|
|
|
Email this citation
The Engines of Integration? Supranational Autonomy
and Influence in the European Community
*
Center for German and European Studies, University of California at Berkeley
November 1996
Abstract
The supranational organizations of the European Community -- the Commission, the Court of Justice, and the European Parliament -- are often depicted as the engines of the integration process, nudging the member states toward ever-deeper integration. This paper suggests that the role of supranational organizations in the integration process is best understood in terms of principal-agent analysis, which suggests that the autonomy and influence of supranational organizations varies as a function of four key variables: the preferences of the member governments, the institutional decision rules governing EC policymaking, the distribution of information between supranational organizations and member states, and the possibility of transnational coalitions between the organizations and interest groups within the member states. Case studies of the Commission's executive functions in structural policy, competition policy and external trade policy suggest that the Commission can serve, and has served, as the engine of the integration process in these sectors, but only within the limits established by the four aforementioned factors, and by its changing relationship with another supranational organization, the European Court of Justice.
The notion that the EC's supranational organizations might act as the "engines" of the integration process is not a new one 2 . Indeed, the phrase derives from the neofunctionalist literature of the 1950s and the early 1960s, which predicted that the EC's supranational organizations--the Commission, the Court of Justice, and the European Parliament--would act as a sort of vanguard, nudging the member states of the Community towards deeper and deeper integration. Later, however, during the late 1960s and 1970s, intergovernmentalists argued that the member states of the Community remained very much in control of the process of integration through the intergovernmental bodies of the EC, namely the Council of Ministers and the European Council, thereby limiting the ability of supranational organizations to drive forward the integration process. During this period, neofunctionalist predictions about the causal importance of supranational organizations seemed to have been falsified, and students of the EC turned their attention to the process of intergovernmental bargaining in the Council.
Since the relaunching of European integration in the I 980s, however, the debate on the causal role of these organizations has been reopened with regard to all three major EC organizations:
- With regard to the Commission, Sandholtz and others have argued that the Commission can play an important leadership function in the EC, while Moravcsik has argued that the Commission is generally causally unimportant, and that the powerful member states continue to set the agenda for the EC 3 .
- Similarly, with regard to the ECJ, a number of authors have argued that the ECJ effected a quiet legal revolution in the 1960s and 1970s, at precisely the height of the intergovernmentalist wave, by creating in effect a system of EC constitutional law which, by the time the member states realized it, was effectively entrenched 4 . Garrett, on the other hand, has argued that the Court's independence was only apparent, and that in fact it followed the preferences of the most powerful member states 5 .
- Finally, there has historically been little controversy over the European Parliament, because until recently there was general agreement that it was causally unimportant. In 1994, however, Tsebelis published a now-famous article arguing that the cooperation procedure established in the SEA had given the Parliament a conditional agenda-setting power, and since then Tsebelis and others have attempted to specify in rational choice terms the precise agenda setting powers of the Parliament under different legislative procedures 6 .
1.1 The Preferences of Supranational Organizations
In retrospect, the neofunctionalist concept of the EC's supranational organizations as engines of the integration process relied on two fundamental assumptions. The first of these assumptions concerns the preferences of supranational organizations. As a first approximation of supranational preferences, a number of theorists have adopted the assumption of supranational organizations as "competence-maximizers" which seek to increase both their own competences and more generally the competences of the European Community 7 ; and a similar view of supranational preferences was implicit or explicit in much of the neofunctionalist literature. Ross puts it most succinctly, arguing that supranational organizations like the Delors Commission seek "more Europe." 8 There are a number of reasons why such organizations might adopt a pro-integration or competence-maximizing agenda, including self-selection of personnel for the European organizations 9 , socialization of members within the organizations 10 , or simply bureaucratic politics 11 . Whatever the source of their pro-integration preferences, however, it seems clear that the Commission, the Court of Justice, and the European Parliament have indeed pursued a broadly pro-integrationist agenda throughout the history of the European Union.
As Hix and others point out, however, actor preferences are multi-dimensional, concerning not simply the question of integration but also other dimensions such as environmental protection, consumer protection, and the neoliberalism vs. interventionism cleavage that has characterized much of the Community's history 12 . On these issues, the preferences of supranational organizations like the Commission are less consistent and less predictable than along the integration dimension. The reason, as a number of authors point out, is that the Commission itself is not a monolithic actor, but itself a complex "multi-organization," consisting of (a) an essentially political college of Commissioners headed by a President as "first among equals" and (b) an essentially administrative civil service divided up into Directorates-General (DGs) and Services, which possess differing and often contradictory preferences on various issues 13 . As we shall see below, for example, the Commission is frequently split between the relatively interventionist Commissioners and DGs responsible for industrial and technology policies, and the more laissez-faire Commissioners and DGs responsible for competition and commercial policy. Under these circumstances, the substantive preferences of the Commission, and of other supranational organizations, is the result of the internal organization and the internal politics of each institution. Hence, theorizing deductively about the preferences of supranational organizations outside the integration dimension is exceedingly difficult, and most case studies have simply treated these preferences as an object of empirical study rather than theoretical prediction, as I will do in the case studies presented below.
1.2 Delegation and Agency
The second assumption underlying the concept of supranational organizations as the engines of integration is the claim that supranational organizations are actually autonomous in the pursuit of their preferences, and can exert an independent causal influence on policy outcomes. The central argument of this section is that the autonomy and the independent causal influence of supranational organizations can best be understood in terms of rational choice models of principal-agent interaction 14 .
In the standard principal-agent model of delegation, an actor or set of actors, known as the principals, may choose to delegate certain functions to another actor or actors, known as agents. However, this initial delegation of authority immediately raises a problem for the principals: 'What if the agent behaves in ways that diverge from the preferences of the principals? Agents might behave in this way, the literature suggests, for two reasons. First, the agent may use its delegated powers to pursue its own preferences at the expense of the principals, a process known as "shirking." Second, the agent may, as a result of the structure of delegation, be subject to perverse incentives to behave in ways contrary to the aims of the principals, a process known as "slippage". Although both shirking and slippage are likely to create losses for the principals, the principal-agent literature in political science has focused primarily on the problem of shirking.
Agency shirking is a problem because, and insofar as, an agent has the ability to pursue its own preferences at the expense of those of the principals. As Terry Moe argues, an agent's incentive and ability to shirk can be quite considerable:
A new public agency is literally a new actor on the political scene. It has its own interests, which may diverge from those of its creators, and it typically has resources--expertise, delegated authority--to strike out on its own should the opportunities arise 15 .In particular, the literature suggests, the agent possesses better information than the principal regarding its area of expertise, its budgetary needs, and its own activities, and this asymmetrical distribution of information may make it difficult for the principals to control agency shirking.
The principals are not, however, helpless in the face of this dilemma. Rather, when delegating authority to an agent, principals can also adopt various administrative and oversight procedures to limit the scope of agency activity and the possibility of agency shirking. Administrative procedures define ex ante the scope of agency activity, the legal instruments available to the agency, and the procedures to be followed by it. Such administrative procedures may be more or less restrictive, and they may be altered in response to shirking or slippage, but only at a cost to the flexibility and comprehensiveness of the agent's activities 16 . In the case of the EC, both the Commission and the Court of Justice have generally been given a broad mandate, while the European Parliament was restricted to a limited institutional role prior to the Single European Act.
Oversight procedures, on the other hand, allow principals ex post to (a) monitor agency behavior, thereby mitigating the inherently asymmetrical distribution of information in favor of the agent, and (b) influence agency behavior through the application of positive and negative sanctions. With regard to monitoring, for example, McCubbins and Schwartz suggest that principals may use any one of a number of oversight mechanisms, including the "police-patrol" method of standing oversight committees, and the "fire-alarm" oversight offered by individual constituency complaints and judicial review of agency behavior 17 . As for sanctioning, the literature points Out that principals enjoy a formidable array of sanctions, including control over budgets, control over appointments, overriding of agency behavior through new legislation, and revision of the agency's mandate. Through the use of such monitoring and the application of sanctions, much of the literature argues, both shirking and slippage by agents can be minimized, if not eliminated 18 .
As Moe points out, however, both administrative and oversight procedures can be quite costly to principals as well as agents, and these difficulties can create some limited room for agency autonomy from principals 19 . Monitoring, for example, can consume significant time and resources, and yet fail to cover more than a small cross-section of an agent's activities. Perhaps more importantly, the application of sanctions by multiple principals is not automatic, but generally requires a positive decision by the principals--and it is here that the agent can exploit conflicting preferences among its principals to escape sanctions and pursue its own preferences, within limited bounds. Thus, for example, if a decision to sanction a supranational agent requires a majority (or a unanimous vote) among the member state principals, then the agent may drift considerably from the Council's collective ideal point, as long as it does not call forth the requisite majority (or unanimity) among the member states required for the imposition of sanctions. Within these bounds, principal-agent theory predicts, an agent may enjoy considerable discretion in its actions, without the fear of sanctions being applied.
The precise extent of an agent's discretion, in this view, depends in turn upon (a) the voting rules and (b) the default condition governing the application of sanctions. With regard to voting rules, the application of sanctions against an agent is uniquely difficult where the decision rule is unanimity, granting any single actor veto power over such decisions. In effect, a single member government can protect or shield an agent from any possible sanctions, thereby increasing drastically the autonomy of the agent. Less demanding decision rules such as a simple or qualified majority of the principals, by contrast, make sanctioning correspondingly easier, and thus restrict the autonomy of the agent. In the EC's "comitology" system of oversight committees, for example, the Council majority required to overrule Commission decisions varies from one type of committee to another, with advisory committees unable to overturn Commission decisions, management committees able to refer decisions to the Council by a qualified majority, and regulatory committees able to refer decisions by only a blocking minority. As a result, the Commission enjoys considerable discretion under the advisory committee procedure, somewhat less discretion under the management committee procedure, and the least discretion under the regulatory committee procedure.
Second, and equally important, Scharpf argues that a status-quo default condition makes institutional reform--and, by extension, revision of the agent's mandate--more difficult, by privileging the existing delegation of authority to the agent. Take, for example, the threat of Treaty revision, in which the member governments might agree to revise the mandate, and reduce the powers, of an agent like the Commission or the Court of Justice. Such a Treaty revision is, in some ways, the ultimate threat, but the institutional barriers to carrying out such a threat are high. Not only would such a revision require a unanimous vote of the member governments, and ratification by national parliaments and electorates, but the default condition for Treaty provisions is the status quo, meaning that in the absence of a unanimous agreement on Treaty revision, the powers of the Commission or the Court stand. For this reason, Treaty revision is essentially the "nuclear option"--exceedingly effective, but difficult to use--and is therefore a relatively ineffective and non-credible means of member state control.
Some of the Commission's powers, however, are established not by Treaty provisions but by Council Regulations with a fixed expiration date, as with the Structural Fund Regulations or the Regulations governing the Community's Framework R&D program. In these cases, the relevant Regulations require periodic revision and readoption in the Council, meaning that the default condition is not the status quo, but the expiration of the program, and with it the Commission's executive powers. The practical upshot of this need for Council revision, then, is that member states are periodically given the opportunity to "clip the Commission's wings" if it acts in a way that diverges from their interests.
Applying the principal-agent model to the EC's supranational organizations yields a number of important hypotheses. First, it suggests that supranational organizations will attempt to "shirk," by pursuing integrative or competence-maximizing strategies which go beyond the integrative aims of their member state principals, but that all of these organizations will be constrained in their ability to do so by the administrative and oversight mechanisms established by the member states.
Second, it suggests that, ceteris paribus, the Commission should be more constrained by the member states than either the European Court of Justice or the Parliament. The reason for this is simple: While the Commission enjoys a large number of substantive responsibilities in areas ranging from trade and competition policy to agriculture and structural policy, it is also subject to a large number of oversight procedures (including the EC's "comitology" oversight committees, judicial review, and monitoring by the European Parliament and the Court of Auditors) and is in many cases relatively easy to sanction or overrule through a qualified majority vote of the member governments. By contrast, the Court and the Parliament are seldom subject to explicit oversight, and can be sanctioned only through the relatively difficult expedient of Treaty amendment, requiring unanimous agreement among the member governments, and ratification by national parliaments and electorates. Individual members of the European Parliament, moreover, are not appointed by the member governments, as are Commissioners and Judges, but directly elected by their electorates, thus granting them an additional degree of independence from their domestic governments 20 . Ceteris paribus, therefore, the Court and Parliament should enjoy greater autonomy than the more constrained Commission.
Thirdly, however, the precise administrative and oversight procedures established for the Commission have varied from issue-area to issue-area and over time, and this suggests that the Commission's autonomy also varies across issue-areas and over time, as a function of these administrative and oversight procedures. In the sections below, I attempt to illustrate the workings of these control mechanisms, and their effects on the autonomy and influence of the Commission in different issue-areas. Before turning to the empirical analysis, however, we need to examine another aspect of supranational autonomy and influence--the question of agenda-setting.
1.3 Formal Agenda Setting
Much of the literature on the EC's supranational organizations focuses on the purported "agenda setting" functions of these organizations. This literature often runs into conceptional confusion, however, because different analysts use the term "agenda setting" to refer to two different types of activities. For the sake of analytic clarity, therefore, I distinguish here between formal and informal agenda setting. Put simply, formal agenda setting consists of the Commission's right (and the European Parliament's conditional right) to set the Council's formal or procedural agenda by placing before it proposals which can be adopted more easily than amended, thus structuring the choices of the member states in Council. Informal agenda setting, by contrast, is the ability of a "policy entrepreneur" to set the substantive agenda of an organization, not through its formal powers, but through its ability to define issues and present proposals which can rally consensus among the final decisionmakers. Let us consider each, very briefly, in turn.
In the terms established by rational choice theories, formal agenda setting power relies most basically on the right of certain institutional actors to propose legislation for a vote in a legislative body. In the American Congress, this power to propose is typically wielded by Congressional committees, and is arguably the source of their disproportionate influence within their respective jurisdictions. In the European Community, by contrast, the Treaties assign the sole "right of initiative" for most (but not all) Community legislation to the Commission, placing the Commission in the role of the Community's formal agenda setter.
The right to propose, however, is not sufficient to assure agenda setting power. The influence of an agenda setter will, ceteris paribus, be greatest where the voting rule is some form of majority vote, and where the amendment rule is restrictive--in other words, where it is easier to adopt the agenda setter's proposal than to amend it. Applied to the Community, this analysis of agenda power with different voting rules and amendment rules yields varying results depending on the rules governing a given piece of Community legislation. Under the consultation procedure, for example, both the amendment rule and the voting rule are unanimity. Thus, although it is difficult to amend a Commission proposal, it is equally difficult to adopt the proposal, and any member state may veto a proposal with which it is unhappy. The Commission's agenda setting power under the consultation procedure are, therefore, minimal or nonexistent.
By contrast, the cooperation procedure established by the Single European Act seems to confer precisely the sort of agenda power that rational choice theorists assign to Congressional committees and other agenda setters: The voting rule in both cases is qualified majority, meaning that the Commission need only put forward a proposal capable of garnering the support of a qualified majority of the member states; while the amendment rule is unanimity, making it quite difficult to amend a Commission proposal. In addition, the cooperation procedure also provides some limited agenda setting power to the European Parliament, which may propose amendments to the Council's draft legislation. These amendments, if accepted by the Commission, then become part of the Commission's amended proposal, and can accordingly be adopted by qualified majority, but amended only by a unanimous vote. In other words, the EP gains some agenda setting power under the cooperation procedure, but the Commission remains the middle-man in the procedure, without whose cooperation EP amendments enjoy no special status.
