Understanding environmental policy convergence the power of words, rules and money

1.3 Selection of policy items according to trade-related 2.1 Environmental policies: adoption rates % for forty setting for the discharge of chromium, copper, lead, zinc 3.2 Phases and m

Understanding Environmental Policy Convergence

Over recent decades national environmental policies have becomeincreasingly alike This book analyses the driving forces of this pro-cess of policy convergence, providing an in-depth empirical analysis

of the international forces at work It does so by investigating howfour countries – France, Hungary, Mexico and the Netherlands – haveshaped their domestic environmental policies in the context of inter-national institutions and relationships, while taking into account vari-ous domestic factors and national conditions Employing a qualitativeapproach, the authors seek to deepen understanding of the processesand mechanisms through which international forces such as legalharmonisation, institutionalised information flows and global tradedynamics affect domestic environmental policy change Together with

its companion volume Environmental Policy Convergence in Europe: The Impact of International Institutions and Trade (2008) this book provides a

‘showcase’ of mixed methodologies, combining quantitative and tative approaches in an innovative way

quali-      ¨     is Senior Lecturer at the Department of Political andSocial Sciences of the Freie Universit¨at Berlin and managing director

of the Environmental Policy Research Centre (FFU)

              is Professor of European Politics in the ment of Social Sciences at Osnabr ¨uck University, Germany, where shealso holds a Jean Monnet Chair and directs the Jean Monnet Centre ofExcellence on European Studies In addition, she teaches at the College

Depart-of Europe in Bruges

                is Assistant Professor in the Department

of Political Sciences of the Environment, Institute for ManagementResearch, at Radboud University Nijmegen, the Netherlands

Understanding Environmental Policy Convergence

The Power of Words, Rules and Money

edited by

Helge J ¨orgens, Andrea Lenschow

and Duncan Liefferink

Published in the United States of America by Cambridge University Press, New York Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning and research at the highest international levels of excellence.

www.cambridge.org

Information on this title: www.cambridge.org/9781107037823

C

 Cambridge University Press 2014

This publication is in copyright Subject to statutory exception

and to the provisions of relevant collective licensing agreements,

no reproduction of any part may take place without the written

permission of Cambridge University Press.

First published 2014

Printed in the United Kingdom by CPI Group Ltd, Croydon CR0 4YY

A catalogue record for this publication is available from the British Library

Library of Congress Cataloguing in Publication data

Understanding environmental policy convergence : the power of words, rules and money / [edited by] Helge J ¨orgens, Andrea Lenschow, Duncan Liefferink pages cm.

Includes bibliographical references and index.

ISBN 978-1-107-03782-3 (hardback)

1 Environmental policy – International cooperation – Case studies.

2 Environmental management – International cooperation – Case studies.

3 Global environmental change – International cooperation – Case studies.

I J ¨orgens, Helge, 1967– II Lenschow, Andrea III Liefferink, Duncan GE170.U535 2013

363.7 ′ 0561 – dc23 2013022105

ISBN 978-1-107-03782-3 Hardback

Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

1 Introduction: theoretical framework and

               ,       ¨     

                

2 Is there convergence of national environmental

policies? An analysis of policy outputs in

4 Taming the ‘tiger in the tank’: explaining the

convergence of limit values for lead in petrol 104

7 Converging ideas about risk regulation? The

precautionary principle in national legal systems 209

                

v

vi Contents

8 From the outside in: explaining convergence in the

legal recognition of the sustainability principle 237

2.1 Environmental policies: early adoptions, 1970–2000 page 51

2.2 Environmental policies: linear adoptions, 1970–2000 522.3 Environmental policies: late adoptions, 1970–2000 53

4.1 Limit values in France, Hungary, Mexico and the

vii

1.3 Selection of policy items according to trade-related

2.1 Environmental policies: adoption rates (%) for forty

setting for the discharge of chromium, copper, lead, zinc

3.2 Phases and mechanisms of policy convergence in France 793.3 Phases and mechanisms of policy convergence in the

3.4 Phases and mechanisms of policy convergence in Hungary 913.5 Phases and mechanisms of policy convergence in Mexico 974.1 Selected international efforts aiming at the reduction of

4.2 Summary of relevant mechanisms in the Hungarian case 1234.3 Summary of relevant mechanisms in the Dutch case 1274.4 Summary of relevant mechanisms in the Mexican case 1304.5 Summary of relevant mechanisms in the French case 1345.1 National noise emission standards for lorries in dB(A) in

5.2 Legally binding international regulations and

viii

List of tables ix5.3 The main impacts and causal mechanisms affecting policy

5.4 The main impacts and causal mechanisms affecting policy

5.5 The main impacts and causal mechanisms affecting policy

5.6 The main impacts and causal mechanisms affecting policy

6.1 Review of the first governmental plan or legislation

specifically on contaminated sites up to 2000 in

6.2 Core dimensions in designing policy to clean up

6.3 Overview of main plans and legislation on contaminated

6.4 Driving forces for policy developments in the Netherlands 1886.5 Overview of legislation and policy plans on contaminated

7.2 The most important legally binding obligations and policy

recommendations on the precautionary principle 2147.3 The main mechanisms causing convergence/divergence in

9.1 Submechanisms of transnational communication

    -          is a postdoctoral researcher in the Department

of Economics and Social Sciences, University of Potsdam

                is a Research Fellow at the Chair of tive Public Policy and Administration, University of Konstanz

Compara-                        is Chair of International Relationsand Conflict Management, University of Konstanz

        ¨      is Senior Lecturer at the Department of Politicaland Social Sciences of the Freie Universit¨at Berlin and managing director

of the Environmental Policy Research Centre (FFU)

                    is Chair of Comparative Public Policyand Administration, University of Konstanz

                    is Professor of European Politics inthe Department of Social Sciences at Osnabr ¨uck University, Germany,where she also holds a Jean Monnet Chair and directs the Jean MonnetCentre of Excellence on European Studies In addition, she teaches atthe College of Europe in Bruges

                  is Assistant Professor in the Department

of Political Sciences of the Environment, Institute for ManagementResearch, at Radboud University Nijmegen

             is a Research Fellow on the ConsEnSus Project,Geography Department, Trinity College Dublin

                   is a Senior Lecturer in Regulation in theSchool of Law, Queen’s University Belfast

                is a Research Fellow in the Department ofPolitical Science, Stockholm University

x

Over the past decades, national environmental policies have becomeincreasingly alike This book analyses the causal mechanisms that drivethis process of policy convergence The book is the result of a collab-orative European research project conducted by Christoph Knill (coor-dinator, University of Konstanz), Bas Arts (University of Wageningen),Katharina Holzinger (University of Konstanz), Helge J ¨orgens (Free Uni-versity Berlin), Andrea Lenschow (University of Osnabr ¨uck) and DuncanLiefferink (University of Nijmegen) Research for this project was fundedunder the Fifth Framework Programme of the European Commissionwithin the RTD programme ‘Improving the human research potentialand the socio-economic knowledge base’, contract no HPSE-CT-2002–

00103 This financial support allowed us to build an excellent team ofsenior and junior researchers who either joined us for the entire researchproject or stayed only for part of the project: Johan Albrecht, Per-OlofBusch, Stephan Heichel, Jelmer Kamstra, Tobias Meier, Jeroen Ooije-vaar, Jessica Pape, Dieter Pesendorfer, Maren Riepe, Thomas Sommerer,Jale Tosun, Sietske Veenman and Natascha Warta

