PDF hosted at the Radboud Repository of the Radboud University Nijmegen

Transcription

1 PDF hosted at the Radboud Repository of the Radboud University Nijmegen This full text is a publisher's version. For additional information about this publication click this link. [ Please be advised that this information was generated on and may be subject to change.

2 Against the Grain

3 [FSC-logo] Gerry Alons, 2010 Cover design and printing: GVO drukkers & vormgevers B.V. Ponsen & Looijen, Ede ISBN: All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission in writing from the author.

4 Against the Grain French and German Preference Formation on Agricultural Trade during the GATT Uruguay Round Een wetenschappelijke proeve op het gebied van de Managementwetenschappen Proefschrift ter verkrijging van de graad van doctor aan de Radboud Universiteit Nijmegen op gezag van de rector magnificus, prof. mr. S.C.J.J. Kortmann, volgens besluit van het college van decanen in het openbaar te verdedigen op donderdag 11 maart 2010 om 15:30 uur precies door Gerry Catharina Alons geboren op 24 januari 1979 te Musselkanaal

5 Promotor: Prof. dr. R.H. Lieshout Copromotor: Dr. J.M. van der Vleuten Manuscriptcommissie: Dr. K.M. Anderson Prof. dr. H.J.M. Goverde (voorzitter) Prof. dr. C.J. van Kersbergen (Vrije Universiteit Amsterdam) Dr. S. Meunier (Princeton University) Prof. dr. T. Risse (Freie Universität Berlin)

6 I dedicate this book to my parents who gently took me by the hand when I was a child who were not afraid to let go when I grew older who always encourage me to follow my dreams

7

8 PREFACE When something happens against the grain it is considered unusual or out of the ordinary. And indeed, liberalization of agricultural trade was not a generally accepted practice in the 1980 s. As a result, the negotiations and the outcome of the Uruguay Round may be considered out of the ordinary. Moreover, doing something against the grain implies reluctantly acting against one s inclinations or preferences. As this book clearly shows, agricultural trade liberalization was only reluctantly accepted in France and Germany, against the natural inclinations and wishes of farmers and parts of the government bureaucracy. While the title is certainly applicable to the subject of this dissertation, I by no means conducted this research against the grain. To the contrary, I enjoyed the extensive reading, writing, re-writing (to a somewhat lesser extent), and in particular perusing through archival material and conducting interviews. I could not have accomplished it, however, without the help, encouragement and support of a large number of people. I would like to take the opportunity to express my gratitude to a number of them. First of all, I am very grateful to my supervisors. Thank you Bob, for always overseeing the dissertation s larger picture, when I was inclined to immerse myself in the exciting peculiarities of one chapter or case. The considerable time and red ink you spent commenting on my work have certainly improved it. I also appreciate the opportunity you gave me to teach courses in Political Science. For it is indeed in explaining concepts, theories and methods to students that one s own grasp of the matter is put to the ultimate test. Anna, I must have been quite a job for you. When I came to Nijmegen my knowledge of research methods was limited and the phenomenon of theory-guided research had not yet entered my vocabulary. It is thanks to your kind and patient guidance that I was able to design this research project and complete it within a reasonable amount of time. You were very good at sensing when I required a deadline or when I needed to be left alone. Thank you for your encouragement and empathy, both professional and personal: they made you more than an ordinary supervisor. I would like to thank the rest of my colleagues at the Political Science department: for the chit-chat in the hallway, for politically and sometimes even socially relevant conversations during lunch, and for your support. In addition, I would like to express my gratitude to the members of the WOIB for the tough but constructive discussions of my work during its various stages. Sabina, thank you for being a friend and for providing me with useful advice along the way. Niels and Kristof, I really enjoyed your company during the ismus-trips to Krakow and Berlin, the chocolate-cake-tasting-exercises and the game-playing-nights together with Bertjan, Agnes and Diede. Of course, I also loved teaching AV together with you. The continuity in excellent student evaluations over the years must undoubtedly be attributed to the fact that a female coordinator was pulling the strings. While conducting my research, I also interviewed people who have taken part in the negotiations my research focused on. I am grateful to these interviewees for welcoming me in their homes and taking the time to share their experiences. I would further like to thank the

9 reading-room employees at the archives in Koblenz, Berlin, Fontainebleau and the Hague for their assistance and advice during my visits. Finally, I would like to thank my family and friends for their interest, encouragement and support over the years. Mom and dad, thank you for the values you brought me up with and for encouraging me to use my God-given talents. I really enjoyed the family gatherings at birthdays, family weekends, mother-and-sister-outings, and other activities. They provided the muchneeded diversion for a busy bee like me. It is thanks to Langs de lijn and the boekenclub that I also read literature outside the field of Political Science in the last few years. And if it were not for Liane and our punctual weekly swims (ahum), I would probably have gained a few pounds due to the delicious chocolate and liquorice provided by Helma and Wilma in the secretraries office. Last but not least, I am very grateful to my loving husband Hein Christiaan. You are the one who encouraged me to take on the challenge of writing a dissertation. You cook delicious dinners, ready to serve when grumpy little me comes home from a long day at work. But most of all, you love me unconditionally, rain or shine. Thank you for being at my side every step of the way. Gerry Alons, January 2010.

