|
|
* This paper was prepared for the seminar organized by Wayne Sandholtz and Alec Stone on "Issues of State Sovereignty and the European Union". We would like to thank the members of that seminar for their input into this project over three very stimulating meetings at Laguna Beach. We would also like to thank Chris Ansell, Ernie Haas, Wayne Sandholtz, and Alec Stone for more detailed comments.
The European Union produced three large institution building projects in the past 15 years: the Single Market, the Single European Act, and the Treaty on European Union. The emergence of these projects has provoked scholars to reexamine the European Union (hereafter EU) as a political system (Sbragia, 1992; Keohane and Hoffman, 1991). Much of the speculation about the European Union concerns the nature of these political arrangements and their trajectories. The most common debate concerns whether or not the EU is most correctly characterized as an intergovernmental organization limited by the agendas of the member state governments or if it acts as a supranational entity that restructures the politics of western Europe in such a way as to eventually result in a single European state or some more complex polity (Moravscik, 1991; Sbragia, 1992; Schmitter, 1992). Stone and Sandholtz (this volume) have proposed that one way to transcend this debate is to recognize that the policy domains in the EU can be viewed as a continuum from those domains that look more like an intergovernmental organization (for example, defense policy) to those that look more like a supranational organization (for example, the Single Market). They argue that three factors condition this process: the various Treaties which determine what policy domains can exist and the voting rules for that domain (rules), the existence of the supranational organizations of the EU (Council, Commission, Court, and Parliament) who push forward aspects of the integration project, and the existence of transnational actors who help constitute the policy domains.
The major purpose of this paper is to explore these insights by considering how policy domains were actually constructed in the EU over time. Policy domains generally form when there exists a constitutional agreement to create legislation, a collective definition of what the issues are and who gets to be an actor, and organizational capacity and procedures to mobilize the production of new rules in the domain (ie. legislation) (Laumann and Knoke, 1989). Policy domains may be entirely constituted by government organizations or may also include nongovernmental groups.
In the context of the EU, we want to explore two related themes. First, we demonstrate that the analytical concept of a policy domain is helpful and appropriate in describing the evolution of the EU as an institutional/legislative terrain. Here, we describe (a) what a policy domain is, and (b) how the EU's institutional terrain does, indeed, appear to be constituted by policy domains. Second, we want to explore the evolution of these domains over time, looking for evidence of change from intergovernmental to supranationalist arrangements.
Our descriptive analysis traces the evolution of policy domains in the EU and concludes that their contours have been strongly shaped by the Treaty of Rome and its subsequent revisions with the Single European Act and the Treaty on European Union. We demonstrate that the Council and Commission are organized into units that directly mirror the policy domains articulated in the Treaty of Rome, the Single European Act, and the Treaty on European Union. The European Union also classifies legislation into these categories suggesting they function as shared meanings. We show how the Treaty revisions have altered the organization of the Council and Commission as the policy domains have been expanded. The Treaty expansions have directly effected the production of legislation. Legislation rose in the 1970s, dropped slightly around 1980, and rose steeply after 1986. We document the relationship between the presence of transnational pressure groups in the EU and the production of legislation. Our analytic results show evidence for both intergovernmental and supranational accounts of the EU. 13 of 17 policy domains had participants from pressure groups by 1990. The largest number of pressure groups existed in the Industry/Single Market domain giving some credence to the view that the Single Market Program had a built-in constituency in Brussels. Some domains (fisheries, financial/institutional affairs, regional policy, and development) remain firmly controlled by the nation states. We have evidence that legislation grew in arenas where competencies of the EU were subsequently expanded or voting rules changed, such as the environment. But, the states also agreed to expand competencies in arenas where there appeared to be no legislative or pressure group presence. And in every case, once competencies were expanded, legislation increased.
Our results imply that the EU is a complex organization with most of it resembling a supranational regime, but important parts firmly controlled by governmental actors. From the perspective of the continuum provided by Stone and Sandholtz, the policy domains of the EU more resembles a supranational governance structure than an intergovernmental bargain in simple numerical terms. One could argue that the domains which remain under the control of the states, particularly the financial/institutional, allow them to retain control over the whole apparatus. We will consider this idea more closely in the conclusion.
This structure resulted partially from the planned purposes of the states. But it is clear that the production of policy domains was affected by the existence of transnational pressure groups which chose to go to Brussels to express their opinions to their governments.
The goal of this section is to discuss both theoretically and empirically the issue of how many policy domains exist in the EU and how they are constituted. A policy domain implies that a group of actors are organized to participate in a collective debate with the goal of affecting the content of legislation or agreements. These actors include government organizations and might include organized interest groups, some of whom might be other government organizations (Laumann and Knoke, 1989). This view of policy domains is very broad and would include phenomena like intergovernmental organizations, supranational polities, and nation states.
The construction of a policy domain requires a number of theoretical features, including agreement that the domain is a legitimate focus of policy oriented actors (ie. a constitutional agreement that allows actors to come to binding agreement on issues), a set of shared meanings that define that area as a "domain" and provide a way to understand who is a "player"; who has power, and why; a set of organized governmental actors who consider the domain their "turf", and if appropriate, other nongovernmental actors who participate in policy construction in the domain. Institutionalization means that these organizations and meanings are shared and operate to structure subsequent cooperation.
