The Hard Core: The Franco-German Relationship and Agricultural Crisis Politics in the European Union

Douglas Webber


© Douglas Webber, December 1998
European University Institute, Working Paper RSC No 98/46


ABSTRACT

This paper analyses two series of conflicts over agricultural and agricultural trade policy in the European Union (EU), the first surrounding the creation of the Common Agricultural Policy (CAP) in the early 1960s and the second the reform of the CAP and the GATT Uruguay Round negotiations in the early 1990s. It argues that, in each case, the stance(s) taken by the French and German governments were decisive for the outcome of the conflicts: where both governments opposed a given project, it was blocked; where both supported it, it was approved; where the two governments were divided, the outcome - until such time as they found a modus vivendi - was crisis and deadlock. Explanations of the conflicts and their outcomes based on alternative models or images of EU politics are discussed, but rejected. Franco-German conflict-mediation and the 'pre-forming' of multilateral decisions through bilateral bargains between Bonn and Paris are not ubiquitous in the EU. However, these practices have been and are still often followed to resolve major crises in the union and are by no means limited to the 'high-political' treaty negotiations typically analysed by intergovernmentalist theorists of EU politics and integration. The multilateralisation of Franco-German bargains may be facilitated by the fact that the two states often represent 'opposed poles' in the union, so that a bilateral Franco-German agreement is likely to lie close to the median position of the member states as a whole.

I. Introduction (1)

1. Models of EU politics

The growing volume of social scientific inquiry into the European integration process and the politics of the EU (European Union)(2) in recent years has produced a growing number of models or images which purport to explain, or to capture the essence of, this process, to identify the most powerful actors in it, and/or to explain patterns of policy decisions or choices.

Roughly seven models may be identified. The first and oldest of these, the neo-functionalist model, argues that the application of supranational problem solutions in certain economic sectors creates inexorable pressures for their extension to others, so that integration 'spills over' from one sector to others. It predicts, as private actors transfer their loyalties from national to supranational political organs, the waning of national governments and the gradual emergence of legitimate and powerful supranational political institutions capable of developing autonomous policies and imposing them on actors in the member states.

A second model, which displays "important affinities" with neo-functionalism, is transaction-based integration or transnational exchange theory (Stone Sweet and Sandholtz 1997: 300). This model departs from the observation that political integration (supranational governance) has advanced further and more rapidly in some policy domains than in others. Differences in the level of supranational governance between policy areas or sectors are a function of differences in levels of cross-border transactions and communications: "As transnational exchange rises, so does the societal demand for supranational rules and organizational capacity to regulate" (Stone Sweet and Sandholtz 1997: 306). Policy areas can be located on a continuum from intergovernmental to supranational governance depending on the value of three variables: transnational society ("non-governmental actors who engage in intra-EU exchanges - social, economic, political"), EU organisations ("governmental structures, operating at the European level, that produce, execute, and interpret EC rules") and EU rules ("the legal, and less formal, constraints on behavior produced by interactions among political actors operating at the European level") (Stone Sweet and Sandholtz 1997: 304). Although intergovernmental decision-making is "ubiquitous" in the EU and "present even at the the far right-hand pole" of the continuum, the model predicts increasing levels of supranational governance as transnational society develops. The three variables above, being "constitutive of supranational politics ... must move together, and disjunctures that do occur in movement are short-lived" (Stone Sweet and Sandholtz 1997: 305). Transnational exchange theorists concur with neo-functionalists in observing the operation of a 'spill-over' process: "Once fixed in a given domain, European rules ... generate a self-sustaining dynamic that leads to the gradual deepening of integration in that sector and, not uncommonly, to spillovers into other sectors" (Stone Sweet and Sandholtz 1997: 299).

The third is the intergovernmentalist model, which developed from the realist stream of international relations theory and in which the dominant actors in the EU remain the governments of the member states or, as in Moravcsik's analysis of the origins of the Single European Act, the governments of the three big member states - France, Germany and Britain (Moravcsik 1991) (3). In the realist tradition, international organisations, such as the EU, are dominated by their biggest members because they are dependent on the latter for resources. Major EU decisions, according to Moravcsik (1991), are essentially lowest common-denominator bargains struck between these three states, whose governments are at pains at protect their sovereignty; the integration process, far from being smooth and unilinear as the neo-functional model seems to imply, is typically slow, contingent and potentially reversible. One other study of the origins of the Single European Act attributes the dominant role of the big member states - in this case, just France and Germany - to their ability to pull along' other member states with them either by threatening explicitly to implement their proposals bilaterally if they should not be accepted by other member states or by offering them side-payments to compensate for the otherwise possibly negative effect of their proposals on the other states or both (Garrett 1992: 543-547). In his analysis of the Single European Act, Moravcsik also identifies side-payments as the principal mechanism whereby big member states 'bought' the compliance of small member states in the act and argues that France and Germany secured British acceptance of the project by threatening otherwise to create a 'two-track' Europe which would have left Britain "without a say in the details of the new agreements" (Moravcsik 1991: 25-26, 36). A fourth model is that of multi-level governance. Proponents of this model accept that member-state governments remain the "most important pieces of the European puzzle" (Marks et al. 1996: 346), but maintain that decision-making competencies are shared by actors at different levels, that supranational institutions exercise an independent influence on policy outcomes, that national governments no longer have a monopoly of the representation of the interests of national groups vis-à-vis EU institutions and that, moreover, the integration process has, to a significant degree, eroded national sovereignty.

A fifth model - or rather group of models - may be labelled institutionalist (see, for example, Scharpf 1988; Bulmer 1993 and 1994; Garrett 1995; Garrett and Tsebelis 1996; Moser and Schneider 1997). Institutionalist models ascribe the balance of power between different actors in the EU and policy patterns to institutional structures or arrangements, which may include informal 'rules of the game' as well as formal legislative or decision-making procedures. In as far as institutional structures vary across issue-areas, so, too, will the balance of power between actors and policy patterns, making it impossible to make general statements about these questions for the EU as a whole. Up until the revision of voting practices in the Council of Ministers through the Single European Act in the late 1980s, institutionalist interpretations of the EU (see, for example, Scharpf 1988) typically came to similar conclusions to those reached by intergovernmentalist theorists inspired by international relations realism, namely that the EU was firmly in the grip of the national governments, albeit, unlike some realists, they did not explicitly make any differentiation between the member states in terms of their capacity to shape or veto EU decisions.

The sixth model identifiable in the literature is that of epistemic communities, networks of "professionals with recognised expertise and competence in a particular domain and an authoritative claim to policy-relevant knowledge", sharing the same normative and principled beliefs, the same causal beliefs, the same notions of validity and a "common policy enterprise" (Haas 1992: 3). Other things being equal, in a political system like that of the EU, which is subject to less democratic control and scrutiny than national political systems, there is likely to be a higher propensity for such groups of 'experts' to hijack the policy process, at least on 'low-political' issues, and to secure the adoption of policies that are more than just the lowest common denominators of member states' preferences. This is alleged to explain in part why, in the areas of health and safety at work and environmental policy, more than merely incremental policy changes have been possible (Eichener 1996: 276).

A seventh, related model or image of the EU political system is that of bureaucratic politics (Peters 1992). According to this model, EU politics is depoliticized, decentralised and sectorally fragmented. Beneath the level of the European Council, the dominant actors in the policy process are not elected politicians, but rather officials from national government departments and the Commission directorates-general (DGs): much of what happens depends on "bureaucratic interaction and bargaining" (Peters 1992: 121) In this conception, government departments or Commission DGs are "quasi-autonomous actors with their own goals" and policies the product of "loosely organized and flexible policy communities" (Peters 1992: 115, 117). These 'policy communities' cluster around the functionally differentiated councils of ministers, whose policies are not centrally coordinated to the same extent as national government policies, since there is no EU government or 'Cabinet'. 'Policy communities' are one kind of policy network, distinguished from other kinds of networks by the high level of their insulation, close and stable relations of mutual interdependence between their members, and stable membership (4). They make for stable public policies (Richardson 1996: 6).

In the period of relative stagnation of the institutional development of the EU from the mid-1960s to the late 1980s, intergovernmentalism became the dominant paradigm among scholars of the integration process. However, during the last decade, as, with the Single European Act and the Maastricht Treaty, this process was 're-launched', intergovernmentalism has been increasingly contested as a model of EU governance. The multi-level governance model is, in effect, nothing other than a critique of intergovernmentalism, which is criticised for its failure to account not only for the new dynamic of the integration process, but also, in its more or less exclusive focus on treaty negotiations, for 'everyday' EU policy decisions that are more than the lowest common denominator among the preferences of (all or a subset of) the member states' governments.

2. Approach

Meanwhile, though, the reputation of the intergovernmentalist model may have become worse than is merited. The present contribution was conceived as a test of the plausibility of the intergovernmentalist model with regard to two series of conflicts over agricultural and agricultural trade policy in the EU, one covering the first half of the 1960s, when the CAP (Common Agricultural Policy) was created and the other the first half of the 1990s, when the hitherto most radical reform of the CAP was adopted and, in the GATT Uruguay Round, the EU was put under intense pressure by its trading partners to reduce its level of agricultural protection. However, not only do I find other models unconvincing when applied to EU agricultural politics, I also make a bolder argument than is normally encountered in intergovernmentalist interpretations of the EU, namely that the outcomes of the major agricultural (trade) policy conflicts in the EU in both the early 1960s and the early 1990s can be explained largely in terms of the preferences and interaction of the French and German governments.

The bilateral relationship between Bonn and Paris is, in its intensity, unique among the EU member states, if not in the wider world as well (de Schoutheete 1990: 109). The two governments expend considerable resources discussing and trying to coordinate their positions on EU issues - whether at formal bilateral meetings of government ministers and leaders, such as those foreseen by the Elysée Treaty, at the 'breakfast' meetings of the German Chancellor and the French President at European Council summits and other informal contacts of the heads of government of both states, in the COREPER (Committee of Permanent Representatives), where the French and German EU ambassadors cooperate closely, or through the day-to-day contacts of French and German civil servants in Bonn, Paris and Brussels (5). Media coverage of EU affairs often assumes, argues or implies that France and Germany and their bilateral relationship have a major - positive or negative - impact on the union. Collectively, the two states are often identified as the 'motor' or 'engine' of the integration process. Nonetheless, few political scientists have explored systematically the nature of the relationship and its impact on the EU. The relative neglect of this topic is difficult to explain in as far as, at least to the extent that the French and German governments do coordinate their positions on EU issues, their combined resources, in terms of the proportion of the votes that they wield in the EU Council of Ministers, the magnitude of their contribution to the EU budget, and the size of their economies, might be expected to enable them to exercise a very significant influence on EU decisions.

The former French president, Valéry Giscard d'Estaing, summed up the role that, in his view, the Franco-German relationship played in the EU in the statement that "where Germany and France are in agreement, Europe [i.e., the integration process] advances; where Germany and France are divided, Europe marks time" (quoted in Schmidt 1990: 288). This observation does not encompass all possible scenarios, in as far as it does not admit the possibility of the two states agreeing to oppose a given initiative or integration project. In the following analysis, I test three propositions:

  1. That when Germany and France have the same stance or have resolved their conflicts on any given issue, their position becomes the common position of the EU;
  2. That when the two states oppose any major policy proposal, it is defeated; and
  3. That when and as long as they are divided over such a proposal, the EU decision-making process is deadlocked.

Agricultural and trade policy constitute ideal cases for testing these hypotheses because they belong to the policy areas where the national governments have forfeited the most decision-making competences to the EU and where, correspondingly, the neo-functionalist dynamic should be most easily observable (see Schmitter 1996: 125-26). Other things being equal, if the variant of the intergovernmentalist model proposed here provides a satisfactory explanation of the outcome of these conflicts, it should be no less useful for explaining outcomes in policy areas which are less strongly 'Europeanised' than agriculture and trade.

Two sets of conflicts, some thirty years apart, are investigated in this study. An analysis of conflicts that occurred in the EU's early history alone might yield misleading conclusions. In particular, they might be biased against the neo-functionalist model, as it is unlikely that the supranational institutions would have supplanted the national governments as prime objects of political loyalty and principal actors in the EU so soon after its foundation. However, if an analysis of the conflicts in the first half of the 1990s reveals that (a subset of) national governments are still the paramount actors in Brussels, this is hard to reconcile with neo-functionalist expectations, since one may reasonably expect the 'spill-over' process central to the neo-functionalist model to have made a substantial impact over a period of some three decades. Rather, by underlining the continuing centrality of national governments in the EU, despite the integration process having entered its fifth decade, such a conclusion vindicates the intergovernmentalist model. Comparing the outcomes of conflicts from two different stages of the EU's history also facilitates an assessment of the impact of the expansion of the EU's membership and changes in voting rules and customs that took place in the interim period.

This focus on substantive policy, as opposed to 'constitutional', conflicts in this analysis is designed to pre-empt a major criticism that is or may be made of the intergovernmentalist model, namely that its proponents are too exclusively pre-occupied with the analysis of the big intergovernmental conferences and treaty, or treaty revision, negotiations, such as those that led to the Single European Act and the Maastricht Treaty, and that they therefore oversee the fact that 'everyday' Brussels politics follows a very different decision-making logic (see, for example, Garrett and Tsebelis 1996: 293). To the extent that this analysis confirms the intergovernmentalist model of EU governance, this indicates that the logic of intergovernmentalism is by no means confined to the big intergovernmental conferences where each member government has a legally-entrenched right of veto and the supranational institutions are not represented.

In one respect, however, the sample of conflicts is arguably biased in favour of the intergovernmentalist model. All of the issues analysed were more or less strongly contested; in some cases, member governments went as far as threatening to veto proposed EU decisions. To the extent that, by their having provoked intense political controversies, these issues are atypical among those that reach the EU agenda, the findings of this study cannot be generalised to the EU policy process as a whole. However, if the intergovernmental model is confirmed by the analysis of these politically relatively explosive conflicts in the EU's history, this is still an important finding, as it suggests that, whenever the political stakes are or are seen to be extremely high (and not only at the intergovernmental conferences where the big 'constitutional' issues are tackled), it is indeed the - or at least some - national governments rather than supranational organs that wield the greatest political power in Brussels.

The following section focuses on the 'landmark' decisions surrounding the creation of the CAP in the first half of the 1960s; the third looks at the decisions concerning CAP reform and agricultural trade liberalisation taken between 1990 and 1993. At the end of each section, I discuss why other models of EU governance do not account for the observed decisions as well as the 'Franco-German leadership' model. In the conclusion, I compare and discuss the findings from my analysis of the two series of conflicts, assess whether and to what extent the Franco-German relationship in EU agricultural politics might differ from that in other policy spheres and EU governance patterns in agricultural (trade) from those in other sectors, and make an attempt to explain why these two states have been able to exercise a determining influence on 'landmark' agricultural (trade) policy decisions in the EU.

II. The creation of the CAP

The CAP was created by a series of decisions stretching from 1960 to 1965, of which the most important were those relating to the organisation of the common agricultural market(s) in 1961/62, a common cereal price in 1964, and the joint financing of CAP expenditure, which became enmeshed - disastrously - with the scheduled transition from unanimous to qualified majority voting in the Council of Ministers, in 1965/66. In the conflicts over all these issues, France and Germany were pitted against each other; in all of them, the two states were or belonged to the most important protagonists. The first conflict appears to have been mediated bilaterally between the French and German council delegations; the second was mediated by the Commission arbitrating essentially between France and Germany; in the third, Franco-German differences were not reconciled and the young Community was plunged into a deep crisis (6).

1. The conflict over the organisation of the common agricultural market 1961-62

Unsurprisingly, given the multiplicity of interventionist agricultural policies in the member states, the idea of including agriculture in the Common Market was controversial from the outset. No other international organisation had hitherto been able to subordinate agriculture or agricultural trade to international disciplines and previous schemes for liberalising or expanding agricultural trade in Europe had come to nothing (Freisberg 1965: 15; Küsters 1982: 173; Milward 1992: 284-306). The conflicts between the six were not mediated until February 1957, shortly before the signing of the Rome Treaty. Even then, the agricultural policy articles of the treaty were kept general and vague - any attempt to formulate them more precisely could have endangered the success of the intergovernmental negotiations (Küsters 1982: 356) (7). In general, the member states with competitive agricultural sectors - notably France and Holland - supported a common agricultural market and policy, while those with weak agricultural sectors, first and foremost Germany, opposed them. There was, however, on the German side, a division of opinion between the Economics Ministry, which wanted to expose German farmers to more intense competition, and the Agricultural Ministry, which, like the DBV (Deutscher Bauernverband - German Farmers' Association), resisted the abandonment of national protectionism, at least after a change in the Agricultural Minister in 1959. Chancellor Adenauer, for his part, was exposed to diametrically opposed pressures. On the one hand, given the negative stance of the farmers' lobby, he feared that a 'Europeanisation' of agricultural policy could cost him critical electoral support (8). On the other, for geopolitical, foreign policy reasons, he attached enormous importance to maintaining good and close relations with France and to promoting the European integration process (Küsters 1982: 320-28). In the final phase of the Treaty of Rome negotiations, to which his talks with the then French Prime Minister, Mollet, gave a decisive impetus, Adenauer decreed that these objectives take precedence over differences on concrete policies (Küsters 1982: 322-23, 328).

For France, by contrast, a common agricultural policy became a sine qua non of its participation in the common market, the only means, so it was thought, of balancing the advantages that would accrue to Germany through free trade in manufactured goods (Couve de Murville 1971: 313-14; Marjolin 1986: 299-300). Without the promise of such a policy, the Rome Treaty, according to Couve de Murville, would never have been ratified by the French Parliament, where the Gaullist MPs in any case opposed the treaty because of the alleged 'weakness' of its agricultural provisions (Couve de Murville 1971: 313-14; Marjolin 1986: 318). De Gaulle himself, after his return to power, made it clear that he would keep France in the EU only if a common agricultural policy were realised: this was one of the messages that he gave to Adenauer at their meeting at Colombey-les-Deux-églises in 1958 (Maillard 1995: 153-54; also Peyrefitte 1994: 67) (9). Holland, with its very efficient and export-oriented agricultural sector, was equally insistent upon the integration of agriculture in the common market, although, given its strategy of building up an agricultural sector based on importing cheap animal foodstuffs, it preferred a more liberal agricultural regime than did France (Küsters 1982: 350-51; Tracy 1989: 251; Lindberg 1963: 224).

In the first major conflict that erupted over agricultural policy after the creation of the EEC, Holland and Germany were actually the principal protagonists, with France assuming a more minor role than was later to be the case (Freisberg 1965: 41-42; Hendriks 1991: 46-47). The issue at stake, which provoked the first 'marathon' council meeting, was whether the dismantlement of agricultural trade barriers between the six member states should be accelerated at the same pace as the reduction of industrial tariffs. Holland demanded synchrony in the reduction of trade barriers between industry and agriculture; only Germany opposed this demand. The "violent controversy" that erupted between Bonn and The Hague threatened to destroy the young Community, until, despite a Bonn Cabinet decision opposing such a development, the German government caved in and accepted the simultaneous reduction of agricultural trade barriers (Freisberg 1965: 41). According to one source, German opposition was weakened by disunity within the delegation between the Economics and Agricultural ministries - an early manifestation of what was to prove a recurrent phenomenon in Brussels agricultural negotiations (Freisberg 1965: 42).

The fundamental decisions over the organisation of the agricultural common market emerged from a series of no fewer than 45 council meetings staged between December 1961 and mid-January 1962 (10). The Commission's proposals, given its right of initiative under the Rome Treaty, formed the basis of the negotiations, but were evidently developed in close collaboration with the national governments and agricultural interest groups (Lindberg 1963: 237). The Commissioners themselves had been divided over the nature of a common agricultural policy, especially over precautions to be taken against possible over-production (von der Groeben 1983: 105-09). In the period from 1958 to 1962, in as far as the Commission did not adopt its proposals unanimously, there was reportedly a stable 'pro-CAP' majority consisting of the two French and two German commissioners as well as the Dutch Commissioner for Agriculture, Mansholt. According to one of the French Commissioners of the era, the German Commissioners supported moves towards a common agricultural policy because of Adenauer's resolve to do nothing to alienate Bonn from Paris (Marjolin 1986: 307-08). The Franco-German solidarity thus displayed in the Commission was not as evident, however, in the Council or in the SCA (Special Committee for Agriculture), the organ for the deliberation of agricultural policy issues between the Commission and the member states created in 1960, where negotiations proved "extraordinarily difficult" and the Commission had an especially tough job trying to reconcile the French and German demands (von der Groeben 1983: 146). The launching of the CAP became extricated with the transition to the second phase of the common market, which was to begin on 1 January 1962, provided all member states agreed. On this occasion, not only the Netherlands, but also France insisted that they would veto the beginning of the second phase unless agreement was reached on the creation of the common agricultural policy. For De Gaulle, the realisation of the CAP was indispensable to the solution of the 'peasant problem' in France, which, against the background of widespread rural unrest in France in the early 1960s, he regarded as a potential "second Algerian question on our own soil" (as quoted in Peyrefitte 1994: 302; see also Hendriks 1989: 77) (11). He attached especial importance to the common financing of such a policy, for otherwise, in his view, faced with increased competition from firms in other EU member states, French industry would be too heavily burdened with the cost of supporting French agriculture (Peyrefitte 1994: 302; Peyrefitte 1997: 266, 372-373; to Adenauer Akten 1964: 1257). De Gaulle told Adenauer in May 1961 that the EU would be imperilled if French demands for the integration of agriculture into the common market were not met. He repeated the message in early December 1961 and once again shortly before Christmas, as the talks over the launching of the CAP in Brussels were temporarily interrupted (Hendriks 1991: 48-49; Maillard 1995: 172-73)(12) .

The linking of the two projects by the Dutch and French governments placed the German government in a quandary. On the one hand, it remained fearful of the electoral-political repercussions of any concessions it might make on agricultural policy. For this reason, it was important for it to delay any decision on the launching of the CAP beyond the federal elections of September 1961. On the other hand, it supported industrial trade liberalisation, closer political integration and especially the enlargement of the Community to include Britain, while, at the height of the Cold War, it was also concerned to maintain good and close relations with France, whose stance on the Berlin crisis Adenauer appreciated more than the British or American. In as far as the German delegation had a joint negotiating strategy or objective, it was to play for time and to delay the introduction of a common agricultural policy for as long as possible (von der Groeben 1983: 147). As usual, there were, though, considerable conflicts of opinion between the Agricultural and Economics ministries. For the latter, a critical consideration was that an agreement over the launching of the second phase of the common market, according to its calculation, would have a positive impact on the British entry issue (Müller-Armack 1971: 232-33). The Agricultural Minister also supported British EU entry, but more ambivalently, fearing negative repercussions rfor German cereal prices (Akten 1963: 366-67). He and his ministry wanted to protect the system of import quotas through which Germany imported cheap agricultural produce from non-EU member states in product segments where there was no substantial domestic production and which constituted the heart of its own administrative powers (Freisberg 1965: 92, 95). They were also hostile to the harmonisation of cereal prices implied by the Commission's proposals and insisted that alleged 'distortions of competition' in agriculture between the member states be eradicated before the introduction of a common agricultural policy (Freisberg 1965: 79) (13).