10 CONTENTS ABBREVIATIONS AND ACRONYMS 1 1. THE PUZZLE OF CROSS-COUNTRY SIMILARITIES IN FARM LOBBY INFLUENCE INTRODUCTION AGRICULTURAL TRADE POLICY AND THE INTERNATIONAL ENVIRONMENT PREFERENCE FORMATION IN FOREIGN POLICY MAKING THE DOMESTIC STRUCTURE APPROACH AND THE PUZZLE OF SIMILARITIES IN FRENCH AND GERMAN FARM LOBBY INFLUENCE DIRECTIONS FOR A NEW THEORETICAL FRAMEWORK OVERVIEW OF THE CONTENT OF THIS DISSERTATION THEORETICAL FRAMEWORK INTRODUCTION THE STATE AND ITS INTERESTS The State State interests The preference-formation process concerning international trade negotiations INTERNATIONAL STRUCTURE The principle by which the international system is ordered International polarity DOMESTIC STRUCTURE The principle by which the state is ordered Domestic Polarity COMBINING THE INTERNATIONAL AND THE DOMESTIC THE PROCESS OF PREFERENCE FORMATION The degree of societal mobilization and governmental sensitivity The potential effects of process variables under different combinations of domestic and international polarity CONCLUSION RESEARCH STRATEGY AND METHODOLOGY INTRODUCTION RESEARCH DESIGN AND STRATEGY Case study method 41

11 3.2.2 Multiple case analysis Case Selection Within-case analysis OPERATIONALIZATION OF VARIABLES Structural variables Process variables Political, economic and ideological costs and benefits CONCEPTUALIZING AND MEASURING INFLUENCE Conceptualizing influence Measuring influence DATA SOURCES AND ANALYSIS HYPOTHESES EVALUATION AND GENERALIZATION CONCLUSION THE URUGUAY ROUND OF GATT NEGOTIATIONS INTRODUCTION AGRICULTURE IN THE GATT THE RUN-UP TO THE URUGUAY ROUND The players at the table Negotiations and Decisions The increasingly attractive prospect of a new round of trade negotiations THE URUGUAY ROUND NEGOTIATIONS Punta del Este 1986: Off to a rocky start The first agricultural proposals ( ) The Mid-Term Review in Montreal and Geneva: (December 1988 April 1989) From Geneva to Brussels: breakdown in Heysel Arthur Dunkel intervenes: the Draft Final Act (December 1991) Bilateral US-EEC negotiations resulting in the Blair House Accord (November 1992) Renegotiating Blair House and reaching a final agreement CONCLUSION DEFENDING VITAL INTERESTS: FRANCE INTRODUCTION THE FRENCH CONTEXT The French decision-making process The French view on agriculture The French government in the 1980s THE FRENCH INTEREST: DOMESTIC CONSIDERATIONS Political interests 121

12 5.3.2 Economic interests Ideological interests Domestic costs and benefits THE FRENCH INTEREST: INTERNATIONAL CONSIDERATIONS Political interests Economic interests Ideological interests International costs and benefits THEORETICAL EXPECTATIONS AND FRENCH PREFERENCES New GATT negotiations including agriculture The start of the Uruguay Round in Punta del Este The first proposals The Mid-Term Review CONCLUSION FOR THE SAKE OF OUR ECONOMY: GERMANY INTRODUCTION THE GERMAN CONTEXT Decision making in Germany The German view on Agriculture The German government in the 1980s THE GERMAN INTEREST: DOMESTIC CONSIDERATIONS Political interests Economic interests Ideological interests Domestic costs and benefits THE GERMAN INTEREST: INTERNATIONAL CONSIDERATIONS Political interests Economic interests Ideological Interests International costs and benefits THEORETICAL EXPECTATIONS AND GERMAN PREFERENCES New GATT negotiations including agriculture The start of the Uruguay Round in Punta del Este The first proposals The Mid-Term Review CONCLUSION 207

13 7. WE ARE NOT ALONE: FRANCE INTRODUCTION THE FRENCH GOVERNMENT BETWEEN 1990 AND THE FRENCH INTEREST: DOMESTIC CONSIDERATIONS Political interests Economic interests Ideological interests Domestic costs and benefits THE FRENCH INTEREST: INTERNATIONAL CONSIDERATIONS Political Interests Economic interests Ideological interests International costs and benefits THEORETICAL EXPECTATIONS AND THE FRENCH PREFERENCES The Commission proposal of Dunkel s Draft Final Act of December The Blair House Accord of November The Final Accord of December CONCLUSION SUPPORTING FRENCH DEMANDS BIS ZUM ÄUSSERSTEN: GERMANY INTRODUCTION THE GERMAN GOVERNMENT BETWEEN 1990 AND THE GERMAN INTEREST: DOMESTIC CONSIDERATIONS Political interests Economic interests Ideological interests Domestic costs and benefits THE GERMAN INTEREST: INTERNATIONAL CONSIDERATIONS Political interests Economic interests Ideological interests International costs and benefits THEORETICAL EXPECTATIONS AND GERMAN PREFERENCES The Commision proposal of Dunkel s Draft Final Act of December The Blair House Accord of The Final Accord of CONCLUSION 309

14 9. FRANCE AND GERMANY: THE POWER OF IDENTITY AND THE INFLUENCE OF FARM LOBBIES INTRODUCTION EVALUATING THE HYPOTHESES Evaluating the case study results Evaluating the hypotheses TAKING STOCK OF THE RESEARCH OUTCOMES Back to the puzzle : the influence of French and German farm lobbies and the theoretical and empirical significance of the research project WIDENING THE DEBATE The influence of ideological interests on foreign policy preferences Objective versus subjective measurement of costs and benefits The role of individual decision makers CONCLUSION 330 APPENDIX I: GUIDE TO THE PRIMARY SOURCES 332 APPENDIX II : LIST OF INTERVIEWEES 333 APPENDIX III: OVERVIEW OF FRENCH, GERMAN AND US MINISTERS DURING THE URUGUAY ROUND 334 BIBLIOGRAPHY 335 LITERATURE 335 PRESS AND PERIODICALS 347 ELECTRONIC SOURCES 348 INDEX 353 SAMENVATTING 359 CURRICULUM VITAE 368

