherent flexibility enabling them to cope with the challenges posed by the EU without being threatened in any fundamental way (Bulmer and Burch 1998). Our interpretation is that changes in forms of organization largely reflect incremental local adjustments rather than a coordinated and planned reorganization imposed from above with the aim of meeting EU demands (Cyert and March 1963). The prediction that we should find major convergence and structural similarity between organizational forms (Rometsch and Wessels 1996) draws little support from our material from the Nordic countries. The changes in established forms of organization are too small and the differences in the adaptations, with Norway at one end of the scale and Finland at the other, are too considerable to allow us to speak of structural fusion between the Nordic central administrations at the EU level (Andersen 2000a). This tends to corroborate the robustness thesis as far as changes in formal organizational design are concerned. European integration takes place within reasonably robust and loosely defined organizational models. Even if the EU has had little influence on formal national arrangements, there have nevertheless been significant changes to existing forms of organization in terms of the establishment of collegial bodies, new work responsibilities, additional training, increased staffing and changes in the distribution of tasks. This implies both continuity and transformation (Eising and Kohler-Koch 1999).
Robust and flexible states
99
Adjustments to the Nordic forms of organization have clearly divergent features. Robustness with regard to higher order arrangements is found alongside substantial difference in the adjustments that have been made. Pressure to adapt to the EU is usually met on the organizational level with an integrated solution that involves delegating EU-related tasks to existing units rather than with a specialized structure of units designed to deal exclusively with this kind of work. At the same time the existing organizational structures are supplemented with more flexible collegial bodies. Pragmatic adjustments are made within the established organizational forms through the initiation of various coordination committees, and at lower levels through the creation of special new posts to deal with tasks relating to the coordination of EU matters. The principal adjustments involve ways of absorbing EU affairs into the existing national forms of administration. Neither has there occurred a centralization of EU affairs within the foreign ministries; as with other matters, EU affairs are primarily the responsibility of the individual ministries, and much of the additional capacity that is being developed is concentrated within units of the outer central administration, the directorates and the agencies. The special position of the foreign affairs ministries as coordinating bodies is now under pressure, against a background in which EU issues tend to be entrusted to professionals and experts in sectoral ministries, central agencies and directorates (Christensen 1996, Nilsen 2001). At the same time, the governmental apparatus assumes a more central role as a coordinating organ at the national level. The picture that emerges is that although the established administrative models face challenges and are being adapted to varying extents, they are not being dismantled. The central administrations are undergoing various pragmatic modifications to meet the challenges posed by increased EU integration. Instead of tailoring and fine-tuning the formal organizational designs to deal with EU integration, the trend is to modify pre-established and only loosely defined organizational forms that have already proved effective and workable and which have reasonable scope for variation in administrative behaviour and work strategies. This proves a sensible approach to administrative design (Olsen 1997). Closer integration between the Nordic countries and the EU has not undermined the characteristic features of the national administrations, but EU affairs have been absorbed into the logic and administrative traditions of the national administrative systems of the respective
100
Robust and flexible states
countries. There is interplay between continuity and change within the existing order (Olsen 2002b). This kind of development, involving increased EU integration without fundamental formal changes in national institutional arrangements, is similar to that of other EU countries (Knill and Lenschow 1998a, Eising and Kohler-Koch 1999). The core of the national administrations is robust and shielded against fundamental EU transformation, while still allowing significant informal adjustments and variations in actual patterns of behaviour. This becomes apparent, among other things, in the complex relationships between contact and political control. Extensive contact does not necessarily imply that politicians exert control, as can be seen in the Swedish case. In Denmark the combination of robustness and flexibility is reflected in the somewhat paradoxical situation that its administrative units find it relatively easy to predict politicians’ wishes in relevant EU matters, even though they have relatively little contact with the political leadership. Moreover, they claim to be fairly successful in persuading the political leadership to accept their views and objectives. The well-established Danish coordination system seems to provide opportunities for administration and policy makers to reach consensus. This suggests that there is an interplay between robust formal organizational structures and flexible informal relations based on anticipated reactions and autonomous adaptations. The most extensive organizational changes in the new member states Sweden and Finland consist of the addition of new posts and further training for existing staff. Sectoral studies of the labour inspection authorities, the competition authorities and the pharmaceutical agencies in Norway and Sweden confirm that the official form of EU affiliation has considerably influenced the changes made to forms of organization (Bue 2000, Middlethon 2000, Dyrdal 2001). The differences between Norway and Sweden in terms of EU adaptation within these three areas have increased significantly since Sweden became an EU member. Although the Swedish model of independent agencies has been declared permanent, it can still accommodate considerable adjustments in patterns of behaviour, and it has been shown that actual practices correspond little with what one would expect on the basis of judicial and legal forms alone. In keeping with the Nordic tradition of involving private interests in the shaping of policy, there is considerable use of collegial bodies also in this policy domain. The kind of collegial bodies involved vary considerably from country to country. Norway, Denmark and Finland
Robust and flexible states
101
have developed internal systems of coordination within their respective central administrations consisting of special committees and coordination commissions, whereas Sweden has opted more for public commissions and various project, work- and reference groups. The central administrations of Sweden and Finland have considerable EU expertise, but in order to assess it one has to remember that the starting point for these countries was very different from Denmark’s. This might also reflect the difference between the east and west Nordic administrative models, although the variations are perhaps not as considerable as one would expect on the basis of the formal distinctions. Although the EU has no direct influence on how national administrative apparatuses are organized, it can still have an indirect impact on organizational forms (Hix and Goetz 2000). Especially in relation to the internal market, communications and regional policy, the EU can spur the abolition of national monopolies, and the establishment of autonomous inspectorates and units for national planning (Moen 1998, Bræin 1999, Andersen 2000b). The main conclusion is that the organization of central administration is characterized by robustness while at the same time permitting flexible adaptation. The established administrative models, as apparent in structures for ministerial control, in the independence of the civil services and in the design of departments and directorates, are relatively stable and allow adaptations to be made within their existing structures.
Robust and flexible administrations Developments involving transnationalization, segmentation, coordination and multiplier effects suggest a pattern of extensive change. Increased integration in Europe has led to significant adjustments to informal structures, rules and regulations, patterns of participation, contacts and coordination, and the activities and relationships associated with the various formal administrative models have changed significantly in response to increased European integration. The Nordic countries still turn primarily to one another in search of inspiration and ideas for how to reorganize their activities. There is also broad scope for changes in the administration to be influenced by national administrative policy and by local, internal domain-specific initiatives.
102
Robust and flexible states
At the same time as the Nordic countries are becoming more tightly involved in European cooperation, they are also experiencing administrative differentiation. Although changes have been extensive, they have not amounted to a fundamental break with established organizational forms. By means of pragmatic modifications to their existing organizational structures, the Nordic administrations are adapting to contend with the new tasks and challenges of European integration. The Nordic administrative models, whether of the Norwegian and Danish kind, organized around departments and directorates, or the Swedish, based on independent agencies, together with the Nordic tradition of cooperation through well-established corporative arrangements, all face challenges and are being modified and supplemented, yet not dismantled. One can speculate about the dynamics of the situation in which the established organizational forms and the traditional administrative models seem robust even while accommodating extensive changes in the day-to-day activities of much of the administration. There can be no doubt that the organizational forms allow broad variation in practice, but the question is how we should conceptualize the dynamics of the relation between form and practice. Is it that the relevant forms allow ‘any practice whatsoever’, in other words, that form and practice develop independently of one another, or is there a connection between form and practice? In other words, do formal arrangements impose requirements and limitations on the activities that can be pursued within them, while at the same time leaving scope for variations in practice? The interpretation offered here is in line with the second possibility. In this study we have seen a dynamic in adapting to Europe that has to be studied in terms of both continuity and change. Although the administrative models allow considerable variations in practice, practice is not independent of the formal arrangements. Denmark offers an illustration of how informal processes can reflect formal mandating procedures (Esmark 2002a, Højbjerg 2002). Informal contacts and forms of participation are built up that reflect the formal mandating procedures (Pedersen 2002). In order to understand these changes within the central administrations we must take three significant temporal factors into account. The first concerns the time period. The fact that we conducted our study of integration processes in a specific temporal setting, namely the late 1990s, when the European Union was being introduced and the internal market extended, means that the effects of Europeanization on
Robust and flexible states
103
the national Nordic administrations display a particular dynamic. The second temporal factor is duration, or the length of time that a country has been a member of the EU. With its long interaction and experience of involvement in the EU, the Danish administration is in a very different class from those still immersed in a transitional phase. In other words, the basic situation differs from country to country. The third temporal factor is simultaneity. In terms of the multiplier effects, we have shown that there is an overlap between the relations that develop to cope with EU concerns within the Nordic, international and national contexts. Transnational European relations tend to provide support for and to supplement existing relations more than they undermine and weaken them. In certain respects our interpretation of these changes conflicts, on the one hand, with the traditional instrumental reform perspective, according to which structure shapes behaviour and organizational design and redesign change behaviour (Goodin 1996, Offe 2001, Stinchcombe 2001), and on the other, with a strong position within neo-institutional theory (Meyer and Rowan 1977, Brunsson 1989), predicting that significant changes can be made to formal structures without necessarily leading to changes in practices. Window dressing and vague contacts result in a mismatch between talk and action. Structures are regarded as fac¸ades and myths that have no particular significance for patterns of behaviour. In this study we have shown, however, that major changes can occur in practice and in day-to-day administrative work, even though the formal organizational design remains fairly robust (Olsen 2002b). The implication of this is that organization matters, but in ways that need further specification of the conditions that must apply and how it matters. To understand the reasons for the emergence of this combination of robustness and flexibility of administrations in relation to the EU integration we have to examine the path-dependency of administrative traditions and translation processes. This is the theme of the next chapter.
5
Translating states
In the previous chapters we have examined what kinds of impact European integrative efforts have had on the organization and work procedures of the Nordic central states. In this chapter, we will try to account for the changes observed. We will use the broad theoretical perspectives that were presented in the first chapter. The first is an environmental-determinist perspective, in which adaptation is traced primarily to external pressure from the EU. Here, we expected the response of the administrations to be influenced mainly by the form and length of affiliation with the EU. The second perspective, the historical-institutional one, considers how administrative traditions and the particularities of national administrative regimes determine modes of adaptation to the EU. A broad distinction is drawn between the east Nordic and west Nordic administrative models. The third perspective focuses on national strategies and conscious choices made by the political leadership. According to this perspective, national EU strategies, as expressed in the organization of national mechanisms for the coordination of EU issues and national administration policy, will be important explanatory factors. Fourthly, we will discuss a translation perspective, which combines elements of the other perspectives. Here, national strategies as well as administrative structures and traditions influence how external demands are interpreted and translated, and shape perceptions of which problems are relevant and what constitutes good solutions. Combining theoretical perspectives which are so paradigmatically different may give rise to problems. One way of dealing with them is to allow the theoretical perspectives to co-exist with each other and to offer different ways of understanding. In Allison’s classical study of the Cuban Missile Crisis, three models are juxtaposed: one rational actor, one organizational process and one bureaucratic politics. These
106
Translating states
models explain events in different ways and each and every model adds new understanding (Allison 1971). In Images of Organization Morgan also argues for a form of multi-perspective view, or in other words that greater understanding can be gained from the use of several different metaphors or ways of thinking (Morgan 1986). In this way theoretical perspectives can offer greater possibilities of understanding without necessarily having to be linked to each other. At the same time varied theoretical starting points may supplement each other and thus provide a more multifaceted understanding of the phenomenon studied than if only one perspective (or one perspective at a time) had been adopted. Campbell and Pedersen (2001) argue in their study of The Rise of Neoliberalism that ‘if you want to understand institutional change in specific empirical settings, then you may have to weave together an analysis that draws together a lot of ideas that are sometimes viewed as competitors’ (ibid: 204). These authors claim that the various institutional perspectives or paradigms that have been developed may overlap and complement each other in ways that can provide significant theoretical payoffs. There are different ways of relating disparate theoretical perspectives to each other (Campbell and Pedersen 2001, Roness 1997). One may, to begin with, link theories to each other by specifying, for instance, scope conditions under which certain theories may be used or by determining certain phases where certain theories are applicable and other phases when others are to be adopted. In doing so, the specificity of the various theories is maintained. Secondly, one may blend insights from the various theoretical perspectives to blur the boundary lines that divide them. Thirdly, one may attempt to identify some common basis – a shared problem or a shared adversary – for the theoretical approaches used. A fourth possibility is to relate the perspectives hierarchically, by ascribing superordinate status to some of them and maintaining that it provides the framework in which all the others can be understood. We argue that Europeanization can be understood to a great extent as a translation process, i.e. a process in which there are certain rules and conditions laid down by the EU but where national strategies and historical traditions in the various member states play a major role in determining the outcome. Adopting the translation perspective will mainly involve blending insights from the three other perspectives. We claim that a) the EU’s method of regulating and organizing, b) national strategies and attitudes, and c) historical traditions in the form of administrative models, for instance, are all important factors for
Translating states
107
understanding the way in which a translation is formulated in practice (Olsen 1992). At the same time, translation can be seen to provide its own theoretical perspective. In this way the translation perspective becomes more procedural and focuses to a greater extent on attempting to understand the ways in which pressures from EU, national strategies and historical traditions are blended. The translation perspective aims to demonstrate through which processes the various factors interact. It stresses the active role of the receiver and that both what is translated and those who translate are altered in the process. Emphasis is placed on the two-way nature of the translation process. We see a co-evolution where European organizations and ambitions as well as national organizations and ambitions are changed in the process. One fundamental premise here is that changes evolve through transnational processes, and it is, therefore, not particularly productive to maintain the boundaries between the national and the European, for instance. The import of what we refer to as EU demands is something that must always be interpreted, and this interpretation involves an exchange between the national, European and other organizations. In an administration that has become transnational to a great extent and where changes evolve in segmented processes there is continual translation of both what is European and national. The extensive scope provided for translation of this kind also makes it less likely that we will find uniformity in the Europeanization of the various states. We shall conclude this chapter with a more extensive discussion of the translation perspective and how it relates to other ways of understanding Europeanization. Initially, however, we shall discuss our observations concerning the Europeanization of the Nordic countries (i.e. those described in chapters 2 and 3) using each of the theoretical perspectives separately. We shall discuss the relevance of the specific explanatory factors we enumerated in the first chapter. In the first of these perspectives, i.e. when pressure from the EU is emphasized, both the form of affiliation to the EU and how long states had been members were considered decisive.
Pressure from the EU: form and length of affiliation One plausible conclusion is that the form of affiliation that Nordic states have with the EU is important for the pattern of adjustment that takes place, even if it is not in itself decisive for the changes that
108
Translating states
emerge. Even though the EU permits considerable variation in individual national arrangements, formal affiliation appears to be of considerable significance (Trondal 1999). Thus, the extent and level of adaptation to the EU is less in Norway than in the three member states. Association with the EEA results in much less coordination than affiliation with the EU. This study does not support the view that the Norwegian administration has undergone excessive adaptation in order to reduce uncertainty and increase legitimacy (Sverdrup 1998). There are significant differences in adaptation to the EU between the new members, Sweden and Finland, and Norway, regarding contact with the Commission and the Directorates General. At the same time, Norway’s participation in comitology committees (i.e., the bodies concerned with formulating administrative legislation in the EU) is on a similar level to Denmark’s. A relatively high level of participation in these committees may be seen as a form of compensation for absence from the Council. The Norwegian delegation to Brussels may function as a lobby organization for Norway in those EU bodies where Norway is not represented (Jeppestøl 1999). Thus, while one might have expected the non-EU member Norway to score lower than Sweden and Finland on most indicators of change, a more surprising finding is the Norwegian administration’s high level of adaptation to these parts of the EU. However, it is not only the form of affiliation that is significant. Length of membership is also important, as is indicated by the general differences between Denmark, on the one hand, and Sweden and Finland on the other. Becoming a member in the 1990s required more dynamic administrative adaptation than earlier accession to the EU. In an effort to catch up with established EU members, new members tended to engage in a high level of activity in the areas of contacts, participation, coordination and expertise. Even though the field of coordination has undergone considerable development and expansion within the EU since the mid-1990s, the established member state, Denmark, is still experiencing a lower level of adjustment and adaptation than the central administrations in Finland and Sweden. One obvious explanation is that the day-to-day running of the Danish central administration, after 25 years of membership, has become integrated into the EU in quite a different way than in the other two countries. Changes that do occur are perceived as less dramatic in Denmark than in Sweden and Finland, for whom almost any degree of integration of their administrations into the EU is new.
Translating states
109
Another way in which the form of affiliation is significant is that the traditional core tasks of the EU linked to the historical areas of integration, namely, the inner market and the four freedoms, clearly emerge as those parts of the administrative system that are most affected and where the greatest adaptations have been made in all countries, something shown by segmentation. There is no determinism in the manner in which these changes in the organization and working manner of the administration is undertaken. Rather, it is the product of interplay between external demands and national adaptation. Those policy areas where the EU has the greatest ambitions and the bestdeveloped and established organizations and bodies of rules are also those areas where national administrative activities must make the greatest efforts to adapt. The redistribution of resources in the EU system occurs first and foremost in agriculture and the structural funds, resulting in strong transnational relations within the agricultural, industrial and commercial sectors. These are also the areas where the EEA Agreement is of greatest relevance. Nevertheless, the impact of the EU’s organizational principles on domestic coordination patterns is not simple and straightforward. It is not the case, for example, that Norway – which is associated with the Commission through its sector logic rather than with the Council through its area logic – has a more developed vertical sectoral coordination than the other countries. This illustrates the point that the link between the demands of the EU and national adaptation cannot be described in terms of simple determinism. The distinction between membership and non-membership is, however, not absolute, as Norway’s intermediate position as an EEAmember illustrates. There are also differences between the various forms of membership, as observed in the many exceptions pertaining to Denmark, particularly in areas outside the first pillar. To a certain extent, the question of membership is more of a continuum than a dichotomy (Trondal 2001a). Even though the formal terms of affiliation and the length of affiliation are of significance, the differences between Norway and the member states are nevertheless not so wide as to suggest the interpretation that non-members follow a far more radical line than members regarding adaptation within the national administrative apparatus (Rometsch and Wessels 1996, Sverdrup 2000). Even if they are important, the form and length of affiliation are not the only elements that may explain variations in administrative adaptation.
110
Translating states
Historical legacy and administrative models In this study we have shown that differences exist between the dualistic east Nordic model adopted by Sweden and the more monistic west Nordic model with ministerial governance that characterizes Denmark and Norway. However, these differences may be more a question of form than of practice. For example, it is not the case that there are significantly fewer administrative units in the central administrations of Norway and Denmark than in Sweden. In spite of the important dualistic element, which suggests that Sweden should have small ministries, they are not that much smaller than in the other Nordic countries. In line with our expectations, we find a much closer and integrated contact network between the ministries and the political leadership in Sweden than in Denmark and Norway. There is close cooperation between the political leadership and the units in the Government Offices (regeringskansliet). But the Swedish central agencies do not have less well-developed contacts with the political leadership than in Norway and Denmark. We do not find stronger vertical coordination regarding EU matters in Denmark and Norway than in Sweden. Whether these differences are the consequence of special relations in EU matters or whether these are an expression of a wider gulf between formal arrangements and practice (Jacobsson 1984) is still an open question. What can be stated, however, is that the clear differences that emerge between Sweden and Denmark/Norway do not correspond to traditional ideas about what characterizes the east Nordic and west Nordic models. One interpretation is that the east Nordic model, with its clear formal differences between the departments in the Government Offices (regeringskansliet) and the central agencies, results in the development of closer, yet less formal contacts in EU matters. By contrast, the west Nordic model largely leads to formal contacts between the administrative levels. Although the central agencies are supposed to be independent under the Swedish model, the Swedish central agencies have just as close (or even closer) informal contact and more explicit guiding signals from the political leadership than the central administrations of the other countries. Even though the Swedish model continues to thrive both formally and rhetorically, there is nevertheless reason to question certain of its aspects. The historical legacy is visible, inasmuch as EU activity is integrated into established organizational forms. Furthermore, procedures
Translating states
111
whereby interest organizations and other affected parties are integrated into the decision-making process through routine participation in different corporate bodies, is a well-assimilated element in the Nordic tradition. New organizational methods through collegial internal bodies develop as a supplement, not as an alternative to the traditional organizational structure and correspond otherwise to the general tendency towards broader collegial organizational forms within the central administration (Christensen and Egeberg 1997). One interpretation of this is that national administrative tradition and administrative models are relatively well adjusted to the EU’s profile and demands, and it is, therefore, relatively easy to incorporate the new into the old. Even though changes have been comprehensive, they have nevertheless been largely in accordance with national traditions. Similar findings have been made in the UK (Bulmer and Burch 1998). The formal arrangements are, however, not immutable and do not permit all forms of practice. Even though it is officially maintained that the administrative models are fixed, our data show that the Swedish dualistic model, in particular, is closer to the European model of ministerial governance than one would expect, given its formal characteristics. It is not the case that a greater proportion of EU matters are assigned to the central agencies in Sweden than in the other countries, or that these agencies are less controlled than the corresponding organizations in the other Nordic countries.