This book is the second of two volumes presenting the findings ofthe ENVIPOLCON project The first book presenting a quantitativeanalysis of environmental policy convergence in twenty-four countrieswas published in 2008 with Cambridge University Press.1 It providedanswers to the questions of how much, in which direction and for whatreasons environmental policies in the developed world had convergedover the past thirty years In this second volume we build on, but alsomove beyond, the initial quantitative analysis Through a set of in-depthqualitative case studies, we shed light on the precise mechanisms throughwhich countries adapt their domestic policies to those already in place

in other countries In particular, the book sheds light on the complexways in which words (transnational communication), rules (international

1 K Holzinger, C Knill and B Arts (eds.) 2008 Environmental Policy Convergence in

Europe: The Impact of International Institutions and Trade Cambridge University Press.

xii

Preface xiiiharmonisation) and money (economic regulatory competition) interact

in the adoption and development of domestic environmental policies

It does so by way of a highly systematic set of case studies, coveringthe convergence of seven environmental policy issues in four countries:France, Hungary, Mexico and the Netherlands

We thank the European Commission for its financial support We alsothank our universities and all other organisations that supported theproject directly or indirectly We are indebted to all members of theproject team who made this research project a success It was a greatpleasure working with you! Two anonymous reviewers at CambridgeUniversity Press provided thorough and very helpful criticism of an earlierversion of this book Finally, we would like to thank Lena Keller for herexcellent assistance in editing the final book manuscript, Pat Harper forher skilled and constructive copy-editing, as well as John Haslam, CarrieParkinson and Mary O’Hara at Cambridge University Press for guiding

us through the publication process

ADEME Agence de l’Environnement et de la Maˆıtrise de

l’EnergieAFSSE Agence Franc¸aise de S´ecurit´e Sanitaire

Environnementale

BRGM Bureau de Recherches G´eologiques et Mini`eres

CALM Network EU-funded network established to develop a

Community Noise Research Strategy PlanCARACAS Concerted Action on Risk Assessment for

Contaminated Sites in Europe

DurableCGPC Conseil G´en´eral des Ponts et Chauss´ees

CITES Convention on International Trade in Endangered

Species

CLARINET Contaminated Land Rehabilitation Network for

Environmental Technologies

xiv

List of abbreviations xv

CPP Comit´e de la Pr´evention et de la Pr´ecautionCRECEP Centre de Recherche, d’Expertise et de Contr ˆole

des Eaux de Paris

DG TREN Directorate-General for Transport and Energy

DRIRE Direction R´egionale de l’Industrie, de la

Recherche et de l’Environnement

Development

FJOKK Fodor Jozsef National Centre of Public Health

IAWPR International Association of Water Pollution

ResearchICPE Installations Class´ees pour la Protection de

l’EnvironnementI-INCE International Institute of Noise Control

Engineering

xvi List of abbreviations

INERIS Institute National de l’Environnement Industriel

et des RisquesINRETS French National Institute for Transport and

Safety Research

ISO International Organization for Standardization

KVVM K ¨ornyezetv´edelmi ´es V´ız ¨ugyi Miniszt´erium

(Ministry of Rural Development)

Gen´eticamente Modificados

Protection of the EnvironmentMEDD Minist`ere de l’ ´Ecologie du D´eveloppement

Durable

Cooperation

DevelopmentOKTVF National Inspectorate for Environment, Nature

and Water

PROFEPA Federal Attorney for Environmental Protection

List of abbreviations xvii

RIVM Rijksinstituut voor Volksgezondheid en Milieu

Water Treatment

SEMARNAP Secretariat of Environment, Natural Resources

and FisheriesSEMARNAT Secretariat of Environment and Natural ResourcesS´etra Service d’ ´Etudes Techniques des Routes et

Autoroutes

TNO Netherlands Organisation for Applied Scientific

Research

Development

Europe

VNO-NCW Confederation of Netherlands Industry and

Employers

VROM Ministerie van Volkshuisvesting, Ruimtelijke

Ordening en Milieubeheer

Development

xviii List of abbreviations

1 Introduction: theoretical framework and

to adopt less demanding levels of regulation?

And perhaps even more important: how do processes of tal policy convergence come about? Some argue that cross-national pol-icy convergence is mainly fuelled by the international trade interests ofindividual states Others emphasise formal policy coordination by, forinstance, European Union law or international environmental treaties

environmen-as the predominant convergence mechanism Yet others argue that theimpact of legal harmonisation is overestimated and that much of themutual adjustment of domestic policies, institutions, and instrumentscan be explained by increasing information flows and cross-national pol-icy learning Finally, one always has to keep in mind the possibility thatthere are no international mechanisms at work at all In this case pol-icy convergence would simply be a matter of similar, but independentresponses to similar problems occurring in different countries

As will be set out in considerable detail in Section 1.2, existing arly literature provides partial, tentative, sometimes even fairly powerfulclues to this major puzzle It has been shown that convergence does takeplace at a surprisingly high pace and in fact also at surprisingly high levels

schol-of regulation Generally speaking, environmental policies do not atically fall victim to international economic competition as ‘race to thebottom’ theories would predict Instead, there is increasing evidence thatlegal harmonisation as well as various types of transnational communi-cation lead countries to mutually adjust their policy goals, policy instru-ments and even their levels of ambition Moreover, this convergence isnot restricted to groups of countries with similar political systems or

system-1

2 Duncan Liefferink, Helge J¨orgens and Andrea Lenschow

similar policy styles, or which stand at similar stages of economic opment It can be observed on a European and in many instances evenglobal scale Less is known, however, about the precise ways in which theprevailing mechanisms work and interact in practice

devel-This book seeks to find answers to these questions by way of a highlysystematic set of cases studies, covering seven environmental policy issues

in four countries: France, Hungary, Mexico and the Netherlands Thisintroductory chapter sets out the analytical framework applied in the casestudies It specifies the research questions and the central theoretical con-cepts, explains the selection of the four countries and the seven cases, anddevelops expectations as to which mechanisms of convergence may applyunder which circumstances In Section 1.2 we fix the point of departurefor the present study by briefly reviewing the scholarly state-of-the-art inthe field of environmental policy convergence Section 1.3 then definesand discusses the basic terminology used in the book, notably the concept

of policy convergence and the main mechanisms behind it In Section1.4 we describe how our empirical case studies build upon the findings

of a large-scale quantitative study of environmental policy convergencecarried out earlier.1 The careful and systematic selection of both policyissues and countries makes it possible to investigate in an unusually thor-ough and comprehensive fashion how the various mechanisms of policyconvergence work in practice, how they reinforce or hinder each other,and how effective they are in making domestic policies more similar overtime Finally in this chapter, Section 1.5 sketches the outline of the rest

of the book

the art and further

Over the last decades, the study of processes of cross-national policyconvergence has become a major concern for political scientists In aglobalising world, increased economic, political and cultural interdepen-dence is assumed to make national policies grow more alike over time(Drezner 2001) This convergence of policies and programmes has beenobserved in virtually all areas of public policy making (for a compre-hensive overview see Heichel, Pape and Sommerer 2005; see also thecontributions in Holzinger, J ¨orgens and Knill 2007) In this section we

1 Both the quantitative study and this book form part of the research project tal governance in Europe: the impact of international institutions and trade on policy convergence’ (ENVIPOLCON) For further details, see Section 1.4 and Chapter 2.