15 LIST OF TABLES TABLE 2.1: DIMENSIONS OF THE INTEREST IN SURVIVAL OF STATE AND GOVERNMENT 21 TABLE 2.2: DOMESTIC POLARITY, INTERNATIONAL POLARITY, AND STATE PREFERENCES 32 TABLE 2.3: THE IMPACT OF PROCESS VARIABLES ON THE EXPECTED EFFECTS OF DIFFERENT COMBINATIONS OF INTERNATIONAL AND DOMESTIC POLIARITY ON STATE PREFERENCES 38 TABLE 3.1: DOMESTIC POLARITY, INTERNATIONAL POLARITY, AND STATE PREFERENCES 42 TABLE 3.2: DOMESTIC POLARITY, INTERNATIONAL POLARITY, AND THE FOUR CASES 46 TABLE 3.3: CORRELATES OF WAR COMPOSITE INDEX OF NATIONAL CAPABILITY AND GDP 49 TABLE 3.4: EXECUTIVE-LEGISLATIVE RELATIONS AND DOMESTIC POLARITY 51 TABLE 3.5: DOMESTIC POLARITY IN FRANCE AND GERMANY 53 TABLE 3.6: MEASUREMENT OF THE EXPECTED COSTS AND BENEFITS OF BEHAVIOURAL OPTION X 56 TABLE 3.7: EXAMPLE OF DEPICTION OF INTERNATIONAL POLITICAL COSTS AND BENEFITS 57 TABLE 4.1: BRIEF CHRONOLOGY OF THE URUGUAY ROUND WITH RESPECT TO AGRICULTURE AND CAP REFORM 78 TABLE 4.2: US-EEC TRADE DIPUTES IN GATT PANELS DURING THE URUGUAY ROUND 87 TABLE 4.3: OVERVIEW OF THE US AND EEC OFFERS BETWEEN 1986 AND TABLE 4.4: THE AGRICULTURAL PART OF DUNKEL S DRAFT FINAL ACT 100 TABLE 4.5: THE BLAIR HOUSE ACCORD 105 TABLE 4.6: THE URUGUAY ROUND PACKAGE ON AGRICULTURE 108 TABLE 4.7: THE INTERNATIONAL POLITICAL ATTRACTIVENESS OF TAKING PART IN THE GATT NEGOTIATIONS AND MAKING A COMPROMISE WITH RESPECT TO AGRICULTURE 111 TABLE 5.1: FRENCH DOMESTIC COSTS AND BENEFITS WITH RESPECT TO THE PROPOSALS DURING THE FIRST HALF OF THE URUGUAY ROUND 135 TABLE 5.2: FRENCH AGRICULTURAL TRADE 140 TABLE 5.3: FRENCH, GERMAN AND UNITED STATES DEPENDENCE ON INTERNATIONAL TRADE 141 TABLE 5.4: FRENCH INTERNATIONAL COSTS AND BENEFITS WITH REPECT TO THE PROPOSALS DURING THE FIRST HALF OF THE URUGUAY ROUND 146 TABLE 5.5A: ISSUE: TAKING PART IN A NEW ROUND OF GATT NEGOTIATIONS 148 TABLE 5.5B: ISSUE: INCLUDING AGRICULTURE IN THE GATT NEGOTIATIONS 148 TABLE 5.5C: ISSUE: ACCEPTANCE OF THE 1986 OPENING DECLARATION 152 TABLE 5.5D: ISSUE: ACCEPTANCE OF THE EEC PROPOSALS IN 1987 AND TABLE 5.5E: ISSUE: ACCEPTANCE OF THE PROPOSALS OF THE UNITED STATES AND THE CAIRNS GROUPS 155 TABLE 5.5F: ISSUE: MAKING CONCESSIONS DURING THE MID-TERM REVIEW 157 TABLE 5.5G:ISSUE: ACCEPTANCE OF THE OUTCOME OF THE MID-TERM REVIEW IN GENEVA 158 TABLE 6.1: GERMAN TRADE IN AGRICULTURAL PRODUCTS 180 TABLE 6.2: GERMAN DOMESTIC COSTS AND BENEFITS WITH RESPECT TO THE PROPOSALS DURING THE FIRST HALF OF THE URUGUAY ROUND 185 TABLE 6.3: GERMAN INTERNATIONAL COSTS AND BENEFITS WITH RESPECT TO THE PROPOSALS DURING THE FIRST HALF OF THE URUGUAY ROUND 192 TABLE 6.4A: ISSUE: TAKING PART IN A NEW ROUND OF GATT NEGOTIATIONS 194