Finally, the new codecision procedure established by the Maastricht Treaty establishes a similar agenda setting power for both the Commission and the Parliament, with one important difference: In its final reading, the Parliament may by an absolute majority of its members re-insert amendments to the Commission's revised draft. The Council may then accept the Parliament's amendments, or else convene a "conciliation committee" bringing together delegations from both the Council and the Parliament to reconcile their differences in a final draft, which then returns to the two bodies for final approval. The net effect of this conciliation process (which took place in 14 of the first 32 co-decision procedures) is to remove the Commission as the middle-man between the Council and the Parliament, allowing for direct bargaining between the two bodies, providing the Parliament with a possible veto of the final product--and removing the formal agenda setting power of the Commission in the process 21 .
In short, the formal agenda setting powers of the Commission depend not only on the Commission's formal right of initiative, but also on the preferences of the member states in Council voting, and in particular on the precise institutional rules regarding the adoption and amendment of Community legislation.
1.4 Informal Agenda Setting, or Political Entrepreneurship
Informal agenda setting, finally, consists basically in the ability of a "political entrepreneur" like the Commission to set the substantive agenda for the decisions taken by the member states, by identifying policy problems, proposing and "selling" policy proposals, and brokering compromises among the member states on the terms of the policies ultimately adopted. This model of informal agenda setting derives largely from Kingdon's seminal work in the American context, which relies fundamentally on the existence of imperfect information among policymakers. That is to say, when policymakers have difficulties identifying policy problems, drafting appropriate solutions, and finding compromises among varying interests, a policy entrepreneur may secure the adoption of a policy, and influence its content, by stepping forth at the right time (a "policy window") with a proposal that identifies a common problem and proposes an acceptable solution 22 .
Applying Kingdon's view to the Community, we are faced with the stark observation that the EC's supranational organizations enjoy no monopoly on informal agenda setting, which depends more on expertise and persistence than on the formal right to propose or amend policies. Anyone can be a political entrepreneur, according to Kingdon, and in the case of the EC Moravcsik has argued convincingly that member governments are fully capable of acting as entrepreneurs, providing ideas and brokering agreement with other member governments 23 . Nor do supranational organizations enjoy a unique incentive to set the Community's political agenda: in Kingdon's model, policy entrepreneurs can be motivated by a variety of motives, including material gain, bureaucratic territoriality, or ideological motives. The Commission and the other EC organizations may be considered to possess all of these motives, but so might private sector actors such as the European Round Table of Industrialists or EC member governments.
On the other hand, EC organizations, if not the only political actors with an incentive to set the Community agenda, are nevertheless often well placed to do so. Kigdon lists three characteristics of the successful policy entrepreneur: (1) the person must be taken seriously, as an expert or a leader; (2) the person must be known for her political connections or negotiating skins; and (3) the successful entrepreneur must be persistent and wait for the opening of a "policy window". In varying degrees depending on particular times and Commissioners, the Commission embodies all three characteristics of expertise, brokering skills and institutional persistence, and has the additional advantage of the formal right of initiative and well-developed policy networks. It is, therefore, true that the Commission has no monopoly over informal agenda setting, but it may nevertheless have a comparative advantage over other potential agenda setters, such as member governments or private actors, and a large number of authors have argued that the Commission has indeed played an entrepreneurial role in the adoption of the ESPRIT program 24 , the Single European Act 25 , and the Structural Funds 26 . Similarly, the European Court of Justice is credited with informal agenda setting powers by Garrett and Weingast, who argue that the Court's doctrine of "mutual recognition" served as the touchstone for the Community's revived internal market program of the 1980s 27 . Finally, Corbett has argued that the European Parliament has served as an informal agenda setter in the negotiations leading to the 1992 Maastricht Treaty, although the empirical record suggests that Parliament's influence on the negotiations was in fact quite limited 28 .
Ceteris paribus, Kingdon's emphasis on expertise and imperfect information leads us to expect that the influence of the Commission (or the Court or Parliament) is greatest where information is imperfect, uncertainty about future developments is high, and asymmetrical distribution of information between the Commission and the member states favors the former.
1.5 Hypotheses
Across all three of these areas, I have argued, four primary factors or independent variables explain the autonomy of supranational organizations, and their ability to influence political outcomes through formal and informal agenda setting. The first, of these, familiar from Moravcsik's intergovernmentalism, is the distribution of preferences among the member states and the organizations. Supranational organizations, I have argued, act within the constraints of member state preferences, which must be taken into account during all supranational executive, judicial, and legislative or agenda-setting functions. However, I have also argued that supranational agents may also exploit differing preferences among the member states, to avoid the imposition of sanctions against shirking, and to push through legislative proposals via their formal agenda setting powers.
Second, the autonomy and influence of supranational institutions depends crucially on the institutional decision rules governing the sanctioning of agents. Thus, as we shall see below, EC decision rules establish differing thresholds for the overruling and sanctioning of supranational agents, and these decision rules directly affect the autonomy of the agents from member governments. In addition, as we have seen, the formal agenda setting powers of the Commission and the Parliament both depend on the specific decision rules established for the legislative process, and the influence of each institution varies from the consultation to the cooperation and codecision procedures.
A third factor, which derives from the principal-agent and entrepreneurship literatures, is the distribution of information, or uncertainty, among the organizations and the member states, respectively. Put simply, the autonomy of a supranational organization is greatest where information is asymmetrically distributed in favor of the organization, and where the member states have difficulty monitoring its activities.
Fourth and finally, the literature on supranational leadership or entrepreneurship emphasizes that the influence of supranational agents is greatest where those agents possess clear transnational constituencies of subnational organizations, interest groups, or individuals within the member states, which can act to bypass the member governments, and/or to place pressure directly on them. Indeed, I would argue, all three EC supranational organizations possess such transnational constituencies: interest groups and multinational firms in the case of the Commission 29 , national courts in the case of the European Court of Justice 30 , and national electorates in the case of the European Parliament. In all three of these cases, national constituencies act both as a constraint on the freedom of action of the supranational organizations (the European Court of Justice, for example, must rely on national courts to accept its jurisprudence), but also as a counterbalance to the influence of the member governments (once national courts accept ECJ jurisprudence, the costs of noncompliance for member governments rise considerably). In other words, all three supranational organizations navigate constantly between two sets of constituents: the intergovernmental principals that created them and may still alter their mandates, and the transnational constituencies that act both as constraint and resource in the organizations' efforts to establish their autonomy and strive for "more Europe."
1.6 The Perils of Empirical Analysis
Unfortunately, testing such hypotheses empirically is far more difficult than it might appear at first blush, and the principal-agent literature is replete with methodological warnings about the difficulties of distinguishing between obedient servants and runaway bureaucracies. In essence, the problem is that agents such as the Commission may rationally anticipate the reactions of their principals, as well as the possibility of actions, and adjust their behavior in order to avoid the costly imposition of sanctions. If this is so, then agency behavior which at first glance seems autonomous may in fact be subtly influenced by the preferences of the principals, even in the absence of any overt sanctions. Indeed, as Weingast and Moran (1983) point out, the more effective the control mechanisms employed by the principal, the less overt sanctioning we should see, since agents rationally anticipate the preferences of the principals and incorporate these preferences into their behavior. In this view, sanctions should take place only rarely, when an agent miscalculates the likely reactions of its principals, or the likelihood of sanctions in response to its actions.
The relevance of these observations becomes clear when we examine the literature on supranational organizations in the European Community. For example, in response to Mattli and Slaughter's claims that the European Court of Justice has independently fostered the development of a supranational constitution for the EC, Garrett has argued that the Court's independence was only apparent, and that the judges actually rationally anticipated the responses of the most powerful member states, and adjusted their rulings accordingly. Similarly, students have differed in their interpretation of the comitology system of committees overseeing the Commission, and the remarkable rarity of negative opinions by these committees. According to Gerus, for example, the management and regulatory committees for agriculture issued some 1894 opinions on Commission actions during 1990--not a single one of which was negative! 31 At first glance, the remarkably low rate of committee referrals to the Council would seem to suggest that committee oversight is perfunctory, and the Commission largely independent in its actions. However, as Gerus points out, rational anticipation of committee action by the Commission may mean that the Commission is effectively controlled by the member states, despite the startling rarity of sanctions against it. As one Commission official explained to me, having one's proposal referred from a committee to the Council can cast a long shadow over the career prospects of a young fonctionnaire--a powerful incentive to rationally anticipate a proposal's reception in the relevant committee 32 .
The point here is not that the Commission and other supranational organizations enjoy no autonomy from the member governments, but rather that such autonomy cannot be easily ascertained from the apparently independent behavior of supranational organizations, and that quantitative measures of comitology votes or legislative sanctions are unlikely to capture the nuances of agency autonomy from, and influence on, member governments. Instead of focusing on such broad aggregate data, I would argue, testing of the above hypotheses should rely on three particular research strategies. The first of these, and perhaps the most important, is to conduct systematic case studies and engage in careful process-tracing, in order to establish the respective preferences of the member states and supranational organizations, and the subtle influences that these actors may exert upon each other.
A second method, recently advocated by Moravcsik for the study of informal agenda setting, is counter-factual analysis, asking what would likely have happened if the Commission, the Court, or the Parliament had not behaved as they did in a given case. If it seems likely that a member state, or some interest group, would have stepped in to fill the breach, Moravcsik argues, then the independent causal role of the supranational organization is clearly less significant.
A third and final way to study the nature, and the limits, of supranational agency is to examine cases open conflict between supranational organizations and one or more member states, which may or may not result in sanctioning of the organization and a change in its behavior. The risk of focusing on such conflicts is that they are, after all, extremely rare, since agents like the Commission typically avoid open conflict with, and sanctioning by, their principals. Despite this risk, focusing on conflicts between member states and their supranational agents has the advantage of revealing the conflicting preferences among the various actors, and illuminating the conditions under which member governments are able--or unable--to rein in their supranational agents, limiting their autonomy and their influence on policy outcomes. Such incidents of open conflict are, furthermore, hard or critical cases for the principal-agent model presented above, according to which agents like the Commission should enjoy autonomy within the confines of member state preferences, and not directly against the member states. Hence, if we find that agents like the Commission enjoy some independent causal influence in cases of open conflict, it is likely that such agents should enjoy as much or greater influence in other, less high-profile cases where member states have little information or only weak preferences.
In keeping with these rough guidelines, I devote the rest of this chapter to a preliminary testing and illustration of the above hypotheses, focusing upon the executive actions of the European Commission in three issue-areas: the administration of the EC's Structural Funds, the conduct of EC competition policy, and the representation of the common EC position in the Community's external trade policy. Within each issue-area, I focus in some detail on a particular instance of conflict between the Commission and the member states, namely: the RECHAR controversy in the Structural Funds, the use of Article 90 and the debate over the Merger Regulation in competition policy, and the treatment of agriculture in the Uruguay Round of the GATT. In all three areas, I argue, member governments have delegated significant powers to the Commission, which the Commission has exploited to pursue its own preferences for "more Europe." By the same token, however, I shall demonstrate how the Commission's efforts in these areas have been constrained in particular by the preferences of the member governments, by the varying possibilities for sanctioning available to dissatisfied member governments, by the information available to the Commission and the member states at different points in time, and by the Commission's ability to strike up alliances with transnational actors and with other supranational organizations such as the European Court of Justice. I begin with the Commission's high-profile role in the administration of the Structural Funds.
In EC parlance, "structural policy" refers to the administration of the Community's Structural Funds--the European Regional Development Fund, the Social Fund, and the European Agricultural Guidance and Guarantee Fund Guidance Section--created primarily to reduce regional disparities in the Community. 33 In the 1970s and the early 1980s, these Funds were essentially a redistributive share-out, agreed as side-payments in larger intergovernmental bargains, and the Commission's executive role in implementing the Structural Funds was a minor one.
By the mid- 1980s, however, the largest contributing member states had become concerned with the efficient expenditure of the increasingly large Structural Funds, especially in the new Southern member states of Greece, Spain, and Portugal, and they began pressing for greater control over, and monitoring of, the use of EC funds. In Kingdon's terms, these calls for greater control created a "window of opportunity" for a new and more ambitious structural policy, with a greater role for the Commission. And it was in this context that an entrepreneurial Delors Commission seized the initiative, proposing a series of Structural Fund Regulations which simultaneously increased Community monitoring of Fund expenditures to include "value for money," while at the same time substantially increasing the Commission's role in both the planting and implementation of the Funds, which would henceforth take on a genuine Community dimension.
The Commission's reforms, which were adopted with very few amendments by the member governments, were based on four principles:
- concentration of the Funds' resources in the neediest areas;
- partnership among the Commission, the member governments and regional authorities in the planning and implementation of the Funds;
- programming, whereby member states would be required to submit comprehensive development programs for each region, rather than individual development projects as in the past; and
- additionally, the principle that any Community funds should be additional to, rather than replace, national development funds in a given area.
2.1 The Commission exploits its new powers: The RECHAR controversy
Between 1988 and 1993, the Commission exercised its new powers vigorously, building strong networks to subnational regions; launching Community Initiatives which reflected the policy agendas of the Commission rather than the member states; and insisting that all member states satisfy the Funds' criteria for additionality, the principle that all EC funds should be additional to, rather than replacement for, national regional funding. Indeed, in the most famous case of conflict under the 1988 reforms, the Commission's insistence on additionality in the granting of aid brought it into direct conflict with the British government of Prime Minister John Major. In that case, the Commission designed a Community Initiative program, dubbed RECHAR, the benefits of which would accrue largely to coal-mining regions in Scotland. In order to secure these benefits, however, the UK government would have to accept the Commission's definition of additionality, as laid down in Article 9 of the new framework Regulation for the Structural Funds. When the UK government refused to do so, Commissioner Bruce Millan froze the UK funds. Finally, in response to pressure from the Commission above and regional governments below, and with a general election looming, Major's government backed down and agreed to the Commission's demands in return for its share of RECHAR funding. For many analysts, this RECHAR incident became emblematic of the Commission's renewed power and influence vis-a-vis even the most powerful member states 36 . In the RECHAR case, as in Envireg and other Community Initiatives, member governments complained that the Commission was either interfering in internal affairs or duplicating efforts already underway within the CSFs; but, faced with the possible loss of EC funding, member governments gave in and participated in these programmes on the Commission's terms.
2.2 The Commission reigned in: The 1993 Structural Fund Reforms
As I have documented elsewhere, however, the Commission's position was fundamentally weakened by the fact that the 1988 Fund Regulations, and hence its own executive powers, were set to expire at the end of 1993, and required a unanimous vote from the member governments for reauthorization. In Scharpf's terms, the default condition for the Commission's powers was not the status quo but expiration, meaning that a positive decision would be required to reauthorize the Commission's powers under the 1988 Fund Regulations. Under the rules of the Single European Act, moreover, the unanimous agreement of the Council would be required for the most important of the new Fund Regulations. These decision rules considerably strengthened the position of member governments--including but not solely the British government--seeking to clip the Commission's wings by demanding substantial changes in the Fund Regulations. Without going into great detail, it is worth noting that the EC Court of Auditors, the European Parliament, and, above all, Member state governments had in the interim expressed a number of complaints about the implementation of the 1988 reforms. These included, inter alia, at least four major issues 37 .
First, a number of Member states complained, and the Commission itself acknowledged, that die three-stage planning system laid down in the 1988 Regulations (namely the drawing up of national development plans by Member states, the adoption of Community Support Frameworks (CSFs) by the Commission in cooperation with the Member states and regions, and the submission and approval of Operational Programmes (OPs) had created unnecessary bureaucracy for member governments, and should be simplified.