National strategies and the organization of EU work National administrative policies enforce variation, ranging from Norway’s ‘No’ to full membership, through Denmark’s many derogations, Sweden’s reservations about the EMU to Finland’s relatively enthusiastic attitude to the EU. During the period when this survey was conducted, Norway had a ‘No to the EU’ government, and this may possibly have contributed to weakening political pressure for close cooperation with EU, which had been the strategy of the previous Labour government following the popular rejection vote in 1994. The Norwegian strategy was essentially to delegate adaptation to the EU to bureaucrats and experts and allow EU matters to become an integrated element in the administration’s everyday work. The collegial coordinating committees based on the Danish pattern, which were established for prospective EU membership, continued to exist even after Norway rejected EU membership, but they had less of a central
112
Translating states
coordinating function than intended (Sætereng 2001). In contrast to the Danish parliament, the Norwegian parliament (Stortinget) was also assigned a more modest advisory role in EU matters (Nordby 2000). The relatively modest scope and depth of Norwegian adaptation to the EU, as it emerges in our material, must also be seen in the light of the considerable opposition to the EU among its citizens, which restricted the latitude accorded to the political leadership and consequently to the civil service. The adaptation strategy Norway adopted vis-a`-vis the EU was only partly successful. This no doubt had something to do with the limitations imposed by links with the EEA but also reflected the changing importance attached to EU matters by different governments. One factor that may explain why Norway consistently emerges with the fewest changes and adaptations along the dimensions studied is its relatively cautious administrative policy. Whereas in the other Nordic countries the momentum for reforms was coordinated by stronger cabinets or coordinating ministries, in Norway it came mainly from sectoral ministries. The tradition of being a reluctant reformer with a segmented administration policy (Olsen 1996) may have affected Norway’s adaptation to the EU. Denmark has adopted a positive strategy towards economic integration as expressed through the four freedoms and the internal market, but it remains sceptical about political integration and the development of the EU in a supranational federal direction. Six referendums have considerably restricted the latitude of the central administration. Although an EU member, Denmark has been granted exceptional status in a number of areas. It is this hesitance towards many aspects of EU policy that places the Danish central administration in an intermediate position between Norway on the one hand and Sweden and Finland on the other. At the same time, Denmark’s coordination of EU matters is fairly centralized. This is seen in the relatively strong position of the parliament through the European Affairs Committee (Europaudvalget), in the central role of the Foreign Ministry in national coordinating committees, and also in the fact that interest organizations are given increasingly integrated participation rights in the same committees. This has helped to make forms of contact on EU issues in Denmark more formal than in Sweden or Finland, where the national coordination systems are structured differently. These trends in cooperation result in Denmark having a relatively strong system of national mandates in EU matters, something which possibly contributes to stronger domestic support for the Danish
Translating states
113
position. This may also be one reason why the Danish administration has a more proactive attitude and has (at least in its own view) achieved a greater degree of success than the administrations of the other Nordic countries. It would not be true to say, however, that the coordination of EU matters in Denmark has recently become more extensive, since coordination activity was already at a higher level than in the other Nordic countries and cooperation is concentrated to a smaller number of units in Denmark. A somewhat more centralized strategy than in Norway corresponds well to the general Danish administrative policy, in which the Ministry of Finance plays a stronger role in coordinating activities than in Norway (Lægreid and Pedersen 1999). In Sweden, there is also relatively strong opposition to the EU among the population. This manifests itself particularly in the Swedish rejection of the EMU. As a new EU member state, however, Sweden has adopted a largely proactive stance – i.e. it has made an active attempt to change the organization and working methods of the EU to bring them more into line with the Swedish administrative tradition. One example is its endeavours to increase the level of openness and transparency of EU procedures and decision-making processes. EU work in Sweden has become largely politicized. Regarding the level of EU coordination, Sweden has adopted a different strategy to Denmark and Norway. Sweden does not have the same system of permanent collegial coordinating committees and has largely left decisions about how coordination should proceed to the individual ministries. This has produced a ‘looser’ and more informal network of advisory groups. Generally speaking, much of the coordination in Sweden is carried out informally. At the same time, the Prime Minister’s Office has taken precedence over the Foreign Ministry where coordination is concerned, and the role of the Riksdag is more modest than that of its Danish counterpart. According to our data, the difference between the Danish parliament (Folketinget) and the national assemblies of the other countries is not as great as previous studies suggested (Norten 1996, Wessels and Rometsch 1996). One indication of this is that the Danish units report a higher level of success in gaining acceptance of their viewpoints and intentions in the Folketinget than their counterparts do in the Norwegian Storting or the Swedish Riksdag. Contact with companies is also somewhat more developed in Sweden, reflecting the differences in the commercial and industrial structure in the different countries. The main
114
Translating states
impression is that in Sweden the administration and the government have extensive and deep relationships with the EU and take a fairly proactive approach even though this seems to yield limited success. The Swedish strategy towards the EU can also be seen in relation to Swedish administrative policy, which has incorporated New Public Management (NPM) models for many years, coupled with elements of structural devolution, privatization and management by objectives and results. While NPM – with its focus on market competition, management orientation and cost efficiency – may be said to be well in line with the EU as a new liberal market development venture, the comprehensive role of informal coordination, both at cabinet level and between the cabinet and central agencies, suggests that EU coordination has largely occurred alongside these other forms of control and coordination. For Sweden, Europeanization constitutes a challenge both to administrative models and to administrative policy, but as yet Europeanization has not led to any profound discussion concerning the organization of the civil service. Finland’s strategy towards the EU must be regarded in the light of the loosening of ties with Russia following the fall of the Berlin Wall and the dissolution of the Soviet Union. Even though there is significant EU scepticism in Finland, the new foreign policy situation has led to a powerful re-orientation towards the EU within the civil service and at the political level. The Finnish strategy has largely been that of the ‘model pupil’, motivated by a desire to be accepted as an equal and full member of the EU and to be regarded as a ‘good European’ as soon as possible. Unlike Sweden, Finland has entered the EU with very few aspirations of changing the Union to fit its own image. Also, at the time the survey was being conducted, the Finnish central administration was preparing itself both organizationally and psychologically for assumption of the EU presidency, even though this was a year prior to the inauguration. The organization of Finnish EU coordination lies somewhere between Denmark/Norway and Sweden. Even though a number of collegial coordinating committees exist, Finnish coordination is more informal in character than in Denmark or Norway. One expression of this is the government’s ‘teˆte-a`-teˆte’, where bureaucrats and members of the government meet once a week for a free and open discussion of current matters (Selovuori 1999). This implies that the Finnish civil service engages in comprehensive coordination while at the same time appearing to have considerable autonomy with regard to the political leadership. However, one should
Translating states
115
remember that the positions of senior civil servants in the Finnish central administration are political appointments, which contributes to politicization of the Finnish bureaucracy. While Sweden emerges as politicized concerning coordination, Finland is more bureaucracycentred. But this does not necessarily suggest that the Finnish civil service is unresponsive to political signals. In Finland there are also clear parallels between the formulation of administrative policy and the intensity of adaptation to the EU. In the same manner as in Sweden, Finland has operated a relatively intensive NPM-inspired administrative policy. An extensive administrative reform process is related to comprehensive adaptation towards the EU. In summary, it may be said that there is a relationship between national strategies as expressed in the formulation of administrative policy as the basis for EU activity and the actual adaptation that occurs within the central administration. Those countries, which have introduced the most drastic administrative reforms, are the very countries, which have engaged in the most comprehensive adaptation to the EU. Conversely, Norway, which has been a hesitant administrative reformer, also emerges as a cautious adapter to the EU compared with the other countries.
Blending the theoretical perspectives EU related adaptation within the central administration cannot be traced back to a single explanatory factor or one basic perspective. The changes which take place in an administration as a consequence of increased integration in the EU are neither purely an adjustment to pressure and demands emanating from the EU nor solely the result of a conscious national strategy or of historic legacy and administrative traditions. Instead, they reflect a complex interplay between external pressure, national strategies and choices and particular path dependencies that can be traced through the administrative history of each country (Campbell and Pedersen 2001, Christensen and Lægreid 2001, Olsen 1992). EU integration is significant, but its significance is not as simple as is often claimed in the literature on Europeanization (Goetz 2000). It is obvious that the administrative apparatus in each country cannot be regarded as a consummate actor with a large degree of homogeneity, such as is maintained, for example, by intergovernmentalists (Moravcsik 1993). The strong segmentation shows that the differences
116
Translating states
between sectors and administrative levels in certain cases can be just as important as the differences between countries. EU activity pays heed not only to national borders but also to sectoral limits and organizational boundaries, something, which is a central feature of transnational administration. The dichotomy between European interests and national interests is doubtful. Identities, as well as loyalties and operations, are created in processes, which traverse national boundaries. The transnational administration, to some extent, is both national and European. To see the state as a coherent and rational actor does not seem to be fruitful. However, this does not mean that states are becoming obsolete, and will eventually wither away. States are still important as organizations, but they must (like all organizations) be understood in relation to a wider environment of organizations, rules and ideas. States as organizations used to be described as regulatory, autonomous and coordinated, but what we see in this study is that they are also transnational, segmented and regulated (Jacobsson 1999). States issue rules, but they are also subject to European rules and ideas. States still produce national strategies, but these strategies are also created in transnational processes. States present themselves as monoliths, ‘weare-speaking-with-one-voice’-actors, but in practice different parts of states may also act loosely coupled from other parts. We propose a more complex model of states, which includes the idea that some parts of states may be less coordinated and more transnational than others. There are no clear distinctions between what may be considered European and what may be considered national and there is interplay between structural constraints and conscious actors. Adjustment impulses, which frequently arise through increased integration, are translated and edited to make them compatible with national administrative traditions and political strategies (SahlinAndersson 1996). A perspective of this type implies that we must combine ‘internal’ and ‘external’ factors in order to understand why EU adaptation varies in content, depth and intensity in the Nordic countries. Demands from the EU are often ambiguous, unclear and multifarious and they are filtered, interpreted and modified in different ways (Christensen and Lægreid 2001), both by the national politicaladministrative history, culture, tradition and decision-making style but also by national administrative policies and EU strategies. Transnationalization is also a twofold process. Influences also travel from the European North to Brussels, and the Nordic countries have
Translating states
117
been influential in the discussions about the environment, public openness, employment and enlargement. A translation perspective of this kind is critical of the idealistic notion that individual states may freely choose their method of adaptation to the EU and that actors have a complete overview and power over reform processes with respect to the EU. It is equally critical, however, of the fatalistic notion that development cannot be influenced at all, that there is no choice whatsoever and that, therefore, states are obliged to adapt to the external demands of the EU. Rather, this perspective offers an intermediate position where political leaders have a certain amount of leeway regarding their own choices and strategies, while their possibilities for manoeuvring are restricted by environmental constraints and national administrative structures and traditions and their attitudes and actions are formed and moulded in transnational relations. Problems and solutions are interpreted, edited, modified and revealed within a process of complex institutional change. This translation perspective adds complexity to the understanding of adaptation to the EU in the national administrative apparatus. This may make the presentation less elegant but hopefully more realistic (Christensen and Lægreid 2001). Pressure from the EU may have considerable effect if it is strong, consistent and tightly woven into the administrative structure and working methods of the country in question and if it receives active support from the political-administrative leadership. Conversely, if it is ambiguous, inconsistent and only tenuously linked to the administration’s day-to-day work, and, furthermore, if it encounters opposition or only negligible support from the national political administrative leadership, its impact on the national administrative apparatus may be marginal. These are extreme positions. As illustrated in this study, adaptation to the EU in practice is more complex, inconsistent and vague in its execution. The demands from the EU are not always clear and consistent. National traditions and structural arrangements represent broad categories which permit correspondingly broad variations in the pattern of adaptation, and there are frequently conflicts and opposing views among national actors about how the strategy for adaptation to the EU should be organized and what its content should be. This results in administrative units discovering values and solutions in transnational relations. These divergent development trends are apparent in the different response patterns of the individual states as expressed in variations in the intensity of change and in broad and converging developmental
118
Translating states
trends in areas where the response pattern is either similar across national borders or proceeds in the same direction. This corresponds with the pluralistic approach, which allows for different coordination models within the national administrative EU adjustment process (Spanou 1998). Increased integration within the EU generates changes in the administrative structures and routines, although the pressure for adjustment varies from country to country, depending on the form and length of affiliation with the EU and the degree of correspondence between what is occurring in the EU and the traditions and institutional structures of the country in question (Jupille and Caporaso 1999). It is our interpretation that the administrative adaptation required by increased EU integration is generally perceived as falling within acceptable and reasonable limits as far as the administrative traditions of the individual countries are concerned (Knill 2001). At the same time, the transnational relations that emerge from this adaptation process must be regarded as something new that cannot be wholly traced back either to national structures, or to demands from the EU. European integration efforts produce a tendency towards homogenization and convergence, but more so in legal texts and policies than in structural arrangements and practices (Olsen 2002b). If the focus is on administrative arrangements, no significant convergence towards a common European administrative model can be seen to have occurred (Hanf and Soetendorp 1998, Harmsen 1999, Olsen 2002b). Demands may be experienced differently depending on how long a state has been a member. Norway, with only one foot in the door, will naturally experience greater pressure from the EU along some dimensions than along others. Similarities between the states can also partly be understood as the product of a learning process. For a long time Nordic states have borrowed from one another and learned from each other’s experiences. One implication of this may be that Europeanization not only occurs via a vertical logic through an adjustment process downwards from the EU, but also via a horizontal logic incorporating learning and model-borrowing across national boundaries. Member states look to each other for inspiration (Ziller 1998, Goetz 2000). At the same time, national political and administrative structures, decision-making styles, traditions and culture will provide restrictions as well as models with respect to external demands and contribute to divergence and variation in the responses of the various countries. There might also be some element of anticipatory and autonomous
Translating states
119
adjustment within the domestic administration to European integration. States that aspire to membership of the EU will adopt a distinct policy and rearrange their organizations and policies before the change in the form of affiliation (Scharpf 1999, Sverdrup 2000). We believe that the transnational processes portrayed in the previous chapters create new identities, loyalties and strategies. Translations, modifications and adaptations occur transnationally. This is a challenge both to the national and to what is genuinely European. To understand these processes, we have developed a multifaceted understanding, blending three different perspectives. However, by regarding administrative adaptation purely as a meeting between external pressure and national constraints and strategies, we lose sight of important aspects of the processes. In the last section, we will be more specific in how this blending takes place. The translation perspective aims to show the processes which enable these factors to interact.
A translation perspective There are a number of traditional ways of understanding EU integration that are always, and sometimes more or less as a matter of routine, normally related to each other. The school of thought that is often referred to as liberal intergovernmentalism places great emphasis on the states and their strategies. The states are regarded as rational actors that first strike a balance between various social interests to determine national interests, which are then safeguarded in negotiations with other states that are also acting rationally. There are often references to a ‘two level game’ (Putnam 1988, Moravscik 1993). Integration in the EU seen from this perspective is the outcome of negotiations between states that have predetermined well-defined interests. In discussions about why the EU has developed in the way it has, explanations are found largely in the actions of the major states (see Milward 1992, for example) and even if the significance of the supranational organizations is not disregarded, they are considered mainly as the devices of (mainly large) states or as arenas in which states uphold their interests. The motive force in the integration process can be found, according to those known as the neo-functionalists, in the joint organizations, the Commission and the European Court of Justice in particular (Haas 1958). The importance of experts and technocrats in reshaping the
120
Translating states
agenda is stressed. Here the emphasis is not placed primarily on the intergovernmental EU but rather on the processes that lead to the spillover of integration from one sector to another. The different theoretical discussions have led to various speculations about the future of the nation-states, with neo-functionalists tending either to tone down their significance or perhaps even to predict that in the future European community there will not be such a great need of states. Partly in connection with these two schools of thought, a third has developed in EU research, which has been labelled multi-level governance (Scharpf 1994, Marks, Hooghe and Blank 1996, Jachtenfuchs and Kohler-Koch 1996, Benz 1998). This maintains that Subnational actors act directly both in national and supranational arenas, creating transnational associations in the process. States do not monopolize links between domestic and European actors; they are one among a variety of actors contesting decisions that are made at a variety of levels. (Marks, Hooghe and Blank 1996: 346) Multi-level governance can be seen as an attempt to grasp the complexity of the EU by indicating that there exist a number of levels at which political decisions are being made, and in which each level has some degree of autonomy but is also subsumed in some form of hierarchy. There is a tendency to shared and interlinked decision making. The multi-level-governance perspective could be said to provide an explanation of the simultaneous occurrence of integration and continued state autonomy. Organizational researchers could maintain that contradictory demands can be taken into account by partially de-coupling the various levels from each other (Cohen, March and Olsen 1972, Weick 1976, Jacobsson 1987, Brunsson 1989). Transnationalization can be seen from this multi-level perspective as low politics or an early phase of the decision-making process that reveals clear signs of integration but which, on the other hand, can lead to a fair degree of limitation when it comes to influencing what is normally termed high politics (Esmark 2002c). It is mainly in the context of approaches that are linked to or identical with multi-level governance that the concept of transnationalization has found currency (Joerges 1999, Wessels 1999).
Translating states
121
Knill (2001) claims that the discussions of how EU integration is to be understood in more general terms are of limited interest for those who are concerned with the impact of the EU on central administrations. But at the same time he traces from the mid-1990s an increased interest in seeing to what extent and in what way the national systems have been influenced by the EU and how far one can speak of greater homogenization of the central administrations. As we pointed out in chapter 1, these studies have produced somewhat contradictory findings, and this was the context in which both our own study of the adaptation of the Nordic states and Knill’s comparative study of the introduction of the EU’s environmental policy in Britain and Germany were devised. We should like, however, to emphasize the importance of providing feedback from the experiences gained from the study described here to the theoretical discussions referred to above. Our arguments for a translation approach are based on the picture we have painted of an emergent transnational administration in which it is both difficult to define what is national and what is European and also where the belief that there are ‘levels’ can be questioned. It would appear that all the theoretical approaches previously accounted for are based on a relatively mechanistic concept of both what constitutes an actor and the form of propagation of regulations and ambitions between them. A translation approach ascribes greater importance to co-evolution and the dynamic interplay of factors that interact reciprocally and plays down a ‘comparative statics’ approach based on a independentdependent variable design. Even though our discussions of translation are clearly based on the perspectives described earlier (which we labelled EU pressure, national strategies, historical traditions) the translation perspective also challenges material aspects of the established way of thinking about Europeanization. We shall conclude by demonstrating how. It could be said that our studies of the Europeanization of the Nordic states confirms something that we knew already, that the way in which the EU’s regulations and ideas are transferred and introduced into the various states and their administrations is not unproblematic. The processes of introducing or implementing and complying with the regulations are affected to a great extent by historical traditions, cultural legacies and national strategies. For this reason we cannot expect them to lead to homogeneity and uniformity. The belief that implementation will be the same in all states is an extreme position
122
Translating states
adopted by some advocates of the ‘EU pressure’ perspective and is based largely on a concept of diffusion. Diffusion, which is a concept often used about the dissemination of innovations, for instance, presupposes that the ‘same thing’ is being spread. Ambiguities at the EU-level lead to fragmented pressure on domestic administrative organizations and policies and provide considerable leeway for domestic variation in the pattern of adaptation (Sverdrup 2000). The administrative adaptations demanded by the EU are often not very specific and can be regarded to some extent as rather like semimanufactured products that have to be finished locally (Røvik 1998). The term diffusion presupposes a physical process, and indeed concepts like friction and inertia are used to describe circumstances in which diffusion does not take place as rapidly as expected. When a study is made in the EU context of the extent to which the states implement EU directives, inertia of this kind is often envisaged. It is assumed that the natural outcome of the process will be one hundred per cent implementation. It should be remembered, of course, that implementation in the world of the EU does not mean that anything is actually put into practice but only that the national regulations are amended to conform with those of the EU. Even so, the concept of diffusion governs the way people think. When we use the concept of translation instead of diffusion we are taking the studies made by Latour (1986) and Czarniawska and Se´von (1996) one step further. Latour says that what characterizes translation is that ‘the spread in time and space of anything – claims, orders, artefacts, goods – is in the hands of people; each of these people may act in many different ways’. In our studies, the EU’s ideas, demands and regulations are in the hands of politicians and officials who can elect to interpret them in widely varying ways, as a result, for instance, of their traditions or the situations in which they find themselves. And, to use Latour’s words again, they can act by ‘letting the token drop, or modifying it, or deflecting it, or betraying it, or adding to it, or appropriating it.’ (Latour 1986: 267). Translation involves, therefore, something more than mere linguistic interpretation. Translating the EU’s ideas and models always involves some reformulation and amendment of the model by the translator. In our studies it is to a very large extent the officials in the administrations of the various states who, with the help of politicians and others, receive and modify the objectives emanating from the EU. And these have, of course, been created in previous processes by the
Translating states
123
EU’s agencies and officials through modification and augmentation of demands and desires that have come from the member states, for instance. Translation processes like this are taking place continually. A translation perspective emphasizes the micro-processes in which regulations and desires are understood, interpreted and communicated. It is the significance of these processes that we have stressed in the previous chapters. Changes evolve across borders and in the everyday exchanges between officials from various central administrations, EU agencies and other organizations. The conception that the EU’s regulations disseminate through some process of diffusion to the member states fails to recognize that states never ‘receive’ regulations but that all rules, desires and demands are translated. And both the demands and the translators are altered by the translation process itself. Is it then possible for those who translate the EU to do so in any way at all? Sahlin-Andersson maintained that the process of translation is restricted by implicit editing rules (Sahlin-Andersson 1996), asserting that this ‘is a process characterized by social control, conformism and traditionalism’. In our studies we can see that both national traditions and the national strategies provide a basis for editing rules of this kind. For example, it may be important for Swedish politicians to assert that the EU involves no challenges to the Swedish administrative model, or in other words that on the whole the existing system can be retained. At the same time, in other states (although none of those we have studied) there may be very good reasons for showing that the EU will lead to changes in the way things are done. In this way translations can be influenced by how the political leadership wants to establish its own legitimacy. We should, therefore, like to stress that the EU’s regulations and desires are not introduced and implemented in the member states by any mechanical process, but that they are always interpreted, modified and revised. This is what we call translation. This translation is the work of people and organizations that have certain interests and are influenced by the traditions and the culture in which they function. This means that translation always helps to produce something that is to some extent new. Rottenburg offers the following description of a translation process: each actor . . . takes the ‘thing’ into his or her own hands and gives it the shape and direction that best corresponds to his/her context
124
Translating states
and intentions. In this way, we move from the transmission of a thing that remains the same to the trans-formation of the thing (Rottenburg 1996: 215) Moreover, and this is a point that we have emphasized in this study, these translation processes also alter the translator, in this case the states. The administrations in the Nordic states are so intricately interlinked in relationships with other actors that we have chosen to refer to the emergence of a transnational administration. And just as these exchanges are influenced by the interests and traditions of the actors involved (among them the states) so the interests and traditions of the states are reshaped by the exchanges. Administrations learn what Danish, Finnish, Swedish or Norwegian interests are from their participation in European exchanges. Translation transforms both what is translated and those who translate.