16 TABLE 6.4B: ISSUE: INCLUDING AGRICULTURE IN THE GATT NEGOTIATIONS 194 TABLE 6.4C: ISSUE: ACCEPTANCE OF THE 1986 OPENING DECLARATION 199 TABLE 6.4C: ISSUE: ACCEPTANCE OF THE EEC PROPOSALS IN 1987 AND TABLE 6.4E: ISSUE: ACCEPTANCE OF THE PROPOSALS OF THE UNITED STATES AND THE CAIRNS GROUPS 202 TABLE 6.4F: ISSUE: MAKING CONCESSIONS DURING THE MID-TERM REVIEW 204 TABLE 6.4G:ISSUE: ACCEPTANCE OF THE OUTCOME OF THE MID-TERM REVIEW IN GENEVA 205 TABLE 7.1: FRENCH DOMESTIC COSTS AND BENEFITS WITH RESPECT TO THE PROPOSALS DURING THE SECOND HALF OF THE URUGUAY ROUND 229 TABLE 7.2: FRENCH INTERNATIONAL COSTS AND BENEFITS WITH RESPECT TO THE PROPOSALS DURING THE SECOND HALF OF THE URUGUAY ROUND 240 TABLE 7.3A: ISSUE: ACCEPTANCE OF THE COMMISSION PROPOSAL IN TABLE 7.3B: ISSUE: ACCEPTANCE OF THE DRAFT FINAL ACT 248 TABLE 7.3C: ISSUE: ACCEPTANCE OF THE BLAIR HOUSE ACCORD 253 TABLE 7.3D: ISSUE: ACCEPTANCE OF THE FINAL GATT ACCCORD 259 TABLE 8.1: THE POPULARITY OF KOHL AND LAFONTAINE IN TABLE 8.2: PREFERENCES FOR CHANCELLOR IN TABLE 8.3: OPINION POLLS SPD, CDU/CSU AND FDP IN TABLE 8.4: GERMAN DOMESTIC COSTS AND BENEFITS WITH RESPECT TO THE PROPOSALS DURING THE SECOND HALF OF THE URUGUAY ROUND 285 TABLE 8.5: GERMAN INTERNATIONAL COSTS AND BENEFITS WITH RESPECT TO THE PROPOSALS DURING THE SECOND HALF OF THE URUGUAY ROUND 293 TABLE 8.6A: ISSUE: ACCEPTANCE OF THE COMMISSION PROPOSAL IN TABLE 8.6B: ISSUE: ACCEPTANCE OF THE DRAFT FINAL ACT 300 TABLE 8.6C: ISSUE: ACCEPTANCE OF THE BLAIR HOUSE ACCORD 303 TABLE 8.6D: ISSUE: ACCEPTANCE OF THE FINAL GATT ACCORD 306 TABLE 9.1: EVALUATION OF THE HYPOTHESES 321 LIST OF FIGURES FIGURE 2.1: THEORETICAL FRAMEWORK 16 FIGURE 2.2: CAUSAL MODEL UNDER CONDITIONS OF HIGH INTERNATIONAL POLARITY COMBINED WITH LOW DOMESTIC POLARITY 33 FIGURE 2.3: CAUSAL MODEL UNDER CONDITIONS OF LOW INTERNATIONAL POLARITY COMBINED WITH HIGH DOMESTIC POLARITY 33 FIGURE 2.4: CAUSAL MODEL UNDER CONDITIONS OF HIGH INTERNATIONAL POLARITY COMBINED WITH HIGH DOMESTIC POLARITY 34 FIGURE 2.5: CAUSAL MODEL UNDER CONDITIONS OF LOW INTERNATIONAL POLARITY COMBINED WITH LOW DOMESTIC POLARITY 35

17

18 1 Abbreviations and Acronyms AGPB AMS APCA BBV BDI BGA BMWI CAP CDU CFDT CNJA CNMCCA CNPF CNSTP COGECA COPA CSU DBV DIHT DREE EEC EMS FDP FFA FNPL FNSEA GATT GDP ICC MODEF MTN NAFTA PC PS PSE RPR Association Générale des Producteurs de Blé et autres céréales Aggregate Measure of Support Assemblée Permanente des Chambres d'agriculture Bayerischer Bauernverband Bund Deutscher Industrie Bundesverband des Gross- und Aussenhandels Bundesministerium für Wirtschaft und Arbeit Common Agricultural Policy Christlich Demokratische Union Confédération Française Démocratique du Travail Centre National des Jeunes Agriculteurs Conféderation Nationale de la Mutualité, de la Coopération et du Crédit Agricole Conceil National de Patronat Français Confédération Nationale des Syndicats de Travailleurs Paysanne Comité Général de la Coopération Agricole de l Union Européenne Comité des Organisations Professionnelles Agricoles Christlich Soziale Union Deutscher Bauernverband Deutsche Industrie- und Handelstag Direction des Rélations Économiques Extérieurs European Economic Community European Monetary System Freie Demokratische Partei Fédération Française de l Agriculture Fédération Nationale des Producteurs de Lait Fédération Nationale des Syndicats d Exploitants Agricoles General Agreement on Tariffs and Trade Gross Domestic Product International Chamber of Commerce Mouvement de Défense des Exploitants Familiaux Multilateral Trade Negotiations North American Free Trade Agreement Parti Communiste Parti Socialiste Producer Subsidy Equivalent Rassemblement pour la République

19 2 SGCI SMU SPD TRIM s TRIP s UDF UDC UNICE USDA USTR Secrétariat Général du Comité Interministériel pour les questions de coopération économique européenne Support Measurement Unit Sozialdemokratische Partei Deutschlands Trade-Related Investment Measures Trade-Related Aspects of Intellectual Property Rights Union pour la Démocratie Française L Union Démocratique du Centre Union of Industrial and Employer s Confederations of Europe United States Department of Agriculture United States Trade Representative For abbreviations of archival sources, see Appendix I.