‘Environmen-Introduction: theoretical framework and research design 3will give an overview of the literature on policy convergence in the field

of environmental policy

Since the late 1960s virtually all countries in the world have createdgovernment institutions for the protection of the environment such asenvironment ministries, national environmental agencies or environmen-tal advisory councils (J ¨orgens 1996; Meyer et al 1997) Basic legislation

in the areas of air pollution control, nature and water protection as well aswaste management has equally been adopted in a large number of coun-tries (Busch and J ¨orgens 2005a) At the instrumental level, the morerecent shift in the prevailing policy pattern from a sectorally fragmentedand largely legally based regulatory approach to an integrated environ-mental policy characterised by the inclusion of softer and/or more flexibleinstruments such as negotiated agreements, eco-labels, emissions tradingschemes, or ecological tax reforms is also proceeding on a global scale(De Clercq 2002; J ¨orgens 2003; De Bruijn and Norberg-Bohm 2005;Daley 2007) Even concrete environmental protection standards such asemission standards have strongly converged over time (Holzinger, Knilland Arts 2008) Overall, a global convergence of governance patterns

in environmental policy has been observed (J¨anicke and Weidner 1997;Meyer et al 1997; Weidner and J¨anicke 2002; Busch and J ¨orgens 2005b;Holzinger, Knill and Sommerer 2008; Knill, Holzinger and Arts 2008).Both comparative policy analysis and the study of international rela-tions have contributed significantly to this growing literature on envi-ronmental policy convergence Although the two subdisciplines differsubstantially in their theoretical expectations as well as in their method-ological approach, their empirical findings have become increasinglysimilar over time, supporting the identification of a strong and stableconvergence trend over the past four decades in the field of environmentalpolicy

1.2.1 Comparative policy analysis

Scholars in the field of comparative policy analysis originally focused onthe national determinants of policy choice and policy change Conse-quently, their theoretical point of departure was a general assumption

of cross-national diversity of environmental policies resulting from ferent national institutional frameworks, actor constellations, regulatorystyles and problem pressures (Lundqvist 1974; Kitschelt 1983; Weale1992; van Waarden 1995) However, in their empirical analyses, theyquickly detected that in spite of widely differing national styles of regula-tion, advanced industrial states had been surprisingly similar in decidingwhich risks required positive state action (agenda setting) and in their

dif-4 Duncan Liefferink, Helge J¨orgens and Andrea Lenschow

successes or failures actually to reduce environmental pollution (policyimpacts) (Badaracco 1985; Brickman, Jasanoff and Ilgen 1985; Vogel1986) While these studies did not directly pose the question of conver-gence or divergence of national environmental policies, their commonfinding of ‘different styles, similar content’ (Knoepfel et al 1987) was afirst and important step in that direction In a summary of the findings ofthis first set of comparative environmental policy analyses Knoepfel et al.(1987: 183) concluded that ‘the hypothesis concerning the long-termconvergence of policy outputs in environmental regulation must be testedand questioned in a more comprehensive analysis’

Building on these early findings, a second wave of studies began tocompare systematically the development of domestic capacities for envi-ronmental policy making throughout the group of Western industrialisedcountries These studies found not only that national environmental poli-cies were determined only in part by domestic factors, but also thatprocesses of imitation and learning among geographically, culturally oreconomically related countries had become important and independentsources of any country’s capacity to address environmental problems(J¨anicke 1996; J ¨orgens 1996) As a consequence, Western industrialisedstates responded in a surprisingly homogeneous way to the environmen-tal challenge that had been placed on domestic and international pol-icy agendas in the late 1960s and early 1970s A systematic in-depthcomparison by J¨anicke and Weidner of case studies of thirty industri-alised and developing countries confirmed these findings and extendedthem beyond the narrow group of industrialised countries It revealed aglobal convergence of governance patterns in environmental policy thatcovered not only domestic institutions but also sectoral environmentallaws, specific instruments, strategies, actor constellations and even thestrengthening of societal capacities (J¨anicke and Weidner 1997; Weidnerand J¨anicke 2002)

However, these findings did not go undisputed In a study on thedevelopment of environmental policies in Western Europe, Hanf andJansen (1998) confirmed the previous findings that countries tended torespond to environmental phenomena ‘by legislation that was relativelysimilar in formal terms’, but added that beneath the level of formal lawsand institutions, domestic environmental policies remained ‘quite dif-ferent in terms of operational goals and instruments’ (Jansen, Oslandand Hanf 1998: 281) Like much of the Europeanisation literature, theirstudy found domestic actor constellations and institutional structures

to be important intervening factors which explain differences betweennational environmental policies and institutions (see also Andersen andLiefferink 1997; Liefferink and Andersen 1998; B ¨orzel 2002; Liefferink

Introduction: theoretical framework and research design 5and Jordan 2005) While most Europeanisation studies agreed that thepowerful economic as well as political homogenising pressures within the

EU did not necessarily lead to uniform action at the level of memberstates, but often produced a quite heterogeneous patchwork of institu-tions, instruments and policy styles (see, for example, H´eritier and Knill2001), they disagreed on the concrete level of policy making where con-vergence and/or divergence could be expected as well as on the underlyingcausal mechanisms For example, while Jansen, Osland and Hanf (1998)had expected diversity to be strongest with regard to operational goalsand targets, Jordan and Liefferink found that it was exactly at this level ofindividual environmental standards and concrete instruments that con-vergence was most pronounced (Jordan and Liefferink 2004; Liefferinkand Jordan 2005) Regarding the mechanisms of environmental policychange, Knill and Lenschow (2005a, 2005b), in a study of the effects

of EU policies on the organisational structure and behavioural patterns

of national administrations, found that ‘soft’ European steering modesbased on competition or communication had led to greater administra-tive convergence than ‘hard’ steering modes based on legal obligation.Focusing on policies and instruments rather than administrative struc-tures, Jordan and colleagues found more convergence in areas where the

EU has the authority to adopt binding supranational regulations than inareas where it has little or no legislative competence (Jordan, Wurzel andZito 2003; Jordan and Liefferink 2004)

In parallel to these studies on Europeanisation and policy convergence,

a second strand of comparative studies began to investigate systematicallyprocesses of transfer, diffusion and convergence of environmental poli-cies beyond the relatively small group of EU member states Rather thanrelying on small to medium-sized samples of in-depth case studies – ashad been the case with the earlier generations of European and inter-national comparisons – these studies began to trace the global patterns

of environmental policy change and convergence across large numbers

of countries, sometimes even on a worldwide scale (Tews, Busch and

J ¨orgens 2003; J ¨orgens 2004; Busch and J ¨orgens 2005a, 2007a; Tews andJ¨anicke 2005) Looking at a wide range of policy items which includedenvironmental institutions, different types of environmental laws (fromconstitutional articles to issue-specific ordinances), environmental pol-icy instruments (regulatory, informational, voluntary or market-based)and general principles and programmes, these studies provided strongevidence of a global convergence in environmental policy making Fur-thermore, they showed that a wide range of causal mechanisms, includ-ing economic coercion, legal harmonisation, and voluntary imitation andlearning, all contributed to this convergence and that the interaction of