Summary We have studied the impact of the regulations and aspirations linked to the EU on the day-to-day work of central administrations. We have indicated a development that is moving towards a transnationalization of the relationships between the states and the European organizations. In certain segments of central administrations in particular there are well established transnational networks that include both central administrations, EU agencies and also many other organizations. In this context, to regard Europeanization as a process in which the EU’s regulations and aspirations are implemented (or not implemented) in the central administrations is in our opinion to fail to recognize important mechanisms in the processes. By advocating a perspective in which translation is emphasized we have tried to cast light on other important characteristics of the adaptation processes. 1
We have stressed the active role of the receiver, and that reception always involves an interpretation in which what is being translated – in this case the EU’s regulations and ideas – is modified and altered. Even though the form taken by affiliation is important, there is never any form of automatic compliance with the EU’s regulations. Obeying regulations always involves a considerable degree of interpretation.
Translating states
2
3
4
5
6
125
We have stressed that translation is governed by various rules of interpretation. Translation never takes place haphazardly, but in a specific historical and institutional context. We have demonstrated how factors such as traditional administrative models, national strategies and the organization of EU activities influence how translation will occur. We have emphasized that both what is translated and those who translate are altered in the process. What is to be conveyed, such as an EU regulation, for instance, will be interpreted differently in different states and, therefore, modified. But not even the translator, in our case the central administrations, will remain unaffected. We have stressed the significance of the transnational networks and the role they play in reshaping identities and loyalties. We have underscored the two-way nature of the translation process. In the same way that European ideas are translated by those who receive them at national level, national standpoints are reshaped when they are received and interpreted at Brussels. There is a reciprocity in translation which in the cases we have studied is augmented by the fact that those who will be complying with the regulations also (as members in the organization that) create the regulations to be complied with. We have stressed that translation does not always influence practice and the presentation of this practice to the same extent. As a result of translation, we can discern a de-coupling between practice and presentation. Sometimes the translation may lead to changes in the presentation while the practice remains the same. Sometimes the translation can result in practices being modified while the presentation remains the same. We have emphasized that there is no certainty that Europeanization – for the reasons cited above – will help to create uniformity. Admittedly translations may in certain cases mean greater homogeneity between central administrations, but there is also scope for the retention of dissimilarities and even for the creation of new forms of variation.
6
Democratic challenges
Introduction One important question raised by this study is the implications for democratic control and accountability of the national administrations adapting to the EU. Transnationalization changes the assumptions on which politics and democracy operate. It implies that parts of the administration are deeply involved in policy processes where civil servants are largely in charge. Transnationalization has made particular inroads into segments of the administration for which everyday business has become European in texture and not merely national. For the same reason these cross-border processes may increasingly mould loyalties and identities even when national identity is relatively robust and strong (Egeberg 1999). The growth of a transnational administration may come about without any especially clear or precise political mandate. Those involved frequently base their activities on the rules of professional conduct of their own administration (Fjær 2000). We consider that a growing transnationalization increases the risk of the establishment of a network of bureaucrats and experts who create a transnational state, change the organization of the central administrations in the member states and shape public policy without elected representatives having much opportunity to influence the transformation of these states and the content of their policies. This may occur particularly in areas where technical and professional expertise is considered important. Although various countries have attempted to respond to this challenge by strengthening coordination, there is actually very little exposure of the political leadership to day-to-day EU affairs. This problem is particularly relevant in the EEA countries, which do not have the right to participate in formal decision making. Our study
128
Democratic challenges
reveals, however, that the EU too is a bureaucrats’ paradise, where officials and civil servants rather than democratically elected representatives are the central participants in transnational networks. Without any particularly precise or strong political mandate bureaucrats largely operate the national EU-related agenda on an autonomous basis. Our interpretation is that the control demanded by the civil servants exceeds what politicians can provide, since politicians are frequently merely ‘tagging along’ in the administration of EU questions. There is a danger that if the administration is not subject to control it will become increasingly dominant and politicians will have to take responsibility for matters over which they have less and less power. In practice, the rift between what national politicians must take responsibility for and what is within their power will continue to widen, which is in itself a major democratic challenge (Jacobsson 1999, Jacobsson, Lægreid and Pedersen 2001b, Pedersen 2002). It is not that civil servants are deliberately attempting to exclude politicians from participation in EU matters – on the contrary, bureaucrats may seek stronger political guidance and clearer instructions from politicians in order to avoid being encumbered with sensitive EU decisions (Veggeland 1999a, Raunio and Wiberg 2000b) – but there is still a danger that politicians will have problems in providing the administration with guidance and instructions. Processes largely divorced from politics shape the administration’s image of the world, and it gives high priority to professional expertise. In EU matters, 58 per cent of the units attach very great importance to professional priorities, while the corresponding proportion for political priorities is 36 per cent. This suggests that the administration’s position has been strengthened through Europeanization just as this study indicates that bureaucrats in the central administration have more expertise and capacity, are better informed and integrated and appear to be more effective in EU matters than the political leadership. Politicians are absent to a very great extent from the three aspects connected with transnationalization – creating a transnational state, changing national administrations and moulding policies. In all three of these areas the administration calls the tune. In Norway, where the administration attaches considerable importance to political evaluation, there are few units, which have contact with the political leadership on EU-related questions. In Finland, too, EU activity has strengthened the position of the administration. Politicians are involved in EU matters to the greatest extent in Sweden, while in Denmark their involvement
Democratic challenges
129
is lower and the level of conflict is possibly even lower than in the other Nordic countries. This may be an overly pessimistic picture of room for manoeuvre open to politicians. An absence of conflict could equally mean that an administration is able to anticipate the will of politicians and can act autonomously, and that this is the reason why fewer conflicts are observed. Although transnationalization is strengthening the administration, comparison reveals that there are significant differences between the four Nordic countries, and that the traditional division of power between parliament and government (including the administration) is influenced by how the national system of coordination is structured; how a country is linked to the EU, and how long it has been a full member. The hierarchical system of coordination that exists in Denmark makes it possible to formalize coordination and provide mandates for the national representation, which restricts the positions the government can adopt in the Council of Ministers. By contrast, the dualistic system of coordination in Sweden and Finland promotes informal coordination between parliament and government, providing the national governments, and the national administrations, with considerable freedom in relation to their parliaments. Norway’s position as an EEA member has not given the parliament the opportunity to manage and control the way in which the government handles EU matters. It may be claimed that as a long-time full member Denmark has managed to develop a national strategy for adjusting the division of power between the various political agencies, something which none of the other Nordic countries – up till now – has managed to do. Consequently, it would not be reasonable to describe the EU policy of the Nordic countries as exclusively and unilaterally managed and determined by civil servants. There are differences in the strategies that have been developed in the individual countries in connection with the control by central government of EU matters, and this study reveals some differences in the extent to which national parliaments are able to control their governments and central administrations, and the methods they employ to do so. In the case of Denmark, the parliament, in cooperation with the government, has developed a strategy that makes the political agencies active partners both in the formulation of policy and in the process of management and control (Pedersen 2002). Intensive contacts exist in all these countries between government and the administrative units; the governments’ management of civil servants may be more or less intense, but there is no case in which
130
Democratic challenges
civil servants are permitted in general (or systematically) to deal with EU matters without the participation of the national government. Finally, Europeanization does not involve any fundamental change in the constitutional division of power between parliament, government and administration. The threefold division of power between the legislative, executive and judicial branches of government has been basically upheld, although the EU has given rise to the introduction of important new developments. Nonetheless, we have shown that the national administrations are becoming transnationalised; that a segmentation of administration is taking place; that efforts to coordinate the national administrative bodies are increasing in intensity; and that the EU segment’s channels of communication with the outside world tend to multiply in parallel with the growth in the number of contacts with EU organizations. There is also evidence indicating that the administration is being transformed, that it is flexibly adjusting to transnationalization, and that this is having an influence on the real distribution of competencies and capabilities between parliament, government and the administrative units. Paradoxically, the Nordic administrations have been characterized by both flexibility and robustness, while the driving forces behind the organizational changes have been of both European and national origin. Everyday integration occurs in interplay between national and European organizations, and permits national central agencies to control the manner and extent to which Europeanization is allowed to influence the division of power between parliament and government and to influence the organization of the national administration. For the same reason, this process of everyday integration produces a number of concrete and specific challenges for the national democracies.
The democratic consequences – a method In this chapter we will evaluate the consequences that everyday integration has for parliamentary government. For this purpose we enter the existing debate on the EU’s crisis of legitimacy,1 but from another angle and from a point of departure other than usual. So far, the debate has been based on the popular scepticism expressed through, for example, the Danish people’s ‘No’ to the Maastricht treaty and its subsequent ‘No’ to the euro, or Norway’s ‘No to the EU’ (Petersen 1993, Franklin, Marsh and McLaren 1994). In this book, however, we take a different approach. The existing popular opposition
Democratic challenges
131
to European integration does not enter into our considerations, and we have not asked our respondents to evaluate the influence that popular scepticism may have had on everyday integration. Up till now, the debate has also concerned the EU as an institutional ensemble, and the demands that this makes on legality and legitimacy. We do not address this issue, either. Rather we examine the ideals of the Nordic countries with respect to the ways in which their national central agencies should be legitimized, and how these ideals have been altered in the light of European integration. Finally, the debate has dealt with considerations of the standards of legality and legitimacy to which the EU should in theory conform, but which it fails to live up to in the current situation. Instead of doing so, we combine historical and empirical analysis in a methodical assessment of the consequences that the transnationalization of the national administrations implies for democracy. We have divided this methodological assessment into two parts. The first is concerned with democracy, and encompasses the question of which democratic ideals form the basis for evaluating the consequences of European integration for the national states.2 This is also based on the claim that European integration has led to changes in the democratic ideals, and, consequently, in the ideals upon which the evaluation is based. The second part is concerned with politics, and encompasses the question of which factors democratic government actually exerts control over, and rests on the claim that transnationalization has changed the scope of politics. Using these two sets of assumptions, we have attempted to define a methodical approach to a study of the democratic consequences of market formation within the EU (Pedersen 1998, 1999). This approach is intended to clarify what parliamentary government has implied in particular countries (i.e. the four Nordic nations), in particular periods (especially since 1993), and the question of what, in the context of empirical research, can be considered politics. Our assumption is that an assessment should consider the nature of democratic ideals from a historical viewpoint, and that such democratic ideals should be appraised in relation to the actual scope of politics. The historical and empirical approach is based on the assumption that the concept of democracy is inextricably linked with the concept of politics, and that it is the actual scope of politics that determines which agencies and processes are subject to demands for legality and legitimacy. In this way, we delineate the democratic demands on an authoritative basis, i.e. on the basis of how the political
132
Democratic challenges
agencies (parliament, government and the courts) customarily interpret these requirements and apply them as the basis for binding decisions. The question of how to define the scope and content of politics is thus an important one, as is the manner in which the democratic ideals are formulated. On the other hand, neither the sphere of politics nor the interpretation of the democratic ideals are static entities. We will begin by examining the ideals of everyday integration. This will be followed by an examination of the real distribution of political authority in connection with everyday integration. In this way, we will carry out two sets of studies: one which deals with democratic ideals, which is formulated in connection with the entry of the Nordic countries into and their adaptation to the EU; and another that deals with the question of to whom political authority has actually been delegated or granted in connection with the performance of European tasks. These two studies then form the basis of a comparison of which agencies are identified as politically authoritative by the democratic ideals, and how political power has actually (from an empirical point of view) been distributed. We compare those who have legally been granted political authority with those who actually exercise political authority. On this basis, we identify a number of challenges to democracy, all of which become clear once it is shown that political authority is being exercised by agencies and organizations that are not subject to the demands of democratic control. Accordingly, we place three requirements on the evaluation of the democratic consequences: 1
2
3
The assessment of democracy should be contextual. It must be based on the fact that the scope of politics, like that of the democratic ideals, will always be specific – i.e. dependent on the country and the period under study. The consequences for democratic government should not be related to abstract, noncontextually related democratic ideals, or to unsystematic studies of where and how political authority is actually exercised. The evaluation should be dynamic. It should take account of the fact that political power, like democratic ideals, is subject to continual change. Accordingly, the consequences of European market formation should not be appraised in relation to a static, idealized image of democracy and politics; both must be viewed in a dynamic context. The evaluation should possess an authoritative basis. It must be based on interpretations of the political freedoms, organizations
Democratic challenges
133
and procedural rules that political agencies – including the national parliaments and courts – have laid down, or which have by some other means become the subject of legitimate collective agreement. The consequences of European integration should not be evaluated in relation to ideological (or personal) interpretations of democratic ideals, or interpretations that do not themselves possess democratic legitimacy.
The ideals of everyday integration Everyday integration has given rise to the development of five ideals, which, in an ideal-typical way, apply in all of the Nordic countries. These are mutually overlapping ideals and they deal with the question of the distribution of authority, and consequently, with the national division of power: 1
2
The ideal of sovereignty, which is based on the idea that national governments alone should determine how the government and the administration should be organized, with the aim of representing national interests in the EU in the most efficient possible way. Neither the EU Commission, the supranational nor the transnational organizations possess such an authority. National sovereignty does not exclude the possibility of demands being made via the EU’s code of practice (treaties, EU Directives, regulations, etc.) for the creation of new administrative units or for alterations in the existing national agencies. Such requirements do not, however, alter the prerogative by which the responsibility for the organization of the national administration falls to the national governments, including the implementation of supranational legislation with respect for the existing administrative model and the existing administrative divisions. The ideal of parliamentary government, which is based on the idea that, while the government has powers to represent national interests within the EU, this must occur within the framework of control that has been assigned to parliament by the national constitution or specific legislation. The ideal of parliamentary government encompasses the question of the loyalty of the administrative units; the idea that while civil servants should represent the national interests, this must occur on the basis of the government’s concrete decisions. In the four Nordic countries,
134
Democratic challenges
governments traditionally have been given a freer hand in their foreign policy than in their domestic policy. The foreign policy prerogative of governments also applies to the representation of national interests in EU matters. With some variations from country to country, this privilege has, however, become more or less restricted through the establishment of independent procedures for parliamentary control in connection with the treatment of EU questions. In Denmark, for example, the prerogative has been restricted via the mandating procedure, which implies that the government is required to present proposals to The European Affairs Committee of the Danish parliament regarding how it intends to safeguard the national interest, and that the Committee must grant a mandate to the government to this end. No similar procedure exists in the other Nordic countries, in which it is the responsibility of governments alone to inform or seek guidance from parliament in advance of their participation in the legislative processes of the EU. There are also contrasts between the Nordic countries in relation to the issue of the loyalty of civil servants. In general, civil servants must display loyalty towards the political agencies and must not allow party politics or personal considerations to influence the performance of their tasks. Similarly, they must base the performance of their duties on professional values. But although it is generally the case that civil servants must be employed according to merit and work according to professional standards, the administration of EU functions has assigned political roles to civil servants. In the Commission, they can function as national experts and be formally independent in relation to their government, while in reality acting as its representatives. They negotiate on behalf of their governments in working groups under the Council of Ministers, and in expert committees under the Comitology but can also decide concrete legislation. They can similarly function as deputies in the Council of Ministers, but formally act as ministers. In all instances, the boundary between their political and professional roles has become fluid. The Nordic countries possess different traditions for handling similar situations. In Sweden and Finland, there is a long tradition of politically appointed civil servants (such as Secretaries of State). Such a tradition has never existed in Denmark, where the principle has been to maintain a clear division of administrative and political functions.
Democratic challenges
3
4
135
The ideal of adaptation, which is based on the idea of ‘adapting’ the national system of coordination to European development through policy processes. In the same way that the four Nordic countries, in advance of their entry into the EU, established a system of coordination in the image of the administrative model that they had built up over several centuries, it was emphasized that the systems should adapt themselves on an ongoing basis. The responsibility for this adaptation was allocated – true to tradition – to the governments, which in all four countries exercised the sole prerogative of organizing the administration and the daily working procedures. What precisely was meant by adaptation is, however, unclear. In all of the Nordic countries, individual proposals from the Commission are subject to various kinds of evaluation in terms of their economic and organizational consequences, and to some extent also in terms of their consequences for national legislation. In no case, however, has there been a possibility to undertake a systematic evaluation of the overall consequences for the national division of power, including the organization and control of the national administrations, and their functions in relation to parliament and government. The responsibility is hence quite imprecisely formulated, and it is unclear how far governments are obliged to systematically collect information that can determine the extent to which adaptation is required. Similarly, no procedures exist to determine when such processes should or must be initiated and when the government should or must enter into public deliberation about plans or blue prints for adaptation. Consequently, no allowance has been made for the possibility of public deliberation on the national administrative and constitutional consequences of European integration. The ideal of neutrality, which is based on the idea that there must be internal coordination within policy areas as well as interministerial, and that this coordination must follow professional standards of political neutrality. Two secondary ideals are associated with this ideal: one dealing with objectivity, and the other dealing with the efficient furthering of interests. The ideal of objectivity implies that the most objectively competent authority should assume responsibility for the coordination. In Denmark and Norway, the special committees handle the relevant preparation, while in Sweden and Finland; this responsibility lies with the individual minister and the ministry’s administrative units. In these
136
5
Democratic challenges
countries, the minister responsible can decide the extent to which other public authorities and external interests should be consulted as part of the coordination process. In Sweden and Finland, the responsibility for overall coordination has been taken over by the Prime Minister’s Office, while in Denmark and Norway it remains with the Ministry of Foreign Affairs. In the proactive phase, during which the Commission determines the wording of formal proposals, it is left up to the individual public authorities to participate on their own initiative. The ideal of efficiency on the other hand implies that the most important public and private interests should be coordinated, and this is expected to result in the efficient representation of national interests in the European legislative process, and on the national level, to the efficient implementation of the adopted legislation. There are variations between the Nordic countries with regard to the number of interests that are generally taken into account. There are also differences in the extent to which rules or norms exist governing how many other public authorities, besides the authority directly responsible, should be consulted, and whether such consultation should involve only national authorities or include the authorities of other countries. There are similar differences in the extent to which public authorities are required to provide information, conduct hearings or take decisions in coordination with national political agencies (including ministers, the government and parliament, or political parties in parliament), as well as in the extent to which private interests should be consulted. In most instances it is left upon the responsible authority itself to determine which interests should be consulted and how much weight they should be allotted. Similarly, it is in most instances left to the individual authority to decide how contacts to political organizations, private interests and other public authorities should be organized. In the case of Denmark, however, there are rules and unwritten norms for when and in how political organizations should be consulted, just as formal rules exist for how coordination should be organized by the special committees, and which special-interest organizations should participate in these or be consulted via hearings. The ideal of transparency and participation, which is based on the idea that the relevant parties should be given sufficient insight and participation to permit their interests to influence the coordination, and that the general public should in a broad sense be given insight
Democratic challenges
137
into the political decision-making processes. A secondary ideal of pluralism is associated with the ideal of openness and participation, based on the idea that all relevant interests should be able to obtain insight and achieve participation. Another secondary ideal of consensus is associated with the ideal of transparency and participation, which implies that interests should be coordinated via negotiation and compromise, rather than by voting and instruction. This ideal also implies that special interests (administrative, political and private) are included in the coordination on a voluntary basis. Interests are free to participate and cannot be instructed. They are also free to enter into deliberations or to withdraw themselves from negotiations and coordination. In Denmark, the coordination process is subject to the mandating procedure, which places pressure on participants to choose between accepting a compromise or withdrawing from the coordination process, but which also obliges the government to consult the relevant parties and permit them to determine the extent to which they wish to compromise. The procedure also lays down norms and rules determining which private parties should be allowed to participate and be consulted via hearings. In Sweden and Finland, it is left open to the responsible authority to determine how far it is willing to go to try to achieve compromise or consensus between the participating interests. Similarly, it is in most cases left to the individual authority to decide how far, when, and if at all the general public is to be informed. In Sweden – which has a long tradition of openness – there are more comprehensive norms and rules governing transparency and the provision of information to the relevant parties than in the other Nordic countries. The five ideals on the one hand show how the number of those who are considered to be legitimate participants in the political and administrative representation of national interests has become extended. On the other hand, they show that only very general and imprecise norms have been developed with regard to the procedures that must be adhered to in connection with democratic government. In sum, this produces a situation in which the group of legitimate participants is under expansion, but in which it remains unclear how democratic government is to be guaranteed. The domain of politics is expanding, without this necessarily implying the concurrent extension of democratic government.