The existence and stability of domains presents difficult empirical problems. Drawing the boundaries and tracking the effects of policy domains is quite problematic. So, if we have two pieces of legislation, how do we tell if they are in the same or different policy domains? Does every group in a given domain have to be equally involved in every piece of legislation? Or alternatively, does every piece of legislation imply a new policy domain that comes together to pass that legislation?
We propose to address these difficult questions by viewing policy domains as broad sets of issue arenas that both state and nonstate actors consider. These domains are do not require every organized group to weigh in on every single issue. The domains evolve over time to specify which concerns are relevant to them and which are not. In this sense, policy domains become umbrella issue arenas. Indeed, there are likely to be turf wars between organized groups in different policy domains over jurisdiction over which particular issues belong in which domain. Part of the ongoing process is about who can control the agenda of a policy domain.
It is possible for groups of organized state and nonstate actors to enter into an arena, exit, or remain inactive over a period of time. Nevertheless, once set into place, the domains act as powerful foci for collective action by defining where actors go to affect policy formation and what kind of language they use to engage in that process.
Actors in domains have different amounts of power. In the EU, for instance, we would expect because of the Treaty agreements that in some policy domains, like foreign and security policy, the nation states would be the most important actors and these domains would remain intergovernmental in their structure. In others, where voting rules have shifted to qualified majority voting and transnational interest groups are present, like setting accounting rules for financial institutions, nongovernmental organizations and the European Commission might be more powerful and those domains resemble supranational arenas. It is possible for groups to be members of more than one policy arena when groups maintain heterogeneous interests in different aspects of policy domains. So, the American Chamber of Commerce in Brussels participates in most policy domains reflecting the varied and broad interests of their membership (Philips, 1994). The definitions and boundaries of certain policy domains which are well-established before some groups come onto the scene, force these organizations to strategically attach their concerns to existing discussions of policy. Groups interested in environmental issues would be naturally led to Directorate General XI, the Commission's directorate for the environment. But groups' interests might not so neatly fit into existing categories. This likely means lobbying in more than one direction, particularly as groups try to increase their influence by discovering how their interests may be most usefully served (Mazey and Richardson, 1993).
Organized groups from more than one policy domain might get involved in a particular policy issue. In the case of the EU, for instance, issues surrounding the internal market might intersect with issues concerning agricultural products. Organized groups from both policy arenas would become likely participants in both discussions. Members from both the Commission and the Council who deal with both sides of the issue are likely to be consulted by whomever is taking the initiative on any given issue. One could hypothesize that actors who were participants in both policy domains would likely have a great deal of power over potential outcomes by virtue of their being "polled" for their opinion twice, and by their standing in separate domains where they have networks of connections in both domains.
Early on, policy domains come into existence as the result of two major forces: a constitutional agreement and by the use of a common language to define how to think about the issue area (the latter is akin to what Geertz called "local knowledge" (1980)). These provide the tools for actors to engage in policy discussions. Once in place, these definitions both enable and constrain action. They enable actors by producing standard understandings about what are legitimate and illegitimate issues and how actors can frame their interests in terms of broader policy concerns. They constrain action in that they foreclose some sets of issues and methods of framing. They affect who can participate in issue area construction and what their roles are.
Institutionalization implies that actors from the nation states, the organizations of the EU, and transnational pressure organizations, come to take advantage of constitutional openings to organize policy domains. Many of the papers in this volume trace out these processes in detail (ie. Sandholtz, Pollack). The Council of Minsters, the Commission, the Court of Justice, the Parliament, and the Economic and Social Committee can be parties to policy domains. In practice, the major actors have been the Council of Ministers, the Commission, and the Court of Justice, although the Parliament has steadily increased its influence (Tsebelis, 1992). The relative ability of these actors to organize policy domains comes from the various Treaties (which function as constitutions in the language used here) and the organizational capacity they muster. Organizational capacity implies having personnel, expertise, and money to bring to bear on any given policy domain.
We argue that rules (constitutions and shared meanings about how to interpret constitutions) and government organizational capacity (bureaucracies dedicated to acting in given arenas) have generally helped nongovernmental organizations to organize to affect policy in the EU. Much of what actors in the EU apparatus do, is facilitate discussion amongst interested parties to any policy domain in order to broker some kind of new collective arrangement. The output of these organized domains is legislation. The purpose of the rest of the paper is to take these insights about the constitution of policy domains and look at various kinds of measures of domain organization over time in order to flesh out a picture of how the EU has expanded.
We first turn to the question of describing the number and scope of policy domains that exist in the EU. Following our earlier argument, it is possible to document how the Treaty of Rome, the Single European Act, and the Treaty on European Union expanded the possible policy domains. Table 1 describes the policy domains as they are defined by the three treaties.[1]
Why are these the policy domains of the EU? The principal Treaties of the EU use a certain kind of language to describe the purposes of the organization (what are called "competencies"). This language offers actors legitimate ways to organize. The foundations of the EU and its policy domains are laid out in Parts Two, Three, and Five of the Treaty of Rome which produce 15 potential arenas for policy making in the original Treaty of Rome (European Community, 1987). The Single European Act modified the Treaty of Rome by adding three distinct policy domains (law relating to undertakings; common, foreign, and security policy; and environment, consumers, and health protection) and expanding EU organizational and institutional capacity in eight arenas that already existed (European Community, 1987: 1007-1094). The Treaty on European Union added two new policy domains (cooperation in the fields of justice and home affairs; and "People's Europe") and substantially added institutional capacity in nine others (European Community, 1992; see in particular p.12-13.). We note that 15 of the 20 potential policy domains were specified in the Treaty of Rome.