6 Duncan Liefferink, Helge J¨orgens and Andrea Lenschow

different mechanisms – for example voluntary diffusion processes pavingthe way for subsequent legal harmonisation – could significantly broadenthe scope and increase the speed of convergence (J ¨orgens 2004; Buschand J ¨orgens 2005c, 2007b) In sum, comparative studies have shown thatnational environmental policies are actually becoming more similar overtime, but that domestic idiosyncrasies constitute an important interven-ing factor which often limits the impact of transnational and internationalconvergence mechanisms

1.2.2 International relations

While scholars in the field of comparative policy analysis focused inantly on the national determinants of policy change and consequentlystarted out from a theoretical assumption of persisting cross-national dif-ferences, international relations scholars focused on international dynam-ics Consequently, they were from the outset more open to theoreticallyderived expectations of cross-national environmental policy convergence.The most widely received of these hypotheses in the environmental fieldwas the prediction of a global race to the bottom regarding standardsfor environmental, consumer or worker protection (Scharpf 1997a).Although the direction of policy change that this hypothesis implied hasrepeatedly been challenged on empirical grounds with numerous studiesshowing that rather than racing to the bottom, domestic environmentalpolicies and standards tend to move steadily towards higher levels of envi-ronmental protection (Vogel 1995, 1997; Botcheva and Martin 2001;Bernauer and Caduff 2004; Holzinger 2007; DeSombre 2008), the basicprediction of a cross-national convergence of environmental standardswas supported by all of these studies

predom-The second big strand of research on environmental policy gence in international relations, but also in international sociology, isbased on a constructivist epistemology Analysing the global prolifera-tion of characteristic elements of modern environmentalism – such asenvironmental ministries, national parks, environmental NGOs or envi-ronmental impact assessments – John Meyer and his colleagues found aworldwide convergence of environmental policies and institutions whichthey interpreted as the domestic implementation of an emerging globalnorm or, in other words, a norm-based ‘world environmental regime’(Meyer et al 1997; see also Frank, Hironaka and Schofer 2000; Hironaka2002)

conver-Most studies on international environmental politics, however, donot deal explicitly with the convergence of national environmental poli-cies International agreements rather than domestic policies are their

Introduction: theoretical framework and research design 7dependent variable (Harrison 2002) The most important strand of thisliterature, empirical research on international environmental regimes,

is predominantly concerned with the development and implementation

of common solutions to transboundary environmental problems vergence, in this literature, is found mainly with regard to the valuestates place on environmental protection and their subsequent willingnessand ability to reach and comply with multilateral agreements Althoughregime studies implicitly assume that domestic policies will converge asmultilateral agreements are being implemented, this assumption doesnot constitute a core concern of the international relations literature and

Con-is hardly ever tested empirically The large body of literature on the tiveness of international environmental regimes illustrates this Focusing

effec-on issues such as oil pollutieffec-on at sea (Mitchell 1994a, 1994b), leffec-ong-rangetransboundary air pollution (Levy 1993), depletion of the ozone layer(Litfin 1994), the transboundary movement of waste (O’Neill 2000) orocean dumping of radioactive waste (Ringius 2001), these studies are pre-dominantly interested in the environmental effectiveness of multilateralregimes Although they often compare systematically how domestic poli-cies change in response to international accords (Miles et al 2002), theirfocus is not on cross-national policy clustering or convergence, but rather

on the specific design features of international institutions that promote

or hinder domestic compliance (Haas, Keohane and Levy 1993).Within this general regime literature, one particular research strandpays greater attention to the diffusion and convergence of domestic envi-ronmental policies Applying the concept of ‘epistemic communities’,Haas (1992) and his colleagues stress the impact of transnationally dis-seminated scientific knowledge They argue that ideas and causal beliefswhich have emerged and were promoted through knowledge-based net-works of experts can shape state interests by ‘framing the issues for collec-tive debate, proposing specific policies, and identifying salient points fornegotiation’ According to Haas, this ‘diffusion of new ideas and infor-mation can lead to new patterns of behaviour’ (Haas 1992: 2–3) Again,the dependent variable is international cooperation rather than domesticpolicy change and convergence However, as the epistemic communityliterature explicitly points out, domestic policies may converge as ‘theinnovations of epistemic communities are diffused nationally, transna-tionally, and internationally to become the basis of new or changed inter-national practices and institutions and the emerging attributes of a newworld order’ (Adler and Haas 1992: 373) Other scholars have taken

up this point, arguing that epistemic communities and other tional actor networks may in fact constitute an important mechanismfor the diffusion and convergence of domestic policies (Finnemore 2003:

transna-8 Duncan Liefferink, Helge J¨orgens and Andrea Lenschow

149–50; Orenstein 2008; Veenman 2008) although their homogenisingimpact will certainly be moderated by domestic factors Thus, in herstudy on the national regulation of the pulp and paper industry inCanada, Sweden and the United States, Harrison shows that the impact

of internationally shared scientific knowledge was ‘undermined by peting domestic interests and different institutional contexts for decision-making’ (Harrison 2002: 65)

com-1.2.3 Bringing the strands together

The research project ‘Environmental governance in Europe: the impact ofinternational institutions and trade on policy convergence’ (ENVIPOL-CON) was developed against the background of the literature reviewedhere.2Our primary aim for this project was to bring together the differ-ent research strands dealing with environmental policy convergence, toovercome their individual shortcomings, to integrate their findings, andthus to ‘advance our theoretical and empirical understanding of causesand conditions of crossnational policy convergence’ (Holzinger, Knilland Arts 2008: 227) To do so, we adopted a mixed-method approach(Heichel and Sommerer 2009) In a first step, we carried out a quan-titative large-n analysis of the extent, the direction and the causes ofenvironmental policy convergence In the light of previous research

on environmental policy convergence, this analysis endeavoured (1) todevelop a coherent analytical framework, (2) to include a large number

of countries (in this case almost all EU and Eastern European countries),and (3) to cover a large number of environmental policy items located

at different levels of the policy process (principles, policies, instrumentsand standards) To this end, the quantitative part of the ENVIPOLCONproject comprised forty environmental policy items in twenty-one Euro-pean countries as well as the USA, Mexico and Japan, over a period

of thirty years In a second step, forming the subject of this volume,the results of the quantitative study were complemented by systematicin-depth case studies

Key results of the quantitative study are summarised and discussed indetail in Chapter 2 of this volume They broadly confirm and in variousrespects refine the main findings of the studies reviewed in the previoussection (Holzinger, Knill and Arts 2008: 228–9) First, they demonstratethat from 1970 to 2000 the environmental policies of the countries under

2 ENVIPOLCON was financed by the EU and carried out by teams at the universities of Berlin (FU), Hamburg, Konstanz, Nijmegen and Salzburg.