20 CHAPTER 1 THE PUZZLE OF CROSS-COUNTRY SIMILARITIES IN FARM LOBBY INFLUENCE 1.1 Introduction The Uruguay Round of the General Agreement on Tariffs and Trade (GATT) negotiations ( ) went down into history as the last, the longest and the most comprehensive trade round ever undertaken before the establishment of the World Trade Organization (WTO). In contrast with prior GATT rounds, negotiations were no longer restricted to trade in goods, but also involved the issues of services and intellectual property. Furthermore, the issue of agriculture figured prominently on the negotiating agenda, while previous rounds of negotiations had never succeeded in placing this sensitive sector under stricter GATT discipline. Unsurprisingly, agriculture turned out to be an important if not the most important bone of contention in the Uruguay Round. The most prominent players in the GATT negotiations, the United States and the European Community (EC), 1 proved to be the prime antagonists on this issue, the former favouring free trade in agriculture, and the latter rejecting it. They crossed swords at several occasions. During this debate, France and Germany, 2 who did not want to give up the protectionist policies in the agricultural sector, under close scrutiny of their farmers, played a leading role in internal EC negotiations on which position the EC should defend in the agricultural chapter of the Uruguay Round. Given these competing pressures at the international and domestic level, how are the French and German preferences on the liberalization of agricultural trade during the GATT negotiations to be explained? It is more or less unproblematic background knowledge among researchers studying European politics that in France and Germany farmers dominate national preference formation not only when it comes to domestic agricultural policy, but also with respect to international agricultural policy. Many an author has argued that French and German farmers have been able to wield considerable influence over the positions taken up by their respective national governments in the European Council of Ministers and have often succeeded in stalling reform of the Common Agricultural Policy (CAP), or trade liberalization in the context of the GATT in the 1980s and 1990s (Andrlik 1981; Philips 1990; Goverde 2000; Keeler 1996; Risse-Kappen 1995). Although this observation is often thus considered a commonplace, it is a most puzzling one in terms of the domestic structure approach to foreign policy (preferences) (e.g., Katzenstein 1 When referring to the organization in general, I will use the denomination EC, but in the empirical chapters describing events before 1993, the historical name European Economic Community (EEC) is used. 2 Throughout this dissertation, I will use the term Germany both when referring to the Federal Republic of Germany in the period up to 1990 and a reunited Germany in the period after October 1990.

21 4 Against the Grain 1976 and 1978; Risse-Kappen 1991). This approach predicts differences in the impact of societal pressure on state preferences as a result of differences in domestic structures. According to this approach, one would expect farm influence to be substantial in Germany, but not in France, since the German domestic structure is considered to be society-dominated, while the French domestic structure is considered to be state-dominated (see further Section 1.4). The domestic structure approach is unable to explain the similarity in outcome (the dominance of societal interests) in both France and Germany, since the domestic structures of these states are dissimilar. The observation therefore poses an empirical puzzle for the domestic structure approach. The claim that farm influence must be related to immaterial factors such as the special ideological nature of agriculture as policy domain does not offer a solution. If these factors indeed play a significant role, then we should still clarify how the immaterial factors proposed should be conceptualized as variables within a theoretical explanation and identify the scope conditions under which these variables may be expected to be of importance. Furthermore, a cursory analysis of German preferences during the agricultural negotiations within the GATT Uruguay Round reveals that the belief in the continuous dominance of the farm lobby in Germany cannot be taken for granted. Why, for example, did the German government, after having defended farm interests in 1990, strongly favour compromising in the agricultural negotiations during 1991, while the German farm lobby continued its vehement opposition to such a policy? If farmers and their interests are considered to be so special, then why do these interests guide state preferences at one moment but not the other? In International Relations (IR) literature it is generally agreed that in order to understand the preferences and behaviour of a state in international relations, we need to take both the domestic considerations and the international considerations of that state into account. As Nettl pointed out as early as 1968, the state is by its very nature Janus-faced (Nettl 1968; cf. Buzan et al. 1993, 120). During the Uruguay Round, France and Germany were simultaneously confronted with international pressure exerted by the United States, the Cairns Group of agricultural-exporting nations and the GATT Secretariat, urging them to agree to the liberalization of trade in agricultural products on the one hand, and with vehement resistance of their farmers against such liberalization initiatives on the other hand. In this introduction I will show that, with respect to the relative impact of international and domestic considerations on state preferences, there is a theoretical lacuna in the existing literature, because the common acknowledgement that state preferences are based on domestic and international considerations notwithstanding, current theories in IR and Foreign Policy Analysis (FPA) are unable to explain under which conditions in cases where those international and domestic considerations create contradictory behavioural imperatives international considerations will be prioritized over domestic considerations or vice versa. It is therefore important, in both empirical and theoretical terms, to undertake a theoretical investigation of the relative influence of domestic and international considerations on state preferences. The theoretical question guiding this research is therefore:

22 The Puzzle of Cross-Country Similarities in Farm Lobby Influence 5 Under which conditions will middle-power states like France and Germany give precedence to domestic considerations over international considerations or international over domestic ones, when forming their foreign policy preferences? 1 In this dissertation, I will claim that the relative importance of domestic and international considerations depends not only on a state s domestic structure, but also on the structure of the international system. By developing a model based on these two intervening variables between the domestic and international considerations on the one hand, and their actual impact on state preferences on the other hand, I seek to explain why states sometimes prioritize domestic considerations and at other times international ones. On the basis of a comparative case study of French and German preference formation during the agricultural negotiations in the GATT Uruguay Round ( ), I will show that only a theory that recognizes both the constraints of the international environment and the constraints of the domestic environment, is fully capable of explaining the formation of foreign policy preferences. In the remainder of this introductory chapter I will first argue that international (economic) agricultural policy constitutes an excellent example of an issue area in which domestic and international considerations play a significant role, and in which the question of the relative impact of domestic as compared to international considerations is particularly pressing (Section 1.2). Section 1.3 will provide an overview of the theoretical debate on preference formation on a state s foreign policy, followed by a more thorough exploration of the domestic structure approach in Section 1.4. On the basis of the shortcomings noted in Sections 1.3 and 1.4, Section 1.5 will provide an outline of the theoretical framework to be developed in Chapter 2. Finally, Section 1.6 will give a brief overview of the contents of the dissertation. 1.2 Agricultural Trade Policy and the International Environment Agricultural trade policy is an issue area in which both domestic and international considerations are expected to influence state preferences. International trade agreements can affect the relative power position of states by increasing or decreasing their economic resources (see for example Grieco 1990). As a consequence, states can be expected to take account of international considerations in preference-formation processes regarding international trade. For the member states of the EC, agricultural trade is an issue area where policies are for the greater part determined at the European level. Since the creation and development of the CAP, decision making on agricultural policy takes place mainly in the Council of Ministers. Decision making on international trade policy is no longer a national prerogative either, and the European Commission negotiates with the EC s trading partners on behalf of the EC as a single entity, on the basis of a mandate provided by the Council of Ministers. When formulating their preferences on the position they wish the EC to defend in the agricultural trade negotiations, France and Germany are therefore subject to international pressure both within the EC and within GATT. Moreover, agricultural trade policy as an issue area was characterized at that time