138
Democratic challenges
An evaluation – six challenges In the following, the main findings of the empirical analysis will be compared with the democratic ideals. On this basis, we will evaluate whether a ‘gap’ has arisen between ideals and reality, and in that case, the consequences that this may imply for democratic government. How does everyday integration challenge democratic government? The first challenge: transnationalization The Nordic comparison leads to the conclusion that national administrations are engaged in building a transnational state. We have shown how relations with EU organizations are routine, as are relations with other national authorities and private interests in connection with the handling of EU matters. We have also shown how national administrations are subject to the multiplier effect. Although the multiplier effect is not equally strong everywhere, there is evidence to suggest that relations with the EU are multiplied in parallel with the growth in the number of contacts to national, European and international players, and that European integration, accordingly, tends to lead to routine interplay with a broader set of surrounding agencies and agents. The claim that a transnational state is being built is based on: (1) the fact that transnationalization can lead to the establishment of a new kind of administration which involves intensive routine contacts between the national administrations and European organizations, (2) the fact that a segment of the national units has got more autonomy in relation to the remainder of the national administration, and has been given tasks and working procedures different from that of the other administrative bodies; and (3) the fact that joint expectations have been established between the units of the EU segment’s and other players concerning the norms and rules to which contacts and negotiations must adhere. The claim emphasizes that a transnational administration is being formed, and that this occurs by what has been termed ‘an open processes’ (Pedersen 2002). The concept of the ‘open process’ emphasizes that the transnational administration is ‘built at sea’, i.e. it is being created without any wish being expressed in advance for this to happen, and without any prior procedures being politically formulated for how it should happen. In this respect, we claim that transnationalization implies a variable geography, variable forms of authority and a variable process, which is quite different from the formal division of power that
Democratic challenges
139
exists within the framework of the national states (Esmark 2002b). It is against this background that the challenge to democratic government arises: Civil servants are empowered to determine the manner and extent to which the administration is involved in the construction of a – new – transnational state. In this connection civil servants take decisions through which the sovereignty and prerogative of governments to determine the organization of the administration is limited, just as the construction of the transnational state occurs without any overall political intention or purpose. Instead, the transnational state is being formed as the uncontrolled consequence of the many contacts and initiatives taken by national civil servants for the purpose of preparing and implementing the participation of the national governments in the European decision-making processes. The second challenge: bureaucratization We have shown that the Nordic administrations have been altered, but that this has occurred within the framework of a traditional administrative model and that the performance of EU matters has been bureaucratized. Bureaucratization is revealed in the establishment of formal as well as informal contacts between parliament and government, between governments and the administration, and between administrative units and private interests. Informal contacts play a substantial role in all four Nordic countries, but the formal and informal contacts are not necessarily utilized in the same way. Neither is it the case that informal coordination is necessarily detached from the formal contacts – they may take place as a ‘shadow’ of the formal process (Scharpf 1997) instead of being fully unconnected with these. The formal meetings and exchanges of written notes can be seen as a form of organization that casts a shadow of informal activity, in which the duty to administer national interests in the EU is undertaken by the same public authorities as in the formal system, but following other procedures and involving a (partially) different group of participants (Pedersen 2002). Conversely, the informal coordination can be seen as a form of organization in which it is to a large extent the administrative units that represents the national interests, in partial autonomy from the formal organization or even political control. In the case of Sweden, Norway and Finland, a considerable surfeit of informal contacts have been made autonomous in relation to their parliaments, and take place between the administration and other
140
Democratic challenges
interests, sometimes detached from the government’s (or the minister’s) participation and insight. The informal contacts do not in this case comprise a shadow to the formal ones, but rather a particular – and informal – form of organization. In the Danish case there are more formal than informal contacts and the informal contacts generally comprise a shadow to the formal ones, though not in every case. Thus there are several examples of Danish coordination that is not oriented to the mandating procedure, and of administrative units handling EU matters independently from parliamentary supervision and control from the side of governments (Esmark 2002b, Højbjerg 2002). The bureaucratization of EU matters is in other words a general phenomenon, which reveals the extent to which it is civil servants that handle EU questions. It also reveals the extent to which governments are responsible for preparing the process of representation of national interests in the EU. In a democratic government, any activity that involves the formulation, ratification or implementation of rules with general application must follow specific legislative procedures, instead we have seen that political-administrative processes takes place autonomous in relation to the processes of public debate, parliamentary preparation and formal decision. We have also seen that the legislative role of the parliaments accordingly has been restricted. The degree of autonomy may be complete, in the sense that the parliaments may not have been informed in the process; or it can be limited, in the sense that parliament has been informed and heard, but that this has not altered the government’s position; finally it can be open, in the sense that parliament’s responsibility is unclear, leaving the issue to the discretion of the government. The same possibilities for autonomy exist in the relationship between the government and the administration. It is against this background that the challenge to democracy arises: The administrative units possess considerable freedom to draft and prepare (and in certain instances even to decide) policies and legislation, and that this autonomy is greatest in countries where no mandating procedure exists and where coordination is not concentrated in formal committees. The democratic challenge is greatest in Norway, Sweden and Finland, but is not inconsiderable in Denmark. The third challenge: incrementalism Our analyses have shown that national administrations are being transformed, but that this is normally not occurring on the basis of any
Democratic challenges
141
overall political decision, or after any comprehensive political deliberations, but rather incrementally, i.e. on the basis of pragmatic and technocratic decisions that do not consider the question of how such decisions are to be deliberated or legitimized. In all four countries, EU matters have been integrated into the existing administration, and the national administrative models have been maintained. At the same time, national systems of coordination have been built up, but only in Denmark – and to some degree in Norway – do these systems comprise a new and independent organizational form. Significant organizational changes are at the same time being introduced: Systems of coordination are being altered on an ongoing basis; and no stable form has yet been arrived at in Sweden, Norway or Finland. Besides this, changes are being introduced to the individual administrative units as the administration becomes subject to segmentation and transnationalization and informal organizational forms are gaining in importance and administrations are being subjected to the multiplier effect. It is unclear who takes the initiative to introduce these changes and why they do so. The ideal of adaptation makes it clear that it is the prerogative of national governments to adjust their national administrations. However, it does not make clear how this should happen, or which aspects of European development should be adjusted to. The transformation thus implies a challenge to democratic government: When the division of power or the organization of the administration is altered, it is important to know why these alterations occur, who is responsible for them, and according to which procedures they take place. As we have seen, none of the political agencies has been allocated a formal responsibility to adjust the division of power. We have also seen that the government is formally charged with the task of adapting the administration, but that in practice, it leaves this task to the individual administrative units. And finally, we have seen that no procedure exist that determines how decisions relating to adaptation should take place. The Danish mandating procedure, for example, has been developed on the basis of a continuous learning process and via sporadic negotiations between the Danish parliament and the government of the day. It is against this background that the challenge to democratic government arises: adaptation of the national central agencies is occurring incrementally, but without deliberations that could secure a clear awareness and knowledge of the challenges that the performance of the European tasks poses for national
142
Democratic challenges
administrations and parliamentary control. Neither the government nor parliament has had any opportunity to formulate the aims of these changes, or to decide how and to what extent transformations should take place. The fourth challenge: politicization A considerable amount of coordination takes place, both within the individual administrative units and across the boundaries of these, but also by involving the public authorities of other countries and external private interests. This coordination is routine, and is the responsibility of the objectively most competent authorities, so as to represent national interests in the EU in the most efficient manner. This coordination has a bureaucratizing effect on politics, but also has a politicizing effect on the administration. There are variations between the Nordic countries with regard to which authorities are in the coordination, but there are particularly large differences with respect to when, how and whether the political agencies are consulted if at all. Similarly, there are large differences – from country to country, and from case to case – in the factors that administration take into account when deciding their positions. In most instances it is left open to the administrative bodies themselves to decide the extent to which the political agencies should be consulted; when it is to take place; and how such contacts should be organized. This also applies to contacts with other administrative authorities and private interests. Only in the Danish case have limits been set to the independence of the administration in this respect. The point at which the line has been crossed between administrative coordination, in which only public authorities are required to be consulted, and political coordination, in which political agencies must be consulted, is something that the administrative units or governments normally decide for themselves. This is where the challenge to democratic government stands out: the coordination process empowers the administration to take political decisions. This may occur when the administration determines who is to be included in the coordination process; which interests are to be taken into account and when, how and whether political agencies are to be consulted. Against this background, the challenge is that political processes are taking place in detachment from parliamentary supervision and government decisions, and that the administrative units, possess considerable autonomy to
Democratic challenges
143
organize working practices and to decide which interests should be consulted, which interests are to be taken into account, and when political control is to take place. The fifth challenge: exclusion The empirical studies show that segments of the administration are opening up towards a broader public. Many interests have access to the national systems of coordination, and the multiplier effect tends to encourage greater participation. However, this opening-up is not systematic, and does not follow procedures laid down in advance. The EU-related administration, on the other hand, seems to be less transparent than other parts of the national administrations. In Denmark and Norway, the special committees are inaccessible to the general public, and the same applies, to some degree, to The European Affairs Committees of the Danish parliament (Folketinget) and the Norwegian parliament (Stortinget). Naturally, the many informal contacts are closed to public insight or participation. The opening up that occurs thus depends on factors that have little to do with the democratization of the administration. The administration is made open as part of an attempt to efficiently represent national interests, not in order to enable the transparency and participation of the general public. At the same time, it is to a large degree left upon the administrative units themselves to decide how open the administration should be. Besides the formal rules giving external interests the right to consultation and hearings, it is left open to the administration to decide who is to be consulted, in which contexts, and with what status. It is against this background that the challenge to democratic government arises: the transnational administration is less transparent than the national administration. There are no procedures by which the general public can achieve insight into or participation in the national systems of coordination or the transnational administration. At the same time, however, the EU administration is becoming more open, but only in relation to particular interests. It is governmental bodies, and private organizations that possess the relevant resources, that participate in the national systems of coordination and in the transnational relations. Equivalence in resources exists between the individual administrative units and the external parties implying that interests who do not possess such resources are excluded.
144
Democratic challenges
The sixth challenge: captivity Empirical studies have shown that national administrations can become dependent on external interests (Esmark 2002b). Equivalent interests can capture the administration. This may occur if the administration is forced to utilize interest organizations as its medium of information from EU organizations or from European interest organizations (Højbjerg 2002). This raises yet another challenge to democratic government: The administration is open towards some organizations, and upon which it can become dependent and be ‘captured’ (Lowi 1979). The mutual dependence that arises between the administration and external interests implies a risk that the coordination system and the transnational state may come to exclude other interests and considerations. This challenge is made more acute by the fact that private interests are able to ‘go around’ or bypass the national public authorities and present their interests directly to the European organizations. In all four Nordic countries, it is the assessment of the national administrations that companies and interest organizations make use of this opportunity, though not on a large scale or in a systematic manner. But the possibility exists, and raises the final challenge: The opportunity for private interests to bypass the national government and go directly to the EU’s organizations in order to further their own interests. This provides organizations with the opportunity to avoid contact with the national authorities, and hence limits governmental control over the interests that are allowed to influence initiatives that later can become national law. This ‘detention’ limits pluralism, just as bypassing alters the conditions for consensus. Private organizations can as a ‘threat’ use their ability to bypass the national administrations (Pedersen 1998).
Conclusion We have shown that the central administration is being segmented, coordinated, involved in transnational relations and simultaneously transformed (Figure 6.1). A new way of differentiation and coupling of the administrative units has occurred. The external sovereignty of the government is being altered as it enters into a close interplay with European and other non-national and non-public organizations. For the same reason, the government’s prerogative to organize the administration’s organization and functions is being restricted, and the
Democratic challenges
145
administration is becoming transnationalized. The internal authority of the government is also being altered. The administrative units capability of determining which contacts the national government should enter into to further its interests in the EU, as well as who should be consulted during the process of national coordination has changed. At the same time, the adaptation of the administration is taking place incrementally and in the absence of procedures that could ensure an awareness and knowledge of the challenges that the performance of the European tasks poses for the national administrations or for parliamentary control. This adaptation is occurring in a manner that allows neither the government nor parliament a clear opportunity to decide how or to what extent it should take place. Thus national units and daily routines are becoming politicized and the building of the transnational state is becoming the task for national bureaucrats which leads to the conclusion that the administration has to some extent itself become capable of determining the extent to which it will be subject to political control, as well as which external parties it wishes to involve. This is occurring in parallel with a considerable degree of closure towards the general public, as well as in the relationship between the administration and special interests, opening up the risk that the latter could ‘capture’ the administration. A comparison of Denmark with Sweden, Finland and Norway reveals that the consequences for democracy are influenced both by the structure of the national system of coordination and by the number of years that a country has been a full member of the EU. The hierarchical system of coordination makes it possible to formalize the coordination and to mandate the national representation. Full membership increases the number of channels of influence, and multiplies the number of contacts and level of participation in a more complex environment. The Norwegian example indicates that partial membership can lead to
Ideal
Challenges
Sovereignty
Transnational State
Parliamentary Government
Bureaucratization
Deliberation
Incrementalism
Neutrality
Politicization
Transparency and Participation
Exclusion and Captivity
Figure 6.1. The transnational administration assessed in relation to democratic ideals
146
Democratic challenges
a relatively closed administration, combined with a relatively limited level of political control. There is, however, no simple solution to the problems for democracy, such as by stipulating more precise and long-term control. We have seen that politicians may become involved in processes and activities that pass as control, without actually having much influence. Irrespective of how much importance is attached to national control and coordination, there is a risk that transnationalization and segmentation will weaken the effect of such strategies. In Sweden, where the rationalist governance model has been strongest, we observe a recurrent use of control signals, but the influence of the civil service is hardly weaker than elsewhere. Political control can essentially become something of a ritual. The question of how politicians are to cope with the challenge of expert control is difficult. It requires discretion and cannot be tackled using over-simplified solutions. The emerging domination of transnational expertise and the problems this raises for politicians also affects the role of parliament. One might say that, in spite of the problems they encounter in entering into the decision-making processes, politicians in government nevertheless have numerous personal contacts in Europe. The problem is greater for the parliaments, which, as a result of the EU, risk relegation into an even more subordinate position (Raunio and Wiberg 2000c). In the Nordic countries, attempts have been made to integrate the popularly elected bodies into the decision-making process by a variety of means, especially in Denmark. But generally speaking, parliaments do not play a particularly important role in the administration of EU functions. A further problem for democracy is the rift between the very positive attitude towards the EU of the various administrations and the deep scepticism of the Nordic populations. There are few administrative units that consider the EU to have had negative consequences (Jacobsson 2001). While there are doubtless many reasons why the administrations see great value in European integration, one contributory factor may be that increased embeddedness and transnational interaction may enhance identity and feelings of loyalty. It is natural that transnational civil servants come to see the world in a new light, and possibly in a different one to the populations of their countries. From a democratic point of view, such a rift between the people and their public servants is a matter of no small concern. However, the extent to which the democratic legitimacy present is sufficient also depends on how the ideals relating to the efficient
Democratic challenges
147
furthering of national interests are weighed in relation to the ideals of democratically controlled participation. Throughout this book, we have seen examples of how national administrations have adapted to European cooperation and have done so for the purpose of efficiently representing their national interests. We have also shown that a powerful and quite independent administration seems to be a prerequisite for the ability to formulate national interests and further these in an efficient manner. The extent to which the democratic legitimacy present is sufficient is thus particularly dependent on which democratic ideals are given most weight. The challenges may be many or few. If the process of integration were to be widened to encompass treaty ratification or the formal transfer of sovereignty via treaties, there would be more, and different, challenges to democracy. The same would apply if our point of departure were the general expectations of the public with regard to how political authority should be distributed and how political processes should be carried on. This chapter has thus not exhausted the discussion of the challenges to democratic government, but has merely approached the question from the narrow standpoint of everyday integration. For the same reason, it is clear that the challenges that arise from the ideals of everyday integration are by no means comprehensive – they encompass only a small number of the expectations that the public may place on the exercise of political authority, not all of the ideals that the executive power should in general live up to. Similarly, it is clear that the ideals of everyday integration may be so few in relation to democratic ideals in general, and may be so restrictively formulated in relation to the expectations of the general public, that this in itself will give rise to challenges to democracy.
7
Understanding transnational states
Introduction In chapter 1, we claimed that it was possible to describe the consequences of Europeanization as the growth of a transnational state. In this chapter, we will sum up what we have learned about transnationalization and the transnational state on the basis of the comparative study of the four Nordic countries, review the literature on transnationalization, and indicate how our study contributes to the further development of theories on this subject. We will end up by presenting a more comprehensive view of transnational states.
Theories of transnational relations There have been considerable developments in recent years in research on transnational relations – both in relation to the EU and in relation to globalization (Held et al. 1999, Woods 2000). This development has occurred in the fields of sociology, history, economics and legal theory, but especially within political science. The results, however, remain fragmented and are based at best on limited empirical research. Systematic comparative projects are rare and there is no widespread agreement concerning how to define transnationalization. Consequently, while we are not the first to utilize the terms transnational and transnationalization to describe the changes taking place in connection with European integration, we are among the first to develop these concepts on the basis of a comprehensive collection of comparative data. Research into the EU was for years mainly concerned with explaining the historical events that over time have altered the EU treaties (high politics). It is only recently that political science has begun to evaluate everyday integration and the ways in which the EU
150
Understanding transnational states
functions as a state-like structure (Rosamond 2000). It is in this context that the use of concepts relating to transnationalization and transnational states has begun to spread in the social sciences. Common to the literature on European integration is the idea that national states and European integration can best be viewed from the perspective of a legalistic approach (Esmark 2002c) and that the adaptation of national states to integration can best be explained with the help of a causal approach, just as the consequences of integration can best be evaluated through a parliamentary approach. These leitmotivs are for the most part taken as given in the literature, and are included as prerequisites for defining European integration. We assume – as does the literature in general – that national states and their sovereignty constitute a fundamental prerequisite for European integration. In addition we assume that transnational states represent something other than the integration of sovereign states. Furthermore, we will demonstrate that the changes in the national states that result from European integration can only to a limited extent be explained on the basis of the causal approach, and that more complex explanations are required. We will also show that the challenges to democracy depend to a certain extent upon which national strategies the individual states choose in order to legitimize their participation in the EU. Transnationalization thus entails some challenges to democracy, but can also bring about instances of democratization. Against this background, we address the three questions that form the basis of the aims of this book: How can we describe transnational states by challenging the idea of the sovereignty of states, and describe transnationalization as a different form of border definition than the traditional ones? How can we explain the adaptation of states by challenging the causal approach and explain transnationalization as a dynamic interplay between multiple factors? And how can we assess the consequences for democracy by challenging the parliamentary approach, and reflect upon the consequences of transnationalization as something more than a challenge to democracy? In the following, we will review the literature on European integration. We will conclude the book by returning to the debate on transnational states. The purpose of this is quite simple. It is not our task to present the literature on European integration, or to divide it into schools of thought. Our aim is a more limited one, namely to contribute to a more precise identification and definition of what can be understood by the terms transnationalization and the transnational state.