Introduction: theoretical framework and research design 9study converged strongly Second, the study shows that the speed of con-vergence increased over time during the period of observation Third, thequantitative analysis makes clear that the degree of convergence decreaseswith the level of specification of the policy dimension Convergence ishighest with regard to the presence of policies in the countries underconsideration and least pronounced for concrete standards, with con-vergence on particular instruments remaining somewhere in between.Fourth, and similar to previous research in the field of international rela-tions, the study finds no evidence of environmental races to the bottom.Rather, the study confirms that between 1970 and 2000, environmentalpolicies in Europe converged in an upward direction Fifth, the studyfinds that environmental policy convergence can basically be attributed

to the effects of two causal mechanisms: international harmonisation andtransnational communication By contrast, regulatory competition seems

to play no significant role as a causal factor of international environmentalpolicy convergence

The remainder of this chapter will further elaborate on these findingsand present an in-depth qualitative investigation of the actual mecha-nisms through which environmental policies converge

We define policy convergence as:

any increase in the similarity between one or more characteristics of a certain icy (e.g policy objectives, policy instruments, policy settings) across a given set

pol-of political jurisdictions (supranational institutions, states, regions, local ties) over a given period of time Policy convergence thus describes the end result

authori-of a process authori-of policy change over time towards some common point, regardless

of the causal processes (Knill 2005: 768)

For a more precise analysis of processes of convergence, we make use

of three different indicators for assessing policy convergence

Conver-gence scope refers to how many and which countries and policies are converging Convergence degree is about the extent to which policies in

the countries at stake have actually become more similar over time The

direction of convergence, finally, deals with the question of whether

con-vergence takes place in an upward or a downward direction, i.e whether

it raises or lowers overall levels of environmental protection (Holzingerand Knill 2008).3 The unit of analysis of our assessment is the state:

3 In practice, the latter indicator is only relevant for numerical standards, e.g limit values for the emission of sulphur dioxide or the maximum concentrations of heavy metals in surface water, where we can really speak of a convergence at a more or less stringent level,

10 Duncan Liefferink, Helge J¨orgens and Andrea Lenschow

scope, degree and direction of convergence all refer to national policies.Consequently, policies adopted at subnational levels do not fall withinthe scope of this study Similarly, rule making by private actors such as,for example, the Forest Stewardship Council (Pattberg 2005; Dingwerth2007) is not included in our study, although – as our case studies show –when adopting or changing domestic environmental policies, govern-ments often respond to the pressure of private actors such as NGOs orfirms (on the role of private actors in domestic and international environ-mental governance, see Wapner 1995; Glasbergen 1998; Newell 2000)

As set out above, the quantitative study preceding this book has givenample evidence of the impressive scope and degree of environmentalpolicy convergence in Europe over the past decades and confirmed thegenerally upward direction of this process (Holzinger, Knill and Arts2008) Using statistical analysis, furthermore, it represented an impor-tant first step towards understanding the causal mechanisms behind thegrowing similarity of national policies – highlighting legal harmonisationand transnational communication as the single most important mecha-nisms through which environmental policy convergence occurred (for asummary of the findings of the quantitative study see Chapter 2) By theirvery nature, however, statistical methods face certain constraints First,they need to be selective in the kind of variables investigated, and despitetheir growing sophistication they face limits in converting a complex andmultidimensional world into a quantifiable scheme This is due to thefacts that (a) they will only test for interaction effects already hypothe-sised and (b) they will underestimate factors that are difficult to quantify.Second, statistical methods focus on aggregate patterns and regularities.Exceptions to the rule are of no particular interest unless they grow to a

‘significant’ number Exceptions, however, may be revealing in exposingnew causal factors or structures that were unknown to existing researchand, hence, untested in the analysis

Implied in the definition of policy convergence is the process of policy change at the domestic level, which follows certain logics and mechanisms

that are to be identified Logically, to be sure, policy change in individualcountries does not necessarily lead to convergence It may also result

in the persistence or even amplification of differences between tries As our quantitative study suggests, however, convergence appears

coun-to be the rule in the environmental field over the past decades, andnon-convergence rather the exception While statistical methods are very

i.e either at the ‘top’ or at the ‘bottom’ (Drezner 2001) For convergence regarding, for instance, the use of certain policy instruments or procedural requirements, it is hard, if not impossible, to decide what is ‘top’ and what is ‘bottom’.

Introduction: theoretical framework and research design 11effective in showing causal effects of the kind ‘if A, then (probably) B’,they will not find factors that were not expected and they are silent onthe specific pathways and mechanisms of domestic policy change leading

up, in most cases, to policy convergence Against this background, it isthe principal aim of this study to systematically analyse the processes

by which individual instances of policy change – which at the gate level resulted in an overall movement towards cross-national policyconvergence – came about To do this, we need to combine the statisti-cal analysis with qualitative studies of individual cases of environmentalpolicy change At the same time, in-depth case studies allow us to lookmore carefully into those (relatively rare) cases where convergence didnot occur or individual countries appear as ‘outliers’ in an overall pattern

aggre-of convergence

Theoretically, our analysis is based on existing studies of cross-nationalpolicy convergence which suggest five basic mechanisms, or drivingforces, that cause domestic policies to grow more similar over time.These mechanisms are: international harmonisation, transnational com-munication, regulatory competition, imposition, and parallel problempressure (see Holzinger and Knill 2008; for similar classifications see

J ¨orgens 2004; Busch and J ¨orgens 2005c; Holzinger, J ¨orgens and Knill2007; Simmons, Dobbin and Garrett 2008) The first four of these mech-anisms refer to international factors behind environmental policy con-vergence and are related either to the impact of international institutions(international harmonisation and transnational communication) or tothe impact of international trade (regulatory competition) Imposition

in the strict sense presupposes the use or threat of physical force, but isoften expanded to conditionality, i.e the situation where countries areobliged to act in a certain way in exchange for, for instance, a loan ormembership in an international institution The fifth mechanism (paral-lel problem pressure) represents the main alternative explanation, i.e thepossibility that the presence of similar environmental problems and/or asimilar political demand for action leads to similar, but essentially inde-pendent responses in different countries

In this book we are particularly interested in the way internationaldriving forces behind policy convergence interact with processes at thedomestic level and eventually lead to domestic policy change – or fail to

do so In analysing these processes we will cover the actions of a widerange of domestic actors – from national ministries to local authoritiesand from scientific experts to companies and NGOs – but only insofar asthey are relevant for the establishment and change of national policies We

do not deny the relevance of subnational and private sector environmentalpolicies and we do not deny that cross-national convergence can also take

12 Duncan Liefferink, Helge J¨orgens and Andrea Lenschow

place there, but in order to avoid further complicating an already complexstory, the present volume – and the entire ENVIPOLCON project, forthat matter (see Holzinger, Knill and Arts 2008) – focuses on policiesformulated by the central state Thus, where our study focuses on privateactors it is more on their ability to push national legislators towardscross-national policy convergence than on their quality as ‘alternative’rulemakers

As a starting point for the case studies that follow in the next chapters,

we will now discuss the five mechanisms in more detail and explore theirexpected impact on the scope, degree and direction of environmentalpolicy convergence For analytical reasons, the mechanisms are presentedhere in isolation, but it must be stressed that in practice they may beexpected to work in various combinations – it is in fact one of the maingoals of the case studies to find out to what extent and how they do so

1.3.1 International harmonisation

International harmonisation occurs when states comply with obligationslaid down in international or supranational law such as, for example,binding multilateral treaties or EU directives or regulations Thus, thismechanism does not include so-called ‘soft’ instruments of internationallaw, such as resolutions and recommendations, as these do not entaillegal obligations for states to act Due to their predominantly informa-tive character, these non-binding instruments are subsumed under themechanism of transnational communication (see Section 1.3.2)