Italy, which was interested in opening up European markets for fruit, vegetables and wine, sided with France and Holland in the negotiations, although it was less favourable to trade liberalisation for other products and critical of some aspects of the Commission's proposals (von der Groeben 1983: 148; Lindberg 1963: 264, 267). According to von der Groeben, Belgium, partly because it wanted to keep the integration process going, also allied itself with France and Holland; according to another, both Belgium and Luxembourg aligned themselves with Germany, but were "undependable allies", on whose support the Germans, by implication, could not necessarily count (Freisberg 1965: 91) (14). The Germans' resistance to the creation of the CAP was, at any rate, the strongest. Their delegation was instructed to 'give up' as little as possible in the agricultural area and to gain as much as possible in economic and political terms (quoted in Hendriks 1991: 50). However, without firm allies and internally divided, and after the EU 'clock' had been stopped to permit continued negotiations at the New Year, the German side eventually acquiesced in the introduction of a common agricultural policy in mid-January 1962. The last obstacle to an agreement - relating to the financing of the policy - was overcome in bilateral negotiations between the heads of the German and French delegations (Lahr 1981: 354; Couve de Murville 1971: 317) (15). The only apparent - and, as was later to transpire, temporary - victory that the German delegation took from the negotiations was the rescinding of the Commission proposal to shorten the transitional period to a full agricultural common market by three years to 1967 (Freisberg 1965: 105). It sacrificed the quota system for imports from non-EU member states dear to the Agricultural Ministry: ultimately, Adenauer and the government attached greater importance to maintaining good relations with France (Die Zeit article quoted in Freisberg 1965: 95). By contrast, the French delegation was by and large successful in achieving its aims: not simply vis-a-vis Germany, but also vis-a-vis the Commission, whose competences - for example, in fixing common prices - were reduced to a minimum to the benefit of the Council and whose principal role seems to have consisted in mediating between the delegations (Freisberg 1965: 97, 100-02; Marjolin 1986: 320) (16). Commission members sceptical about the agreement - such as von der Groeben, who judged it to be much less 'market-oriented' than the Commission's original proposals - felt compelled to accept it so as not to endanger the transition to the second stage, which had to be approved unanimously by the Council, and thus not to jeopardise the integration process itself (von der Groeben 1983: 150-55). The Germans hoped - wrongly, as it turned out - that the German concessions on agriculture would be honoured by the French government's acquiescence in British entry to the EU (Müller-Armack 1971: 232) (17).

2. The battle over the common cereals price 1964

The implementation of the CAP required the setting of common prices for agricultural products, of which the most important, in political as well as economic terms, given the strong influence exercised by cereals farmers in the farmers' organisations, was cereals. Between the milestone decisions on the CAP of January 1962 and December 1964, when a common cereal price was adopted, the EU and the Franco-German relationship endured a number of crises, most notably that unleashed by De Gaulle's rejection of the British entry application in January 1963. Despite the latter, De Gaulle and Adenauer agreed the Elysee Treaty foreseeing close Franco-German cooperation, including on EU issues, the same month. However, a preamble added to the treaty by the German Bundestag substantially devalued its significance, disillusioning De Gaulle, and Adenauer, who had sponsored the treaty against considerable domestic opposition, stood down as Chancellor later the same year, to be replaced by the 'Atlanticist' Erhard (18). A German Foreign Office memo prepared for Erhard on his accession to the Chancellorship contained a lengthy list of issues where the French and German governments were in conflict with each other, ranging from the NATO, relations with the US, nuclear weapons strategy and the 'constitution' of the EU to numerous EU policy issues, including agricultural ones (Akten 1963: 1113). Both Erhard and the Foreign Office viewed De Gaulle as striving to construct a 'small' EU under French domination (see Akten 1963: 942 and Akten 1964: 135). For his part, De Gaulle, less than a month after the approval of the Elysee Treaty, complained that, on EU issues, the French and German governments almost always had opposing positions (Akten 1963: 615).

After De Gaulle's veto of British EU entry, the German Agricultural Minister, Schwarz, strongly opposed Germany making any further unilateral concessions to France at the expense of German farmers (Akten 1963: 365-68). For the same reason, the German Foreign Minister, Schröder, pursued a strategy of 'synchronisation' in Germany's relations with France in the EU, whereby Germany would make concessions only in return for simultaneous French ones. In practice, Bonn insisted that it would bend to French pressure for further progress on the CAP only if France were to cooperate in launching the Kennedy Round of GATT trade liberalisation negotiations. Between these sets of two issues, there was, he argued, an non-negotiable interdependence (Akten 1963: 1637). The CAP and the Kennedy Round were linked in the EU's 'work programme' in May 1963. However, this did not prevent a fresh crisis breaking out between the two governments as the French accused the German agricultural minister of reneging on a pledge allegedly made by Erhard not to oppose the adoption of a common milk price before the end of 1963 (Akten 1963: 1611-14). Again, the battle in the EU was fought primarily between France and Germany and again Paris insisted that a failure of the CAP negotiations would have the "most serious consequences " and affect the "whole future of European integration" (Couve de Murville Akten 1963: 1628-29). Finally, the two sides reached a compromise on the lines foreseen in May - whereby, however, the German conceded the adoption of concrete CAP regulations in exchange for French agreement to a negotiating mandate for EU participation in the Kennedy Round, which was not to begin until some time in 1964.

Cereals, however, constituted a much tougher political proposition - and, in Germany, as Erhard told De Gaulle, an especially "neuralgic point" (Akten 1963: 1459). Cereal prices in the member states diverged widely: for soft wheat, the Italian price was the highest, the German and Luxembourg prices were only slightly lower, and the French price was lowest of all, slightly lower than the Dutch. The Belgian price fell roughly halfway between the German and French prices. It was at this level - "in the middle, between the French and the Germans" (Mansholt 1974: 110) , at DM 425 per ton - that, in November 1963, the Agricultural Commissioner proposed to set the common price.

Initially in favour of a rapid decision on cereal prices (Akten 1963: 1458), and despite the council's resolution in December 1963 to adopt a common cereals price by April 1964, Erhard quickly changed his mind and pledged to resist any such decision, at least during 1964 and 1965, when, in September, federal elections were due to be staged in Germany (19). The German Bundestag approved this stance in March 1964. In January 1962, the Council had decided, at the same time as it adopted the decisions laying the foundations of the CAP, that a common cereals price could be adopted only unanimously before the end of the second stage of the common market in December 1965, but thereafter by a qualified majority vote. Although the German government could expect to secure a price closer to German preferences while its support was required for a common price to be adopted, the option of postponing the adoption of a common price until Germany's opposition was irrelevant may have been politically more attractive, since the government could then have told German farmers and their organisations that it had done its best for them, but had been outvoted. This was the strategy that Adenauer at least was accused of having followed so long as he was Chancellor (Freisberg 1965: 139).

Under other circumstances, this option could also have been appealing for the member states with lower cereals prices than the German, including France, as a price adopted by a qualified majority was likely to be lower than one which required German consent. Already in autumn 1963, De Gaulle, however, pushed Erhard to acquiesce in a common cereal price as quickly as possible; otherwise, he told the German Chancellor, there would be no Franco-German agreement over the Kennedy Round and the EU itself would be jeopardised (Akten 1963: 1459). De Gaulle's tactics may have been motivated by his wish, later manifested in the 'empty-chair crisis', to thwart the scheduled transition to qualified majority voting. The Foreign Minister, Couve de Murville's, statement that, on an issue like the common cereals price, a big member state such as Germany could not be outvoted, points in the same direction (Freisberg 1965: 163-64). After the German Agricultural Minister had blocked a cereal price decision in April, the Agricultural Council decided in June that a decision must be taken by December. After, in May 1964, De Gaulle had told the permanent secretary in the German Foreign Office that "France is not in a hurry" over the cereal prices issue, Bonn was encouraged to believe that it might be possible to postpone a decision beyond the 1965 German elections after all (Akten 1964: 1161-62). By October, however, the French president had changed his mind again. If the CAP was not implemented according to the agreed schedule, he had his press spokesman declare, France would cease to take part in the EU (Akten 1964: 1187). This prospect was so terrifying, according to one observer, that it caused consternation among the other member states (Freisberg 1965: 164; von der Groeben 1983: 235). The threat seems in retrospect not to have been a mere bluff, although the German Foreign Office considered it more likely that the French government would prevent any progress in the Kennedy Round negotiations than withdraw from the EU altogether (Akten 1964: 1188-89) (20).

The French tactic of 'upping the stakes' in the cereal price conflict raised the pressure on the German government and heightened the dilemma Erhard faced. On the one hand, it was clear that, if Germany did not acquiesce in a common cereal price, France would block a successful conclusion of the Kennedy Round as well as German proposals for closer European political integration, if not destroy the EU altogether as well (Freisberg 1965: 164-66). The dangers posed by this scenario to the Franco-German relationship motivated the 'Gaullists' in the CDU/CSU, led by Adenauer, to put pressure on Erhard to cede on the issue. On the other hand, to cave in to French pressure risked incurring dangerous domestic political consequences: the alienation of the DBV, a revolt by the farmers in the CDU/CSU Parliamentary party, a decline in farmers' support for the CDU/CSU at the upcoming federal coalitions and a crisis in the governing coalition. In October 1964, at the same time as De Gaulle threatened to take France out of the common market if there was no agreement on a common cereal price, Erhard's coalition partner, the liberal FDP, threatened to withdraw from the coalition if the government were to give in to French pressure (Freisberg 1965: 171). Moreover, there was no prospect that another member state would rescue the German government from its dilemma by taking over its role as principal opponent of a common cereals price. Italy had just as strong an interest in resisting the Commission's proposal, but it began to signal its preparedness to agree to a compromise on the issue in November 1964, threatening to leave the German government completely isolated among the six (Freisberg 1965: 133-34, 153; Akten 1964: 1361).

As in the conflict over the foundations of the CAP, France, supported again by Holland, was the principal force for agricultural policy integration among the six and Germany the principal opponent. As in 1961/62, the German government found itself exposed to pretty well diametrically opposed foreign and domestic pressures. As in 1961/62 also, it found itself divided, with the Agricultural Ministry defending the interests of its domestic clientele, as defined by the DBV, and the Foreign Office pushing for a German 'climb-down', in exchange of which it hoped to be able to secure concessions for Germany on other issues, including the Kennedy Round (Akten 1964: 1359-61). Again as in 1961/62, the Chancellor, forced to mediate between conflicting intra-governmental positions and concerned to promote a successful conclusion to the Kennedy Round, prioritised foreign policy over domestic political considerations (Osterfeld 1992: 122). Erhard sided with the Foreign Office (Akten 1964: 1199-1200; 1394-96), bypassed his own Agricultural Minister and began to negotiate with the chairman of the German Farmers' Association over the possible terms of German acquiescence in a common cereals price. The farmers' leader, in the light of the government's repeated pledges to prevent a cut in German cereals prices, demanded a high price for his association's acquiescence in a government climb-down: roughly DM 1 billion of additional national agricultural subsidies, almost all of which Erhard ultimately granted. Shortly before the decisive Council meeting, the German government too signalled that it would no longer oppose the adoption of a common price. However, it did not want to accept any per tonne soft wheat price lower than DM 440.

The council negotiations turned into a confrontation primarily between the German and French ministers and were characterised by the most extreme tensions that the Agricultural Commissioner had ever witnessed at an international conference (Mansholt 1974: 110, 114). With the talks at an impasse, the council asked the Commission to formulate and propose a 'package deal', which Mansholt and his advisers drew up in a Brussels restaurant with the aim of offering something attractive to each of the delegations, not least to the French and German (21). Mansholt put forward the package, which contained the same cereal price that the commission had proposed originally, on a 'take-it-or-leave-it' basis; the council accepted it. The leader of the German delegation who accepted the deal was the Economics Minister - at the time the package was accepted, the German Agricultural Minister was back in Bonn. Aided by his farmer supporters in the CDU/CSU Parliamentary party, the Agricultural Minister tried to persuade Erhard to reverse the German acceptance of the package, but to no avail. Before acquiescing in the deal, the Economics Minister Schmücker had consulted the Chancellor and obtained his approval. Both the Agricultural Minister and the DBV had been temporarily shut out of the German decision-making process - Schmücker had been instructed that, if necessary, Erhard's agreement with the farmers' association had to be sacrificed in favour of "European solidarity" (Hendriks 1991: 53; Freisberg 1965: 181-83). For the Italian, Luxembourg and German cereal farmers (who also obtained national subsidies) the losses implied by the common price were 'sweetened' by the provision of temporary EU subsidies (22). Otherwise, however, the only concessions that the German delegation obtained were slight increases in the common prices for rye and hops compared with the Commission's proposal and a one-year delay in the implementation of the common cereals prices - to 1967 (Tracy 1989: 262; Freisberg 1965: 183). Schmücker declared in the decisive council meeting that the German government expected in exchange for its acquiescence in a common cereal price concessions from the other member states in the "expansion of economic and political cooperation" (Akten 1964: 1396). Concretely, this probably meant that it sought French and other member states' support for the initiative for closer political cooperation between the Six that Erhard had launched in November 1964 (Osterfeld 1992: 123).

3. The financing of the CAP & the 'empty-chair' crisis 1965-66

The crises over the CAP in the years 1961/62 and 1964 were resolved in the first case through bilateral Franco-German mediation and in the second through the intervention of the Commission as an 'honest broker' between conflicting, primarily French and German preferences. Confronted in the first case with a joint Franco-Dutch, and in the second case with a French, ultimatum, and after fighting long rearguard actions, the German government chose on both occasions to make major agricultural policy concessions in favour of overarching foreign policy objectives, such as the maintenance of a close Franco-German relationship. The crisis that blew up in connection with the financing of the CAP in 1965 differed from the previous two in as far as Franco-German differences were not mediated, the French government chose to confront the other member states over the EU's future constitution, and the dispute plunged the EU into a deep crisis, the like of which did not recur - rather as a farce - until the very short-lived British 'boycott' of the EU over the 'mad cow disease' in 1996.

According to the January 1962 council decision, a new formula for the financing of the CAP, which was initially funded by variable contributions from the national exchequers, was to be agreed by the end of June 1965 (23). At the cereal price meeting in December 1964, the council asked the Commission to make proposals for a new formula by April 1965. The Commission's proposals were worked out by a small group of officials working with Mansholt (Newhouse 1972: 254). In retrospect, it seems that the Commission was emboldened by the success of its mediation efforts in the cereals price negotiations (Le Monde 1993: 56-57). The proposals, by giving the EU direct control over revenues from import duties and thus increasing its financial autonomy vis-a-vis the member states and by expanding the budgetary powers of the EP (European Parliament), would have substantially strengthened the EU's supranational character and so were bound to antagonise the French government. The Commission had been divided over whether to play the role of a 'mediator' and make modest proposals or to play the role of 'motor' of the integration process that the member states, including, at least in agricultural policy, France, had so far accepted (von der Groeben 1983: 269; Marjolin 1986: 343). Given the nature of the proposals and as De Gaulle himself recognised, the majority of the Commission clearly calculated that the French government would be prepared to sanction a widening of the EU's powers in exchange for a CAP financing regulation that would be in France's financial interests and that, in this way, Gaullist resistance to a stronger EU could be circumvented (Newhouse 1972: 254; Freisberg 1965: 199; Peyrefitte 1997: 296-297; Willis 1968: 343) (24).

The French government was all the more hostile to the proposals because of the Commission's failure to consult the member states about them and the fact that they had been presented to the EP before the Council. It immediately rejected them (see Peyrefitte 1997: 281-285). Opinion among the other member states was more nuanced. In contrast to the French government, the German liked the 'integrationist' components of the package, but disliked the proposals concerning the financing of the CAP, to which it would be a net financial contributor. Following Foreign Minister Schröder's 'synchronisation' strategy, the German government was again reluctant to make anything more than short-term concessions to France over the CAP before France had made concessions to it over the Kennedy Round. which was still in progress (Freisberg 1965: 120; Groeben 1983: 272; Newhouse 1972: 255). De Gaulle had also angered Bonn by rejecting its proposal for a meeting of the foreign ministers of the Six to consider its plan for closer political integration - contrary to a pledge that Erhard believed De Gaulle had made following Germany's acceptance of a common cereal price (Osterfeld 1992: 184-185). For the German government, Schmücker told the Council of Ministers in May 1965 that it would the CAP financing proposal only in exchange for progress in "other areas of the Common Market" (Akten 1965: 1104).

Of the other member states, Belgian and Luxembourg supported France so as not to endanger the integration process, while, without being as hostile to the proposals as the French government, both Italy and Holland wanted to make some changes in them - not all of them the same ones that France wanted (von der Groeben 1983: 273). Italy, as the country which had done most poorly out of the already existing CAP financing arrangements, was actually to become France's strongest opponent in the ensuing council negotiations (Newhouse 1972: 255-56; Peyrefitte 1997: 289).

The Franco-German relationship continued to be beset by the same issues and conflicts that it plagued it, at the latest, since the British entry application crisis in January 1963. At the regular Franco-German summit meeting a few weeks before the decisive council meeting, the two governments failed to reconcile their differences over the Commission's proposals (see Akten 1965: 1039-1044). Essentially, the French side insisted that the financing regulation be adopted by the end of June, as agreed in January 1962, and the Germans that this deadline be extended, because the issues on which they sought concessions from France were too complex to be resolved in the time left. If the French government wanted an agreement on the CAP budget within the month, this could not be for the five-year period it demanded (25). The consultations between the governments were complicated by poor coordination between the Bonn ministries and evident differences between Erhard and Schröder, as to the right strategy to pursue vis-à-vis Paris (Manfraß 1972: 329; Peyrefitte 1997: 260-61; Osterfeld 1992: 206-207) (26). A serious clash in Brussels appeared, however, to have been averted, when, just prior to the scheduled council meeting, fresh talks between leading foreign ministry officials of the two governments produced a bilateral agreement. A German memorandum suggested that they had agreed to limit the expansion of the European Parliament's powers proposed by the Commission, and at least to postpone the direct transfer of revenues from import duties to Brussels. The German Foreign Office understood that Paris would not insist on a definitive resolution of the CAP budget at the forthcoming council and that, in continuing negotiations, it might acquiesce in an initial one-year agreement on the issue (Newhouse 1972: 263; Akten 1965: 1102). Hence it did not anticipate that the meeting would lead to a Franco-German confrontation.

As is well known, the council meeting instead ended in an éclat and was followed by a six-month-long French boycott of the EU institutions. The offical French version of the meeting blamed the German delegation for the meeting's collapse. According to this account, the Germans reneged on the pre-summit bargain reached by the two foreign ministries, first by rejecting a French proposal to detach the agricultural policy components of the Commission's package from the institutional-political components and to deal with the former first and then, following a corresponding Bundestag resolution, by supporting an extension of the powers of the European Parliament. Informally, French officials blamed the crisis on - roughly in this order - the Commission, which eschewed a mediating role and refused to modify its original proposals, and the Italians and Dutch, who insisted on treating the proposals as an indivisible whole (Newhouse 1972: 266-67; Willis 1968: 344) (27). Worried that they were likely to become an ever bigger contributor to the CAP budget, the Italians were prepared to accept only a very temporary agreement on the policy's financing (Vaïsse 1997: 586). The precipitate declaration of the meeting's failure by Couve de Murville, president of the council at the time, astonished the other delegations, which had reckoned with a continuation of the negotiations. According to Mansholt, an agreement would have been possible if the French government had wanted one (quoted in Freisberg 1965: 204). The German delegation argued that progress had been made in the talks and that, if the 'clock' had been stopped as in 1961-62, an agreement could have been reached in 10 to 14 days (Akten 1965: 1109). Bonn surmised that "new instructions from the élysée" (i.e. De Gaulle) had prompted Couve de Murville to disregard the pre-council Franco-German bargain and declare the meeting to have failed (Akten 1965: 1114). In retrospect, it indeed seems very probable that the alleged German 'betrayal' of France in the meeting was used as a pretext to justify the French government's action and that, as Mansholt and others supposed, the aim of the French boycott was rather to curtail the powers of the Commission and thus to stifle a strengthening of the supranational character of the EU (quoted in Freisberg 1965: 204; see also von der Groeben 1983: 277; Manfraß 1972: 329; Newhouse 1972: 270; Lahr 1981: 428-29; Akten 1965: 1114, 1271). This was certainly the ghist of De Gaulle's later - both private and public - remarks on the boycott (Peyrefitte 1997: 296; Tracy 1989: 263) (28). The 'empty-chair' crisis boiled down essentially to a battle between France, on the one hand, and the other five member states and the Commission, on the other, over the EU's future constitution rather than over the financing of the CAP, which became a subordinate issue. Among the five, Belgium and Luxembourg adopted a more conciliatory stance towards French demands, Italy played a mediatory role within the group and Holland and Germany, where, after the September 1965 elections, Erhard and Schröder were still in their respective offices, took tougher positions (Newhouse 1972: 274). The German government calculated that, for economic reasons, France could not afford to leave or destroy the EU and decided in favour of taking a "relaxed" attitude towards the boycott, sticking to its bargaining position hitherto and opposing especially resolutely any French demands to change the Rome Treaty (Akten 1965: 1113-17 and Document No. 314; Vaïsse 1997: 578) (29). Any sign of weakness towards De Gaulle, argued the German ambassador in Paris, would be likely to raise the cost of the concessions that the five would have to pay to secure France's return (Akten 1965: Document No. 280). In the negotiations that finally resolved the crisis in January 1966, Schröder was the principal player on the side of the five. The so-called(30) 'Luxembourg Compromise', which ended the French boycott, documented the disagreement over what to do in the event of major conflicts in the EU rather than resolving it: neither side made any irrevocable concessions to the other. On the financing of the CAP, the agreement reached corresponded fairly closely to the French government's aspirations (Marjolin 1986: 352; Newhouse 1972: 276). There were no changes in the treaty of the kind that De Gaulle wanted to secure at the outset of the conflict. However, in practice, to the extent that unanimous voting on fundamental issues remained the norm in the EU during the following two decades, the French government also won the struggle over the constitution of the EU (31).

4. Discussion

The creation of the CAP in the first half of the 1960s was a politically highly charged process, in which each new policy step unleashed a crisis and France and Germany repeatedly found themselves in opposing camps. In the first two cases portrayed here, the conflicts between France and Germany were mediated and landmark decisions adopted; in the third, the crisis remained unmediated, plunging the young community into a crisis which was defused only by an 'agreement to disagree' after a six-month stand-off between France and the other member states as well as the EU institutions.

What is most striking about the cases is the extent of the divergence of the initial French and German positions and, more so, the extent to which the conflicts were dominated by their antagonism. In the first two conflicts, the two countries (whereby France was supported strongly by Holland) represented the two poles, the most extreme positions on the political spectrum, among the member states; in the third, the cleavage ran rather between France, on the one side, and all the other member states, including Germany, on the other. In the first two cases, once the conflict between France and Germany was settled, so, too, practically, was the conflict within the Community as a whole. When, as in the third case, France and Germany could not reach a modus vivendi, a resolution of the conflict between the Six was not possible either. Thus, the big battles over the creation of the CAP in the first half of the 1960s seem to confirm the hypotheses that when Franco-German divisions on a major agricultural policy issue have not been overcome, the EU is deadlocked and that when they have been settled, the Franco-German position is taken over by the EU as a whole. As there was no case of both states opposing an important agricultural policy proposal, the hypothesis that, in these circumstances, the proposal will fail altogether can be neither confirmed nor refuted.

How plausibly can, by contrast, other models of EU governance account for the pattern of outcomes of the crises concerning the creation of the CAP? The neo-functionalist model emphasizes the centrality of supranational institutions. Of these, however, only the Commission itself played a significant role in these conflicts. As the neo-functionalist theorist, Lindberg, himself points out, the EP (European Parliament) and the ESC (Economic and Social Committee) were marginal to the process by which the founding CAP decisions were taken in 1961/62 (Lindberg 1963: 252-260). Equally, the COPA (Comité des Organisations Professionelles Agricoles), the umbrella association of the national farmers' organisations, whose foundation the Commission had encouraged, had no observable impact on the decisions. Nothing suggests that these organs played a more important role in the battles over the common cereal price and the financing of the CAP later on. The neo-functionalist case rests wholly on the role in these events played by the Commission. However, although, in the case of the 1961/62 decisions, Lindberg highlights how closely the final package corresponded to the Commission's original proposals, it seems clear, reading between the lines of his analysis, that the proposals were tailored with a view to their political feasibility. The Commission did secure the adoption of its proposal concerning the common cereals price, but the fact that this was set halfway between the existing French and German prices suggests that, here too, considerations of political feasibility influenced the Commission's stance more strongly than any 'autonomous' Commission ideas as to the 'best' common price (32). In the 1965-66 conflict with France, the Commission, with its bold proposals, seems to have suffered a comprehensive defeat that braked the integration process for the following two decades. All this is not to say that the Commission was not a critical actor in these conflicts, least of all in the 1964 one, but its role was that of a mediator, arbitrator and broker rather than an autonomous supranational agency and when it became too ambitious, as in 1965, it was cut down to size.