23 6 Against the Grain by its potential for significant conflict with the United States (cf. Woolcock 2000; Butler 1983, 105). During this period Europe and the United States crossed swords over a number of agricultural trade panels in GATT, the most important issue concerning soya. Since the United States was both an important provider of security to the EC and an important trading partner for many European member states, any disagreements and potential conflict with the United States would result in pressure on the European member states to give in to the United States demands for trade liberalization. Considering the pressure from the agricultural sector to withstand these demands, agricultural trade policy was an area in which domestic and international pressures could often pull states in opposite directions. In a policy area such as agricultural trade in which both domestic and international considerations are important and in which domestic and international pressures are likely to pull states in conflicting directions, it is of the utmost importance to apply an explanatory model that does justice to both a state s domestic and its international considerations and is able to predict the relative weight of these domestic and international variables in the formation of state preferences. In the next section, I shall argue that current approaches within IR and Foreign Policy Analysis (FPA) cannot provide the theoretical model that is required. 1.3 Preference Formation in Foreign Policy Making The approaches in political science trying to explain a state s foreign policy preferences may be classified in three groups. The first group emphasizes international-level factors as explanatory variables, while the second group stresses domestic-level variables. The third group applies both domestic and international variables in explaining state preferences. Neo-realism (cf. Waltz 1979; Grieco 1990) and neo-institutionalism (cf. Keohane and Nye 1989) are two examples of mainstream theoretical approaches that focus on the international level alone when explaining state preferences and behaviour. In these theories, the structure of the international system and the role of international institutions are the main explanatory variables. The second-image reversed literature (cf. Gourevitch 1978; for an overview see Almond 1989) studies the effect of international variables on the character of states (institutional structures, regime type) (Koch 1997, 39). However, in the context of this research it is not the construction of such domestic institutions that requires explanation, but the formation of a state s foreign policy preferences. Theoretical contributions emphasizing domestic-level factors as independent variables in explaining foreign policy preferences may be found in FPA, International Political Economy (IPE), as well as Andrew Moravcsik s Liberal Intergovernmentalism. Scholars in the field of FPA focus primarily on domestic variables such as public opinion, domestic institutions, pressure from interest groups and the characteristics of individual leaders as factors explaining state preferences (for an overview see Haney et al. 1995; Hudson and Vore 1995; and Hudson 2007). Haney et al. (1995) reserved the term Foreign Policy Analysts specifically for scholars focusing on domestic determinants of foreign policy, treating the international environment

24 The Puzzle of Cross-Country Similarities in Farm Lobby Influence 7 simply as a contextual variable and not an explanatory one. Valerie Hudson (2007), in her overview of classic and contemporary theory of FPA, does not a priori exclude variables at the level of the international system, but the fact that of this overview of 223 pages only ten are devoted to international variables demonstrates the tendency of FPA to focus on domestic explanations of foreign policy. The IPE literature also emphasizes the domestic sources of foreign economic policy (cf. Milner 1998; Knopf 1998, 5). In explaining a state s openness to trade liberalization, Ronald Rogowski (1989) for example focuses on factors of production (capital and labour), while Jeffry Frieden (1991) emphasizes factor specificity. In a state where labour is abundant and capital is scarce, Rogowski expects that domestic preferences on trade liberalization will vary between the actors providing labour (who will favour trade liberalization) and those providing capital (favouring protectionism). Frieden rather expects differences in preferences of specific industries, depending on whether they primarily depend on capital or on labour. Krasner (1976) argues that the economic power of a state and its economic development explain its attitude towards free trade. Finally, in Moravcsik s Liberal Intergovernmentalist approach, the government appears to be no more than a transmission belt for the preferences of important societal groups. He argues that [p]references reflect the objectives of those domestic groups which influence the state apparatus (cf. Moravcsik 1998, 24). Moravcsik thus provides a perspective on domestic-international interactions that emphasizes how societal actors shape the policies of states (Katzenstein et al. 1998, 668). Neo-realists and neo-institutionalists acknowledge that the systemic factors they emphasize cannot account for all state actions (Keohane 1993, 294). This implies that domestic variables must also be included to explain certain state preferences. However, they provide no indication of which domestic variables may be relevant. At the same time, foreign policy explanations that focus on a multitude of domestic-level variables tend to suffer from a lack of parsimony and run the risk of providing ad hoc explanations, focusing on one variable in one case and another variable in another, without specifying the scope conditions under which one variable can be expected to be more important than another. Unfortunately, these two groups of scholars seem not to communicate with one another and to follow their own course. The explanation of preference formation is in need of theoretical approaches that systematically take both international variables and domestic variables into account. In the words of Peter Evans (1993, 400), the discipline is in need of an integrative approach to thinking about international relations and domestic politics. This brings me to the third group of scholars: those who apply both domestic and international variables to explain state preferences. The principal approaches that combine domestic and international variables are Robert Putnam s two-level games (Putnam 1988; Evans et al. 1993), neoclassical realism (e.g. Wohlforth 1993: Lieshout 1995; Schweller 1998; Zakaria 1998), and James Rosenau s pretheories of foreign policy (Rosenau 1971) and the domestic structure approach (cf. Katzenstein 1987; Mastanduno et al., 1989; Risse-Kappen 1991 and 1995; Skidmore 1994). In his metaphor of two-level games, Putnam (1988) argues that international negotiations can be regarded as games that take place at two levels simultaneously. At the domestic level, societal 1