In principle, international harmonisation is a rather straightforwardmechanism: states jointly commit themselves to common standardswhich are subsequently implemented domestically Logically, depend-ing on the precise character of the policy, this should lead to some sort ofcross-national policy convergence However, at least four factors may beexpected to affect the impact of international harmonisation First, and

in fact rather obviously, international harmonisation only applies to thosestates that have actually subscribed to the policy at stake, either by signingand ratifying the respective international treaty or through membership

in the international institution issuing the policy, for instance the EU.This is a key determinant for the scope of convergence Generally speak-ing, it leads us to expect that countries that are strongly embedded inrelevant international institutions (the EU, OECD, specific internationalenvironmental agreements, etc.) will be subject to convergence throughharmonisation more than those with a lower level of institutional embed-dedness

Introduction: theoretical framework and research design 13Second, the impact of international harmonisation depends on thespecification of the jointly agreed policies The level of specification mayrange from very general and open-ended objectives to concrete mini-mum standards or even a total harmonisation of the level of nationalenvironmental protection standards In the latter case, the degree of pol-icy convergence is likely to be highest.

Third, compliance is an important factor Even if states have signedand ratified a treaty or jointly adopted a piece of EU legislation,correct implementation cannot be taken for granted States may havevarious reasons for dealing light-heartedly with obligations they oncesubscribed to Implementation costs may be higher than expected, theremay be unanticipated domestic conflicts of interest, or the symbolicvalue of signing the treaty may from the beginning have been consid-ered more important than actually working out all the details In thecase of international treaties, enforcement powers are generally morelimited than is the case with the EU (e.g Neyer and Z ¨urn 2001; Miles

et al 2002; Skodvin, Andresen and Hovi 2006) This is due, amongother things, to the supremacy and direct effect of EU law and the activemonitoring and enforcement of member state compliance by the Euro-pean Commission and the European Court of Justice, even includingthe possibility of imposing fines Thus – and even though in practice

a considerable implementation gap exists also here (see Jordan 1999;Knill and Liefferink 2007) – EU legislation may generally be expected

to lead to a higher degree of convergence than other internationallaw

Fourth, the process by which a given international obligation was inally established may provide important clues as to the actual scope,degree and direction of convergence that can later be observed Forinstance, harmonised policies based on broad international agreementabout the character and causes of a problem and its possible solutionsare likely to be implemented better and, thus, to lead to a higher scopeand degree of convergence than more controversial measures For thedirection of convergence, moreover, it is important to know how and

orig-to what extent certain countries have been pressing for stricter policies.National priorities and ambitions together with institutional factors atthe international level (e.g decision-making rules, or the right to estab-lish or maintain stricter standards unilaterally) determine to what extentsuch ‘pioneering’ behaviour can effectively spark off a dynamic towardsthe ‘top’ (Liefferink and Andersen 1998; J¨anicke 2005) While it is diffi-cult to formulate precise expectations here, our choice of the case studyapproach allows us to go into the specific details of international anddomestic processes underlying policy convergence

14 Duncan Liefferink, Helge J¨orgens and Andrea Lenschow

1.3.2 Transnational communication

Under the term ‘transnational communication’ a broad array of relatedmechanisms can be subsumed These mechanisms have in common thatpolicy transfer is based purely on communication and the exchange ofinformation among countries – with ‘countries’ used as a shorthand torefer to a wide range of public and private domestic actors involved

in transnational communication The focus on communication guishes these mechanisms from both international harmonisation, where

distin-a legdistin-al obligdistin-ation is involved, distin-and reguldistin-atory competition, where itive pressure is the key determinant Under the category of transnationalcommunication we distinguish the following four submechanisms:4

compet-r Lesson drawing, in which governments use experiences from elsewhere

to solve a domestic problem It must be noted that lesson drawingdoes not necessarily lead to convergence A foreign model may also beadapted for domestic use, or even be rejected (see Rose 1991; MeseguerYebra 2003)

r Transnational problem-solving, in which experts from different tries jointly develop solutions to a similar domestic problem and thenapply them at the national level.5Transnational elite networks, expertnetworks or epistemic communities play an important role in this pro-cess (Haas 1992; Kern 2000)

coun-r Emulation of policies, in which governments copy a policy used where They may do so for instance because they believe that themost widespread solution is indeed the most appropriate (Meyer andRowan 1977; Baum and Oliver 1992; Levi-Faur 2002), or in order

else-to increase the legitimacy of the policy vis-`a-vis domestic acelse-tors or else-toreduce uncertainty (DiMaggio and Powell 1991), or simply to savethe time and costs of developing a solution themselves (Bennett 1991;Tews 2002; Karch 2007)

r International policy promotion, in which international institutions mote a certain policy approach, for instance through the comparativeevaluation of national policy performance, benchmarking or the dis-semination of best practices (Humphreys 2002; Tews 2002) The inter-national institutions involved may be organisations such as the EU, theOECD or the World Bank, but also NGOs or a transnational interest

pro-4 Our typology is based on the vast literature in this field, to which we can refer only briefly here For a more extensive discussion of this literature, see Busch and J ¨orgens (2007b); Holzinger and Knill (2008); Holzinger, J ¨orgens and Knill (2008) and Veenman (2008).

5 The crucial distinction from the mechanism of international harmonisation is that no international legal obligation is involved in transnational problem-solving (even though,

to be sure, this is a conceivable step at a later stage).

Introduction: theoretical framework and research design 15organisation (Keck and Sikkink 1998) In addition, individual coun-tries may promote their own policies and thus stimulate others to followthem The main driving force in this mechanism is legitimacy pressure:countries not following the model run the risk of losing legitimacy inthe eyes of their (international) peers In this sense, international policypromotion is closely related to emulation, with the exception that themain impetus to adopt a policy results from its active promotion by aninternational organisation.

Although transnational communication comprises a variety of anisms, the general expectation to be derived from them is relativelysimple Lesson drawing, transnational problem-solving, emulation andinternational policy promotion all crucially depend on communicationand the exchange of information International institutions play a key role

mech-in stimulatmech-ing, facilitatmech-ing and organismech-ing communication flows, eitherwith the help of non-binding recommendations, or by producing reports,

or just by providing a platform for exchange Therefore, the scope and thedegree of convergence through transnational communication are likely to

be higher among countries that maintain stronger links to internationalinstitutions In view of the considerable focus on the dissemination of

‘good examples’ and ‘models’ in transnational communication, more, we may generally expect a dynamic of convergence in an upwardrather than a downward direction As with international harmonisation,the case studies will offer the opportunity to study the impact of mem-bership of international institutions, or institutional embeddedness, ontransnational communication at an issue-specific level

further-1.3.3 Regulatory competition

Convergence may occur not only because of international commitments

or transnational communication, but also because countries feel the need

to mutually adjust domestic policies in the face of international itive pressure This mechanism is referred to as regulatory competition.Regulatory competition is supposed to be relevant only for trade-related policies, notably the regulation of production processes and trad-able products This is a restriction to the scope of convergence throughregulatory competition In addition to this, it must be noted that regula-tory competition is not relevant for non-market economies This applies

compet-in particular to the former communist countries before 1990 As gary is one of our case study countries, we will discuss this point in moredetail below