Understanding transnational states
151
Transnationalization Some fairly comprehensive and highly complex literature now exists to explain European integration. This literature has several features in common. Usually, and for the most part implicitly, it starts from the assumption that it is possible to distinguish between state sovereignty on the one hand, and European integration on the other. Integration is viewed as something that involves the transfer of, restrictions on, or abrogation of state sovereignty, i.e. something that occurs at the expense of national sovereignty. Few contributions utilize the concept of transnationalization, and neither is it common to examine issues concerned with the transnational state. Within the framework of the distinction between sovereign states and integration, there are three prevalent ideas about what characterizes European integration: (1) The EU is an inter-state forum for negotiation with certain supranational characteristics; (2) the EU represents the operation of an integration dynamic; and (3) the EU is a complex system of coordination with different and potentially conflicting levels and decision-making processes (Esmark 2002c).1 It is in the literature relating to this third idea that the concepts of transnationalization and the transnational state are gaining ground. The view of the EU as an inter-state forum for negotiation with supranational features is associated with the position that continues to see state sovereignty as both a factual and a desirable scenario. This position is termed intergovernmentalism, as states are assumed to have an interest in maintaining themselves, and are expected to behave in a rational manner. Qualifications to these assertions have been advanced. It has been claimed that the limited transfer of sovereignty can be in accordance with the instinct of states for self-preservation, but that this integration never will (or should) arrive at a stage at which it would be impossible for the states in question to regain their sovereignty (Grieco 1996, Øhrgaard 1997, Wivel 2000). From the intergovernmentalist perspective, the study of the EU is focusing on the study of treaty signings, summit meetings and the legislative process of the Council of Ministers. Elements of the pre-directive phase, such as the Commission’s right to initiate, are ascribed less importance in the political decision-making process (Moravcsik 1995, 1997), and consequently, of less importance for the study of the EU and European integration. Liberal intergovernmentalism is intentionally better equipped to explain high politics and major turning points in European
152
Understanding transnational states
integration than to explain low politics or how the system works on a daily basis (Moravcsik 1998, Egeberg 2002). Everyday integration and transnationalization are not regarded as important subjects. The traditional counterpart to intergovernmentalism is functionalism. EU is viewed as a dynamic of integration. This tradition shares features with the idealistic position (transnationalism) in international politics. Here, too, it is argued that it is possible to establish a ‘moral’ community as an alternative to inter-state anarchy. The assumptions underlying this position are drawn from functionalist sociology and its assertion that potentials for integration can be identified in any context. In contrast to functionalism, neo-functionalism bases its propositions regarding integration on economic forces rather than on intangible values. From the neo-functionalist perspective, there are two possible paths to integration: One via the formation of supranational institutions, and the other via the formation of transnational coalitions (Haas 1958, Lindberg and Scheingold 1970, Streeck and Schmitter 1991, Johansson 1997, Zervakis and Johansson 2002). Once again we recognize the idealistic position, in which the formation of alternative ‘communities’ to the national state is turned into a political project. In addition to this normative ambition, neo-functionalism possesses another more descriptive or explanatory aim, in which supranationalism and transnationalism are seen as mirror images of each other, and in which supranationalism implies integration ‘from above’ via treaties, while transnationalization implies integration ‘from below’ via an interplay between national and international, public and private actors, across the boundaries of national states. Up until now, neo-functionalism has been the most popular perspective on transnationalism. It has also represented almost the only approach in which transnationalization has been seen as something other than a normative ambition. Neofunctionalism has turned everyday integration into an independent issue and opened up for the possibility of describing the transnational state and explaining how and why transnationalization takes place. Today the critical response to intergovernmentalism and neofunctionalism is represented by the position referred to by the title of ‘multi-level governance’ (Scharpf 1994, Jachtenfuchs and Kohler-Koch 1996, Marks, Hooghe and Blank 1996, Benz 1998). Multi-level-governance (MLG) has become a basis for studies in transnationalization, and has made it possible to introduce several neo-institutionalist approaches to the study of European integration (Bulmer 1994, 1998, Jupille and Casparoso 1999, Aspinwall and
Understanding transnational states
153
Schneider 2000, Olsen 2002a). Against this background, transnationalization has become an important issue in the study of European integration. European integration has also become one of the areas in which the application of the various neo-institutionalist schools of thought (rational choice, sociological institutionalism, historical institutionalism and discursive institutionalism) is debated (Dowding 2000, Schneider and Aspinwall 2001, Stone Sweet et al. 2001, Pedersen 2002), as well as one of the areas in which attempts are made to develop a coherent, integrated form of institutionalism (Jupille, Casparoso and Checkel 2002). Everyday integration has become an example of ‘the second movement in institutional analysis’ (Campbell and Pedersen 2001). MLG is an attempt to chart the EU’s institutional complexity by delineating a number of levels at which political decisions take place, each of which is equipped with a degree of self-government, but is also ranked in a kind of hierarchy. MLG draws on organization and network theory, in addition to institutional theory (Keohane and Hoffmann 1991, Risse-Kappen 1994, 1995, 1996, Scharpf 1994, Wincott 1995, Josselin and Wallace 2001). With regard to the question of integration, MLG has taken upon itself to demonstrate how aspects of integration can co-exist with a continuation of state independence. From this perspective, transnationalization is seen as a low political level (low politics) and/or an early phase in the decision-making process; one that on the one hand shows clear signs of integration, but on the other hand implies a reduced ability to exert an influence on the really important political decisions (high politics) (Peterson 1995a). In recent years, several studies have been made of low politics, or of the policy processes taking place within the framework of already ratified treaties and agreements, establishing a third approach to the study of EU integration, alongside intergovernmentalism and supranationalism (Joerges 1999: 16). Within this third approach three readings can be distinguished, each characterized by their separate positions on how the question of state sovereignty underlies the definition of transnational relations (Wessels 1999): the network-oriented approach, the organizational approach and the administration-oriented approach. The network-oriented approach The network-oriented understanding is characterized by the attempt to describe transnational actors as autonomous in relation to their national
154
Understanding transnational states
backgrounds: They enter into special deliberative processes, are equipped with a specific identity, and integration takes place with special – transnational – units that are neither national nor European, but comprise ‘administrations without a state’ (Wessels 1999: 265). This approach relates to the characteristics of the interplay between national and European, public and private actors,2 when this takes place independently of (or partly liberated from) the national state’s distribution of authority and control. Network approaches seek to explain EU policy outcomes on the basis of three characteristics of sector policy networks, namely membership stability, insularity and the relative strength of resource dependencies between actors (Peterson 1995a, 1995b, Jupille and Casparoso 1999). Here it is emphasized that transnationalization implies that ‘the dynamics that inform outcomes are not a reflection of national interests but of functional deliberation and sectoral pressure’ (Weiler 1999: 342). Christian Joerges (1999) has utilized the concept of governance to describe processes that take place on neither a national nor European basis, are neither hierarchical nor formal, and are carried out by deliberation and not through negotiations (p.319). Others have described how such ‘policy networks’ and ‘advocacy coalitions’ contribute to the transfer of ideas (Peterson and Blomberg 1999, Marcussen 2000, 2002). The organizational approach The organizational reading examines the EU organizations, their construction, composition, tasks and routines. This approach is characterized by emphasizing the study of the comitology and of committees and working groups under the Commission. It is also characterized by viewing these as formal organs within the framework of treaties and the EU organizational structure. The emphasis here is placed on the participation of national experts and representatives, and on the ways in which they organize their interplay within the framework of the Commission’s right to initiate. The national participants are regarded as independent (national) experts, but also as representatives of national or European interests. Some studies concentrate on the scope and significance of committees as a general phenomenon (Algieri and Rometsch 1994, Buitendijk and van Schendelen 1995, Falke 1996). Others are oriented towards policy studies, and utilize case studies to reveal the composition of committees, their internal processes and their interplay with other organizations within the EU, particularly the
Understanding transnational states
155
European Parliament, the Commission and its Directorates-General (Blumann 1996, Gehring 1999, Landfried 1999, Esmark 2002c, Højbjerg 2002). Yet others compare committees in order to identify variations in their procedures, efficiency and representativeness, and subsequently construct a typology of these (Wessels 1999). The administration-oriented approach In the administration-oriented approach, the emphasis is on the study of national administrations and on the consequences of the European Union policy process for their organization, functions and tasks.3 In contrast to the network-oriented understanding, the focus here is on the consequences for the national actors of participating in transnational networks; and in contrast to the organizational reading, the European organizations are seen as merely a contextual factor. Most of the studies undertaken within this approach are national studies emphasizing how the distribution of resources between the national central agencies has been affected by European integration.4 Several comparative studies, however, have now also been produced.5 At the same time, the question of whether the national administrative structures are converging or diverging in connection with European integration has become a separate issue (Schneider 2001). At times, the administration-oriented approach can be difficult to distinguish from comparative studies of changes in the national administrations or cross-national studies of state administration.6 This is partly because this reading is characterized in particular by the study of how national administrations transform or adjust under the pressure of external influences – national as well as international, European as well as global – but it is also due to the fact that most of the studies are concerned with changes in the administration circumscribed by national boundaries.7 This literature mainly concludes that we are not witnessing a fundamental change in administrative structure and styles and a convergence towards a common Europe Model. Changes are rather seen as mediated through existing domestic institutions, practices, and strategies that leave considerable discretion to domestic actors and institutions.8 This literature is also grounded in the institutionalist view that political agencies do not adapt quickly to changes in external conditions and human purpose. Within institutionalist theory there is a growing body of work that focuses on the complexity of the explanatory role of institutions (Campbell and Pedersen 2001, Olsen
156
Understanding transnational states
2002a, 2002b, 2002c). In relation to European integration, the explanatory role of institutions is studied by examining the establishment, maintenance and change of European organizations, together with studies on how states both have an impact on and adapt to changes at the European level. This approach is supplemented by studies based on an organizational reading, specifying the organizational components that institutions may contain (Egeberg 1999, 2000, 2002, Trondal 2001b).
Transnationalization and the transnational state This whole body of literature offers only a limited number of systematic – and comparative – studies of why, how and with what consequences the European Union’s process of integration has influenced the national central agencies.9 Transnationalization and organizational changes on the national level are usually not placed in relation to each other. The central question is whether or not national administrations change under the influence of European integration. Answered in the affirmative, it would imply that European development is the independent variable, while national change is the dependent variable. Nonetheless, a number of studies have gradually appeared giving an analytical insight into administrative changes across the boundaries of the individual states,10 as well as a number of comparative studies of administrative reforms.11 In addition, it has become more commonplace for neo-institutional theory to view European integration as a question of interplay between exogenous contingencies and domestic institutional configurations. This has given rise to a number of studies of the effects of transnational participation on the national administrations,12 and on the consequences of European integration in general for the relationship between the legislative and executive authorities of the member states.13 Besides studies of the Europeanization of the administration, it has been more or less explicitly pointed out that a new kind of state is currently under development, on the basis of existing treaties and agreements, implying that the national administrations are taking part in the formation of something new that has neither been decided nor foreseen – namely the transnational state. Peter Nedergaard speaks of a ‘merger bureaucracy’ (see also Wessels and Rometsch 1996) between the Danish administration and the EU administration.14 Wessels talks about the formation of an administration ‘without states’ (1999). As mentioned earlier,
Understanding transnational states
157
Thomas Risse-Kappen’s definition of transnational relations is the one that has been most widely adopted (Risse-Kappen 1994, 1995, 1996). We will return to it later. In what follows, we return to the basic organization of the book and present three issues which are mutually overlapping, but all of which ought to have a central place in future studies of the transnational state and transnationalization. The first issue will be based on how we have described the transnational state, the second on how we have explained transnationalization, and the third on the assessment of the consequences on democratic government on the building of the transnational state. We will draw a line between transnationalization and the transnational state, and distinguish between the dynamic process by which transnational states are formed and the static description of how the transnational state can be characterized. Describing: differentiation and coupling The first issue concerns the question of what is meant by the statement that transnational relations imply the ‘transcendence’ or ‘blurring’ of the outer and inner borders of national states. In the literature on transnational relations, the European level is assigned considerable independence in relation to the national level, just as the national level is assigned considerable self-government in relation to the European level. Against this background, it is assumed that the national administrations are subject to change under the influence of the European project, but are not themselves active in the formation of transnational central agencies. The European process of development is thus viewed as an external reason for the internal adaptations, but the internal alterations are not viewed as an element of an incipient transnational administration. The same applies to changes in the national balance of power. Within public administration research, it is usual to look upon governmental administrative policies as the reason for changes in the organization and function of the individual administrative units, i.e. it is assumed that reform policies provide the independent variable, and the administrative changes are the dependent variable (Lægreid and Pedersen 1994, 1996, 1999). But although studies in public management have to some extent begun to focus on the administration’s transcendence of territorial boundaries in connection with Europeanization and internationalization, it has to a greater extent focused on the blurring of the internal boundary, namely
158
Understanding transnational states
the boundary dividing the administration from various private interests. It is assumed that the state has both external and internal sovereignty, i.e. that it can be described on the basis of a legalistic approach, and that it is within these borders that changes can be detected. On this basis, it is concluded that such changes lead to borders becoming ‘transcended’ or ‘blurred’. Against this background, we wish to point out, not just that transnationalization assumes the sovereignty and unity of states, but also that it implies the formation of another set of differentiations, consisting of couplings between national and European, public and private actors. We have emphasized that transnationalization implies the creation of new internal differentiations within states (such as through segmentation and the construction of coordination systems), as well as new couplings between states (by way of contacts and participation, and their stabilization). These new differentiations mean that the national states can maintain their interior and exterior borders while at the same time being included in a new type of administration in which authority is distributed in a different way than it traditionally has been. The differentiations also imply that the exterior borders between states, and the interior borders between state and society, can become ‘consolidated’ or ‘emphasized’ at the same time as they are ‘transgressed’. Our studies have, for example, shown that the national authorities, when they participate in transnational relations, do so by formulating a ‘national interest’ (Jacobsson 1997a, 1997b, Pedersen 2001). The transnational processes thus end up revolving to a large extent around how national interests are formulated as a prerequisite for national actors to enter into transnational processes. In this way, national civil servants and politicians formulate ‘national interests’ and confront other domestic interests in transnational policy processes. They learn about their own interests and thereby strengthen their own identity as representatives of Danish or Swedish authorities, while at the same time, the idea of what it means to be Danish or Swedish influences the interplay taking place in transnational relations. The same phenomenon occurs with respect to the inner border. Here, the national systems of coordination ‘compel’ the particular administrative units to formulate their own interests. The same applies to private interest organizations and companies. Both the exterior and the interior borders are thus ‘consolidated’ or ‘emphasized’ at the same time as the special interests are advanced in contexts where exterior and interior borders are ‘transcended’ and transnational loyalties are founded.
Understanding transnational states
159
Transnationalization can in other words cause old borders to be consolidated at the same time as new differentiations are formed. Consequently, the transnational administration cannot be understood with the help of the legalistic leitmotiv and it is, by definition, impossible to determine whether transnationalization is a zero-sumgame or a plus-sum-game. The question must be determined empirically. The actual distribution of sovereignty must be examined via concrete empirical studies of how the ‘live’ administration is exercised in practice and how new differentiations and couplings are established. Explaining: causality and complexity The second issue concerns the question of how an explanation can be arrived at for the formation of a transnational administration. We have based our study on the causal approach, and on the fact that, while it is possible to distinguish between state sovereignty on the one hand and European integration on the other, this does not necessarily imply that the European process of development must be seen as the independent variable while the national changes are seen as the dependent variable. We have assumed that it is possible to distinguish between sovereignty and integration, and between ‘external’ and ‘internal’ factors. At the same time, we have shown that EU-related adaptation within the state administration cannot be reduced to a single explanatory factor or to one basic perspective. Instead we have emphasized how several factors have to be blended in explaining transnationalization. Firstly it was demonstrated that the EU comprises a factor to which the national administrations adapt, but also that this adaptation possesses its own dynamic, inasmuch as the Nordic administrations themselves assess that they were more influenced by, and used more resources on, EUrelated tasks in 1998 than in 1993, and that adaptation to the EU constituted an important explanatory factor during this period in relation to organizational changes in the national administrations. Secondly it was emphasized that the actual length of membership is a factor, but also that length of membership comprises an independent dynamic, in the sense that Denmark – as the Nordic country that has been a full member for the longest period – differs in several respects from the more recent member states (Sweden and Finland), and hence that length of membership provides an opportunity for gaining experience and learning. Time, consequently, constitutes a dynamic
160
Understanding transnational states
of its own in the explanation of changes in the national administrations. Thirdly we discovered that the EU decision-making processes are a factor in themselves, but also that these are equipped with a dynamic of their own, in the sense that, as the multiplier effect demonstrates, the performance of EU tasks opens the administration up to an ever wider environment, and consequently paves the way for ever greater diversity in contacts and an ever greater complexity in patterns of interaction. Finally, we have shown that the transnational processes comprise a specific factor in themselves. It is through transnational processes that EU Directives and regulations are formulated to create the rules that national administrations are later obliged to implement and adapt to. Implementation occurs by harmonizing the national legislation and by undertaking the necessary organizational adaptation in the national administrations. In this way, the circle is closed: The transnational networks and decisions that the national authorities have established return in the form of requirements for the harmonization of the national legislation and the integration of national central agencies into the transnational state. Against this background, we have proposed a more complex model of causal factors to explain transnationalization called the blending of factors (Campbell and Pedersen 2001). There are no clear distinctions between what may be considered European and what may be considered national, and coevolution exists between structural constraints and conscious actors. Adjustment impulses, which frequently arise through increased integration, are translated and edited to make them compatible with national administrative traditions and national strategies. Problems and solutions are interpreted, modified and identified within the coevolution of national strategies, external impulses and internal constraints. Against this background, it is clear that European integration occurs via several dynamic processes that are concerned with structural factors, with membership length, and the time factor, but also related to the multiplier effect and the feed-back mechanisms that begin to operate when national administrations enter into transnational networks to formulate requirements concerning how they themselves are later to be further integrated into the European project. Such a transnational perspective places greater emphasis on the coevolution of several different factors, and downplays the significance of simple adaptation as the most important dynamic. It also emphasizes the fact that the national authorities have the opportunity to influence the rules and laws to which they must later
Understanding transnational states
161
adapt, and that this translation, consequently, is not external, but is rather prepared and proposed via transnational processes. Thus editing, modification and interpretation are characteristics of the adaptation process. Assessing: challenges and development The final issue that we deal with concerns the consequences of transnational administration. In this context, we have utilized another leitmotiv from the literature, namely the parliamentary approach, which is based on the idea that there is a certain balance of power between the national political agencies, and that alterations in this pose challenges to democratic government. Here, too, we point to a greater complexity and to the fact that the approach, while it may say something about the actual forms of governance, can nonetheless function only as a heuristic basis. There are several – and at times mutually contradictory – consequences of transnationalization. The national balance of power is altered when the administration gains influence in relation to parliaments and governments. The national power balance can also be maintained, or even enhanced, by the establishment of systems of coordination to ensure the ability of parliaments to control governments. National political agencies can also be opened up towards a wider world and a more comprehensive set of private and public, national and international interests and needs. The multiplier effect paves the way for external interest to gain influence on governments, but also creates the possibility for transnational networks to isolate themselves from broader public and national authorities to become dependent on special interests. Exclusion and captivity can become the consequence of greater openness and the opening up of more channels of influence. We have consequently emphasized that the consequences of transnationalization for democratic government are a matter for empirical research, and that the answers can only be supplied by means of studies that are methodically prepared and systematically carried out.
Summary The existing literature on European integration rests on a fundamental division between integration and state sovereignty. There is, furthermore, a tendency to associate ‘integration’ with something that occurs
162
Understanding transnational states
at the expense of national sovereignty. In this book, we attempt to take a different approach. We have blended a structural perspective with an actor perspective, and a perspective concerned with historical trends (path dependency) with one concerned with historical innovation (national strategies). In this way, we have analysed: (1) the extent to which national administrative units enter into networks with EU organizations, private interests and the administrations of other countries – thereby coupling themselves to such administrations and building the transnational state; (2) the ways in which the national administrations are influenced by European everyday integration – and thereby become differentiated in new ways; (3) the extent to which national strategies, administrative structures and traditions influence the ways in which external demands are translated; and finally, (4) the ways in which several factors can be utilized to explain transnationalization. Against this background, we will now return to the definition of transnationalization and the transnational state. In the literature we can find several definitions of transnationalization. Probably the most commonly accepted is by Risse-Kappen (1995: 1). His definition resembles ours to a great degree but there are some aspects that we would like to underline. There is, to begin with, a risk that this definition plays down the great variety of actors involved in the processes by lumping them all together as non-state. Secondly, we wish to stress that states are not monoliths, but consist of many units and that these units may be relatively loosely linked to each other (segmentation). Thirdly, we emphasize how state agencies adapt over time to environments that in turn adapt to state activities (coevolution). Fourthly, we stress that one should not look upon transnationalization as something that primarily exists ‘outside’ states, but instead that this process is embedded in everyday activities. Fifthly, we emphasize that influence on state activities can be exerted even without regular interactions. Interaction can take place without necessarily being significant, just as actions may be influenced even though there is no regular interaction. Against this background, we choose to view transnationalization as follows: Transnationalization is said to have taken place when the EU affects the administration to a significant degree, and when the EU has a considerable effect on the majority of areas of relevance to the administrative units. At the same time, a transnational state is said to have been created when significantly stable relationships develop in the form of contact networks, participation rights and actual
Understanding transnational states
163
participation in a number of national and European organs, and when such relationships are characterized by weak and ambiguous political mandates and signals from the political leadership at the national level. In sum, this allows us to describe the transnational state as neither national nor European, but both. Just as it allows us to describe the transnational state as neither public nor private, but both, and as neither political nor administrative, but both. It also allows us to highlight that transnationalization can neither be explained by structural factors nor by national strategies alone, but only by both in combination, and that the processes of transnationalization is neither a zero-sum nor a plussum game, but possibly both; just as transnationalization is neither solely the unforeseen consequence of EU decisions nor solely the result of collective deliberation but the coevolution of both. And finally it allows us to estimate that the transnational administration neither involves the undermining nor the enhancement of authority, but possibly both; and neither implies a loss of democracy nor an increase in democracy, but possibly both. The transnational state is thus characterized by a coevolution of factors, as well as by complex and paradoxical consequences for the organization of national administrations and the balance of power between national political agencies. Similarly, transnationalization implies that national state sovereignty (the exterior border) and the unity of the state (the interior border) enter into political and administrative activities, through which a new type of state is created. By stressing the need for a more complex explanation of transnationalization we are introducing the method of blending (Campbell and Pedersen 2001). Blending explanatory factors is coming late to the study of European integration even if it is not new in the social sciences (see ex. Giddens on dualism, 1979) and enables the study of how impulses, innovations and constraints are coevolving. The theory of translation is a possible outcome of this method but the study here presented has so far only cultivated the prospect of such a theory. Our knowledge of the transnational state is, as yet, limited, and there is a clear need for further studies. Up until now, descriptions of transnational relations have mainly been based on national studies or, as in this instance, on comparative studies involving only a few EU member states, and utilizing the questionnaire technique alone. More complete and diverse data are required. Firstly, it will be necessary to involve more member states. There is a lack of data from the larger member states that could enable the comparison, not only of the most
164
Understanding transnational states
similar cases (the Nordic countries), but also of the most disparate cases, as well as the comparison of larger with smaller member states. Secondly, time series should be established which would make it possible to study transnationalization over time, and to reveal in a more systematic manner the dynamic aspects involved in the integration process and the formation of transnational states. Thirdly, detailed process studies are needed to uncover how external impulses; national innovations and internal institutional constraints are translated into institutional changes and the building of a transnational state. Fourthly, comparative case studies should be launched, doubtless including ‘thick descriptions’ of policy areas. Earlier examples of such studies have demonstrated that it is only via in-depth, detailed studies that the most important and systematic differences and similarities (between policy areas, sectors and levels) can be revealed and indicated (Pedersen 2002). Finally, the focus has to be extended from central administrative apparatus to include the executive and the legislative. A theory of translation in other words demands further studies. Instead of the inductive research process based on data collection and description, it will be necessary to formulate more hypotheticaldeductive studies in order to ascertain how various factors (causal, structural, historical, process-related) are blended, and why countries with different administrative traditions nonetheless converge in the formation of transnational states. Finally – and perhaps mainly – assessments of the consequences of Europeanization for democratic government should be organized in a methodical and systematic way. It is striking how only a very few studies have been undertaken of how national central agencies develop strategies to legitimize their participation in the European project. For the same reason, we know only very little about how democratic ideals have been reinterpreted, or new ideals formulated, in association with European integration. Historical studies of the national ideals for democratic government should be carried out, and should lead to cross-country comparisons and studies of the processes by which ideals of this kind are reinterpreted or reformulated. Similarly, methods should be developed to enable systematic comparisons of developments in the balance of power between the national central agencies. There is a striking lack of methodological considerations of how best to assess (or possibly measure) the effects of European integration on democratic government and its legitimacy.
Understanding transnational states
165
Against this background, it is self-evident that research into transnationalization and a theory of translation faces major challenges. However, although there are many fundamental problems to be discussed and a great deal of empirical data to be collected, there can no longer be any doubt that a new kind of state and a new way of organizing state administration is being constructed. Consequently, the study of the transnational state helps to acquire insight into how the current political geography, which has existed in Europe ever since the peace of Westphalia created the basis for the national state, is now being transformed.