The transnational exchange model does not aspire to explain what "specific rules and policies" are adopted in the EU, although the proponents of the model subscribe to the intergovernmentalist view that the larger member states, because they command "greater resources", tend to exercise greater influence on policy outcomes than the small (Stone Sweet and Sandholtz 1997: 310, 314). In general, the model predicts that political integration will take place fastest in sectors which exhibit the highest levels of cross-border transactions (trade). That is why the EU has moved further towards supranational governance in issue-areas relating to the internal market than, for example, in foreign and security policy (Stone Sweet and Sandholtz 1997: 308-309). However, given that agriculture was a sector where there had been very high trade barriers and relatively limited inter-state trade in Europe before the 1960s, the model can not easily explain why political integration should have started earlier and proceeded faster in this sector than in others. Although Stone Sweet and Sandholtz (1997: 311) qualify their overall argument to the extent that they expect integration to "occur most readily in policy domains included in the Treaty", a claim with which the rapid integration of agricultural policy can be squared, this begs the questions of why provision was made in the treaty for a common policy for agriculture, but not for many other sectors, and why a common policy was realised a great deal more quickly in agriculture than in other sectors for which the treaty foresaw common policies (33). Of course, it might be argued that the CAP was engineered by two member states - Holland and France - that wanted to be able to sell their surplus agricultural production on other West European markets and that, at least indirectly, the CAP was the product of pressures exerted by 'transnational society' (in this case, export-oriented farmers in the two countries). But the member states whose agricultural sectors were less competitive were equally hostile to agricultural trade liberalisation, so that what has to be explained in the conflict over the foundation of the CAP, and what the transnational exchange model can not account for, is why the former group of member states got the better of the latter.

The multi-level governance model likewise fails to deliver a plausible retrospective interpretation of the political struggle over the foundation of the CAP. Although the Commission played an important agenda-setting role, given its exclusive right of policy initiation according to the Treaty of Rome, the competence to make decisions lay fairly and squarely in the council, that is to say, it was shared between the national governments and not between national and supranational actors. There is no powerful evidence of the Commission - and much less of other supranational organs - having exercised an independent influence on council decisions. The available literature does not enable a judgement to be made over the extent, if any, to which national interest groups bypassed their own governments in trying to influence the outcome of the conflicts. Lindberg reports that the Commission actively tried to integrate agricultural interests into the process of developing the original CAP. However, it is apparent that the Commission encouraged the formation of a European-level farmers' organisation (COPA), presumably to have a single interlocuteur for agricultural policy issues and to avoid the need itself to have to aggregate the interests of the different national farmers' organisations (Lindberg 1963: 237-238). The national farmers'organisation most hostile to the CAP, the German, clearly - and rationally, given the unanimity decision-making rule in the council - concentrated its efforts to prevent or shape the CAP on its own national government: it was not with Hallstein or with Mansholt, but with Erhard that the DBV chairman negotiated over the common cereal price. Finally, at this early stage of the integration process, national sovereignty had not been eroded to the degree where threats by national governments - or at least France - to withdraw from the organisation were no longer credible. De Gaulle's threats to leave the EU unless the CAP was adopted were both meant and taken seriously.

An epistemic communities-based explanation of the foundation of CAP would have to identify a network of professional experts and show that it had dominated the decision-making process. In the case of agricultural policy, this could only be the profession of agricultural economists. However, the agricultural economists have always been sidelined in the EU agricultural policy-making process. Far from having been the policy's architects, they have typically been critics of the CAP and its 'economic irrationality'. To this extent, there are no similarities between this policy area and others in which scientific experts have been observed to exercise a strong influence on EU policy choices. The bureaucratic politics/policy community model also appears inappropriate. It would hardly be plausible to argue that, in the first half of the 1960s, a tight, closed network of national agricultural ministry and Commission DG VI officials dominated CAP decision-making. The conflicts surrounding the foundation of the CAP were not resolved at the level of the civil service; they could be settled only by politically responsible ministers. And, although an arguably important first step towards the formation of a 'closed' agricultural policy community had been taken with the creation in 1960 of the SCA to prepare the Agricultural Council meetings (in place of the COREPER), the Agricultural Ministers were far from being a law unto themselves at the time that the landmark CAP decisions were reached. The decisive negotiations took place in 'Jumbo' council meetings attended not only by agricultural, but also by other (typically economics and foreign) ministers. It is by no means clear that, in France, with its relatively tight process of inter-ministerial coordination, the Agricultural Minister Pisani always had the last word on agricultural issues when there were inter-ministerial conflicts: in the 1965-66 crisis, for example, he seems to have been surprised by and unhappy with the French negotiating tactics and decision to withdraw from the EU organs (see above and Newhouse 1972: 269). The looser inter-ministerial coordination process in Germany enabled Pisani's German counterpart to operate more autonomously of other ministers and the Chancellor (Peyrefitte 1997: 248). But the German agricultural minister also seems to have been sidelined in the 1965 crisis, in which the dominant Bonn actor was rather the foreign minister Schröder. In the cereals price conflict, he was bypassed in critical phases - by Erhard negotiating directly with the DBV president and the Economics Minister accepting the cereal price agreement with Erhard's backing while he was absent from the negotiations. Altogether, one has the strong impression that French and German behaviour in these three 'high-political' conflicts was strongly shaped by overriding foreign (as well as agricultural) policy priorities determined more by the foreign ministers, the German Chancellor and the French president than by the agricultural ministers themselves.

The primacy of the national governments in the disputes over the foundation of the CAP observed here is compatible with an interpretation based on the institutionalist model. However, whilst the subsequent evolution of the CAP is explicable in terms of this model, in as far as the policy, once launched, could be reformed only with the support of all member governments (see Scharpf 1988), the institutionalist approach can not easily account for the fact that the CAP was launched at all (34). Given the unanimity voting requirement, the German government could have vetoed the start of the policy - why did it not do so? One hypothetical explanation is that, because of its strong initial opposition to the proposals in the first two conflicts and because it could not be outvoted and was therefore in the strongest bargaining position, it was able to mould the policy in its interests and thus had no reason ultimately not to accept it (35). This interpretation is difficult to square with the facts that the common cereal price was set not at the German level, but halfway between the German and French levels and that, quite clearly, important German actors, notably the DBV and the Agricultural Ministry, would have preferred to have no common agricultural policy at all. The CAP, in its original guise, corresponded much more closely to French than to German (agricultural) interests.

It might be possible to reconcile an institutionalist interpretation with what happened by arguing that there was a cross-issue trade-off between France and Germany whereby France obtained the CAP in exchange for conceding industrial trade liberalisation to Germany and that issue-linkage thus circumvented a German veto of the CAP (36). To the extent that the 1961-62 CAP decisions fell at the same time as that to make the transition to the second stage of the common (industrial) market and both France and Holland insisted that they would veto the latter unless the CAP was launched, there was indeed such a trade-off, although the available literature does not show whether fear of the collapse of the common industrial market was in fact the main motive for the German government's ultimate acquiescence in the CAP's launching. The simultaneous agreement of common milk, meat and rice prices and a negotiating mandate for the Commission for the Kennedy Round at the end of 1963 may seem to have been a Franco-German trade-off. However, Erhard's acquiescence in these common prices appears rather to have been a response to the pressure exerted on him by the Francophile current in the CDU/CSU led by the former Chancellor Adenauer (Schwarz 1994: 886) (37). In any case, in as far as there was such a trade-off on this issue, it reflected the normal pattern in Franco-German relations - which was for Germany to make concessions to France in the present in the subsequently disappointed hope or expectation that France would make concessions to it on other issues later. Thus, Germany received no simultaneous concession from France when it agreed to the common cereal price. Rather, it conceded the common price in exchange for the other member states - it meant, of course, France - agreeing to closer "economic and political cooperation" (see above) in the future. In other words, Foreign Minister Schröder's 'synchronisation' strategy repeatedly failed (38). The main reason for this failure must be seen in the divisions in the German government over the relative priority of maintaining good relations with France and the Chancellor's propensity, in case of conflict, to arbitrate in favour of France on EU issues (see below).

A second possible objection to the institutionalist model is that it makes no distinction per se between the member states in terms of their veto power and capacity to shape EU decisions (39). Could Luxembourg, for example, really have resisted the launching of the CAP as Germany did - or persuaded or coerced Germany into acquiescing in it as France did? Hardly. In fact, like Germany, Luxembourg and Italy were opposed to the Commission's cereals price proposal, but they gave up their resistance to it much earlier than the Germans. At the same time, it is questionable (although not certain) whether any state other than France could have cajoled Germany into finally accepting the CAP and the common cereals price (40). Of the other member states, France was the one that could most credibly threaten to withdraw if the CAP was not launched (i.e., the one that could most credibly claim to survive without it) and the one to which Germany attached the greatest importance to maintaining a close bilateral relationship (i.e., the one which, as an opponent, could most damage German interests). Although the three conflicts portrayed here were not exclusively Franco-German battles, the first two were conducted primarily between these two states, although important interests of the others were also at stake, while the 1965-66 crisis divided France from all the other member states, including Germany.

In contrast to the institutionalist model, intergovernmentalism, at least in Moravcsik's incarnation, does distinguish between the 'big', more powerful member states (France, Germany, and, since 1973, the UK) and the 'small', less powerful remaining ones. Critical EU decisions, however, represent the lowest common denominator of the preferences of the big three (in the first half of the 1960s, the big two), unless two of the three can credibly threaten the third that they would, if necessary, pursue a given integration project without it (and the third state fears that its exclusion would negatively affect its interests). The CAP decisions in the early 1960s amounted, however, to more than the 'lowest common denominator' of French and German preferences, between which there was hardly any common ground: the French side got a great deal more of what it wanted than did the German. This outcome can not be attributed to France's having brandished a credible threat of 'excluding' Germany from the CAP, since German participation was indispensable to the project - Germany was the biggest prospective foreign market for French agricultural produce and the biggest prospective contributor to the budget of the CAP, the purpose of which, for De Gaulle, was to socialise the cost of supporting French agriculture across the member states. It was not by threatening Germany with 'exclusion' that the French government was able to secure such a favourable outcome of the CAP conflicts, but rather by threatening itself to 'exit' from the EU if the CAP was not launched. Thus, the outcomes of the CAP conflicts are entirely compatible with Moravcsik's intergovernmentalist model if this is amended to include the possibility of more radical than'lowest-common-denominator' solutions being achieved by the threat of 'exit' rather than 'exclusion'.

The question that remains to be answered is why, in the first two conflicts portrayed here at least, the German leaders of the day submitted to the French exit threats and made major, more or less unilateral concessions to Paris that were strongly contested by an important domestic constituency of the CDU/CSU, namely German farmers. Why, in other words, was it ultimately more concerned than the French to avert a crisis, in the worst-case scenario the collapse of the EU? On account of its size, its legitimacy as a World War II Allied power, and its membership of the UN Security Council, France was the strongest political and military power of continental Western Europe. For De Gaulle, the EU was useful in as far as it could be used as a lever for maximising French influence on European and world affairs and a vehicle for making Europe more independent of the United States. But it was not indispensable, least of all if it could not be put in the service of these goals and encroached on French sovereignty. For West German leaders, the EU was intrinsically extremely valuable. Through its integration in the EU, the Federal Republic, following World War II, could aspire to achieve its "international comeback" and greater national sovereignty (Schwarz 1992: 43, 40). The closer economic and political integration of Western Europe facilitated by the EU helped to bolster West German security in the Cold War, which, in the early 1960s, was at its height. Moreover, the EU provided an appropriate multilateral framework for the consolidation of the Federal Republic's relations with France - partly because a close Franco-German relationship was and would be more acceptable to the smaller West European states when it was multilaterally embedded than otherwise (41).

For Adenauer, forging and maintaining a close relationship with France was, like promoting European integration, an extremely important foreign policy goal in itself. In EU affairs, according to a French Commissioner of the era, he was resolved to "do nothing that could estrange him from Paris" (Marjolin 1986: 307) (42). He was prepared to play junior partner to De Gaulle: because of the Third Reich, Germany could not play a leading foreign policy role in the foreseeable future, he told the French ambassador in Bonn. France had therefore to take over this role and Germany could bring its power to bear and foreign policy ideas to fruition only indirectly via France (Akten 1963: 622; also Schwarz 1994: 747; Lacouture 1993: 335) (43). His principal motive in trying to tie France and Germany together as firmly as possible - as in the élysée Treaty - was to try to avert a revival of the kind of anti-German Franco-Soviet alliance that had been formed before World War I and that De Gaulle had in fact tried to resurrect with Stalin in the 1944 Franco-Soviet pact (Akten 1963: 175; Schwarz 1994: 467, 815, 895). In the last year or so of his Chancellorship, Adenauer also moved closer to France for fear that the US might betray (Federal) German interests (as he interpreted them) in direct talks with the Soviet Union over Berlin and in the hope that De Gaulle would take a tougher stance against the Soviets (Schwarz 1994: 814). In aligning himself more closely with De Gaulle, however, Adenauer alienated himself from the majority of the CDU/CSU, not to mention the other main parties, which prioritised the relationship with the United States and strongly supported British EU entry (Schwarz 1994: 720-722, 763-765). Adenauer's increasing political isolation was illustrated by the struggle over the ratification of the Elysée Treaty. His successor, Erhard, and the German foreign minister since 1961, Schröder, wanted a wider and more liberal Europe than De Gaulle and attached higher priority to the relationship with the United States than with France (44). Erhard showed De Gaulle "the cold shoulder" (Osterheld 1992: 388); he disagreed with almost every item of the French president's foreign policy (Willis 1968: 324).

These disagreements did not prevent Erhard making a "decisive contribution" to settling the cereal price dispute - "largely at our urging", according to the French foreign minister (Couve de Murville 1971: 263 and quoted in Peyrefitte 1997: 278; Osterheld 1992: 141; Vaïsse 1997: 553). By this time, however, Franco-German relations had already deteriorated seriously; in 1965, as De Gaulle set out to destroy the EU's supranational vocation, they grew even worse. For Paris, Bonn "really did the minimum to avoid a rupture" at the June council meeting and nothing to facilitate a settlement of the conflict between France and the other member states in the ensuing crisis (Couve de Murville 1971: 263). For Bonn, the crisis was precipitated exclusively by France, whose boycott represented a "naked treaty infringement" (Lahr 1981: 430).

The divergent paths taken by De Gaulle and Couve de Murville, on the one hand, and Erhard and Schröder, on the other hand, on these issues of 'high politics' could only damage the two governments' capacity to resolve their conflicts over agricultural policy, since this capacity was heavily dependent - especially on the German side - on the respective leaders' and foreign ministers' subordination of sectoral to longer-term macro-political priorities. If the resolution of the conflicts over agricultural policy had been left entirely in the hands of the French and German agricultural ministries, the CAP would presumably never have seen the light of day - the bureaucracy in the Bonn ministry was hostile to the EU (Freisberg 1965: 92-96). The Economics Ministry's attitude to the EU and the Franco-German relationship was ambivalent (see Müller-Armack 1971: 225-241) (45). In Bonn, the strongest supporters of the European integration process and a close Franco-German relationship were to be found at the highest echelons of government: in the Foreign Office and the Federal Chancellor's Office. From the outset, the French and German foreign offices evidently tried to coordinate their positions on EU issues. Without their "close" cooperation, according to the French foreign minister, the creation of the CAP and the abolition of tariff barriers between the member states would hardly have been possible (Couve de Murville 1971: 242). Notwithstanding this, relations between the foreign ministers became more strained after Schröder acceded to the office in Germany and especially after De Gaulle's veto of the British entry bid. Although Schröder defended the élysée Treaty, he thought that Franco-German rivalry was inevitable and that Germany was not suited to being France's junior partner (Kusterer 1995: 201) (46). He advised Erhard not to agree to the common milk, rice and meat prices in 1963 (Schwarz 1994: 886), then pressed him to accept the common cereals price in 1964, then was held responsible by the French at least for the tough stance taken by the German government in the 1965-66 crisis (Peyrefitte 1997: 249, 286-287) (47). In 1965, there was no longer any major political actor on either side of the Rhine with both the political will and political authority required to avert a crisis. Indeed, on this occasion, exceptionally, the clash between the highest echelons of the two governments over institutional issues seems to have been more intense than that between the agricultural ministers over farming issues.

III. The GATT Uruguay Round and the CAP Reform 1986-1993

After the 1965-66 crisis and the 'Luxembourg Compromise', unanimous voting was institutionalised in the Council of Agricultural Ministers. This practice actually benefited the German minister, who wanted the highest guaranteed prices and was primarily responsible for their being set well above world market levels (Tracy 1989: 269-270). The Commission, whose members had accepted the "bad compromises" of the 1960s so as not to endanger the overall integration process, was reduced or confined to the role of an 'honest broker', searching for the lowest common denominator between the member states and unable to bring about major changes in the CAP (von der Groeben 1983: 300-302; Tracy 1989: 265) (48). The Commission's weakness and the importance of French and German attitudes in CAP politics were exemplified in the conflict over the Mansholt Plan, whose aim was to modernise European agriculture by creating a smaller number of larger, more productive farms. Among the member states, France and Germany most strongly opposed the plan, which, after a struggle lasting several years, was finally adopted, but in an extremely diluted form that bore little resemblance to the original proposal, in 1972 (Pinder 1991: 83-85). The guiding hands of the French and German governments were likewise evident in the permanent arrangements for financing the CAP adopted in 1969. These were reportedly based on a broader agreement reached between the French President Pompidou and the German Chancellor Brandt that also embraced the issue of British entry into the EU (Gerbet 1990: 94; Brandt 1993: 454). Germany became by far the biggest net contributor to the CAP budget. Helmut Schmidt, German Chancellor from 1974 to 1982, wanted to reform the CAP and tried, but, by his own admission, had "only minimal success" (Schmidt 1990: 190). His scope to push for and secure major changes in the policy was limited by his dependence on a coalition partner, the FDP, which provided the Agricultural Minister and did not want to antagonise German farmers, and by his fear that any attempt to do so would cause "serious irritations" in the Franco-German relationship, which, in his view, became and remained closer during his Chancellorship than it had been even in the times of De Gaulle and Adenauer (Schmidt 1990: 143, 172) (49). During the 1970s and 1980s, German farmers and their organisations made their peace with the CAP and developed into ardent defenders of the policy, which was viewed very critically, on the other hand, by business interests and the Economics Ministry, which had originally hoped that the Europeanisation of agricultural policy would erode Germany's tradition of agricultural protectionism. In the Agricultural Council, France and Germany formed a "strong mutually protective alliance", whereby France supported the high internal prices demanded by Germany and Germany the export subsidies consequently required by France to be able to sell its agricultural produce on the world market (Moyer 1993b: 11).

For two reasons, one internal and the other external to the EU, reform of the CAP finally became a serious political issue in the second half of the 1980s. First, the policy, with its tendency to produce growing produce surpluses, gradually grew so expensive that it threatened to 'break' the EU budget. Second, important trading partners of the EU began to exert pressure on the EU to change a policy which was held responsible for disturbing world agricultural markets - and enabling EU exporters to gain a growing share of them. With a view to curbing EU subsidisation of its agricultural exports, the US and the agricultural-product-exporting states that later coalesced in the 'Cairns Group' insisted that agriculture be put on the agenda of proposed new GATT trade liberalisation negotiations. In the absence of an agreed multilateral framework for agricultural trade, the EU had to reckon with the possibility of the CAP being attacked through the GATT disputes procedure and, for fear of provoking a spiral of trade wars, being forced gradually to dismantle it. Among the EU member states, only France resisted the launching of a new GATT round (50). It agreed finally to agricultural trade at least being discussed in the round, apparently because it "wanted to avoid a major break with Germany" of the kind that its veto of a new GATT round would have precipitated (Odell 1993: 247). The German delegation at the opening round of the conference honoured the French government's preparedness to compromise by supporting its demands concerning the round's agenda (Odell 1993: 247).

The conflicts over the Uruguay Round and CAP reform took place in an EU that had twice as many member states as it had had at the time of the CAP's creation in the early 1960s and in which, beginning in the mid-1980s, more extensive use was being made of qualified majority voting in the council, so that the capacity of single member states to block the Brussels decision-making process was reduced. In the case of the Uruguay Round, the Commission was supposed to negotiate with other GATT member states on the basis of a negotiating mandate adopted by the Council of Ministers and to consult the '113 Committee', a body consisting of high-level civil servants from the member states' trade and economics ministries, foreseen in the Rome Treaty and designed to oversee the Commission's conduct of trade negotiations with third countries. Any agreement reached by the Commission required the council's approval. Normally, the council could decide by a qualified majority, but the range of issues covered in the Uruguay Round was so great that it was uncertain whether a new treaty required qualified majority or unanimous support in the council. Although the Commissioner for External Relations and the Commission DG I had overall responsibility for the conduct of the negotiations and reported to the General Affairs Council, talks on agricultural trade were conducted for the Commission by the DG VI, headed by the Agriculture Commissioner, who reported to the Agricultural Council.

The successive enlargements had increased the diversity of agricultural interests in the EU. France and Germany no longer occupied the two most extreme positions on agricultural and agricultural trade issues. Rather than from Germany, the strongest support for a reform of the CAP and agricultural trade liberalisation came from Britain, which has a very small and relatively productive agricultural sector, followed by Holland and Denmark, both of which have very competitive, export-oriented agricultural sectors and could survive well with a more liberal and less expensive agricultural policy. The strongest supporters of the status quo was France and Ireland. As in the 1960s, the German government was divided on these issues and its position in Brussels poorly coordinated. On the one hand, the Economics Ministry was extremely critical of the CAP and wanted the EU to make extensive concessions on agricultural trade in order to secure a Uruguay Round agreement. On the other hand, the Agricultural Ministry opposed major changes in the CAP and was very reluctant for the EU to make agricultural trade concessions in the GATT talks. If, at this level, its position coincided closely with that of the French government, on concrete issues there were important differences, which had their roots in the greater inefficiency and weaker world market orientation of German, compared with French, agriculture. On the issue of how production ought to be curbed that lay at the heart of the CAP reform debate, the German minister opposed price cuts, which the French government supported, and instead supported administrative measures to take land out of production ('set-aside'), which his French counterpart opposed. On the issues of market-opening (increasing imports of foreign agricultural produce into the EU) and 're-balancing' (imposing tariffs on hitherto duty-free imports of foreign agricultural produce in exchange for reducing tariffs on imports of produce on which duties were already imposed), there was little difference between the positions of the two agricultural ministries, which were against market-opening and for 're-balancing'. However, compared with Bonn, Paris was more prepared to contemplate cuts in the subsidies paid to EU farmers, while Bonn was more amenable to cuts in subsidies for exports beyond the EU - from which French farmers benefited much more extensively than their German counterparts.

On top of these differences, there was a profound divergence in attitudes of civil society in the two countries towards trade liberalisation and the importance of the protection of agriculture. In Germany, the DBV was the only relevant organised interest that opposed trade liberalisation. Via the Economics Ministry and the Chancellor, organised business interests tried to exert their influence to ensure that the Uruguay Round did not collapse on conflicts over agricultural trade. All the relevant political parties - from the moderate right to the Greens on the left - basically supported trade liberalisation. In France, the situation was reversed. The political Left and Right were divided over the Uruguay Round between strong liberalisation opponents and moderate supporters. The principal peak organisation of business, the CNPF (Confédération Nationale du Patronat Français - French National Employers' Confederation) was split between liberal and protectionist member associations; the trade unions were also split. The farmers' associations, on the other hand, were united in their hostility to agricultural trade liberalisation. Public opinion on trade liberalisation, especially agricultural trade liberalisation, was predominantly hostile, pressing the government to stand firm and to risk, if necessary, the talks' collapse.