Hun-The degree of convergence through regulatory competition may beexpected to be higher if international competitive pressure is more intense

16 Duncan Liefferink, Helge J¨orgens and Andrea Lenschow

and also if higher costs are at stake for the industries involved Themost interesting aspect with regard to regulatory competition, however,

is its impact on the direction of convergence It is generally assumedthat pressure to reduce production costs for domestic industries vis-`a-visforeign competitors will lead governments to lower standards relating to

production processes (Scharpf 1997a; Drezner 2001) This mechanism

may be expected to work particularly if the relevant industrial sector

is highly internationalised In the case of industries operating in small,mostly local markets, different national process standards are more likely

to persist As regards tradable products, international rules, such as those

set by the EU and the World Trade Organization (WTO), often givecountries room for setting stricter standards domestically and to banforeign products not complying with those standards from their markets,for instance for health or environmental reasons This causes marketsegmentation Several studies show that, particularly if international tradeinterests are high, problems related to market segmentation may be areason for other countries to adopt stricter standards as well (Vogel 1995;Scharpf 1997b; Holzinger 2003)

It should be noted, though, that not all environmental policies entailhigher costs Classical theories of regulatory competition start from theassumption of an inherent conflict between economy and ecology This isnot necessarily the case Policies aimed, for instance, at reducing energyconsumption or at the introduction of ‘clean’, more resource-efficienttechnologies may be both environmentally and ecologically beneficial Insuch so-called win–win situations, interest constellations can be entirelyreversed, industry may actually press for stricter policies in order to enjoy

‘first mover’ advantages (Porter and van der Linde 1995), and a ‘race tothe top’ becomes likely

Whether regulatory competition will lead to a race to the bottom orrather to a race to the top, in other words, depends on various factors,including in particular the specific structure and market dynamic of theeconomic sector involved The case studies will give the opportunity toanalyse those factors in detail Moreover, they will allow us to studyhow and through which actors the mechanism of regulatory competitionactually works in practice For instance: how do governments deal withthe argument of competitive pressure, how do they respond to lobby-ing by the affected companies, and what is the role of domestic versusmultinational firms in this regard?

1.3.4 Imposition

Imposition occurs when external actors – by exploiting economic or ical power asymmetries – force governments to implement policy changes

polit-Introduction: theoretical framework and research design 17which they would not have otherwise (Busch and J ¨orgens 2007b) Itincludes both direct coercive transfer, which implies the use of somesort of physical force, and conditionality, typically involving internationalloans to developing countries which are granted only if certain policiesare adopted (Dolowitz and Marsh 1996: 347; Meseguer Yebra 2003) Inaddition, it is sometimes argued that policy adjustments may be imposedupon countries in exchange for membership of international institutions(Tews 2002; Schimmelfennig and Sedelmeier 2004) Among the coun-tries included in the present study, this might be the case for Hungary

taking over the acquis communautaire in the 1990s and early 2000s in view

of EU accession, as well as for Mexico in relation to membership of theNorth American Free Trade Agreement (NAFTA) It is difficult in thiscontext to draw the line between ‘imposed’ adjustment and an essentiallyvoluntary act of ‘anticipated’ harmonisation Nevertheless, in order to be

as attentive as possible to this potentially powerful driver of convergence,

we will treat imposition as a separate mechanism in this study

In the environmental field, imposition of the kind alluded to here cally involves the transfer of policies from relatively highly regulated states(old EU member states, the USA, Canada) to less regulated states There-fore, apart from strengthening the scope and degree of convergence, thephenomenon is expected to contribute to an upward movement of thestrictness of policies

typi-1.3.5 Parallel problem pressure

In the context of this study, parallel problem pressure may be seen as

a null-hypothesis for explaining cross-national policy convergence Theexistence and/or perception of similar problems may lead to similar poli-cies in different countries without any notable influence of internationalfactors such as international harmonisation, transnational communica-tion, regulatory competition or imposition

The basic expectation with regard to parallel problem pressure is thatmore similar countries are more likely to develop similar policies Thequestions are: what produces parallel problem pressure?; which similar-ities are crucial here? Is parallel problem pressure the sheer presence ofenvironmental problems? If so, indicators such as environmental quality,the emissions of certain pollutants, the geographical situation, the pop-ulation density or the degree of industrialisation might be relevant Or is

it rather the way these problems are perceived and translated into ical ‘demand’ for action? In that case, similarity with regard to factorssuch as the level of income, the political and institutional structure, thestrength of green parties and NGOs, as well as cultural traditions presentthemselves (Lenschow, Liefferink and Veenman 2005)

polit-18 Duncan Liefferink, Helge J¨orgens and Andrea Lenschow

Table 1.1 Mechanisms of policy convergence

International harmonisation Legal obligation

through international law

Compliance

Transnational communication

Lesson drawing Problem pressure Transfer of model found

elsewhere Transnational

problem-solving

Parallel problem pressure

Adoption of jointly developed model

Emulation Desire for conformity Copying of widely used model International policy

promotion

Legitimacy pressure Adoption of recommended

model Regulatory competition Competitive pressure Mutual adjustment

Imposition Force, conditionality Adjustment of policy to that

of ‘imposing’ country Parallel problem pressure

Environmental pressure

Political demand

Parallel domestic constellation

Independent but similar responses

Source: adapted from Holzinger and Knill (2005: 780; 2008: 42).

Moreover, assuming that in practice several mechanisms of vergence usually come into play at the same time, the impact ofinternational triggers for convergence (international harmonisation,transnational communication, etc.) is likely to be stronger if the prob-lem at stake occurs and is perceived in a similar vein in a wide range

con-of countries Conversely, domestic problem perceptions may be shaped

by international factors, for example by the specific problem definitionsprovided by transnationally operating epistemic communities The rela-tive importance of the various domestic factors, ranging from general tohighly issue-specific factors, as well as their interaction with the interna-tional mechanisms of convergence will be considered in detail in the casestudies in this book

Table 1.1 summarises the mechanisms of policy convergence identified inthis section and to be used throughout this book It should be emphasisedonce again that these mechanisms by no means exclude each other One

of the main objectives of the case studies will in fact be to explore to whatextent and in which ways they interact in real-world policymaking

Introduction: theoretical framework and research design 19

This book offers an in-depth analysis of individual processes of policychange which, taken together, promoted or hindered cross-national pol-icy convergence As set out above, the large-scale quantitative analysiswhich formed the first part of the ENVIPOLCON study showed us thatthe scope and degree of environmental policy convergence in Europe overthe past decades has been remarkable We also learned about the crucialrole that can be attributed to the institutional and – apparently less so –economic embeddedness of countries in explaining overall convergencepatterns (see Chapter 2; Holzinger, Knill and Arts 2008; Holzinger, Knilland Sommerer 2008) In the quantitative study, the causal role of institu-tional and economic embeddedness was tested (a) at the level of nationaljurisdictions and (b) at policy level Roughly speaking, deeply embed-ded states were expected to adjust their policies more quickly to those

of their political or trading partners than weakly embedded countries

In a similar vein, policies with high economic implications (e.g productstandards) and policies that are subject to international harmonisation orinstitutionalised communication were expected to converge more quicklythan those isolated from international relations The in-depth qualitativeanalysis offered in this book follows the same general research design –i.e beginning with the expectation that international embeddednessshapes processes of national policy adoption and change – and aims

at elucidating the precise mechanisms at work This general researchdesign has important implications for the selection of country studiesand policy items (see Lieberman 2005 on mixed methodologies gener-ally and Heichel and Sommerer 2009 for a more thorough discussion ofthe mixed methodology approach in this project)