Notes
1 Europeanization of Nordic states 1 The complete questionnaire is printed in Jacobsson, Lægreid and Pedersen (2001a) appendix 2. 2 Transnational states 1 The Maastricht Treaty (1992) created a new organisation of the European Union, which was based on Three Pillars: The European Communities, a Common Foreign and Security Policy, and Cooperation in the Fields of Justice and Home Affairs. Pillar 1, the European Communities, was by far the most important pillar since it incorporated most of the EU’s policy responsibilities. 6 Democratic challenges 1 Lodge 1994, Hayward 1995, Laffan 1996, Obradovic 1996, Weiler 1996, Hansen and Williams 1999, Lord 2001. 2 Such a study has been undertaken of the evolution of democratic ideals in Denmark from the 1960s to today (Pedersen 2002). Similar studies have not been carried out in the other three countries. The following examination of ideals relies on secondary literature with respect to the three other Nordic countries, and comprises an ideal-typical description of democratic ideals which is applicable to all four Nordic countries. 7 Understanding transnational states 1 In addition to these descriptive and explanatory approaches, there is a more normative tendency that for many years has claimed that it is possible to create a transnational brotherhood or democracy across the boundaries of the existing competitive conditions between the sovereign states. This ‘idealistic’ approach has been especially prevalent within the sub-discipline ‘international politics’ (IP), where it has given rise to at least two major debates (Keohane and Nye 1973). The first of these took place in the 1940s, when the concept of transnationalism was introduced as a response to the ‘realist’ view of how the ‘international’ dimension could be viewed as a condition of continual anarchy between states (Bull 1977). The second took place in the 1970s, when the question of identity was the central concern, and particular attention was focused on the question of cross-border (transnational) identities, such as gender, race, religion and culture (Wendt 1992,
168
2 3 4 5 6 7 8 9 10 11 12
13 14
Notes
Neumann 1996). In recent years, this debate has been widened, and has burst the boundaries of international politics as a discipline. In the wake of the formation of the European Union, a democratic theoretical transnationalism has arisen, emphasizing the union’s ‘democratic deficit’ and the debate on how the union can be democratized. This debate emphasizes cross-border communication, social learning, socialization and democratization (see for example Eriksen and Fossum 2000), and the question of the transfer of ideas and convergence in policy responses to comparable challenges (Checkel 2001). Joerges and Neyer 1997a, 1997b, Joerges, Schepel and Vos 1998, Joerges 1999, Neyer 1999, Weiler 1999. Andersen and Eliassen 1993, He´ritier, Knill and Mingers 1996, Me´ny, Muller and Quermonne 1996, Schmidt 1996, Trondal 1996, 1999, 2001b, 2001c, Veggeland 1999b, Sverdrup 2000, Marcussen and Ronit 2003. Kooiman 1993, Kohler-Koch 1996, Sandholtz 1996, Kohler-Koch and Eising 1999, Lenschow 1999. Knill and Lenschow 1998b, Bo¨rzel 1999, Kassim, Peters and Wright 2000, Cowles, Caporaso and Risse 2001, He´ritier et al. 2001, Knill 2001. Lægreid and Pedersen 1994, 1996, 1999, Peters 1995, Pierre 1995, Olsen and Peters 1996. Eliassen 1993, Derlien 1995, Pierre 1995, Jacobsson 1997a, 1997b, Kelstrup and Branner 2000, Nedergaard 2000, Bulmer and Bursch 2001. Harmsen 1999, Goetz 2000, Knill 2001, Olsen 2002c, Marcussen and Ronit 2003. Peters 1988, 1996, Cowles, Caporaso and Risse 2001: 3, Knill 2001. Peters 1995, Bekke, Perry and Toonen 1997, Harmsen 1999. Metcalfe 1994, Pierre 1995, Hood 1996, Olsen and Peters 1996. Toonen 1992, Page and Wouters 1995, Flynn and Strehl 1996, Wessels and Rometsch 1996, Wright 1996, Jacobsson 1997a, 1997b, Jacobsson and Sundstro¨m 1999, Kassim, Peters and Wright 2000, Lægreid 2000, Sverdrup 2000, Pedersen 2002. Bulmer 1994, Christensen 1996, Smith 1996, Bulmer and Armstrong 1998, Bulmer and Bursch 1998, Sverdup 1998. Nedergaard 2000: 46, 174, also Egeberg 1999.
Bibliography
Algieri, F. and Rometsch, D. (1994) ‘Europa¨ische Kommission und Organisierte Interessen: Die Rolle des “Ausschusswesens” und Ansa¨tze fu¨r einen strukturierten Dialog’, in V. Eichner and H. Voelzkow (eds) Europa¨ische Integration und verbandliche Interessevermittlung, Marburg: Metropolis Verlag. Allison, G.T. (1971) Essence of Decision: Explaining the Cuban Missile Crisis, Boston: Little and Brown. Anckar, D. (1993) ‘Comparative Research in the Nordic Countries: Overcoming Etnocentrism?’, Scandinavian Political Studies, 16 (2): 107–126. Andersen, S.S. (2000a) ‘Hvordan er EU mulig? Et ny-institusjonelt sosiologisk perspektiv’, Oslo: ARENA, Working Paper 00/3. —— (2000b) ‘Norway: Insider and Outsider’, Oslo: ARENA, Working Paper 00/14. Andersen, S.S. and Eliassen, K. (eds) (1993) Making Policy in Europe. The Europeification of National Policy-Making, London: Sage. Aspinwall, M. and Schneider, G. (2000) ‘Same Menu, Separate Tables: The Institutional Turn in Political Science and the Study of European Integration’, European Journal of Political Research, 38: 1–36. Baldersheim, H., Sandberg, S., Sta˚hlberg K. and Øga˚rd M. (2001) ‘Norden in the Europe of Regions. A Summary of Perspectives and Results’, in K. Sta˚hlberg (ed.) Social Science. The Nordic Countries and Europe Part II, Copenhagen: Nordic Council of Ministers, Nord 2001: 23. Bekke, H.A.G.M., Perry J.L. and Toonen, T.A.J. (eds) (1997) Civil Service Systems in Comparative Perspective, Indiana: Indiana University Press. Benz, A. (1998) ‘Ansatzpunkte fu¨r ein Europafa¨higes Demokratikonzept’, in B. Kohler-Koch (ed.) Regieren in entgrenzten Ra¨umen, Politische Vierteljahresschrift, Sonderheft, 29: 345–368. Bergman, T. and Damgaard, E. (2000) Delegation and Accountability in European Integration, London: Frank Cass. Blumann, C. (1996) ‘Le Parlement Europe`en et la Comitologie: une Complication pour la Confe`rence Intergourvenmentale de 1996’, Revue Trimesterielle de Droit Eruope`en, 32: 1–24. Boli, J. and Thomas, G.M. (eds) (1999) Constructing World Culture, International Nongovernmental Organizations Since 1875, Stanford: Stanford University Press. Bo¨rzel, T. (1999) ‘Towards Convergence in Europe? Institutional Adaptation to Europeanization in Germany and Spain’, Journal of Common Market Studies, 37 (4): 573–596.
170
Bibliography
Brunsson, N. (1989) The Organization of Hypocrisy: Talk, Decisions and Actions in Organizations, New York: John Wiley. Brunsson, N., Jacobsson, B. and associates (2000) A World of Standards, Oxford: Oxford University Press. Bruun, N., Flodgren, B., Halvorsen, M., Hyden H. and Nielsen R. (1990) Den nordiska modellen. Fackfo¨reningarna och arbetsra¨tten i Norden – nu och i framtiden, Malmo¨: Liber. Bryman, A. (2001) Social Research Methods, Oxford: Oxford University Press. Bræin, S. (1999) ‘Europeisering og rettsliggjøring. EØS-avtalen, EU og det norske alkoholmonopolet’, Oslo: ARENA, Report 2/99. Bue, L.A. (2000) ‘Nasjonale konkurranseregimer under press. En komparativ studie av norske og svenske konkurranseregimer’, Bergen: LOS-senteret, Report 0007. Buitendijk, G.J. and van Schendelen, M.C.P.M. (1995) ‘Brussels Advisory Committees: A Channel for Influence?’, European Law Review, 20: 37–56. Bull, H. (1977) The Anarchical Society, Houndsmill: Macmillan. Bulmer, S. (1994) ‘The Governance of the European Union: A New Institutional Approach’, Journal of Public Policy, 13: 351–353. —— (1998) ‘New Institutionalism and the Governance of The Single European Market’, Journal of European Public Policy, 5 (3): 365–386. Bulmer, S. and Armstrong, K. (1998) The Governance of the Single European Market, Manchester: Manchester University Press. Bulmer, S. and Burch, M. (1998) ‘Organizing for Europe: Whitehall, the British State and European Union’, Public Administration, 76 (4): 601–628. —— (2001) ‘The “Europeanization” of Central Government: UK and Germany in Historical Perspective’, in G. Schneider and M. Aspinwall (eds) The rules of integration, Manchester: Manchester University Press. Camillieri, J.A. and Falk, J. (1992) The End of Sovereignty? The Politics of a Shrinking and Fragmenting World, Aldershot: Edward Elgar. Campbell, J.L. and Pedersen, O.K. (eds) (2001) The Rise of Neoliberalism and Institutional Analysis, New Jersey: Princeton University Press. Caporaso, J. (1996) ‘The European Union and Forms of States: Westphalian, Regulatory or Post-Modern’, Journal of Common Market Studies, 34: 29–51. Checkel, J.T. (2001) ‘Why Comply? Social Learning and European Identity Change’, International Organization, 55 (3): 553–588. Christensen, J.G. (2002) ‘Den fleksible og robuste forvaltning’, in M. Markussen and K. Ronit (eds) Internationaliseringen af den offentlige forvaltning i Danmark, Aarhus: Aarhus Universitetsforlag. Christensen, T. (1996) ‘Adapting to Processes of Europeanisation. A Study of the Norwegian Ministry of Foreign Affairs’, Oslo: ARENA, Report 2/96. Christensen, T. and Egeberg, M. (1997) ‘Sentraladministrasjonen – en oversikt over trekk ved departementer og direktorater’, in T. Christensen, and M. Egeberg (eds) Forvaltningskunnskap, Oslo: Tano Aschehoug. Christensen, T. and Lægreid, P. (2001) New Public Management. The Transformation of Ideas and Practice, Aldershot: Ashgate. Christiansen, P.M. (1998) ‘Interesseorganisationer, centraladministration og udviklingen af nye indflydelsesstrategier’, in K. Ronit (ed.) Interesseorganisationer i dansk politik, Copenhagen: Jurist- og Økonomforbundets Forlag.
Bibliography 171 Christiansen, P.M. and Sidenius, N.C. (1999) ‘Venner for altid? Samspillet mellem centraladministration og interesseorganisationer’, in J. Blom-Hansen and C. Daugbjerg (eds) Magtens organisering. Stat og interesseorganisationer i Danmark, Aarhus: Aarhus Universitetsforlag. Cohen, M.D., March, J.G. and Olsen, J.P. (1972) ‘A Garbage Can Model of Organizational Choice’, Administrative Science Quarterly, 17: 1–25. Cowles, M.G., Caporaso J.A. and Risse, T. (eds) (2001) Transforming Europe. Europeanization and Domestic Change, Ithaca: Cornell University Press. Cyert, R.M. and March, J.G. (1963) A Behavioral Theory of the Firm, Englewood Cliffs: Prentice Hall. Czarniawska, B. and Se´von, G. (eds) (1996) Translating Organizational Change, Berlin: De Gruyter. Damgaard, E. and Nørgaard, A.S. (2000) ‘The European Union and Danish Parliamentary Democracy’, in T. Bergman and E. Damgaard (eds) Delegation and Accountability in European Integration, London: Frank Cass. Derlien, H-U. (1995) ‘Public Administration in Germany: Political and Societal Relations’, in J. Pierre (ed.) Bureaucracy in the Modern State. An Introduction to Comparative Public Administration, Aldershot: Edward Elgar. Deutsch, K.W. (1957) Political Community and the North Atlantic Area: International Organization in the Light of Historical Experience, Princeton, N.J: Princeton University Press. DiMaggio, P.J. and Powell, W.W. (1983) ‘The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields’, American Journal of Sociology, 83 (2): 147–160. Dowding, K. (2000) ‘Institutionalist Research on the European Union: A Critical Review’, European Union Politics, 1: 125–144. Dyrdal, M. (2001) ‘Nasjonale legemiddelregimer for fall: en komparativ studie av nordiske legemiddelregimers Europatilpasning’, Bergen: LOS-senteret, Report 0117. Egeberg, M. (1999) ‘Transcending Intergovernmentalism? Identity and Role Perceptions of National Officials in EU-Decision Making’, Journal of European Public Policy, 6: 456–474. —— (2000) ‘How Federal? The Organizational Dimension of Integration in the EU (and elsewhere)’, Journal of European Public Policy, 8 (5): 728–746. —— (2002), ‘An Organizational Approach to European Integration – What Organizations Tell us About System Transformation, Committee Governance and Commission Decision Making’, Oslo: ARENA Working Paper 02/19. Egeberg, M. and Trondal, J. (1997) ‘Innenriksforvaltningen og den offentlige politikkens internasjonalisering’, in T. Christensen and M. Egeberg (eds) Forvaltningskunnskap, Oslo: Tano Aschehoug. Eising, R. and Kohler-Koch, B. (1999) ‘Governance in the European Union: A Comparative Assessment’, in B. Kohler-Koch and R. Eising (eds) The Transformation of Governance in the European Union, London: Routledge. Eitrheim, P. and Kuhnle, S. (2000) ‘Nordic Welfare States in the 1990’s: Institutional Stability and Signals of Divergence’, in S. Kuhnle (ed.) Survival and the European Welfare State, London: Routledge. Ekengren, M. (1998) ‘Time and European Governance. The Empirical Value of Three Reflective Approaches’, Ph.D- dissertation, Stockholm: Department of Political Science, University of Stockholm.
172
Bibliography
Eliassen, K.A. (1993) ‘Europeiske utfordringer for nordisk forvaltning’, Nordisk Administrativt Tidsskrift, 4: 420–431. Eriksen, E.O. and Fossum, J.E. (eds) (2000) Democracy in the European Union – Integration through Deliberation?, London: Routledge. Esmark, A. (2001) ‘Mod en transnational forvaltning’, in B. Jacobsson, P. Lægreid and O.K. Pedersen (eds) Europaveje. EU og de centrale forvaltninger i Norden, Copenhagen: Jurist- og Økonomforbundets Forlag. —— (2002a) ‘At forvalte Europa – europæiseringen af den danske Centraladministration’, Ph.D-dissertation, Copenhagen: Department of Political Science, University of Copenhagen. —— (2002b) ‘Skyggeorganisering i Erhvervsministeriet’, in O.K. Pedersen et al. EU i forvaltingen. Broen fra Slotsholmen til Bruxelles, Copenhagen: Jurist- og Økonomforbundets forlag. —— (2002c) ‘Den a˚bne proces’, in O.K. Pedersen (ed.) EU i forvaltningen. Broen fra Slotsholmen til Bruxelles, Copenhagen: Jurist- og Økonomforbundets Forlag. Esping-Andersson, G. (1990) The Three Worlds of Welfare Capitalism, Cambridge: Polity Press. Etzioni, A. (1965) Political Unification: a Comparative Study of Leaders and Forces, New York: Holt, Rinehart and Winston. 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. Ferlie, E. et al. (1996) The New Public Management in Action, Oxford: Oxford University Press. Fjær, S. (2000) ‘Integrasjon og faglig fellesskap. Europeiseringsprocesser i narkotikapolitikken i Danmark, Nederland og Sverige’, Bergen: Department of administration and organization theory, Report no. 74. Flynn, N. and Strehl, F. (eds) (1996) Public Sector Management in Europe, London: Prentice Hall. Frankel, C. (2002) ‘Fællesskabets markedsdannelse’, in O.K. Pedersen (ed.) EU i forvaltningen. Broen fra Slotsholmen til Bruxelles, Copenhagen: Jurist- og Økonomforbundets Forlag. Frankel, C., Højbjerg, E. and Pedersen, O.K. (1998) ‘EU miljøpolitiske potentialer i europæisk eknisk standardisering’, Copenhagen: COS-centeret, Working Paper 1/1998. Franklin, M., Marsh, M. and McLaren, L. (1994) ‘Uncorking the Bottle: Popular Opposition to European Unification in the Wake of Maastricht’ Journal of Common Market Studies, 33 (1): 27–46. Fudge, C. and Gustavsson, L. (1989) ‘Administrative Reform and Public Management in the United Kingdom’, Public Money and Management, 9 (Summer): 29–34. Gehring, Th. (1999) ‘Bargaining, Arguing and Functional Differentiation of Decision Making: The Role of Committees in European Environmental Process Regulation’, in Chr. Joerges and E. Vos EU Committees: Social Regulation, Law and Politics, Oxford: Hart Publishing. Giddens, A. (1979) Central Problems in Social Theory, Los Angeles: University of California Press.
Bibliography 173 Goetz, K. (2000) ‘European Integration and National Executives. A Cause in Search of an Effect?’, West European Politics, 23 (4): 210–231. Goetz, K.H. and Hix, S. (eds) (2001) Europeanised Politics. European Integration and National Political Systems, London: Frank Cass. Goodin, R.E. (1996) The Theory of Institutional Design, Cambridge: Cambridge University Press. Greve, B. (1996) Comparative Welfare Systems: the Scandinavian Model in a Period of Change, Basingstoke: Macmillan. Grieco, J. (1996) ‘State Interests and International Rule Trajectories: A Neorealist Interpretation of the Maastricht Treaty and European Economic and Monetary Union’, Security Studies, 5 (3): 261–305. Grønbeck-Jensen, C. (1998) ‘The Scandinavian Tradition of Open Government and the European Union: Problems of Compatibility?’, Journal of European Public Policy, 5(1): 185–199. Haas, E. (1958) The Uniting of Europe: Political, Social and Economic Forces, 1950–1957, Palo Alto: Stanford University Press. Hanf, K. and Soetendorp, B. (eds) (1998) Adapting to European Integration, London: Longman. Hansen, L. and Williams, M. C. (1999) ‘The Myths of Europe: Legitimate, Community and the “Crisis” of the EU’, Journal of Common Market Studio, 37 (2): 233–49. Harmsen, R. (1999) ‘The Europeanization of National Administration: A Comparative Study of France and the Netherlands’, Governance, 11 (1): 81–114. Hayward, J. (1995) (ed.) The Crisis of Representation in Europe, London: Frank Cass. Hegeland, H. (1999) ‘Riksdagen, Europeiska Unionen och demokratin. En studie av Riksdagens arbete med EU-fra˚gor’, Department of Political Science, University of Lund. Held, D., McGrew, A., Goldblatt, D. and Perraton, J. (1999) Global Transformations, Oxford: Oxford University Press. He´ritier, A., Kerwer, D. Knill, C., Lehmkuhl, D. and Teutsch, M. (2001) Differential Europe. New Opportunities and Constraints for National Policy-Making, London: Rowman and Littlefield. He´ritier, A., Knill, C. and Mingers, S. (1996) Ringing the Changes in Europe. Regulatory Competition and Redefinition of the State. Britain, France, Germany, Berlin: Walter de Gruyter. Hermansson, J. (1993) Politik som interessekamp: Parlamentariskt beslutsfattande och organiserade interessen i Sverige, Stockholm: Norstedts Juridik. Hermansson, J. et al. (1999) ‘Avkorporativisering och lobbyism – konturerna till en ny politisk modell’, Stockholm: SOU 1999: 121. Hix, S. and Goetz, K.H. (2000) ‘Introduction: European Integration and National Political Systems’, Western European Politics, 23(4): 1–26. Hood, C. (1996) ‘United Kingdom: From Second Chance to Near-Miss Learning’, in J.P. Olsen and B. Guy Peters (eds) Lessons from Experience. Experimental Learning in Administrative Reforms in Eight Democracies, Oslo: Scandinavian University Press. Højbjerg, E. (2001) ‘Euopeiske konstitutionsspil – miljø, marked & styring’, Copenhagen: Copenhagen Business School, Ph.D.serie 12, 2001.
174
Bibliography
—— (2002) ‘Processen om EMAS’, i O.K. Pedersen (ed.) EU i forvaltningen. Broen fra Slotsholmen til Bruxelles, Copenhagen: Jurist- og Økonomforbundets Forlag. Højbjerg, E. and Pedersen, D. (2002) ‘Det danske koordinationssystem’, in O.K. Pedersen (ed.) EU I forvaltningen. Broen fra Slotsholmen til Bruxelles. Copenhagen: Jurist- og Økonomforbundet forlag. Jachtenfuchs, M. and Kohler-Koch, B. (eds) (1996) Europa¨ische Integration. Opladen: Leske+Budrich. Jacobsson, B. (1984) Hur styrs fo¨rvaltningen? Myt och verklighet kring departementens styrning av a¨mbetsverken, Lund: Studentlitteratur. —— (1987) Kraftsamlingen. Politik och fo¨retagande i parallella processer, Lund: Doxa och Studentlitteratur. —— (1993) ‘Europeiseringen av fo¨rvaltningen’, Statsvetenskaplig Tidskrift, 96: 113–137. —— (1997a) Europa och staten, Stockholm: SOU 1997: 30. —— (1997b) Statsforvaltningen og Europeiseringen, Stockholm: SCORE. —— (1999) ‘Europeisering och the state transformation’, in K. Goldman et al. Politikens internationalisering, Lund: Studentlitteratur. —— (2001) ‘Relationer til den politiska ledningen’, in B. Jacobsson, P. Lægreid and O.K. Pedersen (eds) Europaveje. EU i de nordiske centralforvaltninger, Copenhagen: Jurist- og Økonomforbundets Forlag. Jacobsson, B., Lægreid, P. and Pedersen, O.K. (ed.) (2001a) Europaveje. De Nordiske centralforvaltninger i EU, Copenhagen: Jurist- og Økonomforbundets Forlag. Jacobsson, B., Lægreid, P. and Pedersen. O.K. (2001b) ‘Diverging Roads to Europe’, in K. Sta˚hlberg (ed.) Social Science. The Nordic Countries and Europe II, Copenhagen: Nordic Council of Ministers, Nord 2001: 23. Jacobsson, B. and Sundstro¨m, G. (1999) ‘Inva¨vd i Europa’, Score working paper series, 1999: 10. Stockholm, Score. Jeppestøl, G. (1999) ‘Lobbyvirksomhet – et no¨dvendig onde? En studie av norsk lobbyvirksomhet i EU: Utstrekning, begrunnelse og utfordring’, Bergen: Department for administration and organization theory. Thesis. Jessop, B. (2001) ‘Globalization and The National State’, in S. Aaronowitz and P. Bratsis (eds) Rethinking the State: Miliband, Poulantzas and State Theory, Minneapolis: University of Minnesota Press. Joerges, C. (1999) ‘Good Governance through Comitology?’, in C. Joerges and E. Vos: EU Committees: Social Regulation, Law and Politics, Oxford: Hart Publishing. Joerges, C. and Neyer, J. (1997a) ‘From Intergovernmental Bargaining to Deliberative Political Processes: The Constitutionalization of Comitology’, European Law Journal, 3: 273–299. —— (1997b) ‘Transforming Strategic Interaction into Deliberative Problem Solving: European Comitology in the Foodstuff Sector’, Journal of European Public Policy, 4: 609–625. Joerges, C., Schepel, H. and Vos, E. (1998) ‘“Delegation” and the European Polity: The Problems of the Law with the Role of Standardization Organizations in European Legislation’, paper presented at the ‘RSC Conference on The Political Economy of Standards’, Florence, 4–5 June, 1998. Johansson, K. M. (1997) Transnational Party Alliances: Analysing the Hard-Won Alliance between Conservatives and Christian Democrats in the European Parliament, Lund: Studentlitteratur.