25 8 Against the Grain actors pressure the government to adopt certain policies. At the international level, governments have to deal with the demands of other states and seek to minimize the adverse consequences of international developments. The government therefore looks for a deal that is acceptable to the other international negotiators and at the same time likely to be ratified at the domestic level. Although Putnam s metaphor raises some interesting ideas about how the domestic and the international level interact, when it comes to the question of the relative impact of the domestic and international levels on state preferences, he eventually prioritizes domestic factors. His explanatory variables, the most important of which is the potential ratification of the outcomes of international negotiations, remain situated at the domestic level. Neoclassical realists also agree that world leaders can be constrained by both international and domestic politics. What distinguishes their approach is that they regard the incentives and pressures arising from the international system to be translated through unitlevel intervening variables such as decision-makers perceptions and domestic structure (Rose 1998, 152; cf. Lieshout 1995, 177). Unfortunately, these domestic variables are often applied ad hoc, without being systematically integrated into the explanatory model. Furthermore, although they apply both international and domestic variables, neoclassical realists ultimately tend to treat only international incentives as explanatory variables, while domestic variables are seen as intervening variables. Finally, neoclassical realist models also generally fail to make predictions concerning the conditions under which domestic variables will be more or less important. Walt has summarized these drawbacks as follows: Neoclassical realism tends to incorporate domestic variables in an ad hoc manner, and its proponents have yet to identify when these variables will exert greater or lesser effects (Walt 2002, 211). Both Putnam s twolevel games and neoclassical theories thus suffer from two problems: (1) although they both apply domestic and international variables, the theoretical weight lifting is ultimately performed by either the domestic variables (in the case of the former) or the international variables (in the case of the latter); (2) they fail to explain the conditions under which domestic considerations can be expected to prevail over international considerations and vice versa. With respect to the latter, both Rosenau s pre-theories on foreign policy and the domestic structure approach appear to offer a solution. In 1971, Rosenau criticized scholars for articulating the premise that international behaviour results from a combination of many factors, both international and domestic, without indicating how the various factors combine under different circumstances (1971, 107, emphasis in original). In his pre-theory of foreign policy, he introduces five sets of variables underlying foreign policy making, four of which are domestic in nature and one international. With respect to the sets of variables relevant to this research, he argues that the relative importance of international systemic variables is based largely on the size of a state (in terms of its geography and physical resources), small states being more susceptible to systemic influences than larger ones. The relative importance of societal variables is expected to increase in line with a state s degree of democratic and economic development. What is problematic however with respect to Rosenau s pre-theory in the context of this research is that France and Germany are neither large nor small states, but

26 The Puzzle of Cross-Country Similarities in Farm Lobby Influence 9 medium-sized powers. His model does not therefore allow for predictions on the relative importance of international variables in these cases. Furthermore, although Rosenau s pretheory was a genuine attempt at theoretical integration, Hudson (2007, 169) argues that it still does not give us the necessary scope conditions or an understanding of the integration of these variables that Rosenau himself held up as benchmark for success. We can only conclude that Rosenau s claim of a lack of theoretical integration in the explanation of foreign policy still applies today. As Helen Milner argued in 1997 (1997, 2), and Hudson still maintains ten years later (2007, 125): Although many scholars have recognized the interdependence of domestic and international politics, few have developed explicit theories of this interaction. In the next section, I will argue that the domestic structure approach, despite suffering from a number of important drawbacks, does enable predictions on the relative importance of domestic and international variables in explaining national preferences and can therefore serve as a foundation for the construction of a theoretical model that is appropriate within the context of my research The Domestic Structure Approach and the Puzzle of Similarities in French and German Farm Lobby Influence The domestic structure approach (cf. Katzenstein 1976 and 1978; Risse-Kappen 1991 and 1995) argues that domestic structure in essence the power relations between state and society influences the degree to which governments are sensitive to domestic pressure, and consequently the degree to which they are able to take account of international interests. In their model, the domestic structure serves as an intervening variable between societal pressure and state preferences. The domestic structure of a state is classified either as state-dominated, society-dominated, or as democratic corporatist (Risse-Kappen 1991, 486). The authors argue that a government in a state with a state-dominated domestic structure is insulated to a large extent from societal pressure and is thus able to take its international interests into consideration. Alternatively, a government in a state with a democratic corporatist or a societydominated domestic structure can be expected to be heavily influenced by societal actors and therefore less able to take account of its international interests (cf. Katzenstein 1978, ; Risse-Kappen 1991, 492 and 504). Researchers in this tradition usually classify the domestic structure of France as state-dominated and that of Germany as democratic corporatist or society-dominated (Risse-Kappen 1991, 492; Krasner 1978, 58; Mastanduno et al. 1989, 470; Skidmore 1993, 207; Zysman 1977, 852; Müller and Risse-Kappen 1993, 34; Van der Vleuten 2001, 87). In view of their domestic structures, it is to be expected that societal groups will be more influential in Germany than in France, and that France will be in a better position to consider its international interests than Germany. These theoretical expectations obviously clash with the commonplace observation that agricultural groups dominate the decision-making process in both states with respect to their negotiating positions in European and international negotiations on trade in agricultural