1. The failure of the Brussels talks 1990

Very little progress had been made in agricultural trade liberalisation talks by autumn 1990, shortly before the Brussels GATT summit at which, originally, the round had been scheduled to finish. Up to this point, the Brussels Commission had negotiated on the basis of an 'overall approach', practically an informal mandate, worked out between the Commission and the Dutch council presidency in 1985 and so called to "pacify the French" (interview). In September 1990 the Commission finally proposed a formal mandate, which it asked the Agricultural Council to approve (51). Numerous agricultural ministers found the Commission's proposals to be too radical. The council met on no fewer than seven occasions to discuss them. Only Britain, Holland and Denmark gave the proposals their full support (Agra Europe, no. 1410, 12 October 1990: E/4). The strongest opposition to them came in the first two meetings from the German minister, Kiechle, who, under pressure from DBV, was particularly worried about the implications of 'tarification', the conversion of physical and other non-tariff barriers to agricultural imports to the EU into normal tariffs (interview). The intensity of Kiechle's opposition, or at least the 'cover' that he enjoyed from Chancellor Kohl, may be explained by the fact that federal elections, the first in the united Germany, were to take place in early December: "The last thing that Chancellor Kohl wanted to do in such circumstances was to alienate the farming lobby within the Christian Democratic party" (Paemen and Bensch 1995: 178; interviews) (52). Encouraged by the German stance, the French minister added his voice and opposition to Kiechle's. The formation of this Franco-German alliance, which was supported by Ireland, explained why so many meetings were required for the council to approve a mandate, one which was finally "much watered-down", compared with the Commission's proposal, and provoked a "barrage of criticism" among the EU's negotiating partners (Paemen and Bensch 1995: 178; Moyer 1993a: 107-111; Agra Europe No. 1410, 12 October 1990: E/4-5; no.1411, 19 October 1990: P/3 and E/1; no. 1413, 2 November 1990: P/1; and no. 1414, 9 November 1990: P/1-2). Kohl is alleged to have instructed the German delegation in the decisive meeting to support the French demands "to the hilt" (Frankfurter Rundschau, 7 November 1990). Given the extreme divergence between the positions of the EU, which was prepared to offer agricultural subsidy cuts of no more than 15 per cent, and the US, which demands cuts of 75 to 90 per cent, the failure of the Brussels summit was practically a foregone conclusion (53).

2. The CAP reform

1992 Fiscal pressures prompted the first attempts to reform the CAP in the 1980s. In 1988, as the CAP threatened to 'break' the EU budget, a so-called 'stabilisers package' was adopted, foreseeing reductions in support prices if production exceeded agreed limits. According to Patterson, more radical measures were prevented by the French and German governments, in particular by their respective leaders' apprehension that they would be sanctioned by farmers in pending elections if they were to acquiesce in a thoroughgoing CAP reform (Patterson 1997: 147-149) (54). The package failed to break the CAP's spending dynamic, so that, already by 1990, the issue of what to do to curb spiralling CAP costs was back on the EU agenda.

Agricultural Commissioner MacSharry actually published first proposals for a fresh reform of the CAP on the day the Brussels GATT summit opened. The timing was no coincidence. Despite official denials and despite the internal, fiscal pressures for changes in the CAP, the "broad thrust" and, especially, the timing of the reform were dictated primarily by the pressures exerted on the EU in the Uruguay Round (interview; Paemen and Bensch 1995: 194). First discussions about a new, more radical reform took place in the Commission DG VI in early 1990. Subsequently, DG VI conceived and drafted the proposals "in consultation with the important member states" (interview). Between MacSharry and Delors, there appears to have been consensus over the aim of the reform, which was intended to safeguard the European "agricultural model" by keeping small farmers on the land, while, especially by reducing European surpluses, making the CAP more compatible with the "imperatives" of international trade (interviews; Ross 1995: 107-115) (55). MacSharry's ideas were discussed and approved by the Commission in 1991: essentially, for cereals at least, they involved much more radical price cuts than had ever previously been proposed (35 per cent), in combination with the introduction of direct payments to smaller farms in place of price support and making such payments conditional upon land being taken out of production ('set aside'). In the Agricultural Council, the ideas were received correspondingly positively by countries with mainly small farms, including Portugal, Spain, Italy and Greece, and with greatest hostility by Britain, Denmark, and Holland, which had larger or at least more productive farms (Handelsblatt: 6 February and 17 July 1991) (56). The authors of the MacSharry Plan calculated that they could do without the support of Britain, Denmark and Holland, but hoped to gain that of France and Germany. As Moyer points out (1993b: 15), this made good sense in terms of internal EU politics, as the three North Sea states did not have a blocking minority in the council, while France and Germany, if supported by Ireland, as they typically were, did.

The French and German agricultural ministers' initial reactions to the reform proposals were, however, negative, albeit for different reasons, the French minister rejecting the set-aside components and the German minister the price cuts, at least in the foreseen magnitude. Nonetheless, no Franco-German front against the reform developed, although Kiechle reportedly tried to build one (interview). Rather, the debate in the Agricultural Council, until shortly before the decisive council negotiations at least, was "dominated" by the conflict between the two states (interview). The divergence between French and German objectives may have facilitated the eventual adoption of the reform by enabling the Commission to play the two states off against each other. Certainly, the reform was also facilitated, however, by the growing internal and external pressures for change, one consequence of which was that the agricultural ministers' control over agricultural policy was diminishing in both Bonn and Paris. In autumn 1991 - a sign that Kiechle's position in the German government was growing weaker - Bonn signalled that it would agree to farm price cuts, at least provided farmers were compensated for such cuts by direct payments (see below). It is possible that, around the same time, the French government accepted the principle of a CAP reform, although, publicly, the Agricultural Minister continued to oppose it up until the last minute (Le Theule and Litvan 1993: 769; Lemaître 1991).

At the highest political levels in Bonn and Paris, the need for a reform of the CAP had been accepted much earlier. Chancellor Kohl and President Mitterrand discussed CAP reform and the Uruguay Round in the presence of their agricultural policy advisers in February 1991, soon after the publication of the Commission's proposals. According to a French official, the two leaders reached an "informal understanding" on issues of both procedure and policy content. On procedure, they agreed to put the Uruguay Round talks "on hold" and first to reform the CAP. On content, they agreed that, in line with the Commission's (and also the French ) approach, the German government should accept a cereal price cut and that, in exchange, the French government would not ultimately block a Uruguay Round agreement (interview). The subsequent course of events concerning the CAP reform can be reconciled with this informal 'deal' between Kohl and Mitterrand, provided it can be assumed that the French and German governments are not monolithic entities and the conclusion of such a deal is not therefore incompatible with the agricultural ministers' continued opposition to the reform. The existence of a Kohl-Mitterrand 'deal' encompassing the CAP reform is corroborated by reports that ultimately Mitterrand obliged Mermaz and Kohl instructed Kiechle to accept the reform (interviews; Die Welt, 24 May 1992).

The CAP reform was adopted in May 1992. The Commission's basic reform programme survived the final council negotiations, but was diluted in significant details. Over MacSharry's opposition, the Portuguese council presidency proposed that the cereals price be cut by 27 per cent and that large farm(er)s too be compensated for the cut by direct payments. A spokesman for the presidency explained: "We have to distance ourselves from the commission's original proposals in order to get a consensus among the member states" (Financial Times, 29/30 April 1992) (57). The reduction of the price cut proposed by MacSharry represented a concession primarily to Germany. Of all the member states, among which Britain, Holland and Denmark wanted an even deeper price cut than MacSharry and otherwise only Belgium and Luxembourg regarded it as too radical, Germany had opposed this proposal most vigorously (Financial Times, 19 May 1992; Agra Europe, no. 1489, 1 May 1992: E/6 and no. 1492, 22 May 1992: P/12). It had not wanted to consent to a price cut exceeding 15 per cent (interview). It may have acquiesced in a substantially larger cut in exchange for French support for the German position on beef policy, as part of a Franco-German accord on the reform reached shortly before the final council negotiations (Le Theule and Litvan 1993: 776; interview). Two other significant changes were made in the council to MacSharry's concept. First, his proposals concerning set-aside requirements were significantly weakened - this was presumably a concession to France, which was the strongest critic of this component of the reform package. Second, direct financial compensation for price cuts was extended to large as well as small farm(er)s. This amendment was pushed most strongly by the British Agricultural Minister, but was supported as well by the French and German ministers, both of whom had groups of farmers which could benefit from such an amendment - Kiechle, for example, following German reunification, in former East Germany (interviews; Financial Times, 22 May 1992). In the final council vote, only Italy voted against the amended package.

The approval of the CAP reform represented a major political victory for MacSharry and the Commission, which, in collaboration with the Portuguese council presidency, appears to have shrewdly exploited its room for manoeuvre between the member states and their conflicting preferences and to have averted the formation of a blocking minority of reform opponents. The outcome does not constitute proof, however, of the Commission's "autonomy" vis-a-vis the member states or a "defeat" for France and Germany. Rather, the fact that a reform which, in the past, would have been politically impossible had now been adopted reflected changes in the political constellation, both inside the EU (fiscal crisis of the CAP) and in its external relations (Uruguay Round). These changes had been translated - within the respective national governments - into irresistible pressures on the agricultural ministers to contemplate and acquiesce in measures which, in the past, they would certainly have or indeed had blocked. The reform was conceived with a view to winning Franco-German support and, despite a prolonged conflict between the French and German agricultural ministers over it, its adoption was facilitated by a late bilateral deal between Bonn and Paris.

3. The Blair House accord 1992

For the Uruguay Round, which had more or less marked time during the CAP reform debate, the EU's reform decisions were a "real boon" (Paemen and Bensch 1995: 210). They promised to reduce the EU's agricultural production surplus and thus the cause of the trade conflicts between the EU and the other big agricultural-exporting states. Moreover, they were a sign that, within the EU, the political balance had begun to move in favour of making concessions in the Uruguay Round. In particular, following the collapse of the 1990 GATT summit in Brussels, the position of the German government, which at that time had coalesced with the French to oppose major cuts in EU agricultural subsidies, had started to shift. The clearest evidence of this shift was a Cabinet decision in October 1991, which the Liberal Economics Minister, Möllemann, claimed to have instigated and which stated that the German government would encourage the Commission to accept specific, binding decisions relating to internal support, external protection and export subsidies in the Uruguay Round and that it would itself accept price cuts in the CAP reform, provided, however, that farmers were fully compensated for these by direct payments (Deutscher Bundestag 1991: 5188; 3829-3835). Chancellor Kohl, who had aligned himself so closely with German farming interests at the time of the Brussels summit, reputedly argued now that a failure of the Uruguay Round would be a "catastrophe" (Deutscher Bundestag 1991: 3835). The German Cabinet encouraged the Commission to try to bring about a successful conclusion of the round by the end of the year (Deutscher Bundestag 1991: 3829).

The signs of greater German flexibility on agricultural subsidy reductions seems to have induced a similar change in the EU position and given some new impetus to the Uruguay Round talks (Financial Times, 12/13 October 1991). At a European-US summit in The Hague in November 1991, the two sides made significant progress towards narrowing their differences over protection and domestic and export subsidies. The US moderated its demands for cuts in subsidies, while, for the EU, the Dutch Prime Minister, as council president, offered bigger cuts than the EU had rejected the previous year. But the gap between the US and, on the EU side, France in particular, remained unbridgeable.

Any hopes that the CAP reform would facilitate a rapid relaunching or indeed conclusion of the GATT negotiations were dashed as the Uruguay Round became entangled with the burgeoning debate and crisis in the EU over the ratification of the Maastricht Treaty. The French president Mitterrand reacted to the treaty's defeat in a referendum in Denmark in June 1992 by calling for a referendum on the treaty in France in September. Before then, the government in Paris was not going to do anything which might endanger the treaty's ratification. Amongst other things, especially following France's acceptance of the CAP reform, which prompted massive protests in the country, this meant avoiding any decisions that could further antagonise farming interests. The first consequence of Mitterrand's decision was thus to rule out any major French concessions on the Uruguay Round at the Munich G7 summit in July. The summit allegedly came "close to resolving the deadlock, but President Mitterrand was able to persuade his fellow heads of government that a breakthrough in Munich could jeopardise a French 'yes' vote" in the September referendum (Agra Europe, no. 1501, 24 July 1992: P/2).

Once the referendum was out of the way, however, the pressure on the Commission to try to bring about a GATT agreement intensified (58). It came first and foremost from the German government, which was increasingly worried about the economic conjuncture and itself was under pressure from industrial lobbies not to let the Uruguay Round fail (interview). Once the German bandwagon started rolling, the Commission "jumped aboard" (interview). The CAP reform behind him, Commissioner MacSharry was now keen to unblock the log-jam over the GATT. The divergence between his approach and that of the External Relations Commissioner Andriessen was no longer as great as it had been in 1990 (59).

The failure of fresh talks over agriculture in Chicago the weekend before the US presidential elections brought the conflict not only between the EU and the US, but also within the EU, to a head. The issue that almost precipitated a trans-Atlantic trade concerned oilseeds (soya beans). Following a GATT disputes panel recommendation condemning EU oilseeds subsidies that the EU refused to accept, the US insisted that the EU impose an upper limit on the volume of oilseeds it produced. The collapse of the Chicago talks over this issue prompted the American administration to threaten to impose penal tariffs (200 per cent) on a range of EU agricultural exports to the US within 30 days if the conflict was not meanwhile resolved.

The American trade sanction threat, together with the failure of the Chicago talks, led to a bitter conflict within the Commission and to an important Franco-German split in the EU council. Angered at Commission President Delors' interference in his conduct of the Chicago negotiations, MacSharry, who, unlike Delors, would like to have clinched a deal at Chicago, resigned - temporarily, as it transpired - as the Commission's chief agricultural trade negotiator. In the ensuing intra-Commission conflict over the issue, MacSharry emerged as victor, his authority vis-à-vis Delors strengthened and with greater latitude to make a fresh attempt to negotiate a deal with the Americans. At the same time, despite obtaining the support of several other, especially southern European, member states, France did not obtain a qualified majority in the council (of foreign and trade ministers) in favour of its demand for counter-retaliation against the threatened American trade sanctions. According to one source, "France had pulled out every stop to get Germany to back delay on GATT, arguing that the government would fall and rioting by French farmers would ensue if the agricultural subsidy cuts in the Uruguay Round were agreed" (Financial Times, 12 November 1992). The French Foreign Minister, Roland Dumas, argued that the GATT negotiations should be frozen while a study was undertaken of the compatibility of their interim results with the CAP reform (Der Spiegel, no. 47, 16 November 1992: 155). On this occasion, however, the German delegation aligned itself with Britain, Denmark, Holland, Luxembourg and Italy in pleading for a conciliatory approach towards the US. As suspicions intensified in Bonn and London that the French government might be bent on collapsing the Uruguay Round altogether, Kohl had apparently authorised his Liberal Foreign and Economics Ministers to oppose the French position. "For the first time", the German Economics Minister, Möllemann, was quoted as saying, "a decision went against France" (Der Spiegel, no. 47, 16 November 1992: 154). The November meeting of the Agricultural Council confirmed France's growing isolation on the Uruguay Round: support for its hard-line position "fell away, with Belgium and Spain leaving the French camp", leaving Ireland as its sole solid ally (Agra Europe, no. 1518, 20 November 1992: E/1; Financial Times, 17 November 1992 ) (60). The council insisted, however, that any agreement reached with the US be compatible with the CAP reform (Soisson 1993: 72-74).

The break in the Franco-German alliance on the Uruguay Round and the relative isolation of France in the council and Delors in the Commission bolstered MacSharry and Andriessen in their intention to try to reach an agreement with the US on the oilseeds and Uruguay Round agricultural trade conflicts (61). The Commissioners made concessions to the US for which DG VI thought that they would not have got prior approval in the council: "We had to take a lead and pull the council along with us ... We had to put the French under pressure" (interview). This time a deal was clinched (62). Andriessen thought that the Blair House accord was compatible with the most recent mandate laid down for the negotiations by the European Council and that, although it would be a "difficult" task, he would be able to persuade the council to approve it (interview). Given the external pressure being put on the EU and the internal pressure being exercised by business interests in favour of a new GATT treaty, DG VI was "fairly certain" that the member states, including France, would accept it (interview). For those member states which, like MacSharry, were most concerned to protect the EU's capacity to keep on paying domestic subsidies, to protect the "social element" of the CAP, the Blair House accord was, by and large, acceptable. These included Germany. Kiechle said that he would advise Kohl not to reject the accord, conceding that, on this issue, German and French interests were diametrically contradictory and that the government had to ask itself "how far we can take solidarity with France on a concrete issue" (Süddeutsche Zeitung, 30 November 1992). The French government, on the other hand, rejected the accord, which it claimed would require additional EU agricultural land to be 'set aside' and was therefore incompatible with the CAP reform. At the first post-Blair House council meeting, uniting foreign and agricultural ministers, several other member states - Italy, Spain, Portugal, Greece, Ireland and Belgium - also expressed worries that the accord might conflict with the CAP reform (63). None, however, seems to have backed a French demand that the GATT agricultural trade talks be frozen until agreement had been reached on all other sectors - a demand which Andriessen argued would, if accepted, destroy the Uruguay Round altogether (Agra Europe, no. 1521, 11 December 1992: E/1).

As the Blair House accord was whipped - fewer than six months before Parliamentary elections - into the maelstrom of French domestic politics, the $64 question in the European politics of the Uruguay Round became: will France dare to have recourse to the 'Luxembourg Compromise' to try to veto the Blair House accord and, with it, the EU's agreement to a new GATT treaty? Soisson had threatened the use of the veto in a meeting with Andriessen prior to the Blair House negotiations (interview) (64). After the deal had been made, the French government made the threat official, securing Parliamentary approval for a resolution stating that the government would "veto any draft agreement contrary to France's fundamental interest". However, while the government said that it would impose a veto only once a legal text on the agreement had been submitted to the council, the right-wing opposition, under pressure, given the impending elections, to outbid the government, pleaded for a tougher stance, insisting that the government veto the accord at once, although it was not clear how this could have been done.

In Brussels, the French veto threats were discounted. Like Andriessen, most observers seemed to regard them as a "bluff" (interview). They did not imagine that Mitterrand, who would remain the final arbiter of French foreign policy, even under a new, right-wing government, would authorise a French veto, which would plunge the EU into a major crisis and jeopardise the progress in the integration process of which, over the preceding decade, he had been a major co-architect (65).

4. The revision of the Blair House accord 1993

At Brussels and among the other EU member states, the French Socialist government's strategy was seen to be to 'contain' the Uruguay Round conflict and to avoid having to decide whether to try to veto an EU decision before the March 1993 elections. The Commission chose not to bring the issue to a head by submitting the Blair House accord to a vote in the council, fearing that there would be either a blocking minority against it or a constitutional crisis with unpredictable consequences if France were outvoted and had recourse to the 'Luxembourg Compromise' to veto it (interview). In any case, there was no need to bring about a decision on the accord until an agreement had been reached on an overall new GATT treaty, of which the accord, if approved by other GATT member states, would form a part. By and large, in fact, after MacSharry's and Andriessen's activism in late 1992, the Commission adopted a low profile in the ongoing conflict over the accord so as not to exacerbate it and destroy the chances of reaching a compromise within the EU (interview). For his part, apart from warning of the dangers of France retreating behind some kind of economic 'Maginot Line', the Commission president Delors also kept relatively quiet on the issue during 1993, calculating that anything he said in support of the accord might incite the new government in Paris to oppose it (interview).

The French elections brought to power a centre-right government with a massive Parliamentary majority composed of parties which had vociferously opposed the Blair House accord and which counted the farmers among their core electoral clienteles (66). President Mitterrand appointed a moderate and liberal Gaullist, Edouard Balladur, as Prime Minister partly to ensure the continuation of a 'pro-European' foreign policy. However, the new government was under powerful domestic pressure to have the accord renegotiated, if not to destroy it altogether. As one participant in the EU Uruguay Round decision-making process put it: "The right-wing parties had leaned a long way out of the window in the election campaign, so that a way had now to be found to haul them back in" (interview).

The Balladur government chose to pursue a 'double-track' policy on the Blair House accord. On the one hand, as a "pragmatic gesture of goodwill" and because it was a "good deal" for the EU (interview), it said that it would accept the oil seeds component of the agreement. On the other, it declared that the overall accord was "not acceptable in its present form", stressing the need for 're-balancing' in respect of duty-free cereal substitute imports into the EU and its opposition to aspects of the foreseen limits on subsidised agricultural exports (Le Monde, 14 May 1993). Domestically, the government was under, if anything, growing pressure not to accept the accord; internationally, this position isolated it. To try to break out of this isolation, it called for a special meeting of the EU foreign and agricultural ministers to discuss a renegotiation of the accord and began to try to assemble a coalition against the accord among the member states. The primary target of this "intense diplomatic effort" (interview) was Bonn. Balladur evidently travelled to the Franco-German summit in Bonn in August 1993, prior to the scheduled council meeting, to try to turn Kohl around and enlist his support to oblige the Commission to try to renegotiate the accord with the US. Right up to the eve of the summit, however, the German government remained deaf to French pleas for a renegotiation of Blair House. It was preparing to tell Balladur "fairly straight" that the Blair House accord should be left as it was (interview).

The official government position nonetheless masked important interministerial divisions over the stance that the government should take vis-à-vis Paris on the issue and over the issue whether the French government would finally try to veto the Blair House accord and hence a GATT agreement. The Agricultural Ministry was sympathetic to the French position (67). Its attitude, determined by historical experience, was that any constraints imposed by a GATT agreement on subsidised French exports to non-EU states would lead to increased French cereals exports to Germany (interview). It was thus in the interests of German agriculture, especially German cereal farmers, to protect the capacity of French cereal farmers to export to non-EU markets. The ministry also bemoaned the absence of any 're-balancing' provisions in the Blair House accord and feared a sharp rise in American exports of cereal substitutes to Europe (interview). 'Re-balancing' would permit tariffs to be imposed on these products, which currently entered the EU duty-free. By contrast, the Economics Ministry evidently downplayed the threat and argued that the government should stick to its existing policy of not ceding to French demands for the accord's renegotiation. Before the summit, it submitted a paper to the Federal Chancellor's Office pleading in favour of Kohl taking a "really tough" stance against the French demands (member of the presidium of the FDP, quoted Handelsblatt, 31 August 1993).

The Bonn ministries - apart at least from the leadership of the Agricultural Ministry - were astonished when, at the end of the summit, Kohl announced at a press conference that Germany, too, had "problems" with the Blair House accord and, when asked whether Germany now supported the accord's renegotiation after all, the Chancellor replied that: "We must find a compromise acceptable for everyone" (Financial Times, 27 August 1993) (68). Without saying directly that he wanted Blair House to be renegotiated, he appeared thus to support the French position. This impression was strengthened by the announcement that Balladur and he had agreed to try to reach as "similar a position as possible" in time for the upcoming Brussels council meeting (Le Monde, 28 August 1993) (69).

In Bonn, Kohl's remarks produced "alarm and confusion" in several ministries or, indeed, "chaos" (Financial Times, 27 August 1993; Handelsblatt, 30 August 1993) (70). Agricultural Ministry officials explained that the government would not oppose new negotiations on specific issues, but that it did not want to put the whole accord in question (Financial Times, 29 August 1993). The Federal Chancellor's Office clarified that Kohl was not calling for a formal renegotiation of the accord (Financial Times, 31 August 1993). Kohl's coalition partner, the FDP, which had been more critical than the Christian Democrats of French Uruguay Round policy, was also apparently annoyed by the Chancellor's intervention. Foreign Minister Kinkel told the press that the government did not want to mess around with the agreement, although he hinted at the possibility of helping the French by 're-interpreting' it (Handelsblatt, 31 August 1993). What had happened to induce Kohl to modify, if not to reverse, his position on Blair House? Two factors - one related to domestic politics, the other to the Franco-German relationship - appear to have played a role(71). First, there were the Agricultural Ministry's worries about the impact of the Blair House accord on German agriculture, especially cereal farms. Second, and probably more decisively, Kohl was impressed by Balladur's portrayal of the severe domestic political constraints that he was facing in Paris (interview). Evidently, Balladur told him that unless he succeeded in having the accord re-opened, he would be forced to veto it or otherwise his government would fall (72). Either way, the Uruguay Round would have collapsed: an outcome that would have led to a major crisis in the EU as well as in trans-Atlantic relations and, by precipitating an upsurge of protectionism, could have strengthened recessionary tendencies in the world economy and reduced Kohl's chances in the German elections to be held in 1994 (73).