Second, many national policymakers complained about the system for designating the regions eligible for assistance under Objectives 2 and 5b, whereby the Commission identified the eligible regions under the conditions specified in the Fund Regulations. It was argued that area coverage was fragmented, and that lack of coordination between DG XVI (regional policy) and DG IV (competition) led to differences in regions eligible for EC Structural Funds on the one hand, and those regions regarded as eligible for national state aids under EC competition policy. The Member states therefore recommended greater coordination within the Commission, as well as a greater role for Member states--in keeping with the principle of subsidiarity--in the designation of Objective 2 and 5b regions within their respective territories 38 .
Third, in keeping with their on-going concerns with "value for money," Northern Member states, led by the UK, continued to demand more effective ex ante assessment and ex post monitoring and evaluation of programmes.
Fourth, and rather predictably, there was widespread irritation among the Member states (not simply the UK) with the Community Initiatives. Most Member states, for example, argued that there were too many CIs, and that each of these individual CIs spread a small amount of EC funding across a wide area, decreasing their effectiveness. The bureaucratic requirements for these initiatives, moreover, remained equally onerous for the Member states regardless of the size of the programmes, with the result that a large proportion of these funds were spent on administration. Most importantly, however,
...policy-makers believe strongly that there is too little consultation with Member States regarding the introduction of CIs. Indeed, the negotiation process has been described as a "complete sham" with the predominance of self-interest. Member States say that they are often taken completely by surprise when new CIs are launched. However, Member states and regions have a vested interest in receiving as much BC finance as possible, and it is difficult for them to object constructively to Commission proposals without harming their changes of obtaining funding. Policy-makers are frequently under political pressure, especially at regional levels, to apply for and use CI funds regardless of whether the money is limited and the measures are inappropriate or undesirable 39. Thus the Member states, if not the Commission, had a clear incentive to reassert control over this least predictable and controllable aspect of the 1988 reforms--and would indeed do so, as we shall see presently.
2.3 The Commission's proposals
In early 1993, following the adoption of the Delors II budgetary perspectives for 1993-1999 at the Edinburgh European Council, the Commission submitted proposals for the new Fund Regulations, which were described by the Commission as largely a continuation of the principles of the 1988 reforms, with several administrative changes to improve the efficiency of the Funds 40 .
Put simply, the Commission proposed to retain the four basic principles of the 1988 reforms, while proposing slight changes to each of these. Briefly, these principles, and the proposed changes, are as follows:
- Concentration of the Funds' resources on the neediest areas: Here, in keeping with the agreement of the Edinburgh European Council, the Cohesion Four were promised a doubling of their receipts from the Structural and Cohesion Funds combined; within the Structural Funds themselves, the percentage of the Funds directed to Objective 1 regions would pass from 63% to 70% by 1999.
- Partnership: Here the provisions would be altered only slightly, by specifying that the social partners, as chosen by the Member states, could take part in the design and implementation of the Funds, in partnership with the Commission, the regions and the Member states.
- Programming: Largely in response to Member state complaints, the Commission proposed a simplification of the planning procedure, reducing the existing three-stage plan to two stages, by allowing the Member states to submit their national development plans and their applications for Operational Programmes in a single document.
- Additionality: The wording of the additionality provisions, as laid down in Article 9 of the draft Coordination Regulation, would, unlike the 1988 Regulation, specify the criterion for meeting the additionality requirement and the obligations of the Member states in this regard. Thus, for each Objective, "the Commission and Member state concerned shall ensure that the Member state maintains, in the whole of the territory concerned, its public structural or comparable assistance at least at the same level as in the previous programming period taking into account, however, the macroeconomic circumstances in which the funding takes place including the implementation of convergence programmes [for EMU]."
Finally, with regard to the Community Initiatives, the Commission had proposed in its initial communications that 15% of the Funds' resources go to the CIs. In December 1992, however, the Edinburgh European Council indicated that the CIs should comprise between 5% and 10% of the Funds, and the Commission, predictably, proposed the high end of this range, 10% for the CIs from 1994 to 1999. With regard to the working of CIs, the Commission suggested that these would be fewer in number, and organized around a specific set of priorities. The Commission also proposed some minor administrative changes, which would allow a small portion of the funds to be spent outside the traditional Objectives 1, 2 and 5b areas, and allowing two or more Member states to file a single application for cross-border initiatives such as INTERREG.
2.5 The Council clips the Commission's wings
Now, many of these proposals, with the obvious exception of the last, were intended to address the concerns of various Member states with the operation of the 1988 Fund Regulations, expressed both informally and in the conclusions of the Maastricht Treaty and the Edinburgh European Council. In the event, however, these Commission proposals did not go far enough to address the concerns of the Member states 41 --which proceeded to change the substance of the Commission's proposals in several non-trivial ways, so as to respond to concerns about the distribution of funds, efficiency, and Member state control of the Funds' operation 42 . Four specific changes, of varying importance, are worth noting in this regard:
First, largely at the insistence of the UK, the wording of the provisions for monitoring and assessment were strengthened, with more detailed monitoring procedures and more explicit requirements for both prior appraisal and ex post evaluation. With regard to prior appraisal, for example, the Council draft, unlike the Commission version, specifies that "Assistance will be allocated where appraisal shows medium-term economic and social benefits commensurate with the resources deployed." Furthermore, in an interesting twist, whereas the Commission draft had specified that "Assessment shall primarily be the responsibility of the Member states," the Council changed the wording of the relevant Article to associate the Commission with the process of appraisal and evaluation. These changes clearly reflect the concerns of the contributing Member states to ensure rigorous evaluation, and Commission supervision, of expenditures in all the Member states, in order to ensure "value for money."
Second, at the insistence of the UK, France and Germany, Member states were given a more important role in the designation of Objective 2 and 5b regions. Thus, whereas Article 9 of the Commission 5 proposed Framework Regulation simply had the Commission draw up a list of Objective 2 regions, the final version of Article 9 as adopted by the Council has each Member state proposing to the Commission a list of Objective 2 areas, on the basis of which, "in close consultation with the Member state concerned," the Commission then adopts the final list of Objective 2 areas. A similar system was established for Objective 5b areas in Article 11a.
Third, the wording of the Commission's provisions on additionality was altered. The definition of additionality in terms of expenditure during the previous programming period was retained from the Commission's proposal, but whereas the Commission proposal specified that the this definition should take into account macroeconomic circumstance and convergence policy, the Council's final draft added: "as well as a number of specific economic circumstances, namely privatizations, an unusual level of public structural expenditure undertaken in the previous programming period and business cycles in the national economy." The effects of this new definition, and of the Council's amendments to it, remain to be seen, but the Council's amendments would appear to leave the Member states more room to decrease structural expenditure without violating the additionality provisions than the Commission's original draft.
Fourth, the Council amended the Commission's provisions regarding the Community Initiatives. Thus, for example, the amount to be devoted to the CIs was reduced from 10% to 9% of the total Structural Fund budget. More importantly, the Council created de novo, in a new Article 29a of the Coordination Regulation, a Management Committee for the Community Initiatives. Under the Management Committee procedure, the Commission would adopt Community Initiatives which would apply immediately, but these initiatives would have to be submitted to the Management Committee, which would approve or reject these by a qualified majority; if this committee rejected the Commission's proposals, the Council could, acting within a month of the committee vote, take a different decision by qualified majority. Predictably, therefore, the Commission openly "deplored" the creation of the new Management Committee 43 . In the event, the new Management Committee approved the Commission's proposals for new CIs, but the existence of such a committee meant that the Commission could stray only so far from the wishes of the Member states without risking having its decision overturned by the Council of Ministers.
Finally, it is also worth noting that the Commission also took aboard a number of Parliamentary amendments, the most notable of which was that the Commission should also notify the Parliament before the final adoption of Community Initiatives, as well as providing the Parliament with regular information about national development plans, the CSFs, and implementation of the Funds.
As a result of these changes, the Community Initiatives which were adopted for the period 1994-1999 under the new Fund Regulations were subject to an extended consultation with the Member states, the EP, and other interested actors such as regional and local authorities and the social partners. By contrast with the striking independence of the Commission in the selection of the early CIs, the Commission in June of 1993 published a Green Paper on the future of the Community Initiatives, which proposed a trimmed-down series of initiatives concentrated in five priority areas: cross-border and inter-regional cooperation, rural development, outermost regions, employment and vocational training, and adaptation to industrial change. This initial list of objectives and programmes was then modified, however, after consultation with the Member states and the EP, to include new initiatives on fishing (PESCA), on urban problems (URBAN), and--in response to the intergovernmental bargain struck over GATT ratification in December 1993--on a 400 mecu aid programme for the Portuguese textiles industry 44 . These revised proposals, for 13 initiatives spread over seven priority areas, were then approved by the new Management Committee, and formally adopted by the Commission in June 1994 45 .
2.7 Analysis
Summing up the Structural Funds case, the Commission was able in 1988 to capitalize on widespread member-state concerns about "value for money" to receive significant new powers to draw up Community Initiatives, to play a central role in drawing up Community Support Frameworks with the member governments and the regions concerned, and to police the expenditure and the additionality of EC funds. During the five-year lifespan of the 1988 Fund Regulations, Millan and the Commission used its delegated powers aggressively, funding Community Initiatives in line with the Commission's own policy agenda, and coining into direct conflict with the United Kingdom over the issue of additionality. Given the significant planning and policing powers of the Commission, and the requirement of unanimity to alter such powers, the Commission was able effectively to put its own stamp on the previously intergovernmental Structural Funds. In 1993, however, the Commission's delegated powers for structural policy were scheduled to expire, and the unanimity voting rule favored reformers like the UK, which successfully insisted on changes to the 1993 Fund Regulations in line with their own preferences.
The story of Commission autonomy and influence does not end with the 1993 reforms, however. As Gary Marks and others have pointed out, the 1993 Fund reforms still left the Commission with considerable, and in some cases increased, powers in both the planning and implementation of Community Support Frameworks and Community Initiatives 46 . Principal-agent interaction, therefore, is not a one-shot but an iterated game, in which the Commission exploits loopholes in Council legislation, the Council responds (if possible) by sanctioning the Commission, and the Commission begins the cycle again by making the most of its new mandate. The 1993 Fund reforms are merely the latest cycle in this ongoing principal-agent interaction.
In the area of competition policy, a number of analysts have correctly identified the Commission's powers on antitrust and state aids issues as "the first supranational policy" in the EC 47 . The drafters of the EEC Treaty had foreseen the possibility that certain actions by either private industry or by national governments might distort competition within the common market, and accordingly included in the Treaty a chapter on competition policy (Articles 85-94), which laid down the basic competition rules for the Community, and empowered the Commission to enforce these rules. The rules themselves fall into two broad groups: the first, laid down primarily in Articles 85 and 86 and elaborated in Regulation 17 of 1962, concern anticompetitive practices by firms, such as cartels and abuse of dominant positions, while the second, laid down in Articles 92-94, concern the compatibility of state aids with the common market. In each of these areas, the Commission has acted as the Community's competition authority, using its powers with increasing boldness in the 1980s, and using its competition powers to further the development of flanking EC policies in areas such as research and development and the environment. Throughout the area of competition policy, moreover, the Commission is supervised only by advisory committees, limiting the member states' ability to overturn the Commission's competition decisions through the comitology process.
Under the provisions of Article 85(1) and 85(2), inter-firm agreements and concerted practices which might affect inter-state trade or distort competition within the common market are generally forbidden. Article 85(3), however, raises the possibility of exceptions to this rule, provided that these agreements contribute "to improving the production or distribution of goods or to promoting technical or economic progress, while allowing consumers a fair share of the resulting benefit." Under these provisions, the Commission has enjoyed the authority to lay down and enforce guidelines on inter-firm cooperation in the area of research and technological development, maintaining a general ban on anti-competitive practices but also providing in some areas for "block exemptions" under which certain inter-firm agreements are assumed to conform to the criteria of Article 85(3), and are thus automatically exempted from the prohibition of Article 85(1) without the need for individual review. Thus, as early as 1971, the Commission prescribed such a "block exemption" for joint research and development, subject to certain conditions and to thresholds regarding the joint market shares of the participating firms. This 1971 block exemption was then replaced by revised, and somewhat more generous block exemptions in 1982 and 1985, in line with the Commission's general policy on encouraging the development of Community competitiveness in high-technology sectors 48 .
By the late 1980s, therefore, the legal framework for high-tech R&D collaboration in the EC was being set largely by the Commission, in accordance with its broader strategy on European technological competitiveness--one of many instances in which competition concerns are balanced against concerns about industrial policy and the competitiveness of Community firms in international markets.
Article 86 concerns the abuse of a dominant position by a firm or firms which use their market power to distort competition and affect EC trade by either imposing unfair prices, limiting production, or discriminating among consumers. By contrast with its activism in regard to the cartels provisions of Article 85, the Commission has made relatively little use of Article 86 until recent years, in part because the European Court of Justice has set high procedural hurdles for the Commission in demonstrating both the existence of dominance within a specified market and abuse of that dominant position. Since the late 1980s, however, the Commission has enforced the provisions of Article 86 with renewed vigor, imposing large fines on firms found to be engaging in predatory pricing and other anticompetitive practices, as in the famous TetraPak case of 1991.
The structure of the Treaty provisions on state aids are similar to those regarding cartels and anti-competitive agreements, except that they are addressed to the member states. Thus, under Article 92(1) of the Treaty, state aids to industry which serve to distort competition within the Community are generally considered to be incompatible with the Treaty and forbidden. Articles 92(2) and 92(3), however, then list certain categories of aid which either are, or may be considered to be, compatible with the common market, and are thus allowed under the Treaty. Among the many areas of state aids which maybe considered compatible with the Treaty are aids for the purposes of regional development, research and development, environmental protection, and culture. In the years since 1957, and with increasing industry since the 1980s, the Commission has laid down, and enforced, guidelines for state aids in all of these areas 49 .
During the 1970s, a period of both economic crisis and a sclerotic integration process, the Commission's enforcement of its competition powers is widely considered to have been lax, as the Commission tolerated cartels in sectors such as sugar, steel and shipbuilding, and routinely approved sizable state aids to declining industries. In the 1980s, however, under Competition Commissioners Peter Sutherland and Sir Leon Brittan, the Commission took advantage of the neoliberal preferences of the member states and the completion of the internal market to make greater use of its existing powers, cracking down on both cartels and anticompetitive practices, imposing larger fines on firms found to be have violated EC rules, and specifying conditions for state aids to industry. 50 We should be cautious, however, about assigning great causal importance to Sutherland's and Brittan's activism in this area. While both Commissioners were indeed determined to apply Community competition rules with renewed vigor, their efforts also coincided with the liberal turn toward the market among member-state governments, making the causal roles of the member states and the Commission exceedingly difficult to disentangle.
Rather than focusing on the Commission's activities in the traditional areas of competition policy, therefore, I focus on the Commission's recent activities in two new policy areas. During the late 1980s and early 1990s, the Commission aggressively exploited its long-dormant Treaty power to liberalize state monopolies under Article 90, and conducted a long campaign, ultimately successful, to acquire the power to review mergers and acquisitions of Community-level importance. In both of these areas, the Commission has come into conflict with various member states which have limited its influence, but has nevertheless expanded the scope of Community and Commission competence, and contributed to the completion of the internal market. Let us, very briefly, consider each of these two cases in turn.
3.1 Article 90 and national monopolies
The Commission's aggressive use of Article 90 to liberalize the telecommunications sector in the 1990s is one of the most spectacular and conflictual examples of Commission activity in any area of policy, and has been analyzed at some length by many scholars of European integration 51 . Put simply, Article 90 deals with public undertakings, such as state monopolies in telecommunications and energy. Under Article 222 of the Treaty, member states are free to determine the public or private nature of such utilities, and the Commission cannot force member states, for example, to privatize their telecommunications industries. However, at the insistence of Germany and the Benelux countries, which feared possible trade distortions resulting from the extensive public monopolies in France and Italy, the framers of the Treaty inserted Article 90, which allows the Commission to enforce EC competition rules vis-a-vis national monopolies. Furthermore, Article 90(3) contains an extraordinary clause allowing the Commission to issue Directives binding on the member states, without the approval of the Council of Ministers.