These observations accord with our view of the formation of institutions in two ways. First, the big purposes of the EU were laid down in the original documents, particularly the Treaty of Rome. Second, this language became dominant in discussions of the EU and was embedded in all subsequent discussions of the EU. The largest amount of the expansion of institutional capacity came as the Treaties were modified mostly to change the institutional capacity of actors in existing arenas. These changes were of three varieties: opening new areas for discussion, changing the voting rules for the European Council of Ministers, and altering the procedures whereby legislation was produced.
We want to argue that these analytic categories were viewed by the organizations of the EU as potential arenas in which legislation could be written. If these arenas actually exist, it should be possible to document their organizational and legal reality. One of the most important empirical proofs of our argument is that these are the analytic categories that the EU uses to classify legislation (European Community, 1995: p. 13). In other words, policymakers follow these classifications as templates in their organization of legislative activity.
It is also the case that the European Commission and the European Council of Ministers used these categories to define their own organizational structure.[2] Table 2 presents the areas specified by the Treaty of Rome in one column and contains the names of the Directorate Generals (hereafter DG) of the European Commission in 1970. There is a direct one-to-one mapping of the 15 arenas defined by the Treaty of Rome to the organization of the European Commission. This implies that the competencies of the European Union as they were specified in the Treaty of Rome did not just serve as cognitive categories but also constituted organizational realities.
Table 3 documents how the structure of the European Commission changed between 1970 and 1993. In 1980, before either of the two large Treaty modifications, the structure of the Commission remained almost identical. The major change was moving DG XV (the Joint Research Centre) into DG XII (Research, Science, and Education). DG XV became oriented towards Financial institutions and issues of Taxation. This reflected the increased interest in monetary issues in the 1970s. The other major change was DG XIV (Internal Market) was combined into DG-III (previously, Industrial Affairs). DG XIV became the Fisheries Directorate, which was split off from the DG VI (Agriculture). This reflected the expansion of fishery agreements in the 1970s.
The Single European Act expanded the institutional capacity of the European Community by increasing competencies. Many of these were related to the issue of completing the Single Market. Some of these changes can be linked to the purposes of the Single Market and the original purposes of the Treaty of Rome (for instance, regional policy, see Garrett 1992 for an argument about "side payments"), but many are more difficult to connect, ie. foreign and security policy, science, and culture.
These changes are also reflected in the reorganization of the Commission. DG X changed from being concerned with issues of providing the public with information about the activities of the Community to including issues of media and culture. DG XI in 1980 was concerned with issues of external trade. This group was moved over to DG VIII- Development. DG XI became Environment, Consumer Protection, and Nuclear Safety. DG XIII added telecommunications to its activities, while DG XV added Company Law. Two new directorates were created: DG XXI, Customs Union and Indirect Taxation, and DG XXII Coordination of Structural Policies. The former reflected moving functions from DG XV and DG III while the latter reflected changes made by the Single European Act.
The Treaty on European Union brought about more changes in the structure of the European Commission. The Treaty mainly expanded competencies by altering voting rules. The monetary and political union envisioned by the Treaty changed the issues confronted by the European Union. This expansion is reflected in the changes in the Directorate Generals. DG V expanded its purview to include issues of industrial relations as well as employment and social affairs as the social charter was expanded by the Treaty. DG X continued evolving more into a directorate concerned with media policy in the EU as well as providing member state governments with information about the activities of the EU. A new DG was concerned with enterprise policy, distributive trades (services), tourism, and cooperatives.
Table 4 shows the relation between the original arenas defined by the Treaty of Rome and the structure of the European Council of Ministers. Each of the issues raised in the Treaty of Rome are represented in the organizational structure of the Council, mirroring the structure of the Commission. The Council changed its organization with the Single European Act and the Treaty on European Union. Table 5 presents the organization of the Council of Ministers following these changes. The Single European Act caused the Council to reorganize its activities. DG-B became exclusively concerned with Agriculture and Fisheries, while DG-C focussed on Single Market issues. The residual areas of the Treaty expansion (environment, consumer protection, education, health, and culture) were organized into a new directorate, G. The Treaty on European Union did not alter the organization structure of the Council very much. The expansion of competencies into tourism, broadcasting, and telecommunications were reflected in changes in DG-C.
One can observe that basic policy domains were given shape by the Treaty of Rome, the Single European Act, and the Treaty on European Union. These arenas were made concrete by providing organizational capacity in the European Commission and Council precisely as it was defined by these issues. As the Treaty of Rome was revised, the Commission's and Council's structure was changed.
There are other measures one can generate to index the development of policy domains. Table 6 presents a classification used by the Council of Ministers in their annual report on the topics of meetings. It was at these meetings that members of the Council of Ministers would discuss legislative issues. They index how the Council of ministers spent its time. One can observe that between 1970 and 1980, there was an increase in the number of meetings from 43 to 59 and a similarly increase to 80 in 1987. Over the period, the number of meetings of the Council of Ministers nearly doubled.
The topics of the meetings changed dramatically with the expansion of competencies of the EU. Agriculture, foreign affairs, and financial issues were always topics being considered by the Council. The largest change, which is not surprising given the Single Market Program, is in the category of the Internal Market and Industrial Affairs where the number of meetings rose from zero to 11 in 1987 and fell to seven in 1993. Fisheries became a big issue between 1970 and 1980, but it decreased in importance in the 1980s. Transportation, the environment, development, energy, and consumer protection became the focus of more and more Council meetings. The Treaty revisions in 1987 and 1993, which provided increased competencies in these arenas, expanded the number of meetings on these topics.