Bibliography 175 Josselin D. and Wallace, W. (eds) (2001) Non-State Actors in World Politics, Basingstoke: Palgrave. Jupille, J. and Caporaso, J. (1999) ‘Institutionalism and the European Union: Beyond International Relations and Comparative Politics’, Annual Review of Political Science, 2: 409–425. Jupille, J., Caporaso, J.A. and Checkel, J.T. (2002) ‘Integration Institutions: Theory, Method, and the Study of European Union’, Oslo: ARENA Working paper 02/27. Karlsson, M. (1999) ‘EU och organisationsva¨sende’, in K.M. Johansson (ed.) Sverige i EU, Stockholm: SNS Fo¨rlag. Kassim, H., Peters, B.G. and v. Wright, P. (eds) (2000) The National Coordination of EU Policy. The Domestic Level, Oxford: Oxford University Press. Kelstrup, M. and Branner, H. (eds) (2000) Denmark’s Policy Towards Europe after 1945, Odense: Odense University Press. Keohane, R.O. and Hoffmann, S. (eds) (1991) The New European Community – Decision Making and Institutional Change, Boulder: Westview Press. Keohane, R.O. and Nye, J.S. (1973) Transnational Relations and World Politics, Cambridge, Mass.: Harvard University Press. Kjaer, P. and Pedersen, O.K. (2001) ‘Translating Liberalization: Neoliberalism in the Danish Negotiated Economy’, in J.L. Campbell and O.K. Pedersen (eds) The Rise of Neoliberalism and Institutional Analysis, Princeton: Princeton University Press. Knill, C. (2001) The Europeanisation of National Administrations. Patterns of Institutional Change and Persistence, Cambridge: Cambridge University Press. Knill, C. and Lenschow, A. (1998a) ‘European Policies: The Impact of National Administrative Traditions’, Journal of Public Policy, 18(19): 1–28. —— (1998b) ‘Coping with Europe: The Implementation of EU Environmental Policy and Administrative Traditions in Britain and Germany’, Journal of European Public Policy, 5 (4): 595–614. Knudsen, T. (2003) ‘Fra politicus til politicus? Forvaltningsinternationalisering i et historisk perspektiv’, in M. Marcussen and K. Ronit (eds) Internationaliseringen af ˚ rhus: A ˚ rhus Universitetsforlag. den offentlige forvaltning i Danmark, A Kohler-Koch, B. (1996) ‘The Strength of Weakness. The Transformation of Governance in the EU’, in S. Gustavsson and L. Lewin (eds) Essays on Cultural Pluralism and Political Integration, Stockholm: Nerenius and Santerus. Kohler-Koch, B. and Eising, R. (eds) (1999) The Transformation of Governance in the European Union, London: Routledge. Kooiman, J. (ed.) (1993) Modern Governance. New Government-Society Interactions, London: Sage. Krasner, S.D. (1984) ‘Sovereignty: An Institutional Perspective’, Comparative Political Studies, 21: 66–94. Kuhnle, S. (1991) ‘Den skandinaviske velferdsmodellen – Skandinavisk? Velferd? Modell?’, in A.R. Hovden (et al.) Visjoner om velferdssamfunnet, Bergen: Alma Mater. Laffan, B. (1996) ‘The Politics of Identity and Political Order in Europe’, Journal of Common Market Studies, 34 (1): 81–102. Lampinen, R. and Uusikyla¨, P. (1998) ‘Implementation Deficit. Why Member States do not Comply with EU Directives?’, Scandinavian Political Studies, 21 (3): 231–252.
176
Bibliography
Landfried, C. (1999) ‘The European Regulation of Biotechnology by Polycratic Governance’, in C. Joerges and E. Vos: EU Committees: Social Regulation, Law and Politics, Oxford: Hart Publishing. Larsen, G. (2001) ‘Nordiske og internasjonale relationer’, in B. Jacobsson, P. Lægreid and O.K. Pedersen (eds.) Europaveje. EU og de centrale forvaltninger i Norden,Copenhagen: Jurist- og Økonomforbundets Forlag. Larsen, G., Lægreid, P. and Wik, M.H. (1999) ‘EU i Norge. Forvaltningens europatilpasning i Norge’, Bergen: LOS-senteret, Report 9908. Larsson, T. (1986) Regeringen och dess kansli: samordning och byra˚krati i maktens centrum, Lund: Studentlitteratur. Latour, B. (1986) ‘The Powers of Association’, in J. Law (ed.) Power, Actions and Belief, London: Routledge and Kegan Paul. Lenschow, A. (1999) ‘Transformation in European Environmental Governance’, in B. Kohler-Koch and R. Eising (eds) The Transformation of Governance in the European Union, London: Routledge. Lewin, L. (1992) Samha¨llet och de organiserade interessena, Stockholm: Norstedts Juridik. Lindberg, L.N. and Scheingold, S.A. (1970) Europe’s would-be Polity: Patterns of Change in the European Community, Englewood Cliffs: Prentice-Hall. Lindbom, A. (1997) ‘Ministern och makten. Hur fungerer ministerstyret i praktiken?’, SOU 1997: 54, Stockholm: Fritzes. Lodge, J. (1994) ‘Transparency and Democratic Legitimacy’, Journal of Common Market Studies, 32 (3): 342–68. Lord, Chr. (2001) ‘Assessing Democracy in a Contested Polity’, Journal of Common Market Studies, 39 (2): 641–663. Lowi, T.J. (1979) The End of Liberalism. The Second Republic of the United States, New York: W.W. Norton & Co. Lægreid, P. (2000) ‘Europeisering og forvaltningstilpasning i Norden’, in K. Sta˚hlberg (ed.) Kontinuitet och fornyelse. Europeisk integration och nordisk forvaltningstilpasning, Nord 2000: 3, Copenhagen: Nordic Council of Ministers. —— (2001a) ‘Invevd og segmentert’, in B. Jacobsson, P. Lægreid and O.K. Pedersen (eds) Europaveje. EU og de centrale forvaltninger i Norden, Copenhagen: Juristog Økonomforbundets Forlag. —— (2001b) ‘Organisasjonsformer: Robusthet og fleksibilitet’, in B. Jacobsson, P. Lægreid and O.K. Pedersen (eds) Europaveje. EU og de centrale forvaltninger i Norden, Copenhagen: Jurist- og Økonomforbundets Forlag. —— (2001c) ‘Administrative Reforms in Scandinavia – Testing the Cooperative Model’, in B.C. Noland (ed.): Public Sector Reform, London: Palgrave. Lægreid, P. and Olsen, J.P. (1984) ‘Top Civil Servants in Norway: Key Players on Different Teams’, in E.N. Suleiman (ed.) Bureaucrats and Policy Making, New York: Holmes & Meier. Lægreid, P. and Pedersen, O.K. (eds) (1994) Forvaltningspolitik i Norden, Copenhagen: Jurist- og Økonomforbundets Forlag. —— (1996) Integration og Decentralisering. Personale og Forvaltning i Skandinavien, Copenhagen: Jurist- og Økonomforbundets Forlag. —— (1999) Fra opbygning til ombygning i staten. Organisationsforandringer i tre nordiske lande, Copenhagen: Jurist- og Økonomforbundets Forlag. Majone, G. (1996) Regulating Europe, London: Routledge.
Bibliography 177 March, J.G. and Olsen, J.P. (1989) Rediscovering Institutions. The Organizational Basis of Politics, New York: The Free Press. Marcussen, M. (2000) Ideas and Elites. The Social Construction of Economic and ˚ lborg: A ˚ lborg University Press. Monetary Union, A —— (2002) OECD og idespillet. Game over?, Copenhagen: Hans Reitzels Forlag. Marcussen, M. and Ronit, K. (eds) (2003) Internationaliseringen af den offentlige forvaltning i Danmark. Forandring og kontinuitet, Aarhus: Aarhus Universitetsforlag. Marks, G., Hooghe, L. and Blank, K. (1996) ‘European Integration from the 1980s: State-Centric v Multi-Level Governance’, Journal of Common Market Studies, 34 (3): 341–378. Me´ny, Y., Muller, P. and Quermonne, J-L. (eds) (1996) Adjusting to Europe. The Impact of the European Union on National Institutions and Policies, London: Routledge. Metcalfe, L. (1994) ‘International Policy Co-ordination and Public Management Reform’, International Review of Administrative Sciences, 60: 271–290. Meyer, J.W. and Rowan, B. (1977) ‘Institutionalized Organizations: Formal Structure as Myth and Seremony’, American Journal of Sociology, 83: 340–363. Micheletti, M. (1994) Det civila samha¨llat och staten: Medborgarsammanslutningarnas roll i svensk politik, Stockholm: Publica. Middelthon, H. (2000) ‘Europeisering av arbeidstilsynet, velvillig eller motvilling tilpasning? En studie av de norske og svenske arbeidstilsynet 1992–97’, Bergen: LOS-senteret, Report 004. Miles, L. (ed.) (1996) The European Union and the Nordic Countries, London: Routledge. Milward, A. (1992) The European Rescue of the Nation-State, Berkeley: University of California Press. Moen, K. (1998) ‘Fra monopol til konkurranse. EØS, norsk legemiddelpolitikk og norsk medisinaldepot’, Oslo: ARENA, Report 1/98. Moravcsik, A. (1993) ‘Preferences and Power in the European Community: A Liberal Intergovernmental Approach’, Journal of Common Market Studies, 31 (4): 473–524. —— (1995) ‘Liberal Intergovernmentalism and Integration: A Rejoinder’, Journal of Common Market Studies, 33 (4): 611–628. —— (1997) ‘Taking Preferences Seriously: A Liberal Theory of International Politics’, International Organization, 51 (4): 513–553. —— (1998) The Choice for Europe. Social Purpose and State Power from Messina to Maastricht, London: UCL Press. Morgan, G. (1986) Images of Organization, Thousand Oaks, CA: Sage. Mo¨rth, U. (1996) Vardagsintegration – La vie quotidienne – i Europa. Sverige i Eureka och Eureka i Sverige, Department of Political Science, Stockholms Universitet. Nedergaard, P. (2000) Organiseringen af Den europæiske Union, Copenhagen: Handelshøjskolens Forlag. Neumann, I.B. (1996) ‘Collective Identity Formation: Self and Other in International Relations’, European Journal of International Relations, 2: 139–174. Neyer, J. (1999) ‘The Comitology Challenge to Analytical Integration Theory’, in C.J. Joerges and E. Vos: EU Committees: Social Regulation, Law and Politics, Oxford: Hart Publishing. Nilsen, E. (2001) ‘Europeisering og diplomati. En studie av de norske ambassadene i København og Stockholm’, Bergen: LOS-senteret, Report 0102.
178
Bibliography
Nordby, T. (2000) I politikkens sentrum. Variasjoner i Stortingets makt 1814–2000, Oslo: Universitetsforlaget. Norton, P. (1996) National Parliaments and the European Union, London: Frank Cass. Obradovic, D. (1996) ‘Policy Legitimacy and the European Union’, Journal of Common Market Studies, 34 (2): 191–221. Offe, C. (2001) ‘Institutional Design’, in P.B. Clarke and J. Foweraker (eds) Encyclopedia of Democratic Thought, London: Routledge. Ohmae, K. (1990) The Borderless World. Power and Strategy in the Interlinked Economy. Management Lessons in the New Logic of the Global Marketplace, New York: Harper Collins. —— (1995) The Pluralist State, London: Macmillan. Oliver, C. (1991) ‘Strategic Responses to Institutional Processes’, Academy of Management Review, 16: 145–179. Olsen, J.P. (1992) ‘Analyzing Institutional Dynamics’, Staatswissenschaft und Staatspraxis, 3 (2): 247–271. —— (1996) ‘Norway: Slow Learner or Another Triumph of the Tortoise?’, in J.P. Olsen and B.G. Peters (eds) Lessons from Experience, Oslo: Scandinavian University Press. —— (1997) ‘Institutional Design in Democratic Contexts’, The Journal of Political Philosophy, 5 (3): 203–229. —— (2001) ‘Omfordeling av makt i Europa’, Nytt Norsk Tidsskrift, 18 (1): 23–38. —— (2002a) ‘The Many Faces of Europeanization’, Journal of Common Market Studies, 40 (5): 921–952. —— (2002b) ‘Towards a European Administrative Space?’, Oslo: ARENA, Working Paper 02/26. —— (2002c) ‘Reforming European Institutions of Governance’, Journal of Common Market Studies, 40 (4): 581–602. Olsen, J.P. and Peters, B. Guy (eds) (1996) Lessons from Experience. Experimental Learning in Administrative Reforms in Eight Democracies, Oslo: Scandinavian University Press. Olsen, J.P. and Sverdrup, B.O. (1998) Europa i Norden. Europeisering av nordisk samarbeid, Oslo: Tano-Aschehoug. Page, E. and Wouters, L. (1995) ‘The Europeanization of National Bureaucracies?’, in J. Pierre (ed.) Bureaucracy in the Modern State, Aldershot: Edwars Elgar. Pedersen, O.K. (1995) ‘Nye demokratiske udfordringer’, in P. Lu¨bcke (ed.) Miljøet, markedet og velfærdsstaten, Copenhagen: Fremad. —— (1998) ‘Interesseorganisationerne og den parlamentariske styreform’ in K. Ronit (ed.) Interesseorganisationer i dansk politik, Copenhagen: Jurist- og Økonomforbundets Forlag. —— (1999) ‘Den danske stat og Grundloven’, i H. Zahle (ed.) Kommenteret udgave af Grundloven, København: Jurist- og Økonomforbundets Forlag. —— (2001) ‘Interesseorganisationer og den europæiske integration’, in B. Jacobsson, P. Lægreid and O.K. Pedersen (eds) Europaveje. EU i de nordiske centralforvaltninger, Copenhagen: Jurist- og Økonomforbundets Forlag. —— (ed.) (2002) EU i forvaltningen. Broen fra Slotsholmen til Bruxelles, København: Jurist- og Økonomforbundets Forlag. Pedersen, T. (2000) ‘Denmark’, in H. Kassim, B.G. Peters and V. Wright (eds) The National Co-ordination of EU Policy, Oxford: Oxford University Press.
Bibliography 179 Peters, B. Guy (1988) Comparing Public Bureaucracies: Problems of Theory and Method, Tuscaloosa: University of Alabama Press. —— (1995) The Politics of Bureaucracy, New York: Longmann. —— (1996) ‘Theory and Methodology’, in H.A.G.M. Bekke, J.L. Perry and Th. A.J. Toonen (eds) Civil Service Systems in Comparative Perspective, Indiana: Indiana University Press. Petersen N. (1993) ‘Game, Set, and Match: Denmark and the European Union after Edingburgh’, in T. Tillikainen and I.D. Petersen (eds) The Nordic Countries and the EC, Copenhagen: Political Studies Press. Peterson, J. (1995a) ‘Decision Making in the European Union: Towards a Framework for Analysis’, Journal of European Public Policy, 2 (3): 69–93. —— (1995b) ‘Policy Networks and EU Policy Making’, West European Politics, 18 (2): 389–407. Peterson, J. and Blomberg, E. (1999) Decision Making in the European Union, Basingstoke. Petersson, O. (1994) The Government and Politics of the Nordic Countries, Stockholm: Publica. —— (1998) Nordisk politik, Stockholm: Nordstedts Juridik. Pierre, J. (ed.) (1995) Bureaucracy in the Modern State. An Introduction to Comparative Public Administration, Aldershot: Edward Elgar. Pierson, P. (1996) ‘The Path to European Integration: An Historical Institutionalist Perspective’, Comparative Political Studies, 29: 95–146. Pollitt, T.C., Hanney, S., Packwood, T., Rothwell S. and Roberts S. (1997) Trajectories and Options: An International Perspective on the Implementation of Finnish Public Management Reforms, Helsinki: Ministry of Finance. Pollitt, T.C. and Bockaert, G. (2000) Public Management Reform. A Comparative Analysis, Oxford: Oxford University Press. Premfors, R. (1998) ‘Reshaping the Democratic State: Swedish Experience in a Comparative Perspective’, Public Administration 76 (2): 141–160. —— (1999) ‘Organisationsfo¨ra¨ndringar och fo¨rvaltningspolitik’, in P. Lægreid and O.K. Pedersen (eds) Fra opbygning til ombygning i staten, København: Jurist- og Økonomforbundets Forlag. Przeworski, A. and Teune, H. (1970) The Logic of Comparative Social Inquiry, New York: John Wiley. Putnam, R. (1988) ‘Diplomacy and Domestic Politics: The Logic of Twolevel Games’, International Organization, 42: 427–460. Raunio, T. and Wiberg, M. (2000a) ‘Strengthened Parliamentary Accountability: The Impact of EU on Finnish Politics’, Journal of Legislative Studies, 6 (1): 59–80. —— (2000b) ‘Building Elite Consensus: Parliamentary Accountability in Finland’, in T. Bergman and E. Damgaard (eds) Delegation and Accountability in European Integration, London: Frank Cass. —— (2000c) ‘Parliaments’ Adaptation to the European Union’, in P. Esaisson and K. Heidar (eds) Beyond Westminster and Congress. The Nordic Experience, Columbus: Ohio State University Press. Risse-Kappen, T. (1994) ‘Ideas do not Float Freely: Transnational coalitions, Domestic Structures, and the End of the Cold War’, International Organization, 48: 185–214.
180
Bibliography
—— (1995) (ed.) Bringing Transnational Relations Back in: Non-State Actors, Domestic Structures and International Institutions, Cambridge: Cambridge University Press. —— (1996) ‘Exploring the Nature of the Beast: International Relations Theory and Comparative Policy Analysis Meet the European Union’, Journal of Common Market Studies, 34 (1): 53–80. Rometsch, D. and Wessels, W. (1996) The EU and Member States. Towards Institutional Fusion?, Manchester: Manchester University Press. Roness, P. G. (1997) Organisasjonsendringar. Bergen: Fagbokforlaget. Rosamond, B. (2000) Theories of European Integration, Basingstoke: Macmillan. Rothstein, B. (1992) Den korporativa staten: Interesseorganisationer och statsfo¨rvaltning i svensk politik, Stockholm: Norstedts. Rothstein, B. and Bergstro¨m, J. (1999) Korporatismens fall och den svenska modellens kris, Stockholm: SNS Fo¨rlag. Rottenburg, R. (1996) ‘When Organization Travels: On Intercultural Translation’, in B. Czarniawska and G. Sevo´n (eds) Translating Organizational Change, Berlin: De Gruyter. Røvik, K-A. (1998) Moderne organisasjoner. Trender i organisasjonstenkningen ved tusena˚rsskifte, Bergen: Fagbokforlaget. Sahlin-Andersson, K. (1996) ‘Imitating by Editing Success: The Construction of Organization Fields’, in B. Czarniawska and G. Sevo´n (eds) Translating Organizational Change, Berlin: de Gruyter. Sandholtz, W. (1996) ‘Membership Matters: Limits of the Functional Approach to European Institutions’, Journal of Common Market Studies, 34 (3): 403–429. Scharpf, F.W. (1994) ‘Community and Autonomy: Multi-level Policy-making in the European Union’, Journal of European Public Policy, 1: 219–242. —— (1997) Games Real Actors Play. Actor-Centered Institutionalism in Policy Research, Boulder: Westview Press. —— (1999) Governing in Europe. Effective and Democratic?, Oxford: Oxford University Press. Schmidt, V.A. (1996) From State to Market? The Transformation of French Business and Government, Cambridge: Cambridge University Press. Schneider, G. and Aspinwall, M. (eds) (2001) The Rules of Integration. Institutionalist Approaches to the Study of Europe, Manchester: Manchester University Press. Schneider, V. (2001) ‘Institutional Reform in Telecommunications: The European Union in Transnational Policy Diffusion’, in M.G. Cowles, J.A. Caporaso and T. Risse (eds) Transforming Europe. Europeanization and Domestic Change, Ithaca: Cornell University Press. Selovuori, J. (1999) Makt och byra˚krati i Finland, Helsingfors: Edita. Selznick, P. (1957) Leadership in Administration, New York: Harper & Row. Sidenius, N.C. (1998) ‘Interesseorganisationer og Europa. En status’, in K. Ronit (ed.) Interesseorganisationer i dansk politikk, Copenhagen: Jurist- og Økonomforbundets Forlag. —— (1999) ‘Business, Government Structure and the EU: The Case of Denmark’, in B. Kohler-Koch and R. Eising (eds) The Transformation of Governance in the European Union, London: Routledge.