Between the Franco-German summit and the special Brussels council meeting four weeks later, several meetings of high-level officials from the French and German agriculture, economics and foreign ministries and Prime Minister's or Chancellor's offices took place to discuss the French objections to the Blair House accord. The goal, for the German side, was to convince their French counterparts to moderate their demands for changes in the Blair House accord and how important it was for Germany that the GATT talks be concluded successfully; the French side's goal was to enlist German support for their demand for a re-negotiation of the accord. The French side claims to have convinced the German - "informally", at least - that Blair House would or could require cuts in EU agricultural production over and beyond those implied by the CAP reform (interview). The German ministries were actually divided in their appraisals of the implications of Blair House. While the Economics Ministry strictly rejected the French analysis, the Agricultural Ministry shared it at least partially, believing that there "could be problems" (interview). The German side appears to have admitted the validity of some French demands for changes in the accord and to have persuaded their counterparts to drop some others (interview). At least on some "points of details", however, the meetings failed to produce an agreement between the two sides, which seem also to have interpreted the results of the consultations differently (interview). Moreover, no doubt for fear of precipitating the collapse of the Uruguay Round and in line with the government policy that Kohl's press conference remarks had cast into doubt, the German delegation did not agree to the French demand for a formal renegotiation of Blair House (74).

Joint French and German delegations also saw the Belgian presidency and the Commission before the special council meeting. In the talks with the presidency, the French delegation dominated the discussion, stressing especially that, if the Blair House accord were not renegotiated, there would be a vote of no-confidence against the government in the French Parliament and the government would fall (interview). The German delegation, by contrast, remained very quiet. This pattern repeated itself at the Franco-German meeting with the Commission. Here again the Germans "had little to say". The purpose of their presence, according to one participant, was simply to "demonstrate Franco-German cohesion" to the Commission (interview).

The Franco-German attempt to reconcile their differences over Blair House was actually greeted by the Belgian presidency and by the Commission's agricultural directorate (DG VI). The Belgians, who were sympathetic to the French position, explicitly encouraged the two states to reach a bilateral agreement before the special council meeting (interview). The signs of a Franco-German rapprochement had been greeted by DG VI because, "without a coordinated strategy of the big member states", an agreement over Blair House and the GATT would, in its view, have been impossible (interview). Concerned to calm the situation and to avert a crisis in the council and a possible legal challenge to a new GATT treaty, the Commission agreed that the council should take a consensual, rather than qualified majority, decision on Blair House and the Uruguay Round (interview; Woolcock and Hodges 1996: 319). This decision notwithstanding, the Trade Commissioner Sir Leon Brittan, the leader of the Commission's Uruguay Round negotiations, was extremely hostile to the idea of being sent back to renegotiate the accord with the US with a 'shopping list' of new demands from the council. He was prepared to talk with the US again only if he could maintain a "free hand" and was not obliged to raise concrete issues (interview).

Despite the Franco-German bilateral mediation efforts, the situation immediately before the council meeting was still unclear. The Commission itself was divided: the Agricultural Commissioner, René Steichen, told Brittan that he would not be able to maintain his opposition to new talks on Blair House and that he would have to compromise with the French (interview). The German government's position appeared to be open. Whilst having signalled its support for some French demands for changes in the accord, it remained opposed to a formal renegotiation: the Belgian presidency received "very contradictory signals" from the German government, which it perceived to be divided between Kohl and his immediate advisers, on the one hand, and the "hard-liners" in the Economics Ministry, on the other (interview). In the last meeting of the 113 Committee before the council, Ireland was the only state to offer France full support, although more qualified support came from Belgium and Italy (interview). The French government was not completely certain that it could secure a majority for its Blair House renegotiation plea in the council (interview). While the German government kept a very low profile, the French redoubled its efforts to influence the Belgian presidency, which was fearful that the meeting would fail and that this would lead to a major split between 'protectionist' and 'free-trading' member states that could, in turn, cause lasting damage to the integration process itself (interview). France was also busy searching for allies among the other member states by pledging to back them on Uruguay Round or agricultural policy issues in which they had important stakes (interview). It counted on the support of Belgium, Ireland and the southern member states, less on that of Italy than on that of the other three - Spain, Portugal and Greece (75).

The council meeting on 20 September developed quickly into a confrontation between Brittan and the French foreign minister, Alain Juppé. Brittan refused to try to renegotiate the Blair House accord with the US. Although he referred to the possibility of trying to 'clarify' the accord, he was strictly opposed to having to raising precise, concrete issues with the Americans. Juppé took a tough stand against Brittan, insisting that there must be a re-opening of the talks on Blair House. In the earlier part of the meeting, the French position was backed, to a greater or lesser extent, by the southern member states and Ireland, but opposed by Holland, Denmark, Luxembourg and, not least, Britain, whose Foreign Minister, Douglas Hurd, before the council, had obliquely threatened that his government would precipitate a crisis in the EU if France should succeed in its demand to have the Blair House accord renegotiated (76). Whilst the German delegation gave observers the impression of being divided, the delegation leader, the Foreign Minister, Klaus Kinkel, clearly took Brittan's side against Juppé at the outset, stating that while Germany was ready to support consultations designed to 'clarify' the accord, it opposed its being renegotiated as such.

The constellation in the council began to shift, however, after Juppé moderated his demand for a renegotiation of the accord, arguing that terminological issues were secondary. The actual conduct of fresh talks with the US and their results were more important than how such talks were labelled. At this point of the meeting, with Juppé having climbed down from the French demand for a formal renegotiation of Blair House and Brittan continuing to oppose the idea of going back to the US with a "shopping list" of concrete proposals for changes in the accord, the German delegation, apparently fearing that the meeting could end in failure, swung in behind the French (interviews). According to one participant in the meeting, Kohl's economic policy adviser, Johannes Ludewig, was responsible for the German delegation's decision to back the French government against the Commission - he may have been instructed by Kohl to ensure that the Uruguay Round did not precipitate a crisis in Franco-German relations (interview) (77). In any case, the German U-turn left a group of only four member states - Britain, Holland, Denmark and Luxembourg - opposed to instructing the Commission to conduct new talks with the US over Blair House. The critical question now became whether the resistance of Britain and the other three states to new talks on the accord could be overcome. The differences between France and Germany, on the one hand, and Britain and the Commission, on the other, were discussed during two restricted council sessions, in which otherwise only the Belgian presidency participated. In the first of these sessions, the French and German delegations produced a joint paper on Blair House, which, after some modifications proposed by the Commission and the Belgian presidency, was submitted to the entire council in place of the presidency's draft resolution, which the French delegation found too weak (interviews). The paper listed almost all French reservations concerning Blair House as well as numerous agricultural policy proposals that the southern member states wanted to see included and France had promised to support in exchange for their backing the French demands for new talks on Blair House. In effect, if not formally, the paper demanded a renegotiation of the accord (78). Despite this, the British delegation followed the German and acquiesced in the paper, Hurd playing, according to one participant, an "extremely useful" role in bringing about an agreement (interviews). The British delegation saw in the fact that Brittan would conduct - or at least oversee - any new negotiations with the US a safeguard against a possible collapse of the Uruguay Round (interviews) (79). Britain, in turn, took the other three anti-Blair House reform states with it, so that the council could adopt a unanimous decision - albeit not the decision wanted by Brittan. According to numerous participants in the EU decision-making process, the September 20 council meeting was the "turning-point" in the Uruguay Round negotiations (interviews) (80).

5. Discussion

Whether France and Germany were united or divided in their attitudes to agricultural trade liberalisation was decisive for the policy course that the EU followed. Their joint opposition, combined with that of Ireland, was largely responsible for the failure of the EU to make an offer acceptable to the other GATT member states in Brussels in 1990. The joint opposition of the French and German agricultural ministers threatened initially also to defeat the Commission's CAP reform proposals in 1991-92. However, a late Franco-German agreement over the reform appears to have played a critical role in facilitating its approval by the Agricultural Council and to have been facilitated itself by the interventions of Kohl and Mitterrand, who had made a prior deal involving mutual concessions on the CAP reform and the Uruguay Round. When the two governments fell out over the Blair House accord negotiated with the US by the Commission and the French government opposed it, the Uruguay Round policy-making process in the EU ground to a standstill that lasted almost a year and was overcome only after, at the September 1993 council meeting, the German government rallied to the support of the French after the latter had reached from its hitherto 'maximalist' position, taking the remaining 'pro-Blair House' member states and the Commission along with it. Especially in the case of the revision of the Blair House accord, but also in that of the CAP reform, bilateral Franco-German mediation seems to have played a decisive role in shaping EU decisions. It does not seem to be an exaggeration to conclude that, in the European politics of the Uruguay Round, as one participant in the process argued, the Franco-German relationship was "absolutely determining" (interview).

This modified intergovernmentalist explanation seems to account much better for the development of EU policy on the CAP and the Uruguay Round than any based on other models or images of EU politics. For reasons similar to those which applied concerning the conflicts over the foundation of the CAP, a neo-functionalist explanation is hardly plausible. Again, supranational organs such as the EP, the European Court of Justice (ECJ) and the COPA played no or, at best, marginal roles in the CAP reform and Uruguay Round policy-making process. Although EP was regularly briefed by the Commission on the Uruguay Round and had the power to veto a new GATT treaty under Article 228.3 of the Rome Treaty, it had no impact on the decision-making process (81). Nobody in the council expressed the fear that the parliament might reject the treaty if the council did not (interview). In fact, it very quickly approved the treaty. The COPA, which opposed the Blair House accord and contested its compatibility with the CAP reform, was similarly marginal to the Brussels policy process, hamstrung by its incapacity to reconcile the conflicting interests of its national member associations (interviews).

As in the early 1960s, a neo-functionalist explanation of EU agricultural politics in the 1990s would have to be able to show that an autonomous Commission was the principal player in the four conflicts. The Commission was indeed a central actor in the conflicts, given its powers of policy initiation (Article 155 of the Treaty) and conduct of trade negotiations with third countries (Article 113 of the Treaty). It conceived the CAP reform and piloted its proposals through the Agricultural Council without having to make concessions that changed or diluted the reform's basic thrust. It exploited its trade negotiation role to conclude the Blair House accord without securing prior approval of its probable contents from the member states in the council, calculating that even if it would not have found a qualified majority for an accord before it was reached, it would be able to do so afterwards (interview). But in neither case, constrained as it was by the need to obtain council approval for its proposals or agreements, did it sponsor proposals that were conceived independently or autonomously of the member states. As in the early 1960s, its action was strongly conditioned by its perception of what was politically feasible. What was atypical about the conclusion of the Blair House accord was that the responsible Commissioners calculated that they could bypass France - a calculation, however, that turned out to be misplaced, in as far as, at its September 1993 meeting, the council forced the Commission to retreat and to re-open talks on the accord with the US as the French government had insisted (82). The Commissioners' relative boldness in this case was facilitated by their analysis of the balance of power within and between the member states and between the EU and its trading partners, especially the US: there was a "strong tendency" in the council at the time pushing for a successful completion of the round, pressure from European business interests in favour of a GATT agreement and "external pressure" as well (interviews). This constellation encouraged the commissioners to believe that, if the council was presented with a fait accompli on agricultural trade, a qualified majority would accept it.

An important reason why the Commission was arguably unable to pursue a more autonomous strategy vis-à-vis the national governments was that, in contrast to what is implied in some analyses of the EU policy process, it was far from being a unitary actor in the GATT and CAP conflicts. The Commission's GATT policy was coordinated between the Agriculture Commissioner MacSharry (in 1993, the former Luxembourg Agriculture Minister, Steichen), Andriessen (in 1993, Brittan, formerly Competition Commissioner), Delors and the German Commissioner responsible for internal market affairs, Martin Bangemann. In general, MacSharry and Delors were hostile to making concessions to the US on agricultural trade, while Andriessen, Brittan and Bangemann were in favour. However, once the CAP reform had been approved, MacSharry, who was less worried about maintaining export subsidies than domestic subsidies, became more open to a Uruguay Round settlement - a shift of position that precipitated his conflict with Delors in autumn 1992. It was only once Delors had been isolated in the Commission that the Blair House accord could be negotiated. The fact, moreover, that, for most of the period of the Uruguay Round negotiations, the Commissioners' positions corresponded closely to that of the governments that appointed them indicates that, far from standing above the member states and articulating a 'European' interest in the conflict, the Commission was rather a mirror of the divisions between the states. The agricultural policy conflicts took place less between the Commission, on the one hand, and the council or member states, on the other, than between different member-state-based groups, factions or coalitions embracing actors in both the council and the commission (whereby, among the member governments, notably the German, with its notoriously loose system of interministerial coordination on European policy, was characteristically divided) (83).

The above findings are also difficult to accommodate within the transnational exchange model, which posits that, with the growth of transnational society, EU organisations and EU rules, "the influence of EC (or supranational) organizations on policy-making processes and outcomes increases" (Stone Sweet and Sandholtz 1997: 304). The outcomes of the series of conflicts over agricultural (trade) policy in the early 1990s suggest that in these policy domains - despite the facts that, from early on, agriculture and external trade belonged to the most highly 'Europeanised' policy domains and transnational agricultural exchange has certainly grown - the Commission and other supranational EU organs are no more influential, and 'intergovernmental bargaining' is no less ubiquitous, than was the case in the 1960s. The failure of the Commission to win an extension of its trade policy-making prerogatives and the rejection of the EP's demand for powers of agricultural policy co-determination at the 1996-97 Intergovernmental Conference point, at the level of the formal 'rules of the game', to the prevalence of continuity in the balance of forces between national and supranational actors: actually, during the entire period since the adoption of the Rome Treaty, there has been no noteworthy change in the formal procedures for making agricultural and trade policy decisions in the EU (84).

The multi-level governance model likewise over-emphasizes the significance of the role played in agricultural (trade) policy-making by supranational organs. Of course, the Commission, by virtue of its policy initiation and trade negotiation powers, has some independent influence on policy outcomes. The evidence from the Uruguay Round conflict(s), however, suggest that its influence was modest, albeit agriculture and trade are policy areas in which decision-making competencies have been most extensively 'Europeanised' (85). Although the Commission's powers enable it to set the agenda and to structure the decision-making situation, its exercise of these powers seems to be strongly shaped by its reading of the balance of power between rival interests and member states as well as by external pressures, while the council - which is to say, the member states acting collectively - has defended its decision-making prerogatives. Moreover, in the conflict towards the end of the Uruguay Round negotiations, the French government successfully re-asserted its right of veto on issues relating to 'vital' national interests, to the extent that the other member states, led by Germany, were unwilling to call its bluff and risk a crisis in the EU by approving a GATT treaty against prospective French opposition. It is therefore questionable whether the integration process has eroded national sovereignty to the degree that the multi-level governance model contends. Although it may no longer be as credible for a member state such as France to threaten to secede from the EU as it was in the 1960s, the Commission and other member states may nonetheless refrain from outvoting or isolating it for fear that such a step could provoke a crisis that would seriously damage the integration process (86).

The other main claim of multi-level governance proponents is that national governments have ceased to monopolise the representation of organised national interests vis-à-vis the supranational organs, which have developed their own direct links with interest groups in the member states. Again, the issue here is not whether there are any direct links at all between national interest groups and the Commission. The issue is how significant they are, compared with those between organised interests and the member governments and between the latter and the Commission. It is clear that, in the French case, there were frequent consultations between the Agricultural Minister and the French farmers' organisations, although the influence of the farmers' lobby on French GATT policy seems to have diminished towards the end of the Uruguay Round, when other priorities became more important than accommodating the government's agricultural 'clientele' (see Epstein 1997) (87). There is no evidence of comparably close links and consultations between national farmers' groups and the Commission's DG VI, for which the principal interlocuteurs were not the national farmers' associations (or the COPA), but rather the ministers in the Agricultural Council (interview). The role played by DG VI in formulating the CAP reform and negotiating the Blair House accord shows that it is not 'captured' by agricultural lobbies to the same extent as are the national agricultural ministries. That the national agricultural ministers should consult more closely with national farmers' organisations than does the Commission is not surprising in as far as the farmers can sanction national governments by staging protest actions or withdrawing electoral support, whereas they can not sanction the (unelected) Commission.

In as far as most EU agricultural policy decisions - the less controversial ones - are taken at the level of the SCA, which is composed of high-level national civil servants and in which a DG VI official also participates, and then merely 'rubber-stamped' by the Council of Ministers, EU agricultural politics may be described as bureaucratic politics. The same is true of trade policy issues, the bulk of which are resolved in the '113 Committee', also composed of high-level national civil servants (88). However, although this model may convey an accurate picture of who takes part in making 'ordinary' agricultural and trade policy decisions, it tells us nothing about the distribution of influence on decisions as between supranational and national actors or between different national, or coalitions of national, actors. Moreover, it would be an entirely misleading characterisation of the agricultural politics of the Uruguay Round, during which, especially from 1990 onwards, the issues at stake were politically highly contested and could be mediated only by the intervention of actors - elected office-holders - at the very highest political levels. This is not to say that the fine details of agreements such as that of Blair House were not largely 'pre-negotiated' by high-level civil servants on both sides of the Atlantic, but rather that the latter were operating under tight constraints set by political superiors (the agricultural ministers and the EU Agricultural Commissioner) well-versed in the implications of alternative possible outcomes (89). It would be equally misplaced to think that Uruguay Round and CAP reform decision-making in the early 1990s was dominated by a cohesive, tightly-knit agricultural 'policy community', albeit the COREPER, which could bring a non-sectoral perspective to agricultural issues is marginalised in agricultural policy-making, the Agricultural Council is marked by a strong esprit de corps, such that ministers "often regard their Council colleagues as allies and their government colleagues responsible for Finance or Trade ... as potential adversaries", and the ministers are typically very reluctant to relinquish control of agricultural dossiers to, or share it with, other councils (Culley 1995: 200; van Schendelen 1996: 541; interviews). Rather, and despite the fact that the GATT agricultural trade negotiations were conducted for the Commission by the Agricultural DG VI, the history of the first half of the 1990s is one of the progressive subordination of purely agricultural policy to wider budgetary, trade and/or foreign policy priorities, as agricultural ministers first acquiesced in a reform of the CAP and then in a GATT agreement curtailing the EU's scope to subsidise agricultural exports and protect its farmers from non-EU competition (90). As in the early 1960s, no 'epistemic community' appears to have had any major impact on the decision-making process, although, through international organisations such as the OECD, agricultural economists may have exercised some influence on the proposals made for agricultural policy reform (Grant 1997: 171) (91).

Compared with the above models, the institutionalist one comes much closer to providing a satisfactory explanation of the outcomes of the agricultural (trade) policy conflicts of the early 1990s. An institutionalist analysis would emphasize the significance of informal as well as formal 'rules of the game' and argue that both operated to preserve the position of national governments and to constrain the growth of influence of supranational actors. The maintenance, since the adoption of the Rome Treaty, of the consultation procedure, under which the Commission proposes and the Council of Ministers makes decisions, has kept the EP on the sidelines of agricultural (and trade) policy-making. In these areas, there have been no ECJ decisions that have carved out more important roles for the supranational organs as there have been in some others, such as telecommunications (92). Other things being equal, the increased resort from the early 1980s onwards to qualified majority voting in the council could have enhanced the influence of the Commission by removing the need for it to accommodate the interests of every member government in its proposals, so that, in the case of the Uruguay Round, for example, it might have disregarded the French government's opposition and pressed for the approval of the new GATT treaty, incorporating an unrevised Blair House accord, by a qualified majority vote. To have thus isolated France would, however, have violated important informal norms about the council's mode of operation. In the Agricultural Council, these have come to favour decision-making by issue-linking 'package deals' that accommodate the interests of the maximum possible number of member states (Culley 1995: 204-207) (93). Qualified majority voting has increased with the number of member states, but even on issues that could be decided by a qualified majority, ministers typically continue to search for a consensus and avoid a vote (interview; van Schendelen 1995: 541). This practice reflects a general norm in the operation of the council of ministers that member states that are strongly opposed to a given project should not be "ground into the dust" (interview), that an effort should be made to make it acceptable to them (but also that, after they have attained some concessions, they should settle). France benefited from the application of this norm in the conflict over the Blair House accord in 1993 - as well as from legal uncertainty as to whether the council could approve the treaty by a qualified majority vote and the perception, widely shared in the EU at the time of the Maastricht Treaty ratification crises, that a confrontation with France over the Uruguay Round might do lasting damage to the integration process. As in the case of the conflicts of the early 1960s, the principal weakness of the institutionalist model is that it offers no evident explanation of why member states possess differential capacities to shape the EU policy process. Why were France and Germany the two member states whose positions appear to have been decisive for EU policy on the agricultural trade components of the Uruguay Round and the CAP reform? Why could French resistance to agricultural trade liberalisation generate a much more intense and longer crisis in the Uruguay Round than, for example, Portuguese resistance to textile trade liberalisation, which, at the cost of the provision of temporary subsidies, was very quickly broken? Could any other member states than Germany have secured French acquiescence in a new GATT round and treaty? Could any other combination of member states than France and Germany have generated the same 'magnet effect' on the remaining members once they had reached a modus vivendi over the CAP reform or the Uruguay Round? To these questions the institutionalist model does not appear to furnish any answers.

IV. Conclusions

Much changed in the EU between the early 1960s and the early 1990s. In the early 1990s, the union contained 12 member states, twice as many as 30 years earlier. There had been significant changes in formal decision-making rules, especially through the Single European Act. The union was also a much more consolidated entity, and the idea that any member state might decide to leave it much more improbable, than it had been three decades previously. None of these changes seemed, however, to have significantly altered the centrality of France and Germany and their relationship to agricultural 'high politics' in the EU. In both eras, in almost all major agricultural policy conflicts, France and Germany initially took opposing sides. When their differences were not mediated and they remained divided, the outcome was a crisis and deadlock in the decision-making process, as in 1965-66 and following the negotiation of the Blair House accord (94). If they were united in opposing a given project, such as an EU offer of substantial agricultural trade liberalisation prior to the Brussels GATT summit in 1990, their position was determining for the stance of the EU as a whole. If they finally reached a modus vivendi on a given project, such as the launching of the CAP in 1962, the common cereals price in 1964, the CAP reform in 1992 and the revision of the Blair House accord in 1993, the EU took over the common Franco-German position. Thus, whether divided over, or united in opposing or supporting, a given initiative, the role of France and Germany and the Franco-German relationship in EU agricultural crisis politics has been decisive.

The point that, in almost all of these conflicts, the French and German governments were, at the outset, the chief or at least leading protagonists is worth stressing. Franco-German cooperation in agricultural policy-making, to the extent that it exists, is not the cooperation of natural allies with parallel agricultural structures, interests and preferences. On the contrary - France is a net food-exporting country, Germany is a net food-importer; France had and has, overall, a relatively strong and efficient agricultural sector confident of its capacity to compete in a single European market, Germany had and (at least up until reunification) has a comparatively weak and inefficient agricultural sector (95); France was and is a leading beneficiary of the transfers paid through the CAP, Germany is the leading contributor; France is more strongly interested than Germany in exporting agricultural produce to third markets; the two countries have a different spectrum of agricultural products; in France, alternative employment opportunities to agriculture are scarcer in the rural and semi-rural areas than in Germany; and so on. These structural differences translate into different national agricultural policy discourses, with France stressing the importance of its 'exporting mission' (vocation exportatrice) and Germany the need to maintain the 'family farm' (bäuerlicher Familienbetrieb), and into opposed positions on issues like the maintenance or reduction of subsidies for EU agricultural exports and the relative weight assigned to price cuts and administrative 'set-aside' measures as instruments for the containment of EU agricultural production and expenditure (96). Typically, it is only on policies that impose costs on third parties, such as high levels of protection for EU farmers vis-a-vis non-EU competitors, that the French and German agricultural ministers can easily agree, and these have normally been contested by the German Economics Ministry and German manufacturing industry, for fear that they will provoke trade wars and negatively affect the latter's access to the world market. Franco-German agricultural policy cooperation is thus extremely antagonistic or conflictual. That it may nonetheless be close underlines the extent to which it is driven, on both sides of the Rhine, by overarching, long-term foreign policy considerations rather than by common interests on concrete policy issues.