Prior to the 1980s, Article 90 had been invoked only rarely by the Commission. In the late 1980s and early 1990s, however, the Commission began to use Article 90 as a key instrument in its drive to liberalize the telecommunications sector in the EC. The Commission's initiative in this regard began in 1987, when it issued a "Green Paper on the Development of a Common Market for Telecommunications Services and Equipment," laying out its plans for the liberalization of the telecommunications sector, and calling for input from both member states and from interest groups representing the users of telecommunications services, who promptly mobilized in favor of the Commission's actions. In its communication, moreover, the Commission pointed to the 1985 British Telecoms case in the European Court of Justice, in which the Court ruled that telecommunications was a regular economic activity under the Treaty, and that BC competition law was applicable in the telecommunications sector. This and other rulings, according to Schmidt, presented the Commission with a crucial "window of opportunity" to apply the long-dormant provisions of Article 90 to a sector which it was keen to liberalize despite some member-state opposition 52 . The Commission therefore implemented a two-fold strategy, initiating specific enforcement actions against the member states under Article 90(1), while at the same time issuing more general Directives to the member states under Article 90(3).
For its first telecoms Directive under Article 90(3), the Commission chose a relatively uncontroversial Directive in terms of substance, much as the European Court of Justice had chosen small and uncontroversial cases to establish major points of European law in the 1960s. This "Terminals Directive," adopted by the Commission without the approval of the Council in May 1988, would liberalize the market for terminal equipment, a move which few member states opposed and which certain member states, such as the United Kingdom, strongly supported on substantive grounds. Nevertheless, the Commission's use of Article 90 to adopt liberalizing Directives without the approval of the Council would set a dangerous precedent which could be exploited in the future, and so a number of member states (France, Italy, Belgium, Germany and Greece) challenged the Directive before the European Court of Justice, arguing that the Commission should have proceeded under Article 100A, with the Council taking the final decision. In early 1991, the Court ruled in favor of the Commission, upholding the Directive and legitimating the Commission's use of Article 90(3) to issue Directives.
Having established with the Terminals Directive its right to issue Directives under Article 90(3), the Commission then turned to the more controversial problem of opening the market for telecommunication services, including data services. Unlike the terminals case, the liberalization of data services was disputed by member states such as France, which relied on such services to underwrite other telecommunications costs, such as the provision of service to outlying areas. The Commission therefore agreed to negotiate with the Council on a package deal comprising the both the Services Directive adopted under Article 90(3), and a framework Directive on Open Network Provision which was adopted by the Council. The final version of the Services Directive would open up the market for enhanced telecommunications services, including data transmission, from January 1993. This Directive was again challenged by France, this time with the support of Belgium, France, Italy and Spain, but the European Court once again upheld both the Directive and the Commission's use of Article 90(3). The adoption of telecoms Directives under Article 90(3), moreover, has continued into the 1 990s, with new Commission Directives on satellite services and equipment in 1994, the liberalization of cable TV networks in 1995, and the liberalization of mobile communications networks in 1996.
However, as Schmidt has persuasively argued, the telecommunications case presents a particularly favorable setting for the aggressive use of Article 90, featuring a string of favorable ECJ decisions, strong support among powerful interests within the member states, and a clear preference for liberalization among large member states such as the UK and Germany. Hence, while the Commission did enjoy extraordinary success with its use of Article 90 in the telecommunications sector, this initial success should not be taken as a sign that the Commission enjoys carte blanche to apply Article 90 in all sectors and regardless of the preferences of member governments. Indeed, as Schmidt demonstrates, the Commission has proceeded much more tentatively in its liberalization of another national monopoly, electricity.
In the case of the electricity sector, Schmidt demonstrates, the Commission began with a similar approach to the liberalization of the sector, adopting a report on "The Internal Market for Energy," in 1988, and following this up with a number of specific proposals in subsequent years. In contrast to the telecommunications case, however, the Commission could not rely on a clear ECJ ruling that the rules of competition applied to electric utilities. Furthermore, in the electricity case the Commission faced strong opposition from many of the member states, from the utilities, and from the European Parliament as well, leading the Commission to withdraw its plans for liberalization of the Community's energy market under Article 90(3). Instead, the Commission proceeded under Article 100A, requiring an agreement within the Council of Ministers, which took years to reach the awkward compromise contained in the final Directive adopted in June of 1996 53 .
In sum, the Commission was able to use its Treaty powers, conflicting preferences among the member states, and above all the support of interest groups and the European Court of Justice, to force the pace of telecommunications liberalization, which has been more rapid and more far-reaching than would likely have been the case if the various Directives had had to wind their way through tortuous Council bargaining. However, as the electricity example demonstrates, the Commission has been considerably less successful in its application of Article 90 in those areas where it lacked the support of interest groups, member governments, and the Court of Justice. The Commission, therefore, acted "alone" under Article 90 only in the narrow legal sense. More broadly, it relied on the support of a variety of national, transnational and supranational actors in liberalizing the telecommunications sector.
3.2 The Adoption of the 1989 Merger Regulation
The adoption of the 1989 Merger Regulation differs in the details from the Commission's use of Article 90, but here once again the Commission was able to achieve a major victory by rallying and relying upon the support of some member states, a large number of transnational interest groups, and the supranational Court of Justice 54 . In order to understand the nature of the Commission's victory in this area, it is important to note that the Commission's competition powers under the Treaty of Rome, although far-reaching, did not include the power to vet mergers and acquisitions, even under Article 86 on the abuse of a dominant position. As Goyder writes in his authoritative analysis:
... there seems little doubt that those responsible for the drafting of Article 86 did not intend that it should give control over mergers to the Commission. Neither the actual wording of the Article nor the evidence of those who participated in the negotiations leading up to the Treaty or in its early administration support any contrary argument . 55Nevertheless, Goyder continues, the absence of any direct control over mergers and acquisitions was seen as a great weakness by the Commission and its Directorate-General IV in charge of competition, and so the Commission decided in the early 1970s to apply Article 86 to prevent a merger which would strengthen the pre-existing dominance of a firm within a particular market.
In its 1972 decision, the Commission prohibited the Continental Can group from acquiring a Dutch packaging company, TDV, arguing that such a merger would increase the already large market share of the Continental Can in the Benelux countries and in Germany, and thus constitute an abuse of the company's dominant position. Continental Can, however, appealed the Commission decision to the European Court of Justice, whose landmark 1973 ruling overturned the Commission's decision on the facts of the case, while supporting the Commission's interpretation of Article 86. More specifically, the Court ruled that the Commission had not demonstrated that the merged Continental Can/TDV group would indeed acquire a dominant position within the markets specified by the Commission, and therefore allowed the merger to go ahead. The Court also ruled, however, that the Commission was correct in arguing that the Commission could indeed use Article 86 to prevent a firm which already enjoyed a dominant position within a given market from expanding its market share through mergers and acquisitions.
Continental Can was a landmark ruling in terms of its expansive reading of the Commission's powers under the Treaty of Rome, but from the Commission's perspective it was not a satisfactory legal basis for exercising control over mergers within the EC, since the impact of the Court's decision was limited to mergers and acquisitions by firms which already enjoyed a dominant position, and not to mergers which would create such a dominant position. The Court's decision, moreover, would give the Commission only post-hoc jurisdiction over mergers, not the prior notification or control of mergers which it sought. In 1973, the Commission therefore proposed to the Council a Merger Regulation that would give the Commission the power to review Community mergers and acquisitions in all cases where the joint turnover of the undertakings concerned exceeded a threshold of one billion units of accounts (later ecus). In its draft Regulation, the Commission proposed that all proposed mergers above this threshold would be subject to prior notification to the Commission, which would then take a decision within 12 months, subject to judicial review by the ECJ. Unfortunately for the Commission, the Commission's proposed regulation remained deadlocked in the Council of Ministers for 16 years, stymied by fundamental member-state opposition to any increase in the Commission's supranational powers in competition policy.
A major step forward was taken, however, with the 1987 Philip Morris ruling of the European Court of Justice. In the Philip Morris case, the Court of Justice ruled that Article 85, which deals with cartels and other anticompetitive practices, could apply to agreements between two or more companies that allowed one of the companies to obtain legal or de facto control over the other. The practical effect of the Court's decision was to provide the Commission with a back-door means of reviewing mergers and acquisitions, and the Commission, which had not sought the Court ruling, responded by successfully applying Article 85 to a number of high-profile mergers, applying conditions to the takeover by British Airways of British Caledonian, and blocking the acquisition of Irish Distillers by GC and C Brands. As a result, the European business community was left uncertain as to its legal responsibilities, which would emerge only in the incremental case law of the Court of Justice, and began lobbying for an EC Merger Regulation which would spell out clearly the powers of the Commission and the responsibilities of business. As Allen writes:
. In March of 1988, therefore, the Commission introduced a new, amended version of its 1973 proposal for an EC Merger Regulation, which was adopted by the Council in December 1989. Within the Council, negotiations focused on two provisions of the Commission's proposed regulation. The first of these concerned the thresholds above which a merger would become subject to the jurisdiction of the Commission, while the second concerned the balance in the Regulation between strict competition concerns on the one hand and concerns about industrial and social concerns on the other. With regard to the thresholds, the Commission, with the support of some of the smaller member states, proposed that EC jurisdiction apply in all cases in which the combined world turnover of the undertakings involved was one billion ecus, with a Community turnover of 50 million ecus for each of the undertakings. These thresholds were resisted, however, by Britain, France and Germany, all of which proposed a threshold of 10 billion ecus in joint world turnover. In early 1989, the Commission proposed a compromise proposal, specifying thresholds of five billion ecus in world turnover, and 250 billion ecus in aggregate Community turnover for each company; these thresholds, however, would be subject to later review and amendment by a qualified majority vote in the Council of Ministers. The member states, including Britain, France and Germany, agreed to this proposal.From the Commission's perspective the great advantage of this merger regime, using Article 85, was the uncertainty it generated. This served to put pressure on the doubting member states to settle for a better worked-out and potentially more limited merger regulation. By a combination of luck and skill the Commission had managed to create a problem which the Council felt could be eased only by passing the legislation it had previously refused to consider 56
The second contentious issues concerned the balancing of competition and other criteria, which split the member states into two camps. The first camp, led by Germany and Great Britain, argued for the criteria to be strictly limited to competitive issues, in line with their own approach to mergers and acquisitions, while the second group, led by France, wanted to include social and industrial policies among the criteria which the Commission could apply in assessing proposed mergers. The Commission's initial draft had included social and industrial policy issues among the criteria to be considered, but the Council's final version was closer to the German and British position, emphasizing the strict application of competition rules. Nevertheless, Article 2(1) of the Regulation does contain a brief reference allowing the Commission to consider in its decisions "development of technical and economic progress provided that it is to consumers' advantage and does not form an obstacle to competition." 57 How the Commission would interpret and apply these criteria, however, would become clear only after the Merger Regulation had come into effect, in September 1990.
3.3 The Merger Regulation Since 1990
A complete discussion of the Commission's implementation of the Merger Regulation is, of course, beyond the scope of this paper, but a brief discussion is nevertheless in order. Under the 1989 Regulation, the Commission must respond to all prior notifications within one month, at which time it may indicate either that the merger is approved, or that a second-stage investigation is being held into the specifics of the merger. By and large, the Commission has lived up to these deadlines, closing the vast majority of cases within the specified periods. In the first four years, for example, the Commission approved some 256 mergers during the first phase of the process, while some 18 cases were sent on for further investigation in the second phase. Of these 18 cases, four were approved with no further conditions, ten were approved with conditions, two were refused outright, and two were referred to member state competition authorities 58 .
Despite its procedural promptness, the Commission has nevertheless come under assault for its administration of the Merger Regulation, in particular from Germany, which has a tradition of strict competition enforcement by the Bundeskartellamt (Federal Cartel Office). As Wilks and McGowan point out in their excellent review, the criticisms have been threefold. First, it is often argued, the Commission's implementation of the Regulation lacks transparency, with the Commission frequently striking up informal agreements with the companies they are regulating. Second, the Commission is often accused of violating the principle of subsidiarity, by refusing to refer competition cases to national competition authorities which request the right to handle specific cases.
Third, and perhaps most importantly, the Bundeskartellamt, supported by the German government, has argued that the Commission's enforcement of the Merger Regulation is excessively lax and "politicized," with the Commission approving mergers which should have been blocked, and improperly applying social and industrial policy criteria to merger decisions 59 . The heart of the problem, in this, lies in the nature of the Commission's decision-making structure. By and large, DG IV and the Commissioner in charge of competition policy tend to take a hard line on competition policy issues, including mergers as well as state aids and other competition cases. The final decision on mergers cases, however, lies not with DG IV but with the full Commission, where the competition criteria spelled out in the Regulation can be watered down by Commissioners who have either (a) national sympathies for the companies involved, or (b) a functional interest in other policy areas, such as social or industrial policy.
The nature of the problem is illustrated vividly by Ross's depiction of the Commission's decision to block the merger of the acquisition of the Canadian aircraft manufacturer de Havilland by ATR, a Franco-Italian consortium--the first merger rejected by the Commission under the Merger Regulation. According to Ross, Brittan took a hard line on the de Havilland case, writing to Delors that "we must not allow this merger to proceed," and planning to make the de Havilland merger a test case. Within the full Commission, however, Brittan was opposed by Delors, by the other French and Italian Commissioners, and by Industrial Policy Commissioner Martin Bangemann, who argues that the merger would give European aircraft manufacturers the economies of scale they needed to compete on world markets. Finally, after days of lobbying within the Commission by Brittan and hi cabinet,, the Commission voted narrowly, by nine votes to seven, to block the de Havilland merger. Ross's discussion of the vote is particularly telling, and is worth citing at some length:
Beyond the other Commission liberals (Andriessen, Schmidhuber, and Christophersen), Cardoso, MacSharry, Dondelinger, Millan, and Marin all voted with Brittan. For each of these commissioners Brittan's hard politicking had been central. Most said simply that they were giving way to the "commissioner in charge," although others, including Millan (a British Labourite with whom Brittan almost never saw eye to eye) and Dondelinger, indicated that they disagreed with Brittan even though "giving way." Manuel Marin, the swing vote, gave an utterly incomprehensible argument to explain himself, but seemed to have traded his vote for help from Brittan on a pending project to reorganize the European fishing fleet, an essential Spanish dossier. When it became clear that he would lose, Delors, as was his usual practice, abstained. 60At the end of the day, the merger had been blocked, leading to tremendous controversy in France and in Italy, but the de Havilland case had made clear how haphazard the Commission's decision-making process was in the area of competition policy, and it provided grist for the mill of the Commission's critics in Germany and elsewhere.
3.4 The Future of the Merger Regulation
The years since the de Havilland merger have witnessed a double stalemate, with Germany attempting vainly to create a European Cartel Office (ECO) which would enforce EC merger rules in place of the Commission, while the Commission has attempted, equally vainly, to increase the scope of its competition powers by lowering the thresholds established by the 1989 Regulation. With regard to the former, the Bundeskartellamt supported by the German Economics Ministry, has pressed for the creation of an ECO independent of the Commission, which would decide merger cases on the basis of narrowly defined competition criteria, thus insulating the decision-making process from the two-fold politicization found in the college of commissioners. The creation of such a new agency, however, would almost certainly require an amendment to the Treaties, and hence a unanimous agreement from the member states 61 . In recent months, the German government has pressed its case for an ECO within the 1996 intergovernmental conference, attracting the support of Italy, but Germany seems unlikely to gam a unanimous consensus in favor of its proposals from other member states, such as France, which have traditionally been more sympathetic to social and industrial policy concerns 62 . In the case of competition policy, therefore, in contrast to the Structural Funds case discussed above, the status-quo default condition and the difficulty of amending the Treaties has favored the Commission, whose powers in the area of merger regulation remain, at least for the moment, secure.