Another important measure of the ability to generate legislation in policy domains is the number of people who work in an issue area. Legislation in the EU comes from the European Commission. It follows that the more people working on a given set of topics, the more likely that the policy domain will produce legislation. Table 7 presents data on the number of people employed by the Directorate Generals of the Commission in 1989. The largest number of people worked for Translation, Personnel and Administration, and Science and Research. Most of the people employed in the Science and Research Directorate are employed in a research center and are not directly involved in the production of legislation.
The two next largest directorates are Agriculture and Development. Since these Directorates both administer large programs, this is sensible. Telecommunications and Information industries, Internal market and Industrial affairs and Energy follow in size. Many of the Directorates which have been the focus of a great deal of attention, the environment, financial institutions, and regional policies (where the structural funds are administered) are relatively small. The number of people employed to deal with these issues suggests that there are very few people who are writing new legislation in arenas recently expanded by the EU.
Another indicator of what the EU is doing is how it spends its money. Table 8 considers how the EU budget has grown from 1970 to 1993. The budget for the EU has expanded almost 22 times in 23 years. While the budget has increased across the board, the most dramatic shift has been away from agriculture and towards other concerns of the EU. In 1970, 91.2% of the EU's budget was spent on the Common Agricultural Policy. This decreased continuously until 1993 where it stood at 53.7% of the budget. The most dramatic increase in the budget occurred in the categories of structural funds. This category was on the rise from 1970 on and took 16.2% of the budget in 1987. This was right at the beginning of the Treaty revision that doubled the total of structural funds (Marks, 1993). Funds spent on research increased more slowly over the period. Funds mostly for outside development issues (called "external action") increased from zero percent of the budget to 3.4% in 1993. Administrative costs of the EU were a relatively small part of the budget stabilizing from 1980 at about 4.5-5% of the budget. The "other" category of the budget includes all other programs of the EU including those focussed on culture, education, and European networks.
The final indicator of institutional capacity is the total number of employees in each part of the EU. Table 9 shows that staffing at the EU has increased from about 9,000 people in 1970 to over 25,000 in 1993. These steady increases occurred as the budget increased over 20 fold. The bottom of the table shows that the European Commission started out as 84.8% of the total in 1970, but dropped to about 70% from 1970 onward. The biggest increases in employment occurred at the European Parliament, particularly with the enlargement of the EU. The Council, Court of Justice, and Court of Auditors all grew as well.
What do these tables tell us about the organizational capacities of the EU? The EU is a relatively small organization. It manages a 65.5 Billion ECU budget with a staff of 25,000. 84% of the budget is tied up in two large programs: agricultural price supports and structural funds. The European Commission that is responsible for generating legislation has almost no regulatory capacity. It also has a relatively small number of people working for it who are directly engaged in producing legislation. This means that policy domains function not just as arenas to affect legislation through lobbying, but probably actually produce much of the legislation (Peters, 1991; Mazey and Richardson, 1993). This must be particularly true in arenas where the Commission has not grown rapidly such as transportation, the environment, and regional policy.
The picture of the EU that is emerging from our examination of its structure, is that the Treaty of Rome, the Single European Act, and the Treaty on European Union strongly shaped the organization of the European Commission and Council of Ministers. The Treaty changes have mostly been couched in the language of the original Treaty of Rome. This shows the power of preexisting institutions and organizations as they have shaped the debate about what the EU can and should do.
The changes brought about by the two Treaty revisions have been substantial. They have altered the logic of the EU by changing from unanimous voting to qualified majority voting and opened up new arenas for possible cooperation. They have also shifted the activities of the EU away from agriculture and towards using the EU as a redistributive structure across western Europe (and indeed, to a smaller degree, the world). The Treaties have also plausibly helped define policy arenas by defining areas where negotiations would occur. These have been enshrined in the organizational structures of the Commission and Council.
The Treaty revisions did not occur in a vacuum. It is clear that changes in the organization of the Council and Commission and shifts in budget priorities were the end point of long run discussions. So, for instance, negotiations over issues surrounding fishing rights eventually led to the formation of a separate DG to manage fishing. Similarly, discussions over coordinating monetary policies in the 1970s led to the founding of a directorate concerned with financial issues by 1980 and eventually, the decision to begin a monetary union. Finally, the idea of structural funds and a regional and social policy to cope with dislocations due to economic development was already present in the Treaty of Rome.
We began this paper by suggesting that studying the EU as a set of policy domains required considering three elements: the existence of a "constitutional" basis for interaction to define policy domains, organizational capacity that corresponded to organize those domains, and the existence of nongovernmental organizations to affect legislation in those domains. The evidence so far has focussed on demonstrating that the various Treaties have indeed defined domains and that there exists organizations that mirror those domains. It is useful to turn to how nongovernmental organizations relate to the legislative output of policy domains.
Haas, in his seminal work The Uniting of Europe (1957), argued that the expansion of the European Union would come through a process of "spillover". This implied that new areas of policy would come from the transnational organization of interests, the activism of the European Commission, and the pressures these groups brought to bear on the national governments that would lead to more cooperation and competencies for the European Community.