To what extent, however, is this pattern of (conflictual) Franco-German cooperation in agricultural crisis politics repeated in 'ordinary' agricultural policy-making and in other EU policy areas? In fact, Franco-German cooperation is not a ubiquitous trait of EU agricultural policy-making, but is practised primarily on those issues which the one or other government or both regard as important (interview). The practice of trying to agree a common Franco-German position on such issues before they are dealt with in the Agricultural Council originated after the 1965-66 crisis with the aim of averting new crises and has meanwhile been routinised in the Bonn and Paris agricultural ministries (interview). The impetus for such coordination, the concern to ensure that, as far as possible, on important issues, France and Germany find a compromise did and does not arise spontaneously in the agricultural ministries. Rather it emanates from the the "highest levels", that is to say, the respective foreign ministries, the Federal Chancellor's Office in Bonn and the Matignon and Elysée in Paris (interview). The habit of seeking a rapprochement on important issues is by no means restricted to the agricultural ministries. From 1974 on, Helmut Schmidt and Valéry Giscard d'Estaing resolved to reach a common Franco-German position on issues they regarded as important, while retaining a "degree of flexibility" on issues they rated as secondary (Giscard d'Estaing 1988: 130). According to Giscard, they always succeeded. The goal of "closely coordinating" French and German strategies on "all major issues" on the EU agenda appears to have survived to the present day, although it has clearly not always been realised, or in some cases even pursued, in practice (97).

Bilateral Franco-German coordination is arguably more intensive in EU agricultural politics than in most other policy areas (see the individual contributions in Webber 1998) (98). Several factors could account for this phenomenon. First, proportionately more EU agricultural policy than other issues are deemed to be politically important by Paris or Bonn or both. The way in which the French government upped the stakes in the agricultural policy conflicts of the early 1960s and the early 1990s and the fact that these became the most intense disputes in the EU in their respective eras testify to the high political salience of agricultural issues in France. That agricultural policy issues can also be politically explosive in Germany is illustrated by the fact that the one occasion in EU history in which the German government had recourse to the Luxembourg Compromise to try to veto a council decision - in 1985 - related to a proposed cereals price cut. The high political conflict-potential of agriculture may, in turn, be rooted in two factors. For one thing, as agriculture gave birth to the first common policy in the EU, attacks on the CAP were or could be construed for a long time as attacks on the integration process itself. For another, agriculture was and is still - in both countries - a politically highly organised and mobilised profession, which, more so in France than in Germany, has developed and preserved a high capability to exert pressure on the government by means of protests, demonstrations, civil disobedience and violence.

The second, related, factor that might help to account for the apparently more intensive bilateral coordination of agricultural policy compared with others is that agriculture is one of the most highly 'Europeanised' policy areas. In policy areas where the EU's competence is restricted or non-existent, France and Germany may, if they do not try to coordinate their policies or do, but fail to find a common position, continue to pursue their separate national policies. In agricultural policy, by contrast, if they do not reach a common position, there is either a stalemate, because no qualified majority can be found in the council, or one of the two states is outvoted and marginalised, as occurred with Germany in the banana trade conflict - an outcome which could precipitate a crisis in the bilateral relationship and/or the EU if the issue is politically explosive for the majoritised state. However, the degree of 'Europeanisation' of a given policy area can not account by itself for the intensity of bilateral policy coordination between Bonn and Paris. Thus, in trade policy, also one of the most 'Europeanised' policy areas, the intensity of Franco-German coordination does not seem to be as great in general as in agricultural policy. Here Germany, for example, cooperated informally more closely with like-minded member states in the so-called 'game of four' (Germany, the UK, Holland and Denmark) than with France (99).

The third factor that might help to explain the relatively high intensity of Franco-German agricultural policy coordination is the age of the policy area at the European level. No other area has been 'Europeanised' for such a long time as agricultural policy. Arguably, the length of time that French and German agricultural ministers and their bureaucracies have been constrained to deal with each other, on top of the strength of the EU's competencies in this area, has assisted the institutionalisation and entrenchment of practices of coordination and mutual accommodation. In the case of the CAP, the impact of the policy's age is reinforced by the intensity and frequency of the ministers' interaction in the council of ministers. None of the EU councils of ministers has been meeting so often for such long sessions for such a long time as that of the agricultural ministers: annually, only the foreign ministers meet more often (albeit for typically for shorter sessions) than the agricultural ministers and most other councils meet much less frequently than the Agricultural Council (100). The agricultural ministers in general and the French and German ministers in particular have accumulated unparalleled experience in the mediation of inter-state policy conflicts. To be sure, this does not mean that the ministries' coordination activities are necessarily successful in the sense that Franco-German differences are settled. On several occasions, agricultural ministers have had to be brought into line by the respective heads of governments, concerned for "overarching political motives" not to allow the Franco-German relationship to be destabilised (interview) (101). Indeed, the extent to which agricultural policy crises have been arbitrated and mediated by French and German political leaders is a recurrent, although not omnipresent, feature of the policy process in the 1960s and 1990s (102). In general, on both the Seine and the Rhine, sectoral ministries are less mindful of the Franco-German relationship than the foreign ministries and the Chancellor's, Prime Minister's and President's offices, in whose judgment, more often than not, the maintenance and stability of the bilateral relationship is more important as an end in itself than are concrete sectoral policy objectives or interests. It is hardly accidental that the most profound crisis in the Franco-German relationship among those portrayed above took place at a time when the French President and the German Chancellor (and Foreign Minister) no longer attached the same importance to the bilateral relationship as had been case while Adenauer had been Chancellor and as did later - in the 1970s - Giscard and Schmidt and - in the 1980s and early 1990s - Mitterrand and Kohl.

Caution is hence advisable in generalising the experience of agricultural crisis politics to 'ordinary' agricultural policy-making or other policy sectors. It should not be assumed that Franco-German coordination is as intensive in other policy areas, or that the behaviour of these two governments is as decisive for policy choices in these areas, as in agriculture. The modified intergovernmentalist interpretation of the big agricultural policy conflicts of the 1960s and 1990s advanced above may not be transferable to other policy sectors or indeed to other agricultural policy issues. At the very least, however, the preceding analysis indicates, in contrast to much recent theorising about EU politics, that intensive Franco-German bilateral coordination and leadership in EU has not been entirely superseded by the growth of 'supranational governance' and is far from being restricted to the great treaty revision debates in which, compared with 'normal' EU policy-making processes, the member states' political leaders play a dominant role and sectoral ministries are relegated to the sidelines by the Chancellor's, president's, prime minister's and foreign offices. Also, in agricultural policy, it appears to be an established routine for the council presidency to seek an agreement on the basis of a joint Franco-German position when one has been reached (interview). When the two states do act jointly in the council, "it always works", i.e. the council adopts their common position (interview) (103). Other policy areas also yield numerous cases in which Franco-German cooperation appears to have been decisive for EU decisions: for example, in monetary policy, the foundation of the European Monetary System (EMS) in the late 1970s, the European Monetary Union (EMU), and the 'Stability Pact' concerning post-EMU fiscal policy and, relating to internal market issues, the decisions concerning electricity and postal services liberalisation (104). Other member states typically find it "very difficult to resist a combined approach of France and Germany ... The cases in which they fail with common initiatives are few and far between" (interview). In this sense, France and Germany constitute the 'hard core' of the EU (cf. CDU/CSU Parliamentary Group 1994: 5) (105).

Why do other member states find it 'very difficult' , or often prefer not, to 'resist' joint Franco-German initiatives, why do so many of these appear very quickly to be multilateralised in the EU? Moravcsik (1991: 49) and Garrett (1992: 544-45) stress the capacity of the two big states - once they have agreed a joint position - to bring other states 'on board' because of the Benelux states' economic dependence on France and Germany and interest in trade liberalisation (assuming the Franco-German initiatives serve this goal) and because they offer the poorer member states financial compensation or side-payments for their acquiescence. As valid as these insights may be, the analysis here suggests that these are not the only, and may not be the most important, factors explaining the high rate of transposition of Franco-German initiatives into EU decisions. An additional factor is the weight of tradition or routine. Over four decades, Franco-German bilateralism has developed into part of the EU's 'standard operating procedure', accepted partly, more or less unquestioningly, because "it's been there such a long time" (interview) and partly also because (normally) it contributes to more efficient decision-making. A further, critical key to understanding the high capacity of France and Germany to shape EU decisions lies in the fact that the two states, as the above cases illustrate, normally stand for opposing interests, that they represent "to a certain extent the opposed poles of the Community" (Couve de Murville 1971: 262). This means that, once the French and the Germans have found a common position, provided it is located somewhere between their two originally opposed standpoints, it is unlikely to be very far away from what most other members will find acceptable (106). In the early 1960s, except in the 1965-66 conflict, Belgium, Luxembourg and, occasionally, Italy seem to have been close to the German position on the CAP, while Holland and, again occasionally, Italy took up similar positions to France. In the conflicts over the Uruguay Round, Ireland and, to a lesser extent, Belgium sympathised with the French stance; Holland, Denmark and Britain were close - once it resolved to push for EU concessions on agricultural trade and the successful conclusion of the negotiations - to Germany. The Mediterannean states, whose agricultural products were much less affected by the Uruguay Round than those of the northern member states and therefore tended toward neutrality on agricultural trade issues, could be rallied to either camp, depending on the pay-offs they were offered for their support. In the end, in the decisive council meeting of September 1993, they swung in behind France (107).

The latter council meeting actually provides a good illustration of the process by which Franco-German agreements are multilateralised. France won the support of a number of southern member states for its demand for a re-opening of the Blair House accord by pledging to support them on other agricultural and Uruguay Round issues. Seeing that France had built up a coalition that could vote down a new treaty in the council unless the Blair House accord was revised or that it could precipitate a crisis in the EU by trying to invoke the 'Luxembourg Compromise', Germany decided to back France. Once Germany had shifted, the pressure on the remaining states - Britain, Holland, Denmark and Luxembourg - to follow suit became irresistible. If they had opposed the Franco-German-led initiative to re-open the Blair House issue, it is they rather than France that would have been blamed for the ensuing crisis in the EU and the GATT. In general, on agricultural and trade policy issues, Germany has frequently been able to mobilise the northern member states and France the southern member states around a common Franco-German position (108). In the capacity of France and Germany to reach bilateral agreements which facilitate multilateral decision-making and which are close to the 'median' position of the member states as a whole may lie the explanation of an apparent paradox : that the two states may exercise an important collective leadership role in the union, but one which, albeit with varying degrees of ambivalence, most of the other member states accept (109).

Douglas WEBBER
INSEAD (European Institute of Business Administration)
Boulevard de Constance
F-77305 Fontainebleau Cedex
France
Douglas.Webber@insead.fr

Bibliography

Agra Europe (English-language edition), various issues, 1988-1993, Brussels.

Akten zur Auswärtigen Politik der Bundesrepublik Deutschland 1963 (3 vols.). Hauptherausgeber, Hans-Peter Schwarz; wissenschaftlicher Leiter, Rainer A. Blasius; Bearbeiter: Mechthild Lindemann and Ilse Dorothee Pautsch, Munich, R. Oldenbourg, 1994.

Akten zur Auswärtigen Politik der Bundesrepublik Deutschland 1964 (2 vols.). Hauptherausgeber: Hans-Peter Schwarz; wissenschaftlicher Leiter: Rainer A. Blasius; Bearbeiter Wolfgang Hölscher und Daniel Kosthorst, Munich, R. Oldenbourg, 1995.

Akten zur Auswärtigen Politik der Bundesrepublik Deutschland 1965 (3 vols.). Hauptherausgeber: Hans-Peter Schwarz; wissenschaftlicher Leiter: Rainer A. Blasius; Bearbeiter: Mechthild Lindemann and Ilse Dorothee Pautsch, Munich, R. Oldenbourg, 1996.

Balladur, Edouard (1995) Deux ans à Matignon, Paris, Plon.

Barber, Lionel (1996a) "France and Germany struggle to turn their European dreams into reality" Financial Times, 14-15 December 1996.

Barber, Lionel (1996b) "A compromise on stability", Financial Times, 16 December 1996.

Bender, Karl-Heinz (1995) Mitterrand und die Deutschen: Die Wiedervereinigung der Karolinger, Bonn, Bouvier.

Brandt, Willy (1993) Erinnerungen, Frankfurt am Main/Berlin, Ullstein.

Brigouleix, Bernard (1995) Histoire indiscrète des années Balladur : Matignon durant la seconde cohabitation, Paris, Albin Michel.

Bulmer, Simon (1993) "The Governance of the European Union: A New Institutionalist Approach", Journal of Public Policy, 13, 4 (October-December 1993), 351-80.

Bulmer, Simon (1994) "Institutions and Policy Change in the European Communities: The Case of Merger Control", Public Administration, 72, Autumn 1994, 423-44.

CDU/CSU Parliamentary Group in the German Parliament ("Schäuble/Lamers Paper") (1994) Reflections on European Policy. Bonn, CDU/CSU-Bundestagsfraktion im Deutschen Bundestag.

Couve de Murville, Maurice (1971) Une politique étrangère 1958-1969, Paris, Plon.

Culley, Paul (1995) "The Agriculture Council", in Martin Westlake, The Council of the European Union, London, Cartermill, 191-210.

Der Spiegel (1987a) "Landwirtschaft - der alltägliche Irrsinn. Subventionen, überschüsse, Fehlentscheidungen: Die Geschichte des europäischen Agrar-Monsters (II)", no. 46, 9 November 1987, 114-36.

Der Spiegel (1987b) "Landwirtschaft - der alltägliche Irrsin. Subventionen, überschüsse, Fehlentscheidungen: Das schwarz-grüne Bündnis in Deutschland (III), no. 47, 16 November 1987, 96-125.

De Schoutheete, Philippe (1990) "The European Community and its sub-systems", in William Wallace (ed.), The Dynamics of European Integration. London and New York, Pinter/Royal Institute of International Affairs, 106-24.

De Schoutheete, Philippe (1997) Une Europe pour tous: Dix essais sur la construction européenne, Paris, Odile Jacob.

Deutscher Bundestag (1991) 12. Wahlperiode, Protokoll der 46. Sitzung, Bonn, Mittwoch, den 9. Oktober 1991.

Eichener, Volker (1996) "Die Rückwirkungen der europäischen Integration auf nationale Politikmuster" in Markus Jachtenfuchs and Beate Kohler-Koch (eds) Europäische Integration, Opladen, Leske und Budrich, 249-80.

Epstein, Paul J. (1997) "Beyond policy community: French agriculture and the GATT", Journal of European Public Policy, 4, 3 (September 1997), 355-72.

Freisberg, Ernst (1965) Die Grüne Hürde Europas: Deutsche Agrarpolitik und EWG, Cologne and Opladen, Westdeutscher Verlag.

Garrett, Geoffrey (1992) "International cooperation and institutional choice: the European Community's internal market", International Organization, 46, 2 (Spring 1992), 533-60.

Garrett, Geoffrey (1995) "From the Luxembourg Compromise to Codecision: Decision Making in the European Union", Electoral Studies, 14, 3 (September 1995), 289-308.

Garrett, Geoffrey and Tsebelis, George (1996) "An institutional critique of intergovernmentalism", International Organization 50, 2 (Spring 1996), 269-99.

Gerbet, Pierre (1990) "Le rôle du couple France-Allemagne dans la création et développement des Communautés Européennes" in Robert Picht and Wolfgang Wessels (eds), Motor für Europa? Deutsch-französischer Bilateralismus und europäische Integration, Bonn, Europa Union Verlag, 69-119.

Gerstenmaier, Eugen (1981) Streit und Friede hat seine Zeit: Ein Lebensbericht, Frankfurt am Main/Berlin/Vienna, Propyläen.

Giscard d'Estaing, Valéry (1988) Le pouvoir et la vie, Paris, Compagnie 12.

Grant, Wyn (1997) The Common Agricultural Policy, New York, St Martin's Press.

Haas, Peter M. (1989) "Do regimes matter? Epistemic communities and Mediterranean pollution control", International Organization, 43, 3 (Summer 1989), 377-403.

Haas, Peter M. (1992) "Introduction: epistemic communities and international policy coordination", International Organization, 46, 1 (Winter 1992), 1-35.

Hayes-Renshaw, Fiona and Wallace, Helen (1997) The Council of Ministers, Basingstoke and London, Macmillan.

Hendriks, Gisela (1989) "Germany and the CAP: national interests and the European Community", International Affairs, 65, 1 (Winter 1988/9), 75-87.

Hendriks, Gisela (1991) Germany and European Integration. The Common Agricultural Policy: an Area of Conflict, New York/Oxford: Berg.

Hrubesch, Peter (1987) "30 Jahre EG-Agrarmarktsystem. Entstehungsgeschichte - Funktionsweise - Ergebnisse", Aus Politik und Zeitgeschichte, no. 18 (2 May 1987), 34-47.

Jones, Tim (1996a) "Germany's volte-face threatens postal deal", European Voice, 14-20 November 1996.

Jones, Tim (1996b) "Chirac gets tough on post", European Voice, 5-11 December 1996.

Kusterer, Hermann (1995) Der Kanzler und der General, Stuttgart, Neske.

Küsters, Hanns Jürgen (1982) Die Gründung der Europäischen Wirtschaftsgemeinschaft, Baden-Baden, Nomos.

Lacouture, Jean (1993) De Gaulle: The Ruler 1945-1970, New York/London, W.W. Norton.

Lahr, Rolf (1981) Zeuge von Fall und Aufstieg: Private Briefe 1934-1974, Hamburg, Albrecht Knaus.

Legras, Guy (1993) "L'Uruguay Round et la réforme de la PAC", Politique étrangère, no. 2 (Summer 1993), 325-31.

Lemaître, Philippe (1991) "La France ne s'oppose plus à la réforme de la politique agricole commune", Le Monde, 20/21 October 1991.

Le Monde (1993) L'Europe de Yalta à Maastricht 19145-1993, Paris, Editions Le Monde.

Lequesne, Christian (1990) "Formulation des politiques communautaires et procédures de consultation avec la RFA en France" in Robert Picht and Wolfgang Wessels (eds.), Motor für Europa? Deutsch-französischer Bilateralismus und europäische Integration, Bonn, Europa Union Verlag, 123-44.

Le Theule, François-Gilles and David Litvan (1993) "La Réforme de la PAC: Analyse d'une négociation communautaire", Revue française de science politique, 43, 5 (October 1993), 755-87.

Lindberg, Leon (1963) The Political Dynamics of European Economic Integration, Stanford, Stanford University Press/London, Oxford University Press.

Maillard, Pierre (1991) De Gaulle und Deutschland: Der unvollendete Traum, Bonn/Berl Bouvier.

Maillard, Pierre (1995) De Gaulle et l'Europe: entre la nation et Maastricht, Paris, Tallandier.

Manfraß, Klaus (1972) "Das geteilte Deutschland" in Karl Carstens, Dietrich Mende, Christiane Rajewsky and Wolfgang Wagner (eds), Die internationale Politik 1964-65, Munich/Vienna, Oldenbourg, 308-43.

Mansholt, Sicco (1974) La crise, Paris, Stock.

Marjolin, Robert (1986) Le travail d'une vie, Paris, Robert Laffont.

Marks, Gary, Lisbet Hooghe and Kermit Blank (1996) "European Integration from the 1980s: State-Centric v. Multi-level Governance", Journal of Common Market Studies, 34, 3 (September 1996), 341-78.

Milward, Alan S. (1992) The European Rescue of the Nation-State, London, Routledge.

Moravcsik, Andrew (1991) "Negotiating the Single European Act: national interests and conventional statecraft in the European Community", International Organization, 45, 1 (Winter 1991).

Moravcsik, Andrew (1993) "Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach", Journal of Common Market Studies, 31, 4 (December 1993), 473-524.

Moser, Peter and Schneider, Gerald (1997) "Rational Choice and the Governance Structure of the European Union: An Introduction", Aussenwirtschaft, 52, 1/2, 11-26.

Moyer, H. Wayne (1993a) "The European Community and the GATT Uruguay Round: Preserving the Common Agricultural Policy at All Costs" in William P. Avery (ed.), World Agriculture and the GATT, Boulder and London, Lynne Rienner, 95-119.

Moyer, H. Wayne (1993b) "EC Decisionmaking, the MacSharry Reforms of the CAP, Maastricht, and the GATT Uruguay Round" (unpublished paper presented at the Third Biennial Conference of the European Community Studies Association, Washington, D.C., May 27, 1993).

Müller-Armack, Alfred (1971) Auf dem Weg nach Europa: Erinnerungen und Ausblicke, Tübingen and Stuttgart, Wunderlich and Poeschel.

Newhouse, John (1972) "Die Krise der EWG" in Karl Carstens, Dietrich Mende, Christiane Rajewsky and Wolfgang Wagner (eds), Die internationale Politik 1964-1965, Munich/Vienna, Oldenbourg, 249-76.

Odell, John S. (1993) "International Threats and Internal Politics: Brazil, the European Community and the United States" in Peter B. Evans, Harold K. Jacobson and Robert D. Putnam (eds), Double-Edged Diplomacy: International Bargaining and Domestic Politics, Berkeley/Los Angeles/London, University of California Press, 233-64.

Osterheld, Horst (1992) Außenpolitik unter Bundeskanzler Ludwig Erhard 1963-1966: Ein dokumentarischer Bericht aus dem Kanzleramt, Düsseldorf, Droste.

Paemen, Hugo and Alexandra Bensch (1995) From the GATT to the WTO: The European Community in the Uruguay Round, Leuven, Leuven University Press.

Patterson, Lee Ann (1997), "Agricultural policy reform in the European Community: a three-level game analysis", International Organization, 51, 1 (Winter 1997), 135-65.

Peters, B. Guy (1992) "Bureaucratic Politics and the Institutions of the European Community" in Alberta M. Sbragia (ed.), Euro-Politics: Institutions and Policymaking in the 'New' European Community, Washington, DC: Brookings Institution, 75-122.

Peyrefitte, Alain (1994) C'était de Gaulle (vol. I) Paris, Fayard.

Peyrefitte, Alain (1997) C'était de Gaulle (vol. II) Paris, Fayard.

Pinder, John (1991) European Community: The building of a union, Oxford/New York, Oxford University Press.

Richardson, Jeremy (1996) "Policy-making in the EU: Interests, ideas and garbage cans of primeval soup" in Jeremy J. Richardson (ed.), European Union: Power and policy-making, London and New York, Routledge, 1996.

Ross, George (1995) Jacques Delors and European Integration, Cambridge, Polity Press.

Ryborg, Ole (1995) "Ideologies rather than real issues drive ministers through 'No' lobby", European Voice, 12-18 October 1995.

Scharpf, Fritz W. (1988) "The Joint-Decision Trap: Lessons from German Federalism and European Integration", Public Administration, 66, 3 (Autumn 1988), 239-78.

Scharpf, Fritz W. and Mohr, Matthias (1994) "Efficient Self-Coordination in Policy Networks: A Simulation Study" (MPIFG Discussion Paper 94/1), Cologne, Max-Planck-Institut für Gesellschaftsforschung.

Schmidt, Helmut (1990) Die Deutschen und ihre Nachbarn: Menschen und Mächte II, Berl Siedler.

Schmitter, Philippe C. (1996) "Examining the Present Euro-Polity with the Help of Past Theories" in Gary Marks, Fritz W. Scharpf, Philippe C. Schmitter and Wolfgang Streeck (eds), Governance in the European Union, London, Sage, 1-14.

Schwarz, Hans-Peter (1992) Erbfreundschaft: Adenauer und Frankreich, Bonn, Berl Bouvier.

Schwarz, Hans-Peter (1994) Adenauer Band 2: Der Staatsmann 1952-1967, Munich, Deutscher Taschenbuch.

Seydoux, François, Dans l'intimité franco-allemande: Une mission diplomatique, Paris, Albatros.

Soisson, Jean-Pierre (1993) Politique en jachère : octobre 1992-avril 1993, Paris, Albin Michel.

Stone Sweet, Alec and Sandholtz, Wayne (1997) "European integration and supranational governance", Journal of European Public Policy, 4, 3 (September 1997), 297-317.

Tracy, Michael (1989) Government and Agriculture in Western Europe 1880-1988 (third edition). Hemel Hempstead: Harvester Weatsheaf.

Van Schendelen, M.P.C.M. (1996), "'The Council Decides': Does the Council Decide?", Journal of Common Market Studies, 34, 4 (December 1996), 531-48.

Védrine, Hubert (1996) Les mondes de François Mitterrand : à l'Elysée 1981-1995, Paris, Fayard.