However, while Germany has thus far been unsuccessful in its efforts to transfer the Commission's merger powers to a new ECO, the Commission has been equally unsuccessful in its attempts to expand its powers by lowering the thresholds for Commission jurisdiction. In keeping with the provisions of the Merger Regulation, the Commission proposed in 1989 to lower the thresholds from 5 billion ecus for worldwide turnover and 250 million ecus for Community turnover, to 2 billion and 100 million, respectively. According to the rules laid down in the Regulation, the Commission would require only a qualified majority vote in the Council to approve the new thresholds, an easier target than the unanimous vote required to adopt the initial Merger Regulation. Furthermore, the Commission's proposal was broadly backed by European business, including the peak employers' association UNICE, which was eager to expand the "one-stop shop" provided by the Regulation for European-level mergers 63 . An initial survey of member state positions, however, revealed fundamental opposition to the Commission's proposal from Germany and the United Kingdom (which found the Commission's enforcement too lax), and from France (which found it too strict), and so in August of 1993 the new Commission Commissioner, Karel van Miert, withdrew the Commission's proposal, bluntly asking, "what is the point of proposing something if you know it won't be accepted?" 64 Instead, Van Miert proposed to continue with the existing thresholds for three more years, and propose new thresholds in the light of experience in 1996.
In January of 1996, the Commission published a Green Paper on the review of the Merger Regulation, renewing its case for a lowering of the thresholds to 2 billion and 100 million ecus, respectively, and this was followed by formal proposals in July. Once again, however, the Commission has encountered entrenched resistance from the member states, and in particular from Germany, which has linked any lowering of the thresholds to the creation of an independent EC0. 65 The result, under the status-quo default condition of the Merger Regulation, is likely to be a continued stalemate, in which the Commission maintains its powers and discretion in merger control, despite dissatisfaction in Germany and others, but is unable to expand these powers due to the entrenched opposition of several member governments.
3.5 Analysis
In the case of competition policy, as in the earlier case of structural policy, the balance sheet on the Commission's autonomy and influence in competition policy is mixed. One the one hand, I have argued that the Commission was successful in applying Article 90 to the liberalization of the telecommunications sector, which would have been unlikely to take place as rapidly or thoroughly had Council decisions been required for every Directive in that sector. Similarly, we have seen how the Commission exploited the uncertainty of both member governments and private-sector firms after the Philip Morris decision to press the Council to adopt the Merger Regulation after 16 years of stalemate. In this case, as in the telecommunications case discussed above, the Commission and the Court effectively shaped the incentives of the member states by altering the status quo, so that while the Council took the ultimate decision, it dad so in the shadow of the actions of the Commission and the Court. In both cases, finally, the Commission was protected against member-state sanctioning by the status-quo default condition governing its powers under competition policy, and by the difficulty of amending the Treaty to amend the Commission's mandate.
On the other hand, we have also seen clear limits to the Commission's ability to overcome determined resistance among the member states. In the case of national monopolies, for example, the Commission was unwilling to force through the liberalization of energy markets under Article 90(3) in the face of opposition from both member governments and transnational interest groups, and without the explicit support of the ECJ. Similarly, in the area of merger control, the Commission enjoyed a triumph with the adoption of the Merger Regulation in 1989, but has since been unsuccessful in its attempt to have the thresholds of the Regulation lowered in the face of firm member-state opposition. Once again, therefore, in order to understand the Commission's autonomy and influence we need to focus not only on the Commission's formal powers, but also on the preferences of the member governments, the rules for sanctioning the Commission, the information available to the various actors, and, perhaps most importantly, the availability of transnational coalitions with interest groups, and, just as importantly, with the European Court of Justice. I will return to this point in the conclusion of the paper.
The Commission's delegated authority in the area of external trade policy constitutes, alongside competition policy, some of its oldest and most important powers, specified directly in the body of the 1957 EEC Treaty. Under Article 113 of the Treaty, the Community possess exclusive competence in the area of commercial policy, and the Commission is designated as the sole and exclusive negotiator for the Community for all international trade negotiations, at which the member states are forbidden to negotiate independently with third parties. The Commission, however, is not given a free hand to negotiate whatever agreements it likes at the international level. Rather, the Commission begins the process by proposing a negotiating mandate to the member states, who may amend and adopt the Commission's mandate within the so-called "Article 113 Committee," a committee of senior national officials who approve, by a qualified majority, the Commission's negotiating mandate. The Article 113 Committee also monitors the Commission's conduct of the negotiations, and may, in response to a request from the Commission, amend the Commission's negotiating mandate, again by qualified majority. Furthermore, the formal agreement negotiated by the Commission must be ratified by the General Affairs Council on behalf of the member states, imposing a final check on the Commission's negotiating authority 66 .
In theoretical terms, the Commission's role in external trade policy is closely analogous to the position of the chief negotiator in Putnam's two-level games model 67 . In Putnam's model, all international negotiations take place simultaneously at two levels: At the international level, or Level 1, chief negotiators bargain with their foreign counterparts m an effort to reach a mutually beneficial agreement. At the domestic level, or Level 2, the same chief negotiator engages in bargaining with her domestic constituencies, which must ultimately ratify the contents of any agreement struck at Level 1.
For each state, Putnam specifies a "win-set," which is the range of international agreements at Level 1 which are likely to be ratified in a straight up-or-down vote at Level 2. States with small "win-sets," he argues, are likely to enjoy greater bargaining power in international negotiations, since they can credibly claim to be able to ratify' only a narrow range of agreements, although these states also run the risk of failing to reach any agreement, or of "involuntary defection," or failure to ratify agreements reached at Level 1. Each state's win-set, in turn, is determined by three factors. The first, and most obvious, of these is the preferences and power of domestic actors at Level 2, who must ratify any agreement reached at Level 1. The second is the political institutions at Level 2, which establish the ratification procedures specific to any given state; the more demanding the ratification procedure, Putnam argues, the smaller the win-set and the greater the likelihood of involuntary defection. The third factor determining the size of the win-set is the bargaining strategies of chief negotiators, who can direct side-payments or sanctions to their own or their counterparts domestic constituencies in order to manipulate the size and shape of their respective win-sets.
Perhaps most importantly for our purposes, Putnam also examines the role of the chief negotiator, whose preferences (like those of any agent) may diverge from those of her domestic principals, and who may be able to influence the substance of an agreement by virtue of her dual role at both the international and the domestic bargaining tables. At the international level, for example, a chief negotiator may adopt strategies to strengthen her own Level 1 bargaining position, by manipulating or misrepresenting the nature of her own win-set at Level 2, or by manipulating the win-set of her adversaries through the selective application of side-payments or sanctions to key domestic actors in other states. In both of these ways, a chief negotiator may manipulate the size of her own and her counterparts' win-sets, and thus her bargaining leverage at the international level.
Conversely, a chief negotiator may employ international pressures, and her own strategic position at both boards, to manipulate her own domestic constituencies. A chief negotiator may, for example, be eager to effect some domestic policies or reforms, but be unable to do so because of resistance from a coalition of domestic interests. In a two-level negotiation, however, the chief negotiator may plausibly argue to her own domestic constituents that her preferred policies are in fact necessary in order to reach agreement at the international level, and must therefore be accepted in order to enjoy the benefits of the overall agreement. The chief negotiator's domestic position may be further strengthened if, as in the case of the United States' "fast track" authority, the resulting international agreement must be ratified according to a straight up-or-down vote, thus providing the chief negotiator with formal agenda setting power and increasing the likelihood of ratification at the domestic level 68 . In sum, the chief negotiator's central position at both the international and the domestic tables may strengthen her bargaining leverage at both tables simultaneously.
In his liberal intergovernmentalism model, Andrew Moravcsik has adapted Putnam's two-level games approach to the study of the European Community, in which EC member governments act as chief negotiators between their domestic polities and parliaments on the one hand, and their fellow member governments on the other hand. According to Moravcsik, this privileged position has allowed member governments to increase their own autonomy vis-a-vis their domestic constituencies, by concentrating resources--initiative, information, institutions, and ideas--in the hands of the member governments negotiating in Brussels. In Moravcsik's model, national parliaments and other domestic constituencies are simply left to rubber-stamp the decisions taken by member governments m Council, and the net effect of the Community's two-level game has been to strengthen, rather than weaken, the member governments of the EC 69 .
Applying Putnam's model to the external relations of the Community, however, reveals that EC trade negotiations are not a two-level but a three-level game: At Level 1, the Commission negotiates with representatives of the United States and other trading partners, in order to reach international trade agreements. These agreements must then be ratified at Level 2, representing the intergovernmental Article 113 Committee and the Council of Ministers. Finally, at Level 3, national governments seek domestic ratification of decisions taken at Community level 70 .
In this three-level game, the Commission as chief negotiator should theoretically enjoy many of the advantages of Putnam's COG, manipulating and misrepresenting its own win-set to increase its bargaining leverage at the international level, and using external pressures to increase its "domestic" bargaining leverage vis-a-vis both the member governments at Level 2 and national interest groups at Level 3. The possibilities of the Commission 5 role as COG in external trade policy, and its limits, are well illustrated by the negotiation of the Uruguay Round and its most contentious element, agriculture.
4.1 The Uruguay Round, the CAP, and the Commission
Convened in 1986 in the Uruguayan capital of Punta del Este, the Uruguay Round of the GATT was to address a number of new issues in the area of international trade, including most notably trade in services, trade-related intellectual property issues (or TRIPs), trade-related investment issues (or TRIMs), and the creation of a new World Trading Organization to encompass the existing GATT. In each of these areas, the Uruguay Round attempted to establish rules for issues which had previously been outside the domain of multilateral international trade negotiations. Indeed, both services and TRIPs involved areas of so-called "mixed competence", for which the member states agreed to negotiate with one voice, and appointed the Commission as their sole negotiator, but without prejudice to the ultimate distribution of competences between the member states and the Community 71 .
Undoubtedly the most difficult issue, however, and one in which the Community had clear and exclusive competence as a result of the Common Agricultural Policy, was agriculture 72 . Agriculture had been included in previous rounds of the GATT, but various exemptions to GATT rules meant that states were in effect free to adopt national (or Community) systems for subsidizing and protecting national production, and for export subsidies as well. By the late 1 980s, however, the Reagan Administration was determined to secure a substantial reduction in agricultural subsidies, especially in the EC, where both subsidies and exports had grown rapidly in the course of the previous decade. Indeed, it was largely American concerns about agricultural subsidies which led the Reagan administration to press in 1985 for the opening of the Uruguay Round.
Not surprisingly, the initial US and EC negotiating positions were far apart. For its part, the Reagan administration put forward a radical proposal, often called the "zero option," calling for the complete elimination of agricultural subsidies by the year 2000. To the Commission and the member states of the Community, on the other hand, such an approach was anathema, as it would threaten the fundamental principles of the Common Agricultural Policy. Put simply, the CAP is based on a series of guaranteed, Community-wide prices for the various agricultural products. Within the Community, the CAP establishes a so-called "intervention price" for each commodity, beneath which the Community will automatically buy-in any excess production. In addition, the CAP establishes a "threshold price" for each item, which in most cases is considerably higher than the world market price. The Community then imposes a variable import levy, equal to the difference between the Community threshold price and the world price, on all imports into the EC, while at the same time granting export subsidies equal to the difference between the Community's internal market price and the world price. The result of this system, simplifying only a bit, was high and stable agricultural prices for EC farmers, coupled with chronic overproduction of certain products (the legendary lakes of wine and mountains of butter), and increasing levels of EC exports at the expense of the United States and other agricultural producers 73 .
The Uruguay Round therefore presented the Delors Commission with both challenges and opportunities. As chief negotiator for the Community, the Commission would face the challenge of reconciling the far-reaching demands of the United States and other states at the international level with the entrenched resistance to any reform of the CAP among EC farmers and among their representatives in the Council of Agriculture Ministers and the Article 113 Committee. The nature of this challenge is well illustrated by the abortive Brussels meeting of the GATT in December 1990, which was intended to conclude the Uruguay Round in time for the pre-established deadline of 31 December 1990. For this "final" round of negotiations, the Commission would require a revised negotiating mandate by 15 October, but the Council of Ministers was deadlocked for months in approving the Commission's proposals, largely because of German and French reticence to make any agricultural concessions on the eve of the first all-German elections and the intergovernmental conference (IGC) which would lead in 1992 to the Maastricht Treaty. Finally, in mid-November 1990, the "Jumbo" Council of foreign and agriculture ministers approved a new mandate for the Commission, but this mandate was narrow and inflexible, leaving the Commission little room to make concessions in Brussels, and leading US negotiators to observe that "the Commission would not negotiate until it had a mandate from the Council and could not negotiate once it had been given one." 74 .
At the Brussels meeting in December, the United States modified its earlier "zero option" proposal on agriculture, calling instead for a 75 percent reduction in internal subsidies and a 90 percent reduction in export subsidies over a 10-year period beginning in 1990. By contrast, the Community's opening position in the negotiations called for only a 30 percent cut in internal supports from 1986 levels (approximately half of which had already been achieved), and no firm commitments on export subsidies--a position rejected by most of the major parties to the negotiations, including not only the United States but also the Cairns group of agricultural producers led by Australia. In response to these demands, and to a possible breakdown of the Brussels meeting over agriculture, the Commission indicated that it might be willing to be more flexible and undertake more specific commitments on export subsidies in return for concessions from the United States and others on issues such as market access, services, and intellectual property. The Commission's proposal, however, had not been cleared with the EC's national trade and agriculture ministers, who did not take part in the actual negotiations but nevertheless attended the Brussels meeting to monitor the Commission's behavior. Upon hearing about the Commission's offer, French Agriculture Minister Louis Mermaz argued that the Commission had exceeded its mandate, and the Commission was publicly rebuked by the Council, leading to a collapse of both the Commission's credibility and the Brussels meeting, which broke up amid nearly universal condemnation of the Community by the other delegations. Put simply, the Brussels meeting demonstrated that the Commission, despite its key role as chief negotiator, could not press too far ahead of the member states in its proposals without risking a backlash, and a correction, by recalcitrant member states.
On the other hand, as Putnam points out, the Commission's presence at both negotiating tables (the EC and the international) also provided it with the possibility of using external pressure to strengthen its negotiating position internally, and vice-versa. The Commission, and in particular President Jacques Delors and Agriculture Commissioner Ray MacSharry, therefore established a dual strategy, with two central goals. First, Delors and MacSharry would design and steer through the Council a far-reaching reform of the bankrupted CAP, designed to make the CAP sustainable through the long term and avoiding any possible bankruptcy or renationalization of the system, while at the same time making the CAP compatible with the minimum demands of the Community's trading partners. Second, and equally importantly, the Commission would present the newly reformed CAP to its trading partners as the Community's bottom-line offer, beyond which the Commission could argue that its hands were tied. Ross, although eschewing the language of two level games, nevertheless sums up the Commission strategy perfectly in his account of the actions of Delors cabinet official, Jean-Luc Lamarty:
Once on the table, Jean-Luc thought, the reform would almost certainly grant the Commission more maneuvering room on the Uruguay Round front, both internally and externally. The fact that CAP reform was in progress would set limits on external pressures while the need for CAP reform to succeed in the Uruguay Round would work internally. 75This is indeed what took place, but the process was to take several years, leading to open conflicts between the Commission and the United States, the Commission and the EC member states, and within the Commission itself
4.4 Internal Bargaining: The 1992 CAP reform
The debacle at the December 1990 meeting of the GATT in Brussels had made clear, at least to the Commission, the need for substantial GATT reform in order to unblock the GATT negotiations, providing the Commission with a strong external incentive to press for CAP reform. In addition, however, the CAP in the early 1990s faced an internal crisis, with rising stocks of agricultural surplus and a spiraling budget which, according to Commission estimates, was expected to increase by some 30 percent in 1991. Such expenditures would, if left unchecked, lead to the breaching of the Community's agricultural guidelines established by the European Council in 1988, and could well create pressures for the renationalization of agricultural policy in the Community. Fearing the collapse and possible renationalization of the CAP, Directorate-General VI (Agriculture) began work in late 1990 on a series of proposals for a radical reform of the CAP designed to bring spending under control and reduce the need for export subsidies to bring the CAP into line with the demands of the Uruguay Round negotiations 76 . In February 1991, after an extended "seminar" and debate, the full Commission approved a general communication to the Council on the need for reform, followed in August by a detailed reform plan, often referred to as the MacSharry reforms 77 .