Our evidence shows that almost all of the expansion of competencies of the European Union occurred by Treaty modification in arenas already defined as relevant to common cooperation. The main new issue brought to the table by the Treaty revisions concerned military and security cooperation, although even here, there was precedent for this discussion. All of the other issues were in some way linked to the original purposes of the organization: ie. the creation of a "single market" and cooperation in monetary and social policy necessary to attain it. Put another way, the "master frame" of the European Union was set by the Treaty of Rome and all other discussions have linked back to that frame.
Having said that, it is also clear that this "frame" has been used skillfully to make agreements that have substantially expanded the policy domains of the EU. After all, a vague agreement to discuss cooperation in some arena does not commit actors towards creating institutions. For instance, if one reads the original Treaty of Rome, there is a vague passage about cooperation on monetary issues. It would be absurd to argue that this passage led to the agreement on monetary union. The competencies of the European Union expanded along lines set in the original Treaty, but this does not answer the question of who acted, when, and why.
This question can be examined by looking at the timing of the Treaty Revisions and the organization of policy domains. Did a flurry of legislation in policy arenas proceed the Treaty renegotiations? Did transnational groups expand their activities leading up to the Treaty revisions? Did gridlock caused by the unanimous voting rules of the European Union slow down the legislative process leading up to the Single European Act and only after that Act's passage did legislation flow? Did transnational groups precede or follow the openings presented to them by the Single European Act and the Maastricht Treaty? Finally, how many of the policy domains look like supranational politics?
Figure 1 presents the frequency distribution of all forms of secondary legislation (directives, decisions, and regulations; see Appendix) from the beginning of the EU until 1994. The amount of legislation grew steadily in the 1960s and 1970s. It levelled off in the late 1970s and early 1980s. Starting in 1985, it rose rapidly, peaking in the early 1990s at over 1500 pieces of legislation a year.
Two patterns of change are worth highlighting here. First, the rate of growth in legislative output decreased markedly in the late 1970s and early 1980s coterminous with the political crisis of the EU (Moravscik, 1993). Second, once the Single European Act came on the agenda, the amount of legislation took off. This reflects the changes in the voting rules in the EU and the expansion of competencies for the EU. This evidence implies that the political crisis of the EU was being forced to some degree by pressure for more and more cooperation as evidenced by the increasing output of legislation in the 1970s. And when that pressure was relieved by the Single European Act, the legislative floodgates came open.
Figure 2 traces the founding rates of the major transnational pressure groups of the EU from 1950 to 1992.[3] The figure clearly shows that, following the formation of the EU, the number of transnational pressure groups took off. The formation of groups lessened after the "Empty Chair Crisis" and the Luxembourg Compromise. During the 1970s, as the legislative output of the EU began to rise, so did the founding of new transnational groups. The political crisis in the EU of the early 1980s saw a fall-off in the formation of transnational pressure groups, although since 1986 there has been a renewed modest rise.[4]
There are several important conclusions to draw from these figures. First, we see solid evidence that the crisis of the early 1980s in the EU did not follow a moribund period from 1966, as measured by the amount of legislation and the number of transnational groups. Indeed, both legislative output and transnational pressure group representation increased during the 1970s. This suggests that expectations of continued and perhaps increased cooperation were brought to bear on the national governments by transnational pressure groups. The political crisis of the 1980s might best be characterized as a search for acceptable ways to get that cooperation. The slowdown in both legislative output and the formation of transnational groups around 1980 shows that the crisis was real, both in output terms for the EU and from the expectations of nongovernmental organizations.
This is evidence for both intergovernmental and supranational accounts of the EU. There was pressure to produce more cooperation in the EU, as measured by the presence of transnational organizations (what Stone and Caparaso in this volume call transactors) and legislative output, but the problem was discovering how states would choose to deal with it. Once they changed the voting rules and expanded the competencies of the EU, a flood of legislation followed.
It is important to disaggregate these data in order to see what they imply about the nature of policy domains. Table 10 presents a breakdown of the total legislative output within the policy domains specified earlier over the period 1958-94 and the number of organizations that attach to each policy domain. Policy domains for organizations were coded by mapping the Directorate Generals with which organizations maintained regular contact onto corresponding legislative domains established in the EU.[5]
Table 10 does not present causal evidence regarding which came first, the domains or the organizations, but instead examines the degree to which organizations and legislation in domains correlate. The zero order correlation is .41, which indicates a strong positive association between the number of organizations in a policy domain and the corresponding amount of legislation.
Earlier, we argued that some of the policy domains could look more like intergovernmental bargaining while others might look closer to supranational regimes. The relatively high correlation between legislative outcomes and transnational pressure groups suggests that many of the domains of the EU either were or eventually became supranational. For instance, the policy domain of Industry/Internal Market has the third highest quantity of legislation and the largest number of transnational pressure groups.
On the other extreme, the Fisheries policy domain has 474 pieces of legislation and only one identifiable transnational pressure group. It is obvious that this domain more closely reflects intergovernmental bargaining. From table 6, we note that the Council devoted a large number of meetings to the issue of fisheries in the 1970s and through the 1980s. The initiative to engage in a fisheries policy emerged in 1983. Many of the issues in this arena concerned fishing rights of various nations that related most directly to questions of national sovereignty (ie. claims over coastal boundaries).
It is useful to explore these relationships in more detail. Table 11 presents the amount of secondary legislation passed in different historical periods by policy domain. The table contains the total amount of legislation, the average amount of legislation passed per year, and the percentage of legislation in that period in each domain.