27 10 Against the Grain products. The domestic structure approach cannot explain the similarity in outcome (the predominance of domestic considerations) in both France and Germany, since the domestic structures of these countries are dissimilar. Thus, the one approach that has been able to formulate propositions on the relative importance of domestic and international considerations applicable to French and German preference formation lacks predictive accuracy. We are confronted with an empirical problem. It is now important to explore the possible shortcomings of the domestic structure approach in greater detail in order to determine whether these shortcomings do in fact create the empirical puzzle, and which modifications will increase the approach s predictive accuracy. One of the significant drawbacks of the domestic structure approach is that it can only explain differences in the relative impact of societal actors between states that vary with respect to their domestic structure, while it cannot explain similarities in domestic influence in states with different domestic structures. Neither can the approach explain changes in the relative weight of societal actors in any given state over time unless the domestic structure also changes (cf. Mastanduno et al. 1989). An approach that is only based on structural variables cannot get a grip on the dynamics of societal change. Furthermore, Friman (1993) correctly claims that domestic structure approaches tend to disregard the role of international pressure as a direct source of influence, because the degree to which international pressure is effective depends entirely on the degree to which society is able to exert influence which in essence depends on state-society relations. The approach clearly privileges state-society relations over the role of the state as an autonomous actor that weighs both its domestic and international interests (cf. Katzenstein, Krasner and Keohane 1998). In the approaches of Thomas Risse-Kappen and Peter Katzenstein, the role of the state and the potential influence of international considerations thus only come into play if the domestic structure is state-dominated. This underestimates the role of the state and implies that insufficient account is taken of the consequences of the anarchic international structure. It also results in another problem. In the words of Gourevitch (1978, 903), the approach provides no explanation for the orientation of state policy in the supposedly state-dominated countries. For these countries with a state-dominated domestic structure, Risse-Kappen (1991, 504) implicitly admits that international factors will be important, when he observes that the concept of state strength will assume a policy unhampered by domestic constraints and able to concentrate solely on international factors. However, he does not specify which international factors are important and how they may guide state preferences. Katzenstein takes a step further by introducing the concept of the national interest as defining the preferences of the state (Katzenstein 1978, 298). In his analysis, Katzenstein fails to specify this national interest, and only traces policy objectives in various empirical situations by observing several areas of foreign economic policy and by looking not only at policy makers rhetoric but also at their actions (Katzenstein 1987, 298). Nevertheless, these descriptions seem to imply that the national interest has to do with relative power and economic performance vis-à-vis other states, phenomena that are international in nature. Granted, Katzenstein does at least provide some clarification on what a state s international considerations may be, but

28 The Puzzle of Cross-Country Similarities in Farm Lobby Influence 11 labeling these considerations as the national interest and assuming that these will only be taken into account if the domestic structure is state-dominated seems to imply, erroneously, that societal pressure is by its nature inconsistent with the national interest (cf. Knopf 1998). On the basis of my Janus-faced conception of the state, I would argue that the national interest includes both international and domestic dimensions and does not necessarily stand in opposition to societal interests. The mere fact that societal interests rarely are homogeneous indicates that the national interest cannot simply be viewed as the opposite to societal interests. A final drawback of the domestic structure approach is the conceptual and operational underspecification of the variable domestic structure. Katzenstein and Risse-Kappen introduce three indicators to assess domestic structure : the nature of political institutions or state structure (centralized versus decentralized), the structure of society (strong and centralized versus weak and decentralized), and the policy networks linking state and society (statedominated, society-dominated or democratic corporatist) (cf. Risse-Kappen 1991, 486). Risse- Kappen defines these policy networks as the mechanisms and processes of interest representation by political parties and interest groups that link the societal environment to the political system (Risse-Kappen 1991, 485). However, in his further elaboration on how to measure domestic structure, these policy networks do not appear to be a separate indicator of domestic structure, existing alongside state structure and societal structure; rather, the policy network is itself a function of state and societal structure. It is shorthand for particular combinations of these two structures (e.g. a combination of a centralized state and a decentralized and weak society is classified as a state-dominated structure). 3 As a result, in Risse-Kappen s measurement scheme, state-society relations are essentially defined by the degree of centralization of state and society. In addition, neither Katzenstein nor Risse-Kappen provides well-defined operational criteria to measure these two central indicators, precisely with respect to their degree of centralization. Risse-Kappen indicates that the degree of centralization of a state s structure depends on three factors: the locus of executive power, who has the upper hand in the bureaucratic infighting between governmental agencies, and the extent to which the government controls the legislative process. However, when classifying the United States, Japan, France and the Federal Republic of Germany with respect to their domestic structure, the author applies different indicators to the various countries (Risse-Kappen 1991, ). As a result, the classification of states with respect to their domestic structure is based largely on empirical observations that cannot be systematically related to predefined classification criteria. 1 3 In Bringing Transnational Relations Back In: Non-State Actors, Domestic Structures and International Institutions (1995), Risse-Kappen does treat policy networks as a genuinely independent component of domestic structure with a content separate from the content of state structure and the structure of society, distinguishing between consensual and polarized policy networks. Although this distinction was effective in classifying non-democratic states (and contrasting them with liberal democratic states), it provides no additional means of contrasting, for example, the domestic structures of France and Germany and is therefore not a useful means of assessing policy networks for this research.