1.4.1 Selection of countries for our case studies

In order to maintain the prominence of two explanatory dimensionsfor policy convergence, namely international economic and institutionalembeddedness, we selected four countries that can be considered mostdifferent cases with regard to these two dimensions, namely the Nether-lands, France, Hungary and Mexico Hence, we chose countries from theoriginal ENVIPOLCON sample of twenty-four countries according totheir degree of vulnerability to international trade (measured by the ratio

of trade in goods to GDP) and their degree of international institutionalintegration (measured by membership in thirty-five European and inter-national organisations) (for more details on indicators, see Heichel et al.2008) (Table 1.2) The other countries that were included in the original

20 Duncan Liefferink, Helge J¨orgens and Andrea Lenschow

Table 1.2 Case study countries

Degree of trade and

institutional factors

Country selected Expectation High economic and

high institutional

embeddedness

Netherlands Follows international convergence patterns

Low economic and

high institutional

embeddedness

France Low responsiveness to regulatory competition;

high responsiveness to legal harmonisation and international communication processes High economic and

low institutional

embeddedness

(until 1990)

Hungary High responsiveness to regulatory competition;

low responsiveness to international legal harmonisation and international communication processes Low economic and

pol-as EU membership) but differ with regard to their political and economicsystems (e.g liberal market economies versus coordinated marketeconomies) Such country selections are based on a comparative design.However, on the basis of our original research question (policy conver-gence) and the leading hypotheses (international economic and institu-tional embeddedness matter), our choice of countries follows anothersystem, as each country represents one ‘type’ with respect to the degree(maximum/minimum) of international trade and institutional interlink-age As such they are ‘crucial cases’ for a systematic test of hypothesesand assumed regularities In addition, with the choice of two EU memberstates, Hungary as an EU applicant and later candidate state in the 1990s,and Mexico outside the EU, we may gain some insight into the specialeffect of the EU as an economic and institutional linkage-framework.The following short country profiles will serve (a) to provide somebackground information on the Netherlands, France, Hungary and Mex-ico and (b) to illuminate the identification of our case study countries inmethodological terms

Introduction: theoretical framework and research design 21

The Netherlands has a small but traditionally open economy and is one

of the largest exporting countries in the world (OECD 2010) Its maintrading partners are Germany and Belgium The classical approach tooperationalising economic interlinkage is trade dependence (e.g tradevolume in relation to GDP) In order to account for the competitivepressures on product and production created by environmental regula-tion, it seems sensible for the purposes of this study to exclude frommeasurement the trade in services In the Netherlands the ratio of trade

in goods (exports plus imports) to GDP has grown from 97 per cent in

1970 to 116 per cent in 2000 Only Belgium has a higher trade/GDPratio over the period investigated in this study (1970 to 2000)

One option to protect the industry in a country like the Netherlandsfrom competitive pressure might be to reduce the level of (environmen-tal) regulations – in convergence terms this would contribute to a race tothe bottom Alternatively, the Netherlands might push for internationalharmonisation in environmental policy In this regard, the high level ofinstitutional embeddedness of the country is a relevant factor As a found-ing member state of the European Union, which is dealing with a greatnumber of environmental regulatory matters, which has a high obligatorypotential (i.e is in the position to create regulatory harmonisation amongits members), and which is characterised by exceptionally close commu-nicative contacts between its members, the Netherlands is well situated.Yet, the country’s institutional embeddedness is broader and not limited

to European integration The data collected for the quantitative part ofthe ENVIPOLCON study identify the Netherlands in the year 1970 asthe second most highly embedded country in the entire reference group,second only to France; in 2000 it was only exceeded by France, Germanyand Italy (see Heichel et al 2008)

In contrast to the Netherlands, France is among the world’s largest

economies, but relies traditionally much less on foreign trade relations.The French post-war growth strategy after the Second World War was

based on a dirigiste model labelled as ‘´etatisme’ The state ‘aimed to

accelerate the pace of modernization and economic restructuring bychannelling resources to critical industrial sectors’ (Levy 2000: 308)and it pushed so-called ‘grands projets’ by combining ‘subsidies, captivemarkets, and the transfer of technologies developed in public researchlabs’ (Levy 2000: 316) Following recessions in the 1970s and 1980s,economic policy in France gradually became inspired more by liberaleconomic ideas and ‘increasingly based on foreign trade’ (OECD 1997:38) Nevertheless, with a trade in goods to GDP ratio of 50 per cent in

2000, France remains on the low end of economic embeddedness andhence comparatively invulnerable to regulatory competitive pressures

22 Duncan Liefferink, Helge J¨orgens and Andrea Lenschow

Institutionally, however, France is one of the most deeply embeddedcountries in the world The ENVIPOLCON dataset identified France ascontinuously in the lead among the twenty-four countries included in theENVIPOLCON reference group As for the Netherlands, membership inthe European Union accounts for much of this embeddedness, but evencompared to countries such as the UK or Ireland, France ranks highly,revealing a more encompassing international engagement Hence, in theFrench case we hope to observe the effects of institutional factors onenvironmental policy convergence in (relative) isolation from economicpressure

Already in the period of the socialist market economy, Hungary was

characterised by relatively high levels of international trade relations Theprimary framework for the country’s international economic relationsused to be the Council for Mutual Economic Assistance (CMEA orComecon), which was created in 1949 as a response to the Marshall Plan.Until the late 1980s, the state had a monopoly on foreign trade which wasplanned as any other economic activity But economic reforms under theK´ad´ar regime also led to increasing trade relations outside the Easternbloc In the 1980s, Hungary exported into 143 countries and receivedimports from 103 countries (Abraham 1987: 31) After the ‘revolution’

in 1989, small- and large-scale privatisation began almost immediately.Especially with regard to large-scale privatisation Hungary opted for astrong influence of foreign capital The liberalisation of foreign trade wascompleted in 1991 (Lavigne 1999: 122ff) Hence, the ratio of trade ingoods to GDP ratio which was already 63 per cent in 1970 and 80 percent in 1980 – approximately double the value of France – jumped after

a slow-down in the 1990s to a remarkable 108 per cent in the year 2000.While the mere numbers identify Hungary as an economically embeddedcountry, hence presumably vulnerable to regulatory competition, prior

to 1989 the free market dynamics that are thought to trigger regulatoryadaptation cannot be assumed In fact, Hungary was selected less as

a prototypical case than in order to study the effects of the economictransition

Also in terms of institutional embeddedness we need to distinguish theperiods before and after transition In the closed context of the socialistworld, Hungary was a member in only a limited number of internationalorganisations Since the 1990s, its institutional embeddedness has risenquickly Hungary became a full EU member only in May 2004, but thepreparation for accession and hence the pressure to comply with EU rules

as well as the intense communicative contacts with other member statesstarted to build up already from the early 1990s and need to be takeninto account in analysing patterns of environmental policy adaptation and