44.4% of the all legislation passed was in the domain of agriculture, with external relations accounting for the next highest amount (10.3%). All domains witnessed increases in the absolute amount of legislation over time. However, the mix of legislation changed somewhat between the 1960s and 1990s. The most dramatic increases in legislation occurred in the domains of fisheries, external relations, environment/consumer/health, science/information/culture, and regional/structural policies. From the 1970s on, the largest relative decreases in the amount of overall legislation appeared in the domains of customs, employment/social policy, right of establishment of services, transport policy, competition policy, and energy (although the absolute level of legislation was increasing in most of these policy domains).
We can see that the Treaty revisions in the 1980s did result in an expansion of legislative activity. But, expansion of legislation after the Single European Act was not confined to the Industrial Policy/Internal Market arena (which already had a large amount of legislation and did not increase that dramatically). If we examine the increases in the average amount of legislation per year, we see that, during 1980-86, there were huge expansions of legislation before the Treaty revisions, particularly in customs, agriculture, fisheries, external relations, regional/structural policies, and environment/consumers/health. After the Treaty revisions, these policy domains continued their dramatic expansion. We conclude that while the Treaty revisions appear to have caused increases in competencies of the EU, there was pressure in several policy domains to increase these competencies.
One can see this more clearly if one examines Table 1. With the exception of agriculture and fisheries, those policy domains that experienced rapid legislative increases before the Single European Act and the Treaty on European Union were all arenas in which the competencies of the EU were expanded. To some degree, the Treaty revisions were a response to legislative pressure. There are exceptions to this result as changes in the Treaties also produced increases in legislative output in arenas where there this pressure did not exist.
Table 12 summarizes the formation dates of the major transnational pressure groups in the EU across policy domains. The greatest expansion of transnational pressure groups occurred between 1958-1969, followed by 1970-79. The largest number of transnational pressure groups were formed in the Industry/Internal Market policy domain, followed by Agriculture, Science/Information/Technology, External Relations, Employment/Social Policy, and Environment/Consumer/Health. The large number of transnational pressure groups present in the Industry/Internal Market domain suggests that, as the Single Market Program began to evolve in the early 1980s, there was an organized constituency that was mobilized to support it. It also suggests that the ability of actors in and around Brussels to drum up enthusiasm for the Single Market was probably high. While scholars have focussed on the role of business in the production of the Single Market (Ludlow, 1988; Colchester and Buchan, 1986), they have generally focussed on a small set of organizations like the European Roundtable. Our results suggest that there was a wider range of organizations that had interests in the construction of the Single Market (Fligstein and Brantley 1995).
There are also policy domains where transnational group formation increased after 1980. This is seen most strikingly in the fields of employment/social policy, right of establishment of services, economic and monetary policy, and the environment. These were all policy domains that were expanded by the Single European Act and the Treaty on European Union (see Table 1). Most of these organizations were founded before the Single European Act.
Since the expectation that the EU was going to be relaunched was apparent by 1984, one could argue that people were just positioning themselves for the institutional openings. On the other hand, the presence of new organizations in arenas where new competencies were being prepared implies that the organizations might have played some role in lobbying their governments to accept treaty revision. The clearest case of organizational founding after the Single Market initiative began, was in the field of the environment. Here, groups appeared to be responding to the political opportunity to organize.
The conclusion one can draw is that transnational pressure groups came to Brussels beginning in the 1950s. The 1970s witnessed a peak of organizational founding. The transnational pressure groups that attached themselves to the Industry/Internal Market policy domain was the largest contingent. There is evidence that growth in the number of organizations in a number of domains proceeded the Treaty openings. These transnational groups were either anticipating those openings or lobbying for them.
It is useful to consider which of the policy domains look the most dominated by transnational groups in the early 1990s. One rough index of this is to examine the ratio of the legislative output from 1992-1994 and divide it by the number of transnational pressure groups in the domain.[6] A large ratio indicates that there is substantial legislative output with few transnational groups. We take this as an indicator of domains dominated by states. A small ratio indicates the presence of more transnational groups and the possibility that the domain can be characterized as dominated by transnational groups.
This ratio has a bimodal distribution. Domains with a high ratio include financial/institutional matters, fisheries, taxation, and regional/structural funds. These domains are plausibly intergovernmental. The rest of the domains with low ratios include customs, agriculture, employment/social policy, right of establishment, transport policy, competition policy, economic/monetary policy, external relations, energy, industry/internal market, environment/consumer/health, science/information/culture, and law regarding undertakings. These domains look more dominated by transnational groups and are plausibly supranational.[7]
This division is intriguing for several reasons. First, it shows that nation state governments have kept control over the internal financial and institutional arrangements of the EU, issues regarding harmonization of taxes, the spending of structural funds, and fishing. Controlling the financial and institutional arrangements is an obvious way to exert leverage over the EU. Similarly, controlling taxation and the distribution of structural funds also seems to be sovereignty enhancing. Nonetheless, nation state governments have agreed to supranational cooperation over a wide range of important issues including monetary policy, agriculture, the internal market and competition policy. This would seem evidence that some amount of sovereignty has indeed been transferred to Brussels.
It is useful to conclude this paper by trying to capture the main outlines of the institutional terrain of the EU as it has shifted over the past 40 years and then consider some implications for intergovernmentalism and supranationalism. The EU is an organization that has evolved in a path dependent fashion. Most of the arenas in which the EU built competencies were laid down in the Treaty of Rome and the original language and definitions gave rise to organizational structures oriented towards producing legislation in those domains. We have evidence that the Commission and Council were organized around those domains, sifted and sorted information in those terms, and used those categories to classify legislation.