Von der Groeben, Hans (1983) Aufbaujahre der Europäischen Gemeinschaft: Das Ringen um den Gemeinsamen Markt und die Politische Union (1958-1966), Baden-Baden, Nomos.

Webber, Douglas, (ed.) (forthcoming) The Franco-German Relationship in the European Union, London, Routledge.

Westlake, Martin (1995) The Council of the European Union, London, Cartermill.

Willis, F. Roy (1968) France, Germany, and the New Europe 1945-1967, Stanford and London, Stanford University Press and Oxford University Press.

Woolcock, Stephen and Hodges, Michael (1996) "EU Policy in the Uruguay Round" in Helen Wallace and William Wallace (eds.), Policy-Making in the European Union, Oxford, Oxford University Press, 301-24.

List of interviews

1. Council of the European Union

Heaton, Trevor, head of division, DGE External Relations and Development Cooperation, formerly responsible for GATT issues. Brussels, 26 February 1996.

Swift, David, head of division responsible for cereals, sugar, oilseeds, rice and fodder crops, DG for Agriculture and Fisheries. Brussels, 21 February 1996.

Tracy, Michael, director of agriculture 1973-1983. Brussels, 2 July 1997.

2. European Commission

Andriessen, Frans, Commissioner for Foreign Relations and Trade 1989-1992. Brussels, 2 December 1993.

Doran-Schiratti, Maeve, head of Agriculture and Fisheries Unit, DG I (External Relations). Brussels, 2 May 1995.

Ehlermann, Claus-Dieter, former head of the legal services division and former Director-General, DG IV (Competition). Florence, 17 April 1997.

Lamy, Pascal, head of the cabinet of Commission President Jacques Delors. Paris, 19 May 1995.

Legras, Guy, Director-General, DG VI (Agriculture). Brussels, 15 February 1996.

Möhler, Rolf, Deputy Director-General, responsible for International Affairs, DG VI (Agriculture) during the Uruguay Round negotiations. Brussels, 1 July 1997.

Olsen, Soren, Deputy Head of Unit for International Affairs, DG VI (Agriculture). Brussels, 17 January 1994.

Steichen, René, Commissioner for Agriculture 1993-1994 and former Minister of Agriculture in Luxembourg. Brussels, 23 March 1995.

3. European Parliament

Plumb, Lord Henry, MEP for the British Conservative Party, former president of the National Farmers' Union and the COPA. Brussels, 21 February 1996.

4. European farmers' associations

Myllerup, Jens-Peter, director responsible for international relations, cereals, oilseeds and sugar, COPA. Brussels, 2 May 1995.

4. Belgium

De Clercq, Regine, official responsible for the GATT in the Belgian Foreign Minister and Belgian member of the 113 Committee during the Uruguay Round negotiations. Rome, 12 February 1996.

De Schoutheete, Philippe, Belgian permanent representative to the EU 1987-1997. Brussels, 29 July 1997.

5. Federal Republic of Germany

Braun, Sigismund Freiherr von. German Ambassador to France 1968-1970 and 1972-1976. Bonn, 13 December 1993.

Feiter, Franz-Josef, state secretary, Federal Ministry of Food, Agriculture and Forestry and former agricultural policy adviser, Federal Chancellor's Office. Bonn, 20 October 1994.

ammann, Detlev, head of section for foreign trade, money and credit, Federal Chancellor's Office. Bonn, 28 January 1994.

Keferstein, head of section for France and some other West European states, Foreign Office. Bonn, 16 December 1993 and 6 May 1997.

Leroy, Hansheinrich, head of section for bilateral relations with EU member states, Federal Ministry of Food, Agriculture and Forestry. Bonn, 29 November 1996.

Lotz, Klaus Martin, head of department for EU and international affairs, DBV. Bonn, 21 December 1993.

Massow, Hans-Henning von, head of section for institutional issues relating to the EU and EU coordination, Federal Ministry of Economics. Bonn, 23 February 1996.

Schirmer, Ulrich, head of section for trade policy issues, Federal Ministry of Economics. Bonn, 22 December 1993.

Sudhoff, Jürgen, German Ambassador to France. Paris, 18 January 1994.

Witt, Peter Jürgen, head of section for international trade issues, Federal Ministry of Food, Agriculture and Forestry. Bonn, 21 December 1993.

Zenner, Michael, official in the section for trade policy and external relations of the EU, Foreign Office. Bonn, 2 December 1994.

Zirbel, Sigrid, official in the Department of International Trade, BDI (Bundesverband Deutscher Industrie - Federation of German Industry). Cologne, 22 February 1996.

6. France

Andrault, Jean-Pierre, official of the AGPB (Association Générale des Producteurs de Blé - General Association of Cereal Producers). Paris, 19 November 1993.

Carmona, Yves, official in section for external relations of the EU, Division for European and Economic Affairs, Foreign Office. Paris, 27 March 1995.

Demange, Jean-Marie, head of the agricultural and agri-industrial unit, division for external economic relations of the Ministry of Economics and Finance. Paris, 19 December 1993.

De Ricolfis, François, representative of the division for external economic relations of the Ministry of Economics and Finance at the permanent representation at the EU. Brussels, 15 June 1995.

De Silguy, Yves-Thibault, head of the SGCI (Sécretariat-Général pour le Comité Interministeriel pour les questions de coopération économique européenne). Paris, 2 February 1994.

Durant, Hervé, official in the international affairs section, Ministry of Agriculture. Paris, 17 December 1993.

Guérin, Marie-Noelle, head of international affairs, FNSEA (Fédération Nationale des Syndicats d' Exploitants Agricoles). Paris, 15 November 1993.

Lebossé, Jean-Claude, adviser for agricultural affairs, Presidency of the Republic. Paris, 15 June 1994.

Moreau, Dominique, head of European and international affairs, CNJA (Centre National des Jeunes Agriculteurs). Paris, 16 November 1993

. Seillière, Ernst-Antoine. Head of the Economics Commission, CNPF (Confédération Nationale du Patronat Français). Paris, 18 April 1995.

7. Italy

Marchini, official in the economic affairs directorate responsible for external relations of the EU, Ministry of Foreign Affairs. Rome, 14 July 1994.

Nigido, Roberto, Director General of Economic Affairs, Ministry of Foreign Affairs. Rome, 14 July 1994.

8. The Netherlands

De Zeeuw, Aart, Director-General, Ministry of Agriculture 1973-1989, former member of the SCA, chairman of the GATT negotiating group on agriculture 1986-1990. The Hague, 19 February 1996.

Engering. Director-General, Ministry of Economics and Trade. The Hague, 19 February 1996.

9. United Kingdom

Roberts, Christopher, deputy secretary, Department of Trade and Industry and British member of the 113 Committee. London, 8 March 1996.

Rollo, Jim, chief economist, Foreign and Commonwealth Office. London, 17 November 1993.


Notes

(1) I should like to thank INSEAD and the Robert Schuman Centre of the European University Institute, Florence, where I was a Jean Monnet fellow from October 1995 to April 1997, for having made possible the research on which this paper is based. I am also grateful to Susanne Schmidt, Simon Bulmer, Wyn Grant, Peter Katzenstein and two anonymous referees for their constructive and useful comments on the first version of the paper. I alone, of course, bear responsibility for the contents!

(2) For the sake of simplicity, the term EU is also used in this paper to refer to the predecessor organisations (the European Community or Communities) of the current union.

(3) Moravcsik has subsequently adapted and modified his model of EU governance in Moravcsik 1993. In his empirical research, however, he still explains EU decisions in terms of inter-state bargaining between France, Germany and (from 1973 onwards) Britain.

(4) This definition is derived from the work on policy networks and communities of Rod Rhodes and David Marsh and Jeremy Richardson. Peters himself does not define the concept of policy communities in his analysis of the EU's 'bureaucratic politics'. For a discussion of the concepts of policy network and policy communities in the context of the EU, see Richardson 1996: 6-11.

(5) According to a high-level official interviewed by the author, the Franco-German relationship "pervades all levels of the administration" and has "become second nature" for many civil servants in Bonn and Paris.

(6) For a very concise analysis of the history of the CAP, see Grant 1997: 63-82.

(7) The agricultural policy articles of the Rome Treaty prescribed the consultation procedure for the adoption of the CAP, whereby the Council of Ministers makes decisions on the basis of Commission proposals and the European Parliament (and the Economic and Social Committee) may offer only advisory opinions. Interestingly, subsequent treaty revisions have not altered the procedure for the adoption of EU agricultural policy decisions (see section IV).

(8) Adenauer reputedly asked the (German) Commission president, Walter Hallstein, how many votes the CDU might lose at the federal elections if the German government agreed to a cut in German cereal prices. When Hallstein estimated 'only about one per cent', Adenauer replied: "What do I do when exactly this one per cent costs me the absolute majority?" (Der Spiegel 1987a: 117). (The CDU governed in Germany with an absolute majority from 1957 to 1961.)

(9) According to De Gaulle (as cited in Maillard 1995: 154 and 1991: 200), Adenauer, while saying that German opinion was hostile to a common agricultural policy, promised to act in such a way that Franco-German differences over agriculture would not prevent the implementation of the common market.

(10) The Commission president, Walter Hallstein, described the negotiations afterwards as follows: "Forty-five separate meetings, 7 of them at night; a total of 137 hours of discussion, with 214 hours in sub-committee; 582,000 pages of documents; 3 heart attacks - the record is staggering" (quoted in Lindberg 1963: 273). The final negotiating session on 13-14 January lasted 21 hours (Lahr 1981: 353).

(11) According to Willis (1968: 287), French farmers in May 1961 staged the "most extensive and violent jacquerie that modern France has known".

(12) The German Foreign Office feared less that De Gaulle would withdraw France from the EU than that he would prevent the scheduled transition to the second phase of the (industrial) common market. "De Gaulle is counting on the others' greater zeal for Europe. He who loves more strongly is at a disadvantage - an old experience" (Lahr 1981: 352). Lahr was then permanent secretary of the German Foreign Office.

(13) The Agricultural Minister coordinated his European policy closely with the DBV (Deutscher Bauernverband - German Farmers' Association). One manifestation of the extremely close relationship was that the Agricultural Minister, Schwarz's, regular participation in the meetings of the association's national presidium. Indeed, according to Freisberg (1965: 45-47), he preferred taking part in these meetings to those of the Agricultural Council in Brussels!

(14) According to a further account, that of the then French foreign minister (Couve de Murville 1971: 317), Germany, Belgium and Luxembourg wanted to proceed to the second stage of the common market without an agreement having been reached on agriculture and Italy adopted an intermediate position, while France and Holland insisted that there had to be an agreement on agriculture before the second stage could be launched.

(15) Lahr, who was the head of the German delegation at the time, negotiated the final agreement with Couve de Murville, the French foreign minister, and the head of the economics division of the Quai d'Orsay, Wormser. He recounts in his memoirs: "In the final night, in which we had to decide on the financing of the agricultural policy, the tough job of bringing about a Franco-German agreement in the chambre separée [author's own words] fell to me. It had long since emerged that a Franco-German agreement is the most important pre-condition of an agreement among the six; thus, it has become the norm for us to talk bilaterally with the French before or during the Council of Ministers. The others, especially the Italians, suspect us thereby of forming a secret society and hegemony, but in fact they are usually quite happy when the two main combattants, both ruffled and a bit weary, return to the main arena after having achieved a compromise". After the Franco-German agreement had been reached at midnight on 13 January, Lahr took over the chairmanship of the full Council meeting, which carried on for a further five and a half hours: "As important as it is for the Germans and the French to have agreed, nonetheless - or, better, exactly for this reason - the others also want then to emphasize their importance. There are the Italians, who complain at such length and so movingly, the Dutch, who keep their pockets closed, and the Belgians, who feel that their European convictions are being abused. Thus, the council chairman had the job of explaining the Franco-German accord without too much or to little verve, to present it as not being unalterable. but to block any changes, to listen to and to praise alternative proposals without seriously taking them up and - to have more meat on the backside than all the others. A clear head is useful, a solid backside indispensable" (Lahr 1981: 354-55). For Lahr, advance bilateral consultations between France and Germany were a means of moderating conflicts so that the 'shock' when they collided in Brussels was not as great as it would be otherwise and reflected the 'natural law' that the biggest controversies in international organisations always took place between the two biggest states (quoted in: Akten 1965: 1274).

(16) Lindberg (1963: 273-282) argues, in contrast, that the Commission was the decisive actor in the negotiations and that it is "striking … that the final regulations do not differ markedly from the Commission's original proposals" (274). It is clear, however, from Lindberg's analysis that the Commission's primary role was that of mediation and brokerage between the member states and that its success lay in its having been able to find something akin to a compromise between the preferences of the six member states rather than to implement an 'autonomous' agricultural policy conception. He concludes that "to date the prevalent pattern [of conflict-resolution] has been one of 'splitting the difference'", while speculating that the trend was toward "an ever more important mediatory role for the Commission" (281). This prognosis was not borne out, however, by the outcome of the conflicts over the CAP in 1964 and 1965, which Lindberg, writing in 1963, could not analyse in his study.

(17) Müller-Armack, division head from the German Economics Ministry, shared the tasks of heading the German delegation and chairing the Council of Ministers with Lahr during the second half of 1961 and first two weeks of 1962, when Germany had the Council presidency.

(18) The dilution of the Franco-German treaty through the preamble added to it in the Bundestag seems to have greatly disappointed De Gaulle, who was inclined to respond by forging closer ties with the Soviet Union. See Peyrefitte 1997: 223-228, 231.

(19) It was widely feared in the CDU/CSU that the government would lose the 1965 elections if, beforehand, it agreed to a common EU cereals price. See, for the view of the then Parliamentary party chairman, Krone, Gerstenmaier 1981: 614. Gerstenmaier (1981: 521) also reports that, in order to ensure his election as Chancellor in 1963, Erhard had apparently promised the farmers in the CDU/CSU Parliamentary party that he would not take any decision "against or without them". After unfavourable local election results for the Christian Democrats in autumn 1964, Erhard told Dutch political leaders that it would be "political suicide" for him to accept a common cereal price before the 1965 federal elections (Akten 1964: 1112). Schröder pleaded with Couve de Murville in August 1964 that the government was in a "very difficult situation" because of the imminent elections. The cereal price was a decision that the government could not simply "decree", but required the farmers' support (Akten 1964: 611).

(20) In January 1963, at least, De Gaulle told his then press spokesman, Alain Peyrefitte, that the thought of the collapse of the common market did not perturb him: "France has existed for centuries without the common market, it can live without it" (Peyrefitte 1994: 350; see also Peyrefitte 1997: 250-51, 253, 256, 282).

(21) In a long interview (Mansholt 1974: 112-13), Mansholt described what happened as follows: "I left the meeting ... with only two civil servants ... I still had the file where all the positions had been marked. Schwarz [the German Agricultural Minister] wants that, Pisani [the French Agricultural Minister] wants this, etc. It was a chart with the fifteen unresolved issues. At that moment, of course, the prices played the biggest role. We withdrew to a little Provençal restaurant ... In a half an hour, we managed to draw up an overall scheme and looked at how we could give something, something attractive to everyone; every minister had to have something to show in his country, so that he could say: I have won this. I still had some coloured crayons and drawings of little flags: blue, white and red for Pisani, red, black and gold for Schwarz, that I put on my table, right in front of my eyes. Everyone could succeed in winning on some issues at the price of making concessions. The main thing was to make a good job of combining and presenting the package. But always in such a way that we reached our objective: the good functioning of the common market. I never made a concession on this point. Naturally, the Commission had taken a stance on and decided the cereal price. We hadn't changed anything. But we gave something to Schwarz and to Pisani, etc. After an hour, we entered the Council room ...". The Commissioner's propensity, in illustrating his points, to refer to the German and French ministers may be taken as an indication of the extent to which he was pre-occupied with reconciling the conflicts between these two, rather than some other combination of, member states.

(22) In the German case, these subsidies amounted to half of the volume demanded (Akten 1964: 1396).

(23) The initial scheme, agreed in January 1962, massively favoured France, which contributed 25% of the agricultural budget, but, thanks to its large volume of exports to non-EU states, benefited from 85% of the expenditure (Akten 1965: 1101).

(24) According to Lacouture (1993: 361), the Commission calculated that, on the eve of presidential elections in 1965, De Gaulle would refrain from any action that could jeopardise the CAP for fear that this would alienate French farmers on whose electoral support he depended. The farmers' organisations did criticise and oppose De Gaulle's subsequent boycott of the EU institutions, but to little effect. De Gaulle mocked the other member states and the Commission: "They thought that we would accept the extravagant powers of the Commission and a federal budget, since we wanted so much to see the agricultural financing regulation adopted. They thought that they could catch us like that and that we would be afraid of the peasants, or of the next election, or I don't know what - things from the Fourth Republic" (Peyrefitte 1997: 297).

(25) Thus, Erhard told the French Prime Minister Pompidou at the summit that no long-term agreements on agricultural policy could be reached before summer 1966, by which time the issues of interest to Germany would also have to be resolved (Osterfeld 1992: 205).

(26) Peyrefitte describes how, at the Bonn summit, the German Secretary of State for Foreign Affairs suddenly tabled a proposal to settle the conflict that neither the French nor Chancellor Erhard had seen before - prompting De Gaulle to remark to him that "il n'y a pas de gouvernement allemand, mais seulement des tendences contraires" ("there is no German government, but only opposing currents") (Peyrefitte 1997: 287). De Gaulle saw this as part of a plan by Schröder to embarrass Erhard and succed him as Chancellor in a Grand Coalition government with the Social Democrats after the September 1965 elections.

(27) In his account of the meeting to the French Council of Ministers, Couve de Murville praised the attitudes taken by Belgium and Luxembourg and said that the Dutch negotiators' hands had been tied by their Parliament, there had been no "fundamental discussion" with the Germans and the Italians "have been the big obstacle". The Commission, he added, was the "big loser. It made absurd proposals, that have not been accepted by anybody. It was absent in the following discussion ... The professional 'Europeans' were ... the major obstacle to European progress" (Peyrefitte 1997: 289). In his memoirs, Couve de Murville (1971: 263) attributes the failure of the Council meeting "partly" to the failure of the Franco-German summit in Bonn three weeks earlier. According to the French ambassador in Bonn, the crisis in the EU originated from a "Franco-German divergence" (Seydoux: 57).

(28) De Gaulle told his press spokesman: "What has to be destroyed above all else is the majority vote" (Peyrefitte 1997: 294). One reason why De Gaulle wanted to prevent the introduction of qualified majority voting in the council was, as he stated in a September 1965 press conference, to pre-empt any, for France unfavourable, changes being made in the CAP. In this sense, constitutional and agricultural policy issues may have been linked in the French government's strategy. See Marjolin 1986: 349 and Peyrefitte 1997: 373. The boycott, however, was unpopular among French farmers' organisations (see Peyrefitte 1997: 298, 300-301). Peyrefitte's account of the French Council of Ministers' meeting on 1 July 1965, the day after the Brussels éclat, gives the clear impression that, among the French ministers present at Brussels, it was Couve de Murville who had orchestrated the decision. He quotes the "sombre" Agricultural Minister, Pisani, as having said: "I have nothing to say" (Peyrefitte 1997: 289). In fact, both Pisani and the French finance minister, Giscard d'Estaing, had offered the Italian delegation "significant" concessions during the final night of the talks, suggesting that they favoured a compromise and might have been surprised by Couve de Murville's declaration of the negotiations' failure (Lahr 1981: 426). Indeed, Couve de Murville had discussed the idea of boycotting the EU organs with De Gaulle several weeks before the Council meeting and had obtained the president's approval for such a step (quoted in Vaïsse 1997: 555). The German president of the European Commission, Hallstein, had told the government in Bonn that the French government was split between 'hardliners' and others and that De Gaulle belonged to the former group (Osterheld 1992: 209-210).

(29) According to Osterheld's account of the German Cabinet's discussion of the French boycott decision (Osterheld 1992: 210), Foreign Minister Schröder and Agricultural Minister Schwarz adopted a relatively relaxed attitude to the boycott, while Erhard and the Economics Minister Schmücker regarded it more seriously.

(30) Lahr, still the permanent secretary in the German Foreign Office, quotes the Belgian Foreign Minister Spaak as having exclaimed at the meeting: "But it is the Germans who are doing everything!" (Lahr 1981: 438).

(31) No significant use of qualified majority voting appears to have been made in the council before the early or mid-1980s. Even nowadays there is a "continuing reliance on consensus, even where QMV is possible", with only one quarter of Council decisions being contested (Hayes-Renshaw and Wallace 1997: 53).

(32) The Dutch government criticised the Commission's price proposals precisely because, in its view, they were based "on a compromise between national price levels, rather than an autonomous concept of what agricultural prices should be" (Lindberg 1963: 268).

(33) The striking contrast case with agriculture is transport, for which the Rome Treaty foresaw a common policy, on which, however, no progress was made for almost three decades.

(34) It is worth noting also that the CAP was fully implemented by mid-1967, three years earlier than the Commission had envisaged in its initial proposals.

(35) Scharpf (1988: 252) argues that the original CAP was a Franco-German compromise in which France achieved the opening of European markets for its agricultural producers and Germany price levels close to its own. It may be that, aided by the continuation of unanimous decision-making in the council after the French boycott, Germany managed gradually to increase the common prices to German levels. However, the critical common cereals price was not set originally 'close' to the German level, but rather halfway between the French and German prices. Overall, France seems to have got a great deal more of what it wanted in the original CAP conflicts than did Germany. Scharpf refers in fact to De Gaulle's having achieved "significant policy changes" through the threat of secession from the EU in the 1960s, but without attempting to explain why De Gaulle's tactics were so effective (1988: 259).

(36) According to von der Groeben (1983: 149), German acquiescence in the CAP in 1961-62 was also facilitated by French acceptance in December 1961 of German demands concerning EU competition policy. At least some elements in the German government expected that France would honour German acceptance of the CAP by agreeing to British entry into the EU (see above).

(37) In as far as there was a Franco-German trade-off on these issues, the outcome was clearly biased in favour of France, as Germany made concrete concessions while France made no commitment to accept the results of the Kennedy Round negotiations.

(38) Of course, the frustration of this expectation by De Gaulle's opposition to the transition to qualified majority voting in the Council of Ministers helps to account for the break between France and Germany in the 1965-66 'empty-chair' crisis.

(39) Interestingly, Scharpf makes at least an implicit such distinction in his analysis of CAP politics, in which he discusses only French and German interests and positions (Scharpf 1988: 251-254).

(40) It was under Dutch, rather than French, pressure that Germany agreed to the principle of liberalising agricultural trade at the same pace as industrial trade (see above).

(41) De Gaulle himself shared this analysis. During the 1965-66 crisis, he remarked to his press spokesman that Germany could not do without the Common Market and would therefore end up giving in to him (Peyrefitte 1997: 300).

(42) Of the period from 1955 to 1957 in which the Treaty of Rome was negotiated, Marjolin wrote (1986: 278-79) that "I never saw the German delegations' instructions in the negotiations ..., but I am sure that they called upon them never to move too far away from the French".

(43) Comparing Adenauer and Erhard on this issue, two German state secretaries who accompanied Erhard on his first visit to Paris as Chancellor in November 1963 confided to Peyrefitte (De Gaulle's press spokesman): "The relations between de Gaulle and Adenauer could be excellent only because Adenauer had resolved to keep behind De Gaulle every time, to concur with him, to let him take the initiatives and then to rally to his support. Chancellor Erhard doesn't have the same attitude. He no longer wants to be an outstanding second ... He thinks that the two countries, the two governments, the two heads of the executive must advance alongside each other" (Peyrefitte 1997: 247).