MacSharry's message to the Council about the state of the CAP and the need for reform was unequivocal:
It appears in these conditions that the Community's agricultural policy cannot avoid a succession of increasingly serious crises unless its mechanisms are fundamentally reviewed so as to adapt them to a situation different from that of the sixties...In response to this crisis, MacSharry proposed a sweeping set of reforms, the heart of which consisted of a shift from price support to direct payments to farmers. More specifically, the prices of a number of agricultural products would be cut severely, most notably for wheat, which would be cut by some 35% to a target price of 100 ecus per ton, near world market levels. Farmers would then be compensated for their loss of income through a system of direct payments linked to the total acreage of each farm. Finally, compensation would be "modulated," so that small farmers would receive greater compensation than large farms, and in all cases compensation would be linked to a commitment to set aside acreage to avoid overproduction in the future. Although the Commission plan would not save money in the short term, and indeed might lead to a slight increase in agricultural spending to finance direct payments to farmers, in the long term the plan would reduce the CAP's incentive to overproduce, and hence the CAP's persistent pressures on the EC budget 79 . A final consideration, unspoken but implicit in the Commission's proposals, was that the proposed reforms would bring EC agricultural prices closer into line with world prices, and thereby reduce the trade-distorting effects of the CAP and the need for export subsidies which were the most sensitive issue in the GATT negotiations. In short, MacSharry's proposed reforms would increase the size of the Community's win-set, and hence the Commission's negotiating mandate, while at the same time drawing a clear and conspicuous bottom line beyond which the Community could refuse to go.The Commission wishes to emphasize that the status quo is the one option that it considers not to be viable. If the present policy is not changed rapidly, the situation on the markets and, as early as the current year the budget position, will become untenable
78 .
In Kingdon's terms, the stalled Uruguay Round negotiations and the budgetary pressures in early 1991 provided the Commission an important "window of opportunity" to press for a far-reaching reform of the CAP, which had been rejected or watered down by a coalition of farm lobbies and agriculture ministers in previous years. Nevertheless, MacSharry's proposals faced unanimous opposition among both the agriculture ministers and EC farm groups when he introduced them to Council in February 1991, and passage was by no means assured. Some member states, such as Great Britain, Denmark and the Netherlands, supported MacSharry's case for CAP reform, but opposed his plans for modulated payments, which they argued would benefit inefficient small farmers at the expense of the larger and more efficient British, Danish and Dutch farmers. At the other extreme, France initially resisted the move from price support to direct payments, while Germany fought for price cuts considerably less draconian than the 35 percent cut in wheat prices sought by MacSharry. Community farm groups, finally, joined the agriculture ministers in their hostility to MacSharry's proposals, which they argued would decrease farm incomes and were being proposed only in response to bullying from the United States.
The twin pressures of the Uruguay Round and the budget, however, eventually led the member governments to support the broad lines of MacSharry's proposals, although several specific provisions of the plan were altered in Council bargaining during the first half of 1992. More specifically, in order to reach agreement, the Portuguese Presidency of the Council proposed a series of compromises in the Commission proposal, including the abandonment of the Commission's proposed modulation scheme to benefit small and medium-sized farmers, which MacSharry reluctantly accepted. More contentious, however, was the Presidency's proposed compromise on the cuts in the price of wheat, which were the linchpin of MacSharry's reform proposals. Whereas MacSharry had proposed a 35 percent cut in the price of wheat (from 155 ecus per ton to 100 ecus), Portuguese Agriculture Minister Arlindo Cunha in April proposed a compromise cut of 27 percent (112 ecus per ton), largely to appeal to Germany, a high-cost producer. MacSharry, however, reportedly dug in his heels and refused to agree to such a cut, which would lead to continuing overproduction and sabotage the Commission's negotiating position within the GATT. Instead, he persuaded Cunha to propose a new price of 110 ecus, "a 29 percent cut close to the 30 percent the Commission had always set out to achieve, with a 1 percent psychological sweetener to enable Germany to feel it was in the '20s'" 80 . The Germans accepted the proposal and the Council, after a last-minute snag on the question of Italian milk quotas, adopted the most radical reform of the CAP since the policy's inception in the 1960s. The Commission had compromised on modulation and on wheat prices, but the central, radical element of the Commission's reform--the shift from price support to direct payments--remained intact in the final Council bargain, leading one of the Commissioner's aides to label the reforms "son of MacSharry, definitely." 81 Remarkably, the member states had adopted in its essentials a reform plan which they, together with the EC's farmers, had been unanimous in rejecting only 18 months earlier.
4.5 External Bargaining: Negotiating Blair House
Having secured the passage of CAP reform, MacSharry and the Commission returned to the agriculture negotiations with the United States, armed with the MacSharry reforms as the Community's new bottom-line negotiating position. The GATT negotiations had dragged on during 1991 and early 1992, including a desperate attempt to broker agreement by GATT director-general Arthur Dunkel, but the negotiations had once again been deadlocked over agriculture, as the French and the Irish objected to the agricultural provisions of Dunkel's Draft Final Act. The talks were further delayed in the summer of 1992, moreover, by Mitterrand's reluctance to alienate French farmers on the eve of the French referendum on the Maastricht Treaty, which was ratified by only a narrow margin on 20 September. Finally, in late September of 1992, US-EC negotiations on agriculture resumed.
The Uruguay Round negotiations were further complicated, however, by the emergence of a new agricultural dispute between the EC and the United States, involving EC subsidies to Community oilseeds producers. Under a 1962 GATT agreement, the EC had agreed to grant US oilseeds duty-free status, yet beginning in the 1970s the Community offered subsidies to European oilseeds processing, contributing to a significant decline in the US share in the European oilseeds market. The US government accordingly took the oilseeds dispute to a GATT arbitration panel, which ruled in 1990 and again in 1992 that the EC subsidies were illegal, and in April 1992 the US announced its intention to impose punitive tariffs on $1 billion worth of EC agricultural imports. Thus, while technically distinct from the Uruguay Round talks, the oilseeds dispute became linked for bargaining purposes with the outcome of the Uruguay Round, and led to hard bargaining between American and EC negotiators in October and November of 1992.
These negotiations led to conflict within the Commission when, on the eve of the US Presidential elections in early November, MacSharry and Trade Commissioner Frans Andriessen traveled to Chicago, along with British Agriculture Minister John Gummer as president-in-office of the Council, for last-minute talks with US Trade Representative Carla Hills and Agriculture Secretary Edward Madigan. In Chicago, MacSharry and Andriessen came close to reaching a global agreement with Hills and Madigan on the oilseeds dispute as well as the outstanding Uruguay Round issues of internal supports and export subsidies, when Commission President Jacques Delors again raised the issue of the Commission's negotiating mandate. As Ross tells the story,
Delors, with the French at his back... telephoned MacSharry to warn him that the proposed deal went beyond CAP reform and the Commission's negotiating mandate. Delors also announced that he would oppose the deal in the Commission and was confident of winning, and that were the deal to go forward, it would be vetoed by at least two member states. MacSharry promptly resigned from his role as oilseeds negotiator and, with Andriessen, went back to Brussels to confront the Commission President, with whom neither was on cordial terms.... Delors was outvoted in the Commission on the issue of the negotiating mandate 82 .On November 10th, five days after handing in his resignation, MacSharry therefore returned as the Commission's chief negotiator on agricultural issues, and resumed the agricultural negotiations with the lame-duck but activist team of Hills and Madigan, who increased the pressure on the EC by announcing the imposition of $300 million in retaliatory tariffs aimed at European agricultural products. The EC, in turn, considered counter-retaliatory tariffs, which were blocked only by the opposition of the British and the Germans. For the first time in years of negotiation, the United States and the European Community seemed to be on the brink of a trade war.
On 20 November, however, the Commission and the Americans, meeting at Blair House in Washington, D.C., signed the so-called Blair House "Pre-Agreement," resolving both the oilseeds dispute and the Uruguay Round agricultural issues. On oilseeds, the Community agreed to curtail domestic production by 10-15 percent in terms of acreage, responding to a key US demand. The Uruguay Round portion of the deal contained agreements on internal supports, specifying which types of support would be eligible for inclusion in the "green box" of non-trade-distorting supports allowed by the GATT, as well as on export subsidies, where the volume of exports receiving subsidies would be cut by 21 percent, rather than the 24 percent demanded by the Americans or the 18 percent offered by the EC. The Commission, finally, also obtained a so-called "peace clause," under which the United States agreed not to challenge EC agricultural subsidies for a period of six years 83 .
4.6 Reneging, renegotiating, and wrapping up: From Blair House to Marrakesh
At Blair House, the Commission had reached the long-sought-after agreement on agricultural issues with Washington, thereby clearing the way for the conclusion of the Uruguay Round, which would follow roughly a year later in December 1993. It had done so, however, by agreeing to a package which had not been explicitly agreed to by the member states, raising the problem of ratification by the Council. As Woolcock and Hodges note in their excellent analysis:
Blair House... illustrates the risks inherent in a strategy in which the Commission "gets out ahead" of a consensus among the member states. Such a strategy may make for flexibility, but when major member states are not "implicated" in the deal they are free to attack it and thus undermine EC credibility. 84In particular, the Blair House agreement came under persistent attack from France, where farmers burned US flags in the streets of Paris and Agriculture Minister Jean-Pierre Soisson argued that the Commission had exceeded its mandate, negotiating an agreement with the United States which went beyond the CAP reforms agreed to in May. The French Socialist government, moreover, was worried about the upcoming French legislative elections, which the Gaullist majority was heavily favored to win, and resisted taking any action that might alienate the powerful farm vote. 85 .
The British and Danish Presidencies avoided putting the Blair House agreement to an immediate vote in the Council, which in any case could be taken by qualified majority vote over French objections, but successive French governments carried on a year-long assault against the agreement even so. In early March of 1993, just prior to the elections, France announced in the Council that it would oppose approval of the Blair House agreement "by all means," including the invocation of its right of veto under the Luxembourg compromise. The new Gaullist Balladur government which took over after March of 1993 took a more cautious approach, calling for a renegotiation of the Blair House agreement without forcing the collapse of the overall Uruguay Round, from which France stood to benefit. On 8 June, Balladur agreed to ratify the oilseeds part of the Blair House agreement, which was the basis for the threatened white wine tariffs against French farmers, but he continued to demand a renegotiation of the Uruguay Round side of the agreement. Such a renegotiation was resisted, however, by the new Trade Commissioner, Sir Leon Brittan, and by a majority of the member states, which were concerned about re-opening the difficult and delicate package agreed to at Blair House.
The French position, however, was strengthened by two factors. First, the member states agreed in the autumn of 1993 to ratify the final Uruguay Round package by consensus, and not by qualified majority vote as specified in Article 113, thus giving France a potential veto over the results of the Round. In fact, however, Balladur preferred not to veto the overall results of the Round, from which France stood to benefit, and he still hoped to force a renegotiation of Blair House prior to the final vote on the Uruguay Round. Second, and more importantly, Balladur prevailed on German Chancellor Helmut Kohl at their 28 August summit meeting to support the French position in the interests of the wider Franco-German relationship. This German change of position was crucial, and in September the Council of Ministers instructed Brittan to ask for "amplifications or additions" to the Blair House agreement with Washington 86 .
Against this European background, the Clinton Administration, eager to reach agreement before the expiration of the US "fast-track" authority, agreed to a series of "clarifications" of the Blair House agreement at a meeting in Brussels on 1-3 December 1993, which went a considerable way toward responding to French demands. As Preeg writes,
...the Americans made very significant concessions that altered the Blair House concessions to the advantage of European--especially French--farmers. The 21 percent reduction in subsidized exports was spread evenly over the six-year implementation period and, more importantly, the base period from which to measure the reduction was moved up from 1986-90 to 1991-92, thus allowing a higher level of subsidized exports throughout. The French wheat farmers gained in the order of 8 million tons over six years. A formula was also included for the EU to dispose on world markets its 25 million tons of grain stocks. Another US concession extended from six to eight years the period of the so-called peace clause... In sum, the French had achieved a significant watering down of the agreement made by the EC negotiators a year earlier, while the basic framework for reversing the upward trend of subsidized exports and cutting them back by about a fifth over six years was maintained 87 .With the agricultural issue out of the way, the contracting parties of the GATT completed the final package of the Uruguay Round negotiations on 15 December 1993, and, in a complex intergovernmental bargain involving side-payments to France and Portugal, the Council of Ministers unanimously approved the outcome of the negotiations on the following day 88 . Formal signing of the Final Act took place on 15 April 1994, in Marrakesh.
In a postscript to the Round, however, a dispute arose regarding the respective competences of the Commission and the member states in the ratification of the Round. The member states had agreed to allow the Commission to act as the exclusive EC negotiator during the Round, but without prejudice to the final distribution of competence between the two levels. As ratification approached, the member states insisted on the right to ratify individually the sections of the Round dealing with new trade issues such as services and intellectual property, while the Commission argued that the entire agreement should be ratified by the Community under Article 113 of the Treaty. Concerned about the considerable difficulties that such individual ratifications might pose in future trade negotiations, the Commission appealed the question of competence to the European Court of Justice.
Rather surprisingly given our assumptions about the integrationist preferences of the supranational organization, the Court in November 1994 handed down a decision which largely supported the position of the member graistates. While the Community did indeed possess exclusive competence to negotiate on trade in goods as well as on non-tariff barriers to such trade, the Court held that in the areas of services and intellectual property rights, the Community and the member states were jointly competent to negotiate agreements with third parties. The Court acknowledged in its ruling that such mixed competence would create problems in future trade negotiations, but held that the problem was to be resolved between the Commission and the Council 89 . This adverse ruling has led Ludlow, among others, to conclude that the Court is now under pressure from the member states "to act as a restraint on the central institutions as much as, if not more than, a catalyst of their advance." 90 Regardless of the Court's motives in this particular ruling, it is worth noting that the Santer Commission has, in its proposals to the 1996 intergovernmental conference to revise the Maastricht Treaty, proposed modifying Article 113 to give the Community exclusive competence to negotiate in the areas of services and intellectual property, although it is as yet unclear whether the Commission will be able to overcome the resistance of sovereignty-conscious member states 91 .