Most of the budget of the organization was originally devoted to agriculture, and later, regional/structural funds and development. Most of the personnel worked for the European Commission. The organizations of the EU (the Council, Parliament, Commission, ESC, and the Court of Auditors) are relatively small. They have very little direct regulative capacity and they rely heavily on the nation state bureaucracies to enforce directives. Besides managing the three large programs, most of what the Commission does is to try and broker agreements between states and nongovernmental actors (Mazey and Richardson, 1993).
Policy domains tend to be constituted with transnational groups, representatives of states, and the appropriate parts of the Commission. There also is a strong association between the presence of transnational pressure groups and the production of legislation. Most policy domains (13 of 17) are political arenas where there exist transnational pressure groups. But there remain policy domains (4 of 17) that are mainly organized by and for the states.
Our most suggestive evidence was obtained by looking at the creation of transnational pressure groups and legislation over time. We showed that the 1970s witnessed an expansion in the amount of legislation and the growth of transnational pressure groups across the EU. Most intriguingly, transnational groups and legislation appeared in arenas where competencies were expanded by the Treaty revisions in the 1980s and 1990s. Yet, here there was some contradictory evidence as well. Some policy domains that witnessed openings by the Treaties did not experience increases in either transnational pressure groups or legislation before the Treaty openings. There was also a great increase in legislation across the board following the Treaty changes, particularly the Treaty on European Union where changes in the Council voting rules affected all policy domains.
Another important result was that the 1970s was a very active period in the EU both legislatively and in terms of the founding of transnational pressure groups. The Treaty revisions of the 1980s took place in the context of a demand for more cooperation at the European level. One can infer that the crisis of the 1980s was partially precipitated by the limits of the institutions and organizations of the EU to deal with these activities. Finally, the largest number of transnational groups were concentrated in the Industry/Internal Market policy domain before 1980. This implies that there were a great many organizations that were mobilized in favor of the Single Market initiative and this explains part of the attractiveness of that initiative.
It is useful to return to the intergovernmentalist-supranationalist debate over the EU. Our evidence could provide grist for both mills. Clearly, the Treaties and their revisions had a great deal of impact over what policy domains were possible, the organization of the EU, the production of legislation, and the way in which the budget was spent, giving support to the intergovernmentalist perspective. But equally clearly, some of the policy domains became organized by the Commission and transnational pressure groups from across the EU and this affected the output of legislation and produced pressure for increased cooperation in those domains. And the Treaties as often facilitated collective action as limited it. The Single Market initiative certainly gained greatly from the network of transnational pressure groups interested in that issue.
There is an obvious way to reconcile these theoretical traditions that is suggested by our data. "Spillover" (in the Haasian sense) mostly took the form of using competencies hinted at in the original Treaty of Rome to create supranational policy domains (ie. organized arenas with state and nonstate actors). In essence, the Commission and transnational actors were able to organize to put pressure on the nation state governments to produce more legislation in policy domains hinted at in the Treaties. As a result, one can assume that nation state actors found themselves being lobbied both in their home capitals, but also in Brussels.
Arguments from both the supranational perspective and the intergovernmentalist perspective imply that nation state actors have interests and act on them (sometimes, intergovernmentalists want to deny this about supranational arguments). The main difference in the arguments is really about where transnational pressure groups, when they exist, lobby their governments, ie. in their home capitals or Brussels.
Our data (consistent with the papers in this volume) imply that supranational arrangements are influenced by transnational pressure groups (indeed part of our definition of supranational policy domains is the existence of such groups). Those groups pressed their governments for more cooperation both at home and in Brussels. This means that intergovernmentalism, which emphasizes the interests of nation state leaders as a cause of cooperation, is compatible with a supranationalism that emphasizes how the interests of nation state leaders was affected by transnational pressure groups both at home and in Brussels. This leaves the issue of how the preferences of governments are formed somewhat open. Here, there is disagreement amongst scholars. It may be the case that those preferences changed as transnational interests groups pressured their governments towards embracing more cooperation as both intergovernmental and some supranational accounts would suggest.
But there is an alternative story one can tell. The leaders of the nation states had to decide how to deal with pressure from transnational groups in the context of trying to preserve their national sovereignty. It may have been that the leaders of the governments were unsure about what to do in the early 1980s. Indeed, the EU was in multiple crises at that moment and the decision to cooperate more was not necessary or preordained. From the perspective of 1981, an observer could have easily concluded that the EU was caught in a bargaining trap from which there was no exit (Fligstein and Mara-Drita, 1996).
Several scholars have argued (Fligstein and Mara-Drita, 1996; Ross, 1994; Pierson, 1996) that the preferences of governments during this period were in fact reshaped by policy entrepreneurs in the Commission and this changed the terms of discussion so that there became momentum towards more cooperation.
But even this position is not diametrically opposed to either a intergovernmentalism or supranationalism based on preferences of governments. Nation state leaders had to agree with whatever changes produced more supranational cooperation. While their conception of their interests were reorganized, their decisions on expanding those competencies were not inconsistent with the interests of their most organized transnational groups who were in Brussels.