Bibliography 181 Sidenius, N.C., Einersen, B. and So¨rensen, J.A (1997) ‘The European Affairs Committee and Danish European Union Politics’, in M. Wiberg (ed.) Trying to Make Democracy Work – The Nordic Parliaments and the European Union, Stockholm: Gidlunds/Riksbankens Jubileumsfond. Siedentopf, H. and Ziller, J. (1988) Making European Politics Work. The Implementation of Community Legislation in the Member States, London: Sage. Smith, E. (1996) National Parliaments as Cornerstones of European Integration, London: Kluwer Law International. Spanou, C. (1998) ‘European Integration in Administrative Terms: A Framework for Analysis and the Greek Case’, Journal of European Public Policy, 5 (3): 467–484. Statskontoret (2000) ‘Den svenska fo¨rvaltningsmodellen i EU-arbetet’, Stockholm: Statskontoret, Report 2000: 20 A. Stinchcombe, A.L. (2001) When Formality Works. Authority and Abstraction in Law and Organizations, Thousand Oaks, CA: Sage. Stone Sweet, A., Sandholtz, W. and Fligstein, N. (eds) (2001) The Institutionalization of Europe, Oxford: Oxford University Press. Streeck, W. and Schmitter, P.C. (1991) ‘From National Corporatism to Transnational Pluralism: Organized Interests in the Single European Market’, Politics and Society, 19 (2): 133–164. Sundstro¨m, G. (1999) ‘Att tala med en ro¨st – En studie av hur EU-medlemskapet pa˚verkar ordningen inom regjeringskansliet’, Stockholm: SCORE, Working Paper 1999: 8. —— (2001) ‘Relationer innom staten’, in B. Jacobsson, P. Lægreid and O.K. Pedersen (eds) Europaveje. EU og de centrale forvaltninger i Norden, Copenhagen: Juristog Økonomiforbundet Forlag. Sverdrup, U. (1998) ‘Norway: An Adaptive Non-Member’, in K. Hanf and B. Soetendorp (eds) Adapting to European Integration, London: Longman. —— (2000) ‘Ambiguity and Adaptation. Europeanization of Administrative Institutions as Loosely Coupled Processes’, Oslo: ARENA, Report 8/2000. Sætereng, T.R. (2001) ‘Ett felles nasjonalt standpunkt? En studie av rammenotatsord¨ S-saker i norsk sentralforvaltning’, ningens effekter pa˚ koordineringen av EU/EO Bergen: LOS-senteret, report 0104. Tallberg, J. (2000) ‘Supranational Influence in EU Enforcement: the ECJ and the Principle of State Liability’, Journal of European Public Policy, 7 (1): 104–121. Toonen, Th. A.J. (1992) ‘Europe of the Administrations. The Challenges of 1992 (and Beyond)’, Public Administration Review, 52 (2): 108–15. Trondal, J. (1996) ‘Tilknytningsformer til EU og nasjonale samordningsprocesser. En studie av norske og danske departementer’, Oslo: ARENA Working Paper, no. 15. —— (1999) ‘Europeisering av sentraladministrative organer: Om tilknytningsfomer til EU og departementale koordineringsformer’, Norsk Statsvitenskaplig Tidsskrift, 15 (1): 40–74. —— (2000a) ‘Mellom politisk styring og faglig uavhengighet. En studie av norske direktoraters europeisering’, Nordiske Organisasjonsstudier, 1(2): 31–58. —— (2000b) ‘Multiple Institutional Embeddedness in Europe. The Case of Danish, Norwegian and Swedish Officials’, Scandinavian Political Studies, 23 (4): 311–341.
182
Bibliography
—— (2001a) ‘Beyond the EU membership – Non-Membership Dichotomy’, Oslo: ARENA, working paper 0115. —— (2001b) ‘Administrative Integration Across Levels of Governance – Integration through Participation in EU-Committees’, Oslo: ARENA Report 01/07. —— (2001c) ‘Is there any Social Constructivist – Institutionalist Divide? Unpacking Social Mechanism Affecting Representational Roles Among EU Decisionmakers’, Journal of European Public Policy, 8 (1): 1–23. Trondal, J. and Veggeland, F. (1999) ‘Norske myndigheter og EØS, mellom utenrikspolitikk og innenrikspolitikk’, Sosiologi i dag, 29 (4): 57–78. Vad, T. (1998) ‘Europeanisation of Standardisation – European Institution Building and National Persistence in the Area of Technical Standardisation’, Ph.D-dissertation, Copenhagen: Institute for Political Science, University of Copenhagen. Veggeland, F. (1999a) ‘Delegering, læring og politisk kontroll i EØS-arbeidet: Norsk deltakelse i EU-komiteer pa˚ veterinær- og næringsmiddelomra˚det’, Internasjonal Politikk, 58 (1): 81–112. —— (1999b) ‘Institusjonelle tilpasninger til europeisk integrasjon. En gjennomgang av forskning om europeisering av nasjonale institusjoner’, Norsk Statsvitenskapelig Tidsskrift, 15 (1): 3–39. Von Dosenrode, S.Z. (1998) ‘Denmark: The Testing of a Hesitant Membership’, in K. Hanf and B. Soetendorp (eds) Adapting to European Integration, London: Longman. Von Sydow, E. (1999) Na¨r Luther kom till Bryssel. Sveriges fo¨rsta a˚r i EU, Stockholm: Bokfo¨rlaget Arena. Weick, K.E. (1976) The Social Psychology of Organizing, Reading, MA: AddisonWesley. Weiler, J.H.H. (1996) ‘Legitimacy and Democracy of Union Governance: The 1996 Intergovernmental Agenda and Beyond’, Oslo: ARENA Working Paper, no. 22. —— (1999) ‘The Constitution of the Common Market Place: Text and Context in the Evolution of the Free Movement of Goods’, in P. Craig and G.D. Bu´rca (eds) The Evolution of EU Law, Oxford: Oxford University Press. Wendt, A. (1992) ‘Anarchy is what the State Makes of it: the Social Construction of Power Politics’, International Organization, 46 (2): 391–425. Wessels, W. (1999) ‘Comitology as a Research Subject: A New Legitimacy Mix?’, in C. Joerges and E. Vos: EU Committees: Social Regulation, Law and Politics, Oxford: Hart Publishing. Wessels, W. and Rometsch, D. (1996) ‘German Administrative Interaction and the European Union: the Fusion of Public Policies’, in Y. Me`ny, P. Muller and J.-L. Quermonne (eds) Adjusting to Europe. The Impact of the European Union on National Institutions and Policies, London: Routledge. Wincott, D. (1995) ‘Institutional Interaction and European Integration: Towards an Everyday Critique of Liberal Intergovernmentalism’, Journal of Common Market Studies, 33 (4): 597–609. Wivel, A. (2000) The Integration Spiral: International Security and European Integration 1945–1999, Copenhagen: Department of Statskundskab, Copenhagens University. Woods, N. (ed.) (2000) The Political Economy of Globalization, Basingstoke: Macmillan.
Bibliography 183 Wright, V. (1996) The National Co-ordination of European Policy-making, London: Routledge. Zervakis, P. and Johansson, K.M. (ed.) (2002) European Political Parties between Cooperation and Integration, Baden-Baden: Nomos Verlag. Ziller, J. (1998) ‘EU Integration and Civil Service Reform’, Sigma Papers no 23, 1998a: 136–154. Paris: OECD (CCNM/SIGMA/PUMA(98)39). Øhrgaard, J.C. (1997) ‘Less than Supranational, more than Intergovernmental: European Political Cooperation and the Dynamics of Intergovernmental Integration’, Millenium, 26 (1): 1–29.
Name index
Algieri, F. 154 Allison, G.T. 106 Anckar, D. 4 Andersen, S.S. 98, 101, 168 Armstrong, K.168 Aspinwall, M. 152–3 Baldersheim, H. 76 Bekke, H.A.G.M. 168 Benz, A. 120, 152 Bergman, T. 77 Bergstro¨m, J. 65, 97 Blank, K. 120, 152 Blomberg, E. 154 Blumann, C.155 Boekaert, G. 75 Boli, J. 8 Bo¨rzel, T. 168 Branner, H.168 Brunsson, N. 18, 20, 30, 103, 120 Bruun, N. 74 Bryman, A. 5 Bræin, S. 101 Bue, L.A. 100 Buitendijk, G.J. 154 Bull, H. 167 Bulmer, S. 3, 87, 98, 11, 152, 168 Burch, M. 3, 87, 98, 111, 168 Camillieri, J.A. 6 Campbell, J.L. 106, 115, 153, 155, 160, 163 Caporaso, J.A. 3, 20, 118, 152–4, 168 Checkel, J.T. 153, 168 Christensen, J.G. 73 Christensen, T. 19, 75, 99, 111, 115–17, 168
Christiansen, P.M. 65 Cohen, M.D. 120 Cowles, M.G. 3, 20, 168 Cyert, R.M. 98 Czarniawska, B. 19, 122 Damgaard, E. 77, 90 Derlien, H-U. 4, 168 Deutsch, K.W. 74 DiMaggio, P.J. 12 Dowding, K. 153 Dyrdal, M. 100 Egeberg, M. 37, 87, 98, 111, 127, 152, 156, 168 Eising, R. 2, 98, 100, 168 Eitrheim, P. 76 Ekengren M. 41 Eliassen, K. 168 Eriksen, E.O. 168 Esmark, A. 28, 35, 37–9, 47, 102, 120, 139–40, 144, 150–1, 155 Esping-Andersson, G. 74 Etzioni, A. 74 Falke, J. J. 6, 154 Ferlie E. 76 Fjær, S. 127 Flynn, N. 168 Fossum, J.E. 168 Frankel, C. 30, 32 Franklin, M. 130 Fudge, C. 76 Gehring, Th.155 Giddens, A. 163 Goetz, K. 3, 86, 101, 115, 118, 168
186
Name index
Goodin, R.E. 103 Greve, B. 74 Grieco, J. 151 Grønbeck-Jensen, C. 84 Gustavsson, L. 76 Haas, E. 119, 152 Hanf, K. 3, 87, 118 Hansen, L.167 Harmsen, R. 87, 118, 168 Hayward, J. 167 Hegeland, H. 92 Held, D. 149 He´ritier, A. 20, 168 Hermansson, J. 65 Hix, S. 3, 86, 101 Hoffmann, S. 153 Højbjerg, E. 30, 37, 52, 90, 102, 140, 144, 155 Hood, C. 168 Hooghe, L. 120, 152 Jachtenfuchs, M. 120, 152 Jacobsson, B. 17–18, 20, 22, 28, 30, 41, 60, 63, 75, 94, 110, 116, 120, 128, 146, 158, 168 Jeppestøl, G. 108 Jessop, B. 6 Joerges, C. 120, 153, 154, 168 Johansson, K.M. 152 Josselin, D. 153 Jupille, J. 118, 152–4 Karlsson, M. 92 Kassim, H. 57, 60, 168 Kehoane, R.O. 153, 167 Kelstrup, M. 168 Kjaer, P. 19 Knill, C. 3, 4, 20, 100, 118, 121, 168 Knudsen, T. 10 Kohler-Koch, B. 2, 98, 100, 120, 152, 168 Kooiman, J. 168 Krasner, S.D. 15 Kuhnle, S. 74, 76 Lægreid, P. 16–17, 19, 22, 28, 31, 37, 46, 53, 73–7, 88–9, 98, 113, 115–17, 128, 157, 168 Laffan, B. 167 Lampinen, R. 12 Landfried, C. 155
Larsen, G. 22, 66–7, 84 Larsson, T. 57 Latour, B. 122 Lenschow, A. 100, 168 Lewin, L. 65 Lindberg, L.N. 152 Lindbom, A. 17, 75 Lodge, J. 167 Lord, C. 167 Lowi, T.J. 144 Majone, G. 86 March, J.G. 15, 98, 120 Marcussen, M. 73, 154, 168 Marks, G. 120, 152 Marsh, M. 130 Mclaren, L. 130 Me´ny, Y. 3, 86, 168 Metcalfe, L. 168 Meyer, J.W. 20, 103 Micheletti, M. 65 Middelthon, H. 30, 100 Miles, L. 15, 82 Milward, A. 2, 119 Mingers, S. 168 Moen, K. 101. Moravcsik, A. 2, 14, 115, 119, 151–2 Morgan, G. 106 Mo¨rth. U, 2 Muller, P. 3, 168 Nedregaard, P. 35, 156, 168 Neumann, I.B. 168 Nilsen, E. 99 Nøgaard, A.S. 90 Nordby, T. 112 Norton, P. 113 Nye, J.S. 167 Obradovic, D. 167 Offe, C. 103 Ohmae, K. 2, 6 Øhrgaard, J.C. 151 Oliver, C. 20 Olsen, J.P. 3, 13, 15–16, 73–5, 89, 99, 100, 103, 107, 112, 115, 118, 120, 153, 155, 168 Page, E. 76, 86, 168 Pedersen, D. 30, 90 Pedersen, O.K. 16–17, 19, 30, 38, 43, 54, 61, 65, 68, 74–5, 88, 97, 102, 106,
Name index 113, 115, 128 –9, 131, 138–9, 144, 153, 155, 157–8, 160, 163–4, 167–8 Pedersen, T. 82 Perry, J.L. 168 Peters, B.G 57, 60, 168 Petersen, N. 130 Peterson, J. 153–4 Petersson, O. 17, 74, 79 Pierre, J. 168 Pierson, P. 15 Pollitt, T.C. 75 Powell, W.W 12 Premfors, R. 16, 75 Przeworski, A. 4 Putnam, R. 119 Quermonne, J.L. 3, 168 Raunio, T. 83, 91, 128, 146 Risse-Kappen, T. 3, 8–9, 20, 153, 157, 162, 168 Rometsch, D. 3, 98, 109, 113, 154, 155, 168 Roness, P.G. 97 Ronit, K. 73, 168 Rosamond, B. 150 Rothstein, B. 65, 97 Rottenburg, R. 123–4 Røvik, K.A. 19, 103, 122 Rowan, B. 20 Sætereng, T.R. 61, 112 Sandholtz, W. 168 Sahlin-Andersson, K. 19, 116, 123 Scharpf, F.W. 2, 119–20, 139, 152–3 Scheingold, S.A. 152 Schepel, H. 168 Schmidt, V.A. 168 Schmitter, P.C. 152 Schneider, G. 153, 155 Selovouri, J. 40, 114 Selznick, P. 15 Sevo`n, G. 19, 122 Sidenius, N.C. 65, 92 Siedentopf, H. 14, 82
187
Smith, E. 168 Soetendorf, B. 3, 87, 118 Spanou, C. 118 Stinchcombe, A.L. 103 Strehl, F. 168 Stone Sweet, A. 153 Streeck, W. 152 Sundstro¨m, G. 22, 28, 39, 57, 61, 168 Sverdrup, U. 12, 74, 86, 108–9, 119, 122, 168 Tallberg, J. 12 Teune, H. 4 Thomas, G.M. 8 Toonen, T.A.J. 168 Trondal, J. 37, 56, 70, 82, 87, 108–9, 156, 168 Uusikyla¨, P. 12 Vad, T. 30 Veggeland, F. 56, 128, 168 Van Schendelen, M.C.P.M. 154 Von Dosenrode, S.Z. 82 Vos, E. 168 Von Sydow, E. 83 Wallace, W. 153 Weick, K.E. 20, 120 Weiler, J.H.H.154, 167–8 Wendt, A. 167 Wessels, W. 3, 98, 109, 113, 120, 153–6, 168 Wiberg, M. 83, 91, 128, 146 Wik, M. H. 22 Williams, M.C. 167 Wincott, D.153 Wivel, A. 151 Woods, N. 149 Wouters, L. 76, 86, 168 Wright, P. 57, 60, 168 Zervakis, P. 152 Ziller, J. 14, 82, 118
Subject index
adaptation, 5, 12, 19, 56–7, 70, 73, 78–9, 81–2, 84, 86, 89, 98, 109, 112, 117–19, 122, 135, 141, 145, 157, 159 adjustment 119 administration-oriented approach 155 administrative differentiation 102 administrative levels 10, 41, 93 administrative models 16–17, 73, 93, 97, 101–2, 106, 110–11 administrative policy 88–9, 101, 111–12, 115 administrative tradition 99, 111 affectedness 28 agricultural and fisheries administration 23, 52, 55 anticipation, anticipated reaction 40, 100, 118 area logic 109 attitude to the EU 14 autonomous reaction 7, 100, 118 bureaucratization 139–40 bypass strategy 69, 77, 144 captivity 144, 161 causal approach 150, 159 central agencies 53, 70, 110 challenge to democratic government 144 coevolution 9, 107, 121 160, 162–3 collegial arrangements 61, 89 comparative approach, design 3, 24 comparative statistics 121 commitology committees 35, 47, 108 conflict 78, 129 contacts with EU organization 34 convergence 86, 98, 118
coordination 32, 40–1, 46, 49, 57–61, 68–70, 77–8, 80–1, 89, 94, 101, 109, 113, 125, 141–2 administrative 142 horizontal 58–60, 70 internal, 62–3 national 58–60, 64, 68 political 142 vertical 58–60, 70, 109 coordination committee 90–1 coordination segment 61, 63–4, 68 Coreper 37 core segment 50–2, 57, 62–3, 65, 68–70, 81 cultural collision 81 Danish coordination system 97, 100 de-corporatization 65, 79 de-coupling 20, 25, 125 defence 23, 52, 56 democratic challenge 128, 140 democratic consequences 130 democratic ideals 131–2 Denmark 4, 82, 90, 108, 112 diffusion 122 divergence 117 dualism 75, 86, 96 east Nordic model 16–17, 74–5, 83, 93, 95, 105, 110 energy sector 52, 55 environmental-determinist perspective 105 environmental policy 63, 121 environmental sector 52, 55 EU Administration 50–1, 57, 65, 70 EU Committee 90
190
Subject index
EU Court of Justice 28, 47 EU Parliament 28, 47, 155 EU sections 91 European Affairs Committee 90, 112 European Council 13, 27–8, 33, 36–7, 47, 78, 92, 108–9, 134 European Commission 6, 13, 27–8, 34, 34–8, 47, 77–8, 109, 133–6, 155 European Economic Area (EEA) agreement 4, 13, 24, 29, 31, 37, 79–80, 108–9, 127 European integration 73, 118–19, 131, 150, 160 Europeanization 2–3, 7, 17, 76, 82, 85, 102, 106, 118, 124–5 128, 130, 151 everyday integration 1, 133, 152–3 exclusion 143, 161 expertise, experts 47, 78, 85, 92, 99, 128, 146 expert committees 34, 37 external explanation 12 fatalistic notion 117 Finland 4, 83–4, 91 Finland’s strategy 114 foreign and security policy 29 foreign affairs 52, 55 formal contact 96 form of affiliation 13, 33, 108, 109 framework guideline arrangements 61, 90 functionalism 152 go-through strategy 69 guidelines, guidance 39–41, 47, 78 high politics 149, 153 historical institutional perspective 15, 105 historical legacy 110 historical links 15–17, 24 horizontal logic 118 idealistic notion, position 117, 152 institutional theory 155 implementation 121 incremental adjustment, incrementalism 98, 140–1, 145 independent judgement 39 influence 78 informal contact, relationship 96, 97 139–40
inner central administration 93 instructions 39–40 instrumental perspective 103 integrated approach 87 intergovernmentalism 151 intergovernmental cooperation 67, 69 intergovernmental organizations (IGOs) 18 interest organization 7, 42–6, 48, 62, 64–6, 69, 70, 77, 80, 92, 93, 97 internal market 29, 46, 55, 77, 80, 101 judicial and police cooperation 29 justice sector 52, 56 learning 118 legalistic approach 150, 158 legislation 32 legitimacy 131, 133, 146–7 length of membership 13, 108–9, 159 liberal intergovernmentalism 119, 151 local adjustment 98 low politics 120, 152–3 mandating procedure 90, 102, 140–1 ministerial rule, governance 93, 111 ministries 53, 70 Ministry of Finance 17, 91 Ministry of Foreign Affairs 52, 63, 56, 83, 91, 99, 136 monism, monistic model 75, 96 most similar system approach 4 multi-level-governance 120, 152 multiplier effect 11, 25, 49, 61–2, 64–6, 68, 101, 138 multiplier thesis 68 national experts 36, 47, 77 national interest 158 national strategies 14, 18, 24, 105, 111, 115 neo-functionalism 119, 152–3 neo-institutional theory 103, 156 network-oriented approach 153, 155 networks 2, 18, 62 neutrality 135 New Public Management (NPM) 75–6, 114 Non Governmental Organizations (NGO) 6–8 Nordic cooperation 68, 79 Nordic Council 67
Subject index Nordic model 73, 74, Nordic relations 66–8 Norway 4, 80–2, 90, 108, 111 open process 138 organizational approach 154 organizational forms 98–9, 102 outer central administration 93 parliament 15, 91–3, 112–13, 129, 139, 141, 143, 146 parliamentary approach 150, 161 parliamentary government 133 participatory activity 35 path dependent 15, 103, 162 pluralistic approach 117 political control 40–1, 100, 146 political leadership 41–2, 47–8, 63, 69–71, 77–8, 81, 84, 94–5, 100, 110, 114, 128 political mandate 47, 127–8 political segment 63–4, 68–9 politicians 41, 97 128–9 politization 142 policy areas 52 power 128–30, 161 Prime Minister’s Office 91–2, 113, 136 private companies 42, 45, 62, 65, 77, 92, 97 proactive activity 38 public commission, commissions 80 reactive activity 38 regulation 19–20, 31, 79, 123 reorganization 88 reparliamentarization 79 robustness 85–6, 98–9, 101, 130 robustness and flexibility 73, 85–6, 100–1 103, 130 sector logic 109
191
sectors 10, 61 segmentation 9, 11, 49, 54, 56–7, 68, 70, 101, 109, 115, 129, 162 sovereignty 133 special committees 90–2, 97, 101, 135, 143 standardization 29–30 success rate 38 supranationalism 153 Sweden 4, 13–14, 83–4, 91, 92, 113 Swedish administrative model 94, 97, 100, 110–11 Swedish strategy 114 temporal factors 102–3 Three Pillars 29 trade and industry 52, 55 transformation 98 translation 11, 17, 20, 26, 119, 121–5, 163 translation perspective 17, 25, 105–7, 117, 119, 123, 160 transnational administration 5, 7, 8, 24, 62, 69, 127 transnational states 46, 116 139, 156, 163 transnationalization, transnational relationship 6–9, 27–8, 46, 49, 54–5, 57, 60–1, 68–71, 76, 78, 80–1, 84–5, 101, 109, 116–18, 120–1, 124, 138, 149, 151, 156, 158–9, 162–3 transparency 136–7 travel activity 37 vertical logic 118 welfare sectors 55 welfare states 4 west Nordic model 16–17, 75, 93–5, 101, 105, 110 work groups 47