4.7 Analysis
The case of the Uruguay Round agriculture negotiations, in sum, presents another mixed picture of Commission influence. On the one hand, I have argued that the Commission was both purposeful and successful in harnessing external US pressures and internal budgetary pressures to produce and steer through the Council a far-reaching reform of the CAP designed to put the policy on secure financial footing, as well as making it compatible with the demands of the Uruguay Round. To be sure, it is difficult to argue counter-factually that CAP reform would not have been adopted in the absence of an entrepreneurial Commission; indeed, the external pressures of the GATT and the internal budgetary pressures on the CAP in the early 1990s made such a reform imperative. Despite these pressures, however, the Council of Agriculture Ministers during this period was notable for its absence of entrepreneurship and reluctance to challenge the entrenched interests of its agricultural clientele, which were likewise united in opposition to any fundamental reform of the CAP in 1991. Indeed, at the beginning of 1991 not a single member government had proposed, or supported, the shift from price supports to direct payments that was at the center of MacSharry's reforms. Thus, although some reform of the CAP was arguably inevitable given the external pressures from the United States, MacSharry and Delors nevertheless played a key role in shaping the content of the 1992 reform, and shepherding it through a reluctant Council of Ministers.
This being said, however, it must also be admitted that the Commission was less successful in securing member-state ratification of its GATT negotiations at Level 1, and in particular of the Blair House agreement, on which the Commission was forced into an embarrassing involuntary defection. In order to understand the Commission's lack of success in this area, consider the four factors mentioned in the introduction as the determinants of the Commission's autonomy and influence: preferences, decision rules, information, and transnational coalitions. In the case of the Uruguay Round agriculture negotiations, all four of these factors worked against the Commission's efforts to shape an agreement with the United States on agriculture: the preferences of the member states, and in particular of France, were clear and intensely opposed to any major agricultural concessions; the decision rule for ratification of the final agreement, although legally QMV, was in practice unanimity, providing France with an effective veto over the Blair House agreement; with regard to information, the member states monitored Commission behavior closely through the Article 113 committee, providing the Commission with few informational advantages over recalcitrant member governments; and finally, the transnational coalition of agricultural interests was largely against any further concessions to the United States on agriculture, and opposed rather than supported the Commission vis-a-vis the member states. In the absence of these four factors, the Commission's delegated powers as the Community's chief negotiator were insufficient to enable it to push through its preferences on agriculture in the GATT negotiations.
Finally, however, it should be pointed out that the agriculture case is not necessarily typical of Community trade policy as a whole. Because of the high political salience of agriculture within the member states (including, but not only, France), member governments provided the Commission with a narrow and detailed negotiating mandate, monitored the Commission closely through the Article 113 Committee, and were willing to risk a breakdown in the Round in order to respond to domestic pressures from politically powerful farmers. It seems likely that in other areas of less political salience, the Commission is granted greater discretion in defining the Community's negotiating position, and that ratification in the Council is less problematic than in the case of Blair House. Dusek's account of the negotiation of the EC-Czech Association Agreement suggests that this is indeed the case, but more research is clearly called for in this area 92
.
In this paper, I have theorized the Commission's role in the European Union in terms of a principal-agent relationship between the Commission on the one hand, and the member states on the other hand, and I have briefly examined the Commission's executive powers in the areas of structural policy, competition policy, and external trade policy in order to shed light upon the workings of this principal-agent interaction. The findings of these brief case studies are suggestive rather than definitive, but they do provide some preliminary support for the hypotheses presented above, and suggest further avenues for empirical research. For the sake of brevity, I focus here on four conclusions.
First, the three cases examined above suggest that the Commission does indeed have independent preferences, and is in fact a competence-maximizer along the lines suggested by Majone, Cram and others. Across all of the areas surveyed, the Commission has attempted, in some cases successfully, to increase both EC and Commission competence in the planning and administration of the Structural Funds; in the establishment of Community-wide criteria for cartels, concentrations, state aids, and, especially, mergers; in the aggressive use of Article 90 to liberalize telecommunications in the post-1992 internal market; and in the Commission's claim to exclusive Community competence to negotiate on the member states' behalf in the new areas of services and intellectual property rights. Along the integrative dimension, therefore, the Commission's preferences have largely conformed to the predictions of neofunctionalist and institutionalist theorists.
Along other dimensions, on the other hand, and particularly along the left-right split mentioned earlier, the Commission has often been internally divided, and its preferences have been less predictable. In trade, for example, Commissioners were openly divided on agriculture, with MacSharry more willing than Delors to make concessions to the United States. Similarly, in the area of competition policy the Commission has been split between neoliberals such as Leon Brittan who advocate the strict competition criteria championed by DG IV, and other Commissioners like Delors who have been willing to weight competition criteria against other, social and industrial policy criteria.
Second, all three cases suggest that the Commission enjoys considerable autonomy and influence in its implementation of Community policies. In structural policy, for example, the Commission successfully set the agenda for major 1988 reforms which increased its own powers, and it was subsequently able to build direct networks with subnational governments, and stand up to powerful member states like the United Kingdom on the issue of additionality, at least in the short term. Similarly, in the area of competition policy, the Commission was able, with the support of the Court of Justice, to apply Article 90 to the liberalization of the telecommunications sector, and to secure its long-sought goal of jurisdiction over European-level mergers and acquisitions. In the case of external trade policy, finally, Delors and MacSharry were able to exploit the external pressures from the United States and other EC trading partners to push through the Council a fear-reaching reform of the Common Agricultural Policy, despite the initial resistance of all of the major EC farm groups and the Council of Agriculture Ministers.
Third, however, the cases examined above suggest that the Commission's ability to act on these preferences, and to press for "more Europe," should not be overstated, and varies widely across issue-areas and over time as a function of the preferences of the member states, the rules governing the sanctioning or overruling of the Commission, the information available to both the Commission and the member states, and the Commission's ability to strike up alliances with important transnational actors. In the case of the Structural Funds, for example, the Commission exploited member state concerns about efficiency, and its asymmetrical access to information, to press for important new powers in the administration of the Funds--powers which it then used aggressively to pursue a Commission policy directly at odds with the concerns of the various member states, including the UK. By 1993, however, the Commission's informational advantage had dissipated, and the decision rules favored those member states seeking a revision of the Commission's powers. The resulting 1993 Fund Regulations did not remove all Commission discretion, but the ability of the Commission to move aggressively, and against the preferences of the member states, has been substantially curtailed.
In the competition policy cases examined above, by contrast, the Commission's powers were laid down in Article 90 of the Treaty and in the 1989 Merger Regulation, respectively. In both cases, therefore, the default condition for Commission authority was the status quo, making it more difficult for member states to sanction Commission behavior of which they disapproved. Nevertheless, a lack of member-state and interest-group support has led the Commission to resist using Article 90 to liberalize the energy sector, and the determined opposition of the UK and Germany has thus far prevented the Commission from mustering a qualified majority in the Council in order to lower the thresholds under the Merger Regulation.
Finally, in the area of trade, the Commission was generally unable to translate its role as chief negotiator into leverage vis-a-vis member states such as France on the substance of the Uruguay Round agreement. Facing strong member-state preferences, a demanding de facto consensus rule for ratification, close monitoring from the Article 113 committee, and intense opposition from agricultural lobbies, the Commission was forced to back down at the 1990 Brussels summit and again on the Blair House agreement in 1992-93. Once again, the agriculture negotiations of the GATT were an unusually controversial issue in EC trade policy, and are not necessarily representative of the Commission's influence in trade policy more generally. Nevertheless, these cases suggest that the Commission's formal Treaty powers are not sufficient to predict or explain actual Commission autonomy and influence, and that we need to look as well at preferences, decision rules, information, and the availability of transnational coalitions in any given case.
A fourth conclusion which emerges from the case studies presented above is that the Commission's autonomy and influence also depends crucially on its rather complex relationship with the European Court of Justice 93 . The Commission and the Court are, in Martin Westlake's fortuitous phrase, both "partners and rivals" in the policy process 94 . On the one hand, the Court shares with the Commission a broad or teleological reading of the Treaties, and a longstanding preference for deeper integration, which has led the Court to support the Commission's efforts to expand Community competence, as in the cases of merger control and telecommunications deregulation discussed above. On the other hand, however, the Court has also sought to defend the overall "institutional balance" among the various EC institutions, and to ensure that the Commission carries out its functions in a clear and transparent manner, and so the Court has often ruled against the Commission in the area of competition policy, and in the Commission's bid for greater negotiation powers in the area of external trade policy. The Commission and the Court may, therefore, be the engines of integration, but as we have seen, the two engines do not always pull in the same direction.
Allen, David, 1983. "Managing the Common Market: The Community Competion Policy," in Helen Wallace, William Wallace and Carole Webb, eds., Policy-Making in the European Community (London: John Wiley and Sons), pp. 290-36.
Allen, David, 1996. "Competition Policy: Policing the Single Market," in Helen Wallace and William Wallace, eds., Policy-Making in the European Union (Oxford: Oxford University Press), pp.157-183.
Alter, Karen J. 1996 The European Court's political power: The emergence of an authoritative international court in the European Union, West European Politics, July 1996 (forthcoming).
Assemblee Nationale, Delegation de l'Assemblee Nationale pour les Communaute's Europeennes, 1993. Rapport d'information, No.221.
Bonhearn, 1978.
Bradley, Kieran St. Clair... Bulmer, Simor, 1994. "Institutions and Policy Change in the European Communities: The Case of Merger Control," Public Administration, Vol.72, No.3 (Autumn), pp.423-44.
Burley, Anne-Marie, and Walter Mattli, 1993. "Europe Before the Court: A Political Theory of Legal Integration," International Organization, Vol.47, No.1 (Winter), pp.41-76.
Commission of the European Communities, 1993a. Community Initiatives (Luxembourg: OOPEC, 1993).
Commission of the European Communities, 1993. Community Structural Funds 1994-99: Regulations and Commentary (Luxembourg: Office for Official Publications of the European Community).
Commission of the European Communities, 1994a. The Future of Community Initiatives under the Structural Funds, COM(94)46 final of 25 March.
Commission of the European Communities, 1994b. Guide to the Community Initiatives first edition (Luxembourg: Office for Official Publications of the European Community).
Corbett, Richard, 1994. The Treaty of Maastricht ( ).
Cram, Laura, 1993. "Calling the Tune Without Paying the Piper? Social Policy Regulation: The Role of the Commission in European Community Social Policy," Policy and Politics, Vol.21, No.2, pp.135-46.
Cram, Laura, 1994. "The European Commission as a Multi-Organisation: Social Policy and IT Policy in the EU," Journal of European Public Policy, Vol.1, No.2, pp.195-217.
Devuyst, Youri, 1995. "The European Community and the Conclusion of the Uruguay Round," in Carolyn Rhodes and Sonia Mazey, eds., The State of the European Union, Vol.3: Building a European Polity? (Boulder: Lynne Werner), pp.449-67. Alter, Karen J. 1996 The European Court's political power: The emergenc Docksey, Christopher, and Karen Williams, 1994. "The Commission and the Execution of Community Policy," in Geoffrey Edwards and David Spence, eds., The European Commission (Harlow: Longman Current Affairs), pp. 117-45.
Dusek, Daniel, 1995. "The Negotiation of the Czech-EC Association Agreement," unpublished Harvard University undergraduate senior thesis, Cambridge.
Earnshaw, David, and David Judge, 1995. "Early days: the European Parliament, co-decision and the European Union legislative process post-Maastricht," Journal of European Public Policy, Vol.2, No.4 (December 1995), pp. 624-49.
Fuchs, Gerhard, 1995. "The European Commission as a Corporate Actor? European Telecommunications Policy After Maastricht," in Sonia Mazey and Carolyn Rhodes, eds., The State of the European Union, Vol. III (Boston: Lynne Werner), pp.413-429.
Garrett, Geoffrey, 1995. "The Politics of Legal Integration in the European Union," International Organization, Vol.49, No.1 (Winter), pp.171-181.
Garrett, Geoffrey, and George Tsebelis, 1996. "An Institutional Critique of Intergovernmentalism," International Organization (forthcoming).
Garrett, Geoffrey, and Barry Weingast, 1993. "Ideas, Interests, and Institutions: Constructing the European Community's Internal Market," in Judith Goldstein and Robert 0. Keohane, eds., Ideas and Foreign Policy (Ithaca: Cornell University Press), pp.173-206.
Gerus, Victoria, 1991. "Comitology within the European Community's Policy-Making Process: A Mechanism of Political Control in the Inter-Institutional Relations of the Council of Ministers and the Commission," unpublished manuscript, Harvard University, September.
Goyder, D.G., 1993. EC Competition Law, Second Edition (Oxford: Clarendon Press.
Greaves, Rosa, 1994. EC Block Exemption Regulations (New York: John Wiley & Sons).
Hanscher, Leigh, Tom Ottervanger, and Piet Jan Slot, 1993. EC State Aids (Colorado Springs: Jobn Wiley & Sons).
Hix, Simon, 1994. "The Study of the European Community: The Challenge to Comparative Politics," West European Politics, Vol.17, No.1 (January), pp.1-30.
Hix, Simon, and Christopher Lord, 1996. "The Making of a President: The European Parliament and the Confirmation of Jacques Delors as President of the Commission," Government and Opposition, Vol.?, No.?, pp.62-76.
Hözer, Heinrich, 1990. "Merger Control," in Peter Montagnon, ed., European Competition Policy (London:Royal Institute for International Affairs), pp.9-30.
Hooghe, Liesbet, and Gary Marks, 1996. "The Birth of a Polity: The Struggle Over European Integration," paper prepared for Herbert Kitschelt, Peter Lange, Gary Marks and John Stephens, eds., The Politics and Political Economy of Advanced Industrial Societies.
House of Commons, Select Committee on European Legislation, 1993. Twenty-sixth Report, Session 1992-93.
House of Lords, Select Committee on the European Communities, 1982. Revision of the European Regional Development Fund, Session 1981-82, 12th Report (London: HMSO).
Kiewiet, Roderick, and Mathew D. McCubbins, 1991. The Logic of Delegation: Congressional Parties and the Appropriations Process (Chicago: University of Chicago Press).
Kilroy, Bernadette, 1995. "Member State Control of Judicial Independence: The Integrative Role of the European Court of Justice, 1958-1994," paper prepared for delivery at the 1995 Annual Meeting of the American Political Science Association, Chicago, August 31 - September 3.
Kingdon, John W., 1984. Agendas, Alternatives, and Public Policies (Boston: Little, Brown).
Majone, Giandomenico, 1994. The European Community as a Regulatory State (Lectures given at the Academy of European Law, European University Institute).
Mancini, G. Federico, 1991. "The Making of a Constitution for Europe," in Robert 0. Keohane and Stanley Hoffinann, eds., The New European Community (Boulder: Westview Press), pp. 177-94.
Mattli, Walter, and Anne-Marie Slaughter, 1995. "Law and Politics in the European Union: A Reply to Garrett," International Organization, Vol.49, No.1 (Winter), pp. 183-190.
Marks, Gary, 1993. "Structural Policy and Multilevel Governance in the EC," in Alan Cafruny and Glenda Rosenthal, eds., The State of the European Community 2: The Maastricht Debates and Beyond (Boulder, Lynne Rienner), pp. 391-410.
Marks, Gary, 1996. "Decision Making in Cohesion Policy: Charting and Explaming Variations," in Liesbet Hooghe, ed., Cohesion Policy. the European Community and Subnational Government (Oxford: Oxford University Press, forthcoming).
McAleavey, Paul, 1993. "The Politics of European Regional Development Policy: Additionality in the Scottish Coalfields," Regional Politics & Policy Vol.3, No.2 (Summer), pp. 88-107.
McCubbins, Mathew, Roger Noll and Barry Weingast, 1987. "Administrative Procedures as Instruments of Political Control, Journal of Law. Economics. and Organization, Vol.3, No.2, pp. 243-77.
McCubbins, Mathew, Roger Noll and Barry Weingast, 1989. "Structure and Process, Policy and Politics: Administrative Arrangements and the Political Control of Agencies," Virginia Law Review, Vol.75, pp.431-82.