One could argue that the nation state governments realized that their sovereignty was already under attack in that governments found it difficult to make economic, and to a lesser degree, social and military policy independent of one another. The nation states agreed to collectively bargain in Brussels and in many cases, let transnational pressure groups and the Commission control the content of legislation. They accept the results of that bargaining as binding, even when they may have been outvoted by qualified majority voting. To the degree that their largest organizations are represented in these discussions, their "interests" can be served. By using their own regulatory apparatuses to enforce those agreements, they prevent Brussels from usurping the power of nation state bureaucracies and at least keep those bureaucracies at work. In this somewhat roundabout way, more cooperation at the EU level could be though to help preserve what is left of nation state sovereignty.
The EU is an organization that has many faces as a result of its uneven institutional history and the complex nature of the Treaties that created it. In some policy domains, we observe transnational groups and Commission to be equal partners to the nation state governments, and in others, their remain situations that appear to reflect intergovernmental bargaining. If we are right, 13 of the 17 most important policy domains are now influenced by transnational groups. The creation of policy domains, the production of legislation, and the growth of transnational pressure groups have all contributed to the expansion of EU competencies in explicable ways.
Bibliography
Anderson, Svein and Kristen Eliassen (1991). "European Community lobbying." European Journal of Political Research 20: 173-87.
Bulmer, Simon and Wolfgang Wessels (1987). The European Council. London: MacMillan Press.
Colchester, Nicholas and David Buchan (1990). Europower. London: Economist Books.
European Commission (1988). The Community Budget: The Facts in Figures. Luxembourg: Office of the Official Publications of the European Community.
-------------------- (1993). The Community Budget: The Facts in Figures. Luxembourg: Office of the Official Publications of the European Community.
-------------------- (Serial). Directory of the Commission of the European Community. Luxembourg: Office of the Official Publications of the European Community.
--------------------- (1987). Treaties Establishing the European
Communities. Luxembourg: Office for Official Publications of the European Community.
------------------- (1992). Treaty on European Union. Luxembourg: Office for Official Publications of the European Union.
------------------- (1995). Directory of Community Legislation in Force. Luxembourg: Official Publication of the European Communities.
European Council of Ministers (Serial). Report on the Activities of the European Council of Ministers. Luxembourg: Office for Official Publications of the European Community.
----------------------------- (Serial). Guide to the Council of Ministers. Luxembourg: Office for Official Publications of the European Community.
Fligstein, Neil and Peter Brantley (1995). "The Single Market Program and the interests of business." In Barry Eichengreen and Jeffrey Frieden (Ed.) Politics and Institutions in an Integrated Europe. Berlin, FRG.: Verlag-Springer.
------------- and Iona Mara-Drita (1996). "How to make a market: Reflections on the European Union's Single Market Program." American Journal of Sociology 102: 1-33.
Garrett, Geoffrey (1992). "The European Community's internal Market." International Organization. Spring: 533-60.
Geertz, Clifford (1980). Local Knowledge. Princeton: Princeton University Press.
Haas, Ernst (1958). The Uniting of Europe: Political, Social,
and Economic Factors. London: Stevens and Sons.
Hay, Richard (1989). "The European Commission and the administration of the community." European Documentation. Luxembourg: Office of the Official Publications of the European Community.
Keohane, Robert and Stanley Hoffmann (1991). "Institutional change in Europe in the 1980s." In Robert Keohane and Stanley Hoffmann (Ed.) The New European Community. Boulder, Co.: Westview Press.
Kirchner, Emil and Konrad Schwaiger (1981). The Role of Interest Groups in the European Community. Aldershot, Eng.: Gower Publishing.
Laumann, Edward and David Knoke (1989). The Organizational State. Chicago, Il.: University of Chicago Press.
Ludlow, Peter (1988). Beyond Europe 1992: Europe and Its Western
Partners. Brussels, Belgium: Center for European Policy Studies.
Marks, Gary (1993). "Structural policy and multilevel governance in the European Community." Pp. 390-410 in Andrew Cafruny and Gerald Rosenthal (Ed.) The State of the European Community. Boulder, Co.: Lynee-Rienner.
Mazey, Sonia and Jeremy Richardson (1993). Lobbying in the European Community. Oxford: Oxford University Press.
Moravcsik, Andrew (1991). "Negotiating the single market act: National interests and Conventional Statecraft in the European Community." International Organization. 45:19-56.
Peters, B. Guy (1991). "Bureaucratic politics and the institutions of the European Community." In Alberta Sbragia (Ed.) Europolitics. Washington, D.C.: Brookings Institution.
Philips, Alan (1994). Directory of Pressure Groups in the European Community. Essex, Eng.: Longman.
Pierson, Paul (1996). "The path to European integration: a historical institutionalist approach." Comparative Political Studies 29: 123-63.
Pollack, Mark (1994). "Creeping competence: the expanding agenda of the European Community." Journal of Public Policy 14: 95- 145.
Sandholtz, Wayne and John Zysman (1989). "1992: Recasting the European bargain." World Politics 42:95-128.
Sbragia, Alberta (1992). "Thinking about the European future." Pp. 257-292 In Alberta Sbragia (Ed.) Europolitics, Washington, D.C.: Brookings Institution.
Schmitter, Phillippe (1992). "Interests, powers, and functions: emergent properties and unintended consequences in the European polity." Unpublished manuscript. Stanford University.
Tsebelis, George (1992). "The European Parliament as a conditional agenda setter." American Political Science Review 88: 128-42.
Methodological Appendix
Table 1 is based on the Treaty of Rome and its two substantive revisions, The Single European Act, and the Treaty on European Union. The categories that appear in the first part of Table 1 come directly from the title and chapter headings from the Treaty of Rome (European Community, Treaties Establishing the European Community, Luxembourg: Office