(44) De Gaulle saw in the transition from Adenauer to Erhard a watershed in the orientation of German European and foreign policy, involving changes in government attitudes towards nuclear weapons, reunification and the Oder/Neisse border with Poland. "The Germans have forgotten quickly", he told his press spokesman in 1965, "You can not count on them. They had been my big hope. They are my big disappointment" (Peyrefitte 1997: 305). At a press conference in July 1964, De Gaulle stressed the differences in French and German attitudes towards the NATO, the USSR, Central and Eastern Europe, China, defence and agricultural policy (Osterheld 1992: 105). Peyrefitte's account of his conversations with De Gaulle during this era (Peyrefitte 1997: 225, 228, 254, 268) show how quickly De Gaulle began to contemplate forging closer relations with the Soviet Union after he had become disillusioned with the development of Franco-German relations. Of Erhard, he said: "If we can't do anything with him, we have no reasons ... to neglect the good relations that we can establish with the East. Why should we restrain ourselves? It will never go very far ... of course, but, who knows, it can get Erhard worrying. It is always useful to have a means to worry one's partner" (Peyrefitte 1997: 263). At the very least, Adenauer's fear of a Franco-Soviet 'encirclement' of Germany as a consequence of Franco-German antagonisms was not completely unfounded. Erhard, however, was convinced that an alliance with the Soviet Union was not a realistic alternative for De Gaulle (Osterheld 1992: 313).

(45) Erhard, German Economics Minister until he became Chancellor in 1963, had originally opposed the foundation of the EU.

(46) On a memo from a Foreign Office official arguing that France was in a stronger position than Germany because it needed French help more than France needed German, Schröder remarked: "Rather muddled thinking!" (Akten 1964: 924).

(47) Of Schröder, De Gaulle said (Peyrefitte 1997: 249) that he "is the man of the Anglo-Saxons ... He has only one idea: to counter me".

(48) It is interesting to contrast the CAP's evolution with the objectives that the Commission originally attached to the policy. According to Lindberg (1963: 243), these objectives were "to concentrate on structural reform and modernization rather than market policy; to maintain a relatively low price level for the Community in order to prevent overproduction ... and to evolve a system which would offer a degree of Community preference, but which would still be relatively liberal toward imports from third countries". The divergence between the Commission's objectives and the 'real' CAP does not suggest that the Commission was able to exercise a decisive influence over its development!

(49) Similarly to Adenauer, Schmidt also deferred to French leadership in the EU, albeit in "some everyday conflicts" and on issues concerning the integration process, "the compromise proposals and initiatives - and often enough also the bigger financial sacrifices - had to come from Bonn" (Schmidt 1990: 173).

(50) The following analysis of the 'EU politics' of the Uruguay Round rests to a considerable extent on 43 interviews of participants in the agricultural trade policy-making process conducted in EU organs and the Belgian government in Brussels (14), in Bonn (13), in Paris (10), in London (2), in Rome (2), and in The Hague (2). The interviewees included officials from the national agricultural, economics and foreign ministries as well as the French Prime Minister's and German Federal Chancellor's Office, representatives of farmers' and business organisations, and officials of the directorates-general for agriculture and external economic relations in the European Commission. Two former European Commissioners were also interviewed by the author.

(51) The Commission itself was divided over the issue of agricultural trade liberalisation: the Dutch Commissioner for External Relations, Frans Andriessen, favoured more far-reaching concessions than the Irish Agricultural Commissioner, Ray MacSharry, and the French Commission President, Jacques Delors.

(52) Chancellor Kohl phoned the Commission president, Jacques Delors, to make it clear that the Commission's proposals were unacceptable to the German government (Agra Europe, no. 1411, 19 October 1990: P/3).

(53) The Brussels summit is discussed - from the Commission's perspective - in Paemen and Bensch 1995: 181-188.

(54) Patterson's study, which looks at why the 1992 CAP reform was more radical than the 1988 one, also focuses strongly on the roles played by the French and German governments - without, however, trying to explain why these two member states should have exercised a stronger influence on the outcome of the reform conflicts than the others. Agra Europe reports (numerous issues, September 1987-February 1988) nonetheless support her interpretation that the French and the Germans were indeed in the vanguard of the opposition to a more radical CAP reform.

(55) Mitterrand, too, strongly supported this aim of the reform. He was evidently very concerned to keep farmers on the land and to prevent a 'desertification' of the French countryside in regions where there was little other economic activity (interview).

(56) The southern European member states did not play a central role in the conflict over the CAP reform, as their principal agricultural commodities were not as strongly affected by it as the 'northern' or 'continental' products: cereals, meat and milk (Moyer 1993: 15).

(57) A cereals price cut of 29 per cent was adopted in the final council decision.

(58) Ostensibly owing to their anger over the CAP reform, French farmers voted strongly against the Maastricht Treaty, a factor which encouraged the French government to take a tough stance in subsequent Uruguay Round talks. See Soisson 1993: 19.

(59) Even now, however, the cooperation between the two Commissioners and their respective DGs clearly left a great deal to be desired. Thus, Andriessen did not receive a copy of MacSharry's negotiating paper for the Blair House negotiations until just one hour before his flight to Washington! (Interview).

(60) The French minister Soisson gives a very different account of the meeting, denying that France was isolated at all (1993: 72-74). However, he refers to the discomfort of numerous ministers concerning the likely consequences of the agreement "desired by their government", thus suggesting that these ministers were no longer in control of their governments' Uruguay Round policies.

(61) A curious characteristic of the Blair House negotiations is that they were conducted by participants who could hardly be held responsible for their decisions. They were conducted on the EU side by two commissioners who were to retire the following month and on the US side by an outgoing administration on behalf of an incoming one. It is hard to say, however, whether their 'political irresponsibility' affected the outcome of the talks.

(62) The agreement foresaw a 20 per cent cut in domestic subsidies (from which direct payments of the kind introduced in the CAP reform were exempted), a 21 per cent cut in the volume of subsidised farm exports and a 36 per cent cut in the value of export subsidies. The terms of the settlement of the oilseeds conflict were more favourable to the EU than those which the US had offered at Chicago. The agreement contained nothing, however, on the issue of 're-balancing'.

(63) The two governments' contrary assessments of the Blair House accord and its compatibility with the CAP reform were confirmed at the Franco-German summit meeting staged in Bonn in December 1992 (see Soisson 1993: 100).

(64) Soisson's account of the meeting (1993: 34) makes no mention of his veto threat.

(65) Mitterrand was indeed reluctant, it seems, to authorise the use of the veto. According to Soisson (1993: 84), he said at the meeting of the French Cabinet on 25 November 1992 that the accord was unacceptable and that, if it were put to a vote in Brussels, the French government would have to oppose it, but he did not want any explicit reference to be made to the Luxembourg Compromise and warned of the dangers of becoming involved in a trade war with the US.

(66) The then chairman of the Gaullist RPR, Jacques Chirac, also a former Agriculture Minister and Prime Minister, described the Blair House accord in the election campaign as an "agricultural Munich", comparing it thus with the policy of appeasement pursued by the French and British governments towards Hitler on the eve of the Second World War.

(67) Although there had been a change of Agricultural Minister in Bonn in January 1993, the Agricultural Ministry had not fundamentally changed its attitude to the Blair House accord. However, it was unhappy with certain aspects of it. The strong French opposition to the accord created an opportunity for it to secure some welcome changes to the accord under French 'cover'.

(68) According to Le Monde (28 August 1993), Kohl said that Germany had an "enormous problem" with the Blair House accord. Le Figaro (27 August 1993) quoted Kohl as having spoken of Germany having "some problems" with the accord.

(69) The agricultural press (Agra Europe, no. 1561, 24 September 1993: P/1) speculated that the German government's newly-discovered sympathy for French objections to Blair House related to the semi-collapse of the European Monetary System in July/August 1993, as a result of which, unless remedial action was decided by the EU, German farmers would suffer income cuts. France did pledge to support Germany in the Agricultural Council in a bid to have measures taken to pre-empt such cuts, but, when the issue came up for decision in the council in December 1993, despite French support, there was no qualified majority for such a measure in the council.

(70) Paemen and Bensch (1995: 239-240) observe similarly that before Kohl met Balladur at the end of August, "the unanimous advice of his counsellors and Ministers was to 'give nothing away on the GATT'. Imagine the surprise of those same counsellors, therefore, when on 27 August, they read a widely published report in the French press to the effect that Chancellor Kohl had said that the (Blair House) pre-agreement posed problems even for Germany and that it would be necessary to find a compromise acceptable to all concerned ... Whatever the reason, by the end of August, the Paris-Bonn axis was very definitely alive and well at the very highest level". According to Brigouleix (1995:133), Balladur told his collaborators on the way back to Paris from Bonn that "you have just witnessed the end of a dogma: that of the inalterability of Blair House ... This day will mark a turning point in the crisis. You will see".

(71) Bender (1995: 119) suggests another possible explanation for Kohl's behaviour: he pledged to support France on the Uruguay Round in exchange for a French commitment to support Germany on the issue of the location of the future European Central Bank, which the European Council indeed decided in October 1993 to locate in Frankfurt.

(72) It may be relevant in this context that the German Embassy in Paris was constantly advising the government in Bonn that it should take French threats to veto a GATT agreement seriously and, moreover, that it was important to facilitate Balladur's acceptance of an agreement so as to keep open his chances of succeeding Mitterrand as French President. The more 'pro-European' Balladur was, it was thought, would be a more congenial partner for Germany than the less 'pro-European' Chirac. Balladur writes in his own memoirs that the majority of the RPR Parliamentary party was hostile to the Blair House accord and that he had been warned that if he were to accept it, his government would be overthrown. He says that he told Kohl and other EU heads of government that he wanted a GATT agreement, but not at any price and that he would have recourse to the 'Luxembourg Compromise' to veto an agreement that his government found unacceptable. He claims that "if necessary", he would indeed have been prepared to veto a new GATT treaty (Balladur 1995: 140-141, 145). President Mitterrand did not oppose his using the veto threat (Balladur 1995: 140; Védrine 1996: 568).

(73) One of the points made by the German embassy to political leaders in France at this time was that there would be 19 (European, federal and state) elections in Germany in 1994 and that Kohl could not afford to go into this year with the GATT negotiations having just failed, given the impact this might have on the economic conjuncture.

(74) The issue of how an EU proposal to conduct new talks with the US on the Blair House accord was labelled was important because the US administration had emphasized very strongly that it would not agree to renegotiate it.

(75) Of these states, Ireland was the only one that actually shared France's objections to the Blair House accord.

(76) The Financial Times (21 September 1993) quoted Hurd as saying: "I cannot conceive how the EC will continue with the normal transaction of business if it were seen as causing the collapse of the Uruguay Round". A British participant in the EU Uruguay Round policy-making process described this retrospectively as being the kind of threat "about which you don't want to be too precise" (interview).

(77) This interpretation is supported by the fact that, while the council meeting was actually in progress, Kohl, in Paris to see Mitterrand, said of the meeting: "There will be a compromise ... because I want one" (quoted in: Die Zeit, 24 September 1993).

(78) The council decision called on the Commission to 'clarify, amplify and interpret' the Blair House accord in talks with the US. According to several newspapers, this formula had been invented in Paris (Le Monde, 6 September 1993; Handelsblatt, 9 September 1993; Libération, 20 September 1993). One participant in the council meeting said that it was developed in the restricted council sessions by the French delegation and the Commission. However, the German Foreign Minister, Kinkel, had spoken of the possibility of helping out France by having recourse to such a formulation well before the council meeting (see p. 30). In the Franco-German meeting with the Commission prior to the council meeting, Kohl's adviser, Ludewig, had also referred to the possibility of demanding "something other than a renegotiation" of Blair House (interview). According to the Financial Times (16 September 1993), the French Agricultural Minister Jean Puech and the Agricultural Commissioner Steichen discussed the possibility of 'completing and modifying' Blair House before the council meeting.

(79) A British participant in EU Uruguay Round policy-making said that the British government was a "bit concerned" about how the US would respond to the council decision, but that its "best guess" had been that there would be some "minor cosmetic changes of a face-saving nature" to the accord and these would enable France to acquiesce in a settlement (interview).

(80) Following a Congress vote to accept the NAFTA agreement in November 1993, the US government agreed to new talks with the EU over the Blair House accord. These talks resulted in some modifications to the accord, which was then subsumed in the overall GATT agreement of December 1993. Although these modifications did not change the accord fundamentally, some of them were quite significant, especially changes in the base years for calculating required export subsidy cuts - which permitted, according to Paemen and Bensch (1995: 243) a "substantial increase" in the EU's [subsidised] agricultural export capability. In the Agricultural Council, the French minister obtained a decision that if the implementation of the GATT agreement required measures to reduce EU agricultural production, prices would be cut rather than increasing the area of agricultural land 'set-aside'. For its acquiescence in the new GATT agreement, the French government also secured a change in council voting procedures making it easier for the EU to impose anti-dumping measures. The French Parliament massively approved the final agreement - although Balladur had been warned by RPR chairman, Jacques Chirac, in early December that the RPR Parliamentary group would probably oppose it (Balladur 1995: 143). Overall, the Balladur government seems to have been extremely skilful in negotiating a very narrow path between contradictory external and domestic political exigencies on the GATT issue.

(81) For a similar judgment over the general role of the EP in EU agricultural policy-making, see Grant 1997: 174-76.

(82) The study by Paemen (the head of the Uruguay Round unit in the Commission's DG I) and Bensch testifies to the constraints imposed on the Commission by the member states. Thus, they write (1995: 95) that: "In the area of trade policy, negotiations between the Member States can sometimes be far more gruelling than negotiations with third countries. Inevitably, proposals intended to reflect the collective position - i.e., the Community interest - are amended to take account of disparate national views until, in many cases, all that is left is the 'lowest common denominator'. During the Uruguay Round negotiations, this fundamental institutional flaw was cruelly exposed from time to time by the lacklustre performance of the European Community".

(83) The oft-cited 'Delors-Mitterrand axis', to the extent that it existed, is an example of this phenomenon. Certainly, Delors displayed great sensitivity to French interests in the Uruguay Round conflicts. In the period 1993-94, he also visited the French Prime Minister virtually every month and displayed a "great deal of understanding" for Balladur's positions (Brigouleix 1995: 140). In general, however, Delors, as Commission president, may have been less a vehicle of French interests than a moderator between France and Germany. He functioned, according to one interviewee, as a "go-between" between the two governments; according to another, his cabinet was staffed with people who enjoyed the confidence of both Bonn and Paris so that it could perform such a role.

(84) The Commission wanted the IGC to include services and issues relating to foreign direct investment and intellectual property rights in the common trade policy, so that its powers in these areas would be extended and trade agreements relating to them could be approved by a qualified majority, rather than unanimous, vote in the Council of Ministers. However, its proposal was opposed by a number of mainly large member states, including first and foremost France, but also Germany. The Commissioner responsible for institutional reform described the failure of the Commission's initiative at the IGC as "extremely disappointing", adding: "We have suffered a failure whose consequences for the effectiveness of our [i.e., the Commission's] future action in international negotiations may be extremely negative. The states have erred through lack of vision and have shown quite deplorable distrust towards the institutions ... Given this situation, the Commission has considered it preferable to withdraw the text proposed, in order to prevent a real step backwards in this field" (Marcelino Oreja, quoted in: Agence Europe, 24 June 1997, p. 3). The EP's demand for an extension of its co-decision powers to include agricultural policy was opposed by the French and German governments (Agra-Europe, German edition, no. 25, 23 June 1997). The IGC also obliged the Commission to make proposals to change the mode of decision-making in the CAP management committees - in particular the German Agricultural Minister was reported to be unhappy with the status quo, according to which Commission recommendations can be overturned only if a qualified majority of member states votes against them. The overall impression that emerges from the IGC is that the influence of supranational actors in agricultural and trade policy-making is more likely to wane in the future than to grow.

(85) A Commission official interviewed by the author said that if, instead of the Commission, the French and German governments had negotiated the agricultural trade components of the GATT Uruguay Round treaty with the US, the result would "not have been very different".

(86) It should be remembered that the Uruguay Round conflict came to a climax at the same time as the conflicts in France and some other member states over the ratification of the Maastricht Treaty, that is to say, at a time when the next major integration project - monetary union - looked like it could fail. The concern not to undermine the Maastricht Treaty may have been one reason why some actors were hesitant about confronting France over the Uruguay Round issue.

(87) Epstein records (1997: 362) that the French Agriculture Minister from October 1992 to March 1993, Jean-Pierre Soisson, for example, met the two main farmers' associations, the FNSEA and CNJA, before every major EU meeting at which the Uruguay Round was discussed and that, after the change of government in March 1993, Prime Minister Balladur and his Agriculture Minister, Jean Puech, met the associations monthly, without, however, granting them much influence on government policy. In his own book on his period as Agriculture Minister, Soisson recounts (1993: 25-26 and 31) how he met the president of the FNSEA, Luc Guyau, the same day he was appointed to the post to discuss the Uruguay Round dossier with him and the entire executive board of the FNSEA two days later.

(88) A former member of the SCA interviewed by the author estimated that about 90 per cent of the EU's agricultural issues are resolved below the level of the Agricultural Council. According to a Council of Ministers official, however, this proportion is declining as a consequence of the greater heterogeneity of interests among the member states brought about by successive enlargements. Another Council official estimated that between 70 and 80 per cent of external trade issues are settled in the 113 Committee.

(89) The final GATT agricultural trade agreement bore a close resemblance to the contents of the 'Dunkel Paper' drafted by the GATT Secretariat and proposed in December 1991. However, the agricultural trade provisions of this paper, named after the then GATT secretary-general, did not reflect an autonomous conception of the GATT staff, but followed closely the terms of a draft agreement that the EU and US came near to achieving at the EU-US summit meeting in The Hague in November 1991 (interview).

(90) After the "catastrophic" experience of autumn 1990, when it asked the Agricultural Council to approve a negotiating mandate for the agricultural trade aspects of the Uruguay Round and the council took seven meetings to agree upon a (very restrictive) mandate, the Commission never sought a negotiating mandate again from the Agricultural Council (interview). The Commission - also DG VI - was happier dealing with the General Affairs Council, in which the agricultural ministers, according to a Commission official, were more tightly bound by instructions from their governments than in the Agricultural Council.

(91) In general, Grant (1997: 148) estimates the influence of agricultural economists on EU agricultural policy to be minimal.

(92) There have, of course, been milestone ECJ judgments on internal market disputes concerning agriculture and processed foods and drinks, most notably the verdict on Cassis de Dijon, which provided the legal underpinning of the '1992' liberalisation programme.(on the role of the ECJ in EU agricultural politics, see Grant 1997: 179-81).

(93) One such 'package deal' was adopted by the Agricultural Council in connection with the EU's approval of the new GATT treaty in December 1993. It contained measures to facilitate France's acceptance of the treaty.

(94) A comparable, indeed classic, case of irreconcilable Franco-German conflict from the early 1990s is that of the EU banana trade, which had to be liberalised as part of the '1992' programme. Here France, which has overseas territories heavily dependent on (uncompetitive) banana production, and Germany, which imports cheaper bananas from Latin American states, failed to reach an agreement before the decisive Agricultural Council meeting in December 1992, at which Germany was finally majoritised by an Anglo-French-led coalition of mainly southern European member states (see Soisson 1993: 119-20). Under strong pressure from banana trading interests, the media and public opinion, the German government unsuccessfully contested the legality of the regulation at the ECJ, while some German banana importers tried to have the regulation overturned by the German courts. The case has meanwhile reached the German Federal Constitutional Court and, as it may lead to the undermining of the hitherto sacrosanct principle of the primacy of EU over member state law, has legal commentators speculating whether 'Europe might slip on a banana skin'. The WTO (World Trade Organisation) declared the regulation to be illegal in 1997. The EU subsequently amended the regulation, but not to the satisfaction of the US government, which referred the case back to the WTO for further adjudication.

(95) The German government attributed this Franco-German contrast to France's natural comparative advantage in farming. Thus, Erhard told De Gaulle in 1963 that France's "better climate and soil", compared with Germany, could not simply be ignored and that a way had to be found to preserve German agriculture in a common European market (Akten 1963: 714).

(96) The stances of the two governments on the latter issue show Germany, on balance, to be a stronger advocate of a 'planned' agricultural economy than France. The German attitude was neatly expressed by the former German Agricultural Minister, Ignaz Kiechle, in the statement: "You can't compensate for the difference between the Paris basin and the [German] alpine foothills through the market" (interview, "'Ich bin gegen den Markt'", in: Der Spiegel, no. 31, 27 July 1992, pp. 85-87).

(97) Quotes from a statement made by the German government's press spokesman , describing an agreement reached between Helmut Kohl and the new French Prime Minister, Lionel Jospin during the latter's first official trip to Bonn in August 1997 ("'Zeitplan für Euro einhalten'", in: Süddeutsche Zeitung, 29 August 1997).

(98) Lequesne (1990: 132) goes further, arguing that the French Ministry of Agriculture conducts an exchange of information across the entire spectrum of agricultural issues with its German counterpart that is "without any equivalent" in other issue-areas.

(99) The 'game of four' expanded after the 'EFTA enlargement' of the EU in 1995 to include Sweden, Finland and Austria and is now known as the 'little G7'. Following German reunification, however, according to an official of a smaller EU member state, the German government's attachment to this informal grouping has weakened as it has sought to increase its latitude to negotiate and forge agreements with France (interview). Trade policy also differs from agricultural policy to the extent that there is no Council of Trade Ministers as such.

(100) Between 1977 and 1993 inclusive, the foreign ministers met 236 times and the Agricultural Council on 226 occasions. Of the other councils, only that of the Economics and Finance Ministers (158 meetings) met more than 100 times (Westlake 1995: 60 ).

(101) Jean-Pierre Soisson records that, during his period as French Agricultural Minister, he always tried to reach an agreement on EU issues with his German counterpart(s) and failed only on the issue of banana trade (Soisson 1993: 208). Soisson was in office, however, for less than six months.

(102) In his memoirs, Helmut Schmidt (1990: 220) recounts a comparable case from his Chancellorship, relating to the foundation of the EMS (European Monetary System). As a Franco-German agri-monetary dispute threatened to hold up the launching of the EMS in February 1979, Giscard sent his Prime Minister, Raymond Barre, to see Schmidt in Bonn: "With two or three assistants each, we resolved the squabble of the farmers and their ministerial subordinates within two hours, got our solution through the Agricultural Council ten days later without any difficulties, and on 19 March 1979 the EMS entered into force".

(103) In this respect, nothing appears to have changed since the early 1960s, when the French Agriculture Minister Pisani told the French council of ministers that "when we, my colleague Schwarz [his German counterpart] and I, are in agreement, everything is fine. If not, the whole machine is jammed" (quotation from the council of ministers' meeting on 27 November 1963 in Peyrefitte 1997: 248).

(104) On the EMS, see Schmidt 1990: 219-240; on the EMU, the chapter by Jonathan Story in Webber 1998; and on electricity liberalisation, the chapter by Susanne Schmidt also in Webber 1998. For accounts of the French and German roles in the negotiation of the 'Stability Pact' and postal services liberalisation, see respectively Barber 1996a and 1996b and Jones 1996a and 1996b.

(105) For the authors of the CDU/CSU paper, France and Germany constitute the 'core of the hard core' of the EU, the latter comprising also the Benelux states.

(106) As a high-ranking official of a small EU member state told the author: "If it's good for France and Germany, it must be good for us too basically".

(107) As a French official told the author, "on a fait des linkages" before the council meeting, i.e. the French government had promised to support various demands raised by the southern states in relation to the CAP and the Uruguay Round in exchange for their support for the French campaign to re-open the Blair House accord.

(108) This remark may not apply on agricultural policy issues to Britain, which has long been very critical of the CAP and appears to be more isolated in the Agricultural Council than any other member state. Between December 1993 and September 1995, Britain cast eight 'no' votes in this council, more than any other member state except Denmark, which cast the same number. Holland cast five 'no' votes, Germany four and France, by contrast, none, suggesting that it was very successful in the council. Britain abstained from voting on eight occasions, Germany on five and France twice (Ryborg 1995). The high number of Danish 'no' votes may be related to the relatively strict Parliamentary control of Danish government voting (which may reduce its flexibility in the council) as well as to the in any case rather critical Danish attitude towards the CAP.

(109) Hence a high-level official from a small member state interviewed by the author declared that, in many years in Brussels, he had "never heard strong voices complaining" about the role of the Franco-German relationship in the EU. For an interpretation of the EU as an international organisation without a hegemonic power or powers, see de Schoutheete 1997: 28-32.


Prepared for the Robert Schuman Centre by Valentina Cagnole Lafond, August 1999.

Working Papers of the Robert Schuman Centre
Robert Schuman Centre