McCubbins, Mathew, and Talbot Page, 1987. "A Theory of Congressional Delegation," in Mathew McCubbins and Terry Sullivan, eds. Congress: Structure and Policy (New York: Cambridge University Press), pp. 409-425.
McCubbins, Mathew, and Thomas Schwartz, 1984. "Congressional Oversight Overlooked: Police Patrols versus Fire Alarms," American Journal of Political Science, Vol.28, pp.165-179.
McGowan, Lee, and Stephen Wilks, 1995. "The First Supranational Policy in the European Union: Competition Policy," European Journal of Political Research, Vol.28, pp. 141-69.
Moe, Terry M., 1987, "An Assessment of the Positive Theory of Congressional Dominance," Legislative Studies Ouarterly, Vol.12 (November), pp. 475-520.
Moe, Terry M., 1990. "The Politics of Structural Choice: Toward a Theory of Public Bureaucracy," in Oliver WIlliamson, ed., Organization Theory from Chester Barnard to the Present and Beyond (New York: Oxford University Press), p. 116-153.
Montagnon, Peter, ed., 1990. European Competition Policy (London: Royal Institute for International Affairs).
Montagnon, Peter, 1990. "Regulating the Utilities," in Montagon, ed., European Competition Policy, pp.52-74.
Moravcsik, Andrew, 1993. "Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach," Journal of Common Market Studies, Volume 31, No.4 (December), pp. 473-524.
Moyer, H. Wayne, 1993. "The European Community and the GATT Uruguay Round: Preserving the Common Agricultural Policy at All Costs," in....
North, Douglas C. 1990. Institutions, Institutional Change and Economic Performance (New York:Cambridge University Press).
Pan, Eric, 1996. "Europe and the World: External Relations, Internal Dynamics," Proceedings of the Thirteenth Annual Graduate Student Conference, Institute on Western Europe, Columbia University, New York, March 28-30.
Pierson, Paul, 1996. "The Path to European Integration: A Historical Institutionalist Analysis," Comparative Political Studies, Vol.29, No.2 (April), pp.123-163.
Pollack, Mark A., 1995. "Regional Actors in an Intergovernmental Play: The Making and Implementation of EC Structural Policy," in Sonia Mazey and Carolyn Rhodes, eds., The State of the European Union, Vol. III (Boston: Lynne Rierner), pp. 361-90.
Pollack, Mark A., 1997. "Delegation, Agency and Agenda Setting in the European Community," International Organization Vol.51, No.1 (Winter), forthcoming.
Preeg, Ernest H., 1995. Traders in a Brave New World: The Uruguay Round and the Future of the International Trading System (Chicago: University of Chicago Press).
Pumam, Robert D., 1988. "Diplomacy and Domestic Politics: The Logic of Two-Level Games," International Organization, Vol.42, No.3, pp. 427-60.
Ross, George, 1995. Jacques Delors and European Integration (New York: Oxford University Press). Sandholtz, Wayne, 1992. "ESPRIT and the Politics of International Collective Action," Journal of Common Market Studies, Vol.30, No.1 (March), pp.1-22.
Schmidt, Susanne K., 1996. "The European Commission's Ability to Act under European Competition Law. The Example of Telecommunications and Electricity Policy," paper presented at the German & American Young Scholars Institute on "The Political Economy of European Integration" at the University of Bremen,August 4-16.
Stein, Eric, 1981. "Lawyers, Judges, and the Making of a Transnational Constitution," American Journal of International Law, Vol.75 (January), pp. 1-27.
Stephen, Roland, 1995. "Interests, Issue Linkage and the Power of the European Parliament," paper prep eared for the annual meeting of the European Community Studies Association, Charleston, SC, 11-14 May.
Stewart, Terrence, ed., 1993. The GATT Uruguay Round: A Negotiating History. Vol.1: Commentary (Deventer: Kiuwer Law).
Stone, Alec, 1995. "Constructing a Supranational Constitution: The Reception and Enforcement of European Community Law by National Courts," proposal to the National Science Foundation, Global Perspectives on Socio-Legal Change Program.
Swinbank, Alan, 1993. "CAP Reform, 1992," Journal of Common Market Studies, Vol.31, No.3 (September), pp. 359-372.
Thomas, Ian C., 1992. "Additionality in the distribution of European Regional Development Fund grants to local authorities," Local Economy, Vol.6, No.4 (February), pp.292-310.
Thomas, Kenneth P., 1996. "Evaluating the EU State Aid Regime," paper presented at the Conference of Europeanists, Chicago, March.
Tsebelis, George, 1994. "The Power of the European Parliament as a Conditional Agenda Setter," American Political Science Review, Vol.88, No.1 (March), pp.129-142.
Tsebelis, George, 1995. "Conditional Agenda Setting and Decision Making Inside the European Parliament," Journal of Legislative Studies 1 (Spring), pp. 65-93.
Tsebelis, George, and Amie Kreppel, 1995. "The History of Conditional Agenda-Setting in the European Union," paper presented at the 91st Annual Meeting of the American Political Science Association, Chicago, 31 August-3 September.
Weingast, Barry R., and Mark Moran, 1983. "Bureaucratic Discretion or Congressional Control? Regulatory Policy-making by the Federal Trade Commission," Journal of Political Economy, Vol.91, No.5, pp. 765-800.
Westlake, Martin, 1994. A Modem Guide to the European Parliament (London: Pinter).
Wilks, Stephen, and Lee McGowan, 1995. "Disarming the Comnussion: The Debate over a European Cartel Office," Journal of Common Market Studies, Vol.32, No.2, pp.259-273.
Woolcock, Stephen, and Michael Hodges, 1996. "EU Policy and the Uruguay Round," in Helen Wallace and William Wallace, eds., Policy-Making in the European Union (Oxford: Oxford University Press), pp.301-24.
Yulll, Douglas, Kevin Allen, John Bachtler, Keith Clement and Fiona Wishlade, eds., 1993. European Regional Incentives, 1993-94, 13th Edition (London: Bowker-Saur).
Zysman, John, and Wayne Sandholtz, 1989. "1992: Recasting the European Bargain," World Politics, Vol. 41, No. 1, pp. 1-30.
Note *: Mark A. Pollack, Dept. of Political Science, University of Wisconsin Back.
Note 1: The author would like to thank the members of the Laguna Beach Group for comments on an earlier draft of this paper, Michael Mosser for invaluable research assistance, and the World Affairs and the Global Economy (WAGE) initiative of the University of Wisconsin for research support. Back.
Note 2: Throughout this paper, I distinguish between EC institutions, which establish the general decision rules for policymakng and institutional change, and EC supranational organizations, which are collective actors operating within the Community's institutional system. For a good discussion, see North 1990, especially p. Back.
Note 3: See Sandholtz 1992, Moravcsik 1995. Back.
Note 4: See e.g. Stein 1981; Weiler 1991; Burley and Mattli 1993; Stone 1995; Alter 1996; and Mattli and Slaughter 1995. Back.
Note 5: See Garrett 1992, 1995. Back.
Note 6: See e.g. Tsebelis 1994; Tsebelis and Kreppel 1995; Stephen 1995; Garrett and Tsebelis 1996. Back.
Note 7: See e.g. Majone 1992; Cram 1993; Pollack 1994; and Garrett and Tsebelis 1996. Back.
Note 8: Ross 1995, p.14. Back.
Note 9: See e.g. Bonham 1978 for a good discussion of self-selection to the European Parliament. Back.
Note 10: See e.g. Mancini 1991 for a good discussion of socialization within the European Court of Justice. Back.
Note 11: For a good general discussion, see Peters 1992. Back.
Note 12: See Hix 1994; Stephen 1995. On the neoliberalism/interventionism cleavage, see Hooghe and Marks 1996. Back.
Note 13: See Cram 1994; Ross 1995. Back.
Note 14: For a more thorough theoretical treatment of the issues discussed here, see Pollack 1997. Back.
Note 15: Moe1990,p. 121. Back.
Note 16: See e.g McCubbins and Page 1987; McCubbins, Noll and Weingast, 1987, 1989. Back.
Note 17: McCubbins and Schwartz 1994. Back.
Note 18: For good discussions, see McCubbins and Page 1987, pp.21-22; and Kiewiet and McCubbins 1991, pp.27-34. Back.
Note 20: As Hix and Lord (1996) point out, however, the future careers of MEPs can often depend on their standing with their national parties back home, offering governments some degree of leverage over MEPs from their own party. Back.
Note 21: For details, see Earnshaw and Judge 1996; Tsebelis and Garrett 1996. Back.
Note 22: See Kingdon 1984. Back.
Note 23: Moravcsik 1995. Back.
Note 24: See e.g. Sandholtz 1992. Back.
Note 25: See Zysman and Sandholtz 1989; and Ross 1995. Back.
Note 26: See Marks 1993. Back.
Note 27: Garrett and Weingast 1993. Back.
Note 28: See Corbett 1994. Back.
Note 29: See e.g. Zysman and Sandholtz 1989; Sandholtz 1992. Back.
Note 30: See e.g. Burley and Mattli 1993; Alter 1996; Stone 1995. Back.
Note 31: Gerus 1991, Table 2. Back.
Note 32: Interview, March 1995. Back.
Note 33: This section draws largely on Pollack 1995. Back.
Note 34: For background on the Community initiatives adopted under the 1988 Regulations, see Commission of the European Communities 1993. Back.
Note 35: Commission 1993, sheet on Envireg. Back.
Note 36: For excellent discussions, see McAleavey 1993; Thomas 1992; Marks 1993. Back.
Note 37: For an excellent and thorough discussion of Member state concerns about the 1988 reforms, see Yuill et al. 1993, pp.63-68. Back.
Note 38: Yuill et al. 1993, p.74. Back.
Note 39: Yuill et al. 1992, p.77. Back.
Note 40: The Commission's initial ideas for the Fund reforms were expressed in its mid-term evaluation, Community Structural Policies: Assessment and Outlook, COM(92)84; and in COM(92)2000. For the Commission's formal proposals, see COM(93)67 final and COM(93)124 final of 12 March and 7 April 1993, respectively. Back.
Note 41: For Member state responses to the Commission's proposals, see e.g. House of Commons 1993; and Assemblee Nationale 1993. For a more general view of problem issues, see Agence Europe, 7/8 June 1993. Back.
Note 42: For more on the Community Initiatives, see Commission of the European Communities 1993. Back.
Note 43: European Report, No.1876, 17 July 1993; and Agence Europe, 15 July 1993. Back.
Note 44: Commission of the European Communities 1994a. Back.
Note 45: Commission of the European Communities 1994b. Back.
Note 47: For good general discussions of EC competition policy, see Allen 1983, 1996; Montagnon, ed. 1990; Goyder 1993; and McGowan and Wilks 1995. Back.
Note 48: See Greaves 1994, pp. 97-101. Back.
Note 49: For a good discussion, see Thomas 1986. Back.
Note 50: For a good discussion, see Montagnon 1990; and Allen 1996. Back.
Note 51: This section draws heavily on the following sources: Montagnon 1990; Fuchs 1995; Schmidt 1996; and Sandholtz in this volume. Back.
Note 52: Schmidt 1996, p.12. Back.
Note 53: Ibid, pp. 7-10. Back.
Note 54: The discussion in this section draws largely on Hözer 1990; Goyder 1993, pp. 387-407; Bulmer 1994; and Allen 1996, pp. 169-75. Back.
Note 55: Goyder 1993, p.386. Back.
Note 56: Allen 1996, p.17l. Back.
Note 57: Goyder 1993, p. 398. Back.
Note 58: Allen 1996, p.174. Back.
Note 59: See Wilks and McGowan 1995. Back.
Note 60: Ross 1995, p.178. Back.
Note 61: Wilks and McGowan 1995. Back.
Note 62: Neil Buckley, "Brussels Seeks More Power on Mergers," Financial Times, 10 July 1996. Back.
Note 63: "Council Looks at Commission Merger Regulation Report," European Report, 29 September 1993 Back.
Note 64: Quoted in "Commission Backs Off Decision to Enhance Merger Review Powers," International Securities Regulation Report, 10 August 1993. See also "Commission Decides to Leave Mergers Regulation Unchanged," Agence Europe, 29 July 1993. Back.
Note 65: Neil Buckley, "Brussels Seeks More Power on Mergers," Financial Times, 10 July 1996. Back.
Note 66: For a good summary of the EC trade-policy process, see Woolcock and Hodges 1996, p.305. Back.
Note 67: The following section draws on Putnam 1988. For elaboration and empirical applications of Putnam's arguments, see Evans, Jacobson and Putnam, eds., 1993. Back.
Note 68: 68 Ibid, p. 448. Back.
Note 69: Moravcsik 1994. Back.
Note 70: For good discussions of EC trade negotiations as a three-level game, see Dusek 1995, and Pan 1996. Back.
Note 71: For good discussion of mixed competence, see Sbragia in this volume. For a discussion of mixed competence in the Uruguay Round, see Woolcock and Hodges, esp. p.302. Back.
Note 72: The account of the Uruguay Round negotiations and the 1992 CAP reform presented here is necessarily brief, focusing primarily on the role of the Commission. For more complete discussions, see the excellent account by Eric Pan (which also applies a three-level analysis to the negotiations), and the accounts by Ross 1995; Woolcock and Hodges 1996; Stewart 1993; and Preeg 1995. Back.
Note 73: For a brief, clear discussion of the pre-1992 CAP, see Stewart 1994, pp.145-49. Back.
Note 74: Quoted in Woolcock and Hodges 1996, p.314. For excellent discussions of the Brussels meeting and the intra-EC bargaining which preceded it, see Woolcock and Hodges, pp.314-15; Moyer, pp.106-10; and Preeg 1995, pp.116-22. Back.
Note 75: Ross 1995, p.113. Back.
Note 76: For an excellent discussion of the early discussions within DG VI during 1990, see "continuing Confusion on Commission's CAO Reform Plans," Agra Europe, 15 February 1991. Back.
Note 77: For good discussions of the initial debates on CAP reform within the Commission, see Moyer, pp. 112-13; and Ross 1995, pp.109-14. Back.
Note 78: For the text of the Commission Communication, see Europe Documents, No.1689, 8 February 1991. Back.
Note 79: For details of the Commission's proposals, see Swinbank 1993 Back.
Note 80: David Gardner, "Basis of a Deal on Radical CAP Reform in Sight," Financial Times, 21 May 1992. Back.
Note 81: Quoted in David Gardner, "Reforms with a Grain of Sense: The EC's Agricultural Package, Brokered by Ray MacSharry, Will Cut Output and Food Prices," Financial Times, 22 May 1992. Back.
Note 82: Ross 1995, p. 211-12. Back.
Note 83: For a good discussion see Preeg 1995, pp.144-47. Back.
Note 84: Woolcock and Hodges, p. 318 Back.
Note 85: David Buchan, "French Government Faces Fury at Home," Financial Times, 21 November 1992. Back.
Note 86: Preeg 1995, pp. 163-64. Back.
Note 87: Ibid, pp.166-67. Back.
Note 88: For an excellent discussion of the final agreement, emphasizing the complex intergovernmental bargaining required for Community ratification of the Uruguay Round, see Devuyst 1995. Back.
Note 89: See ibid, pp. 460-62; and Pan 1996, p.20O. For the text of the Court's ruling, see Common Market Law Reports. 1995:1. Back.
Note 90: Quoted in Devuyst, p.462. Back.
Note 91: Lionel Barber, "Brussels Seeks Control of Trade Deals," Financial Times, 5 September 1996. Back.
Note 92: See Dusek 1995. Back.
Note 93: By contrast, the European Parliament, which is a key player in legislative matters, has been of little importance in relation to the Commission's executive activities -- a shortcoming which has led the Parliament to demand a greater role in the comitology process. For a good discussion, see Bradley 1992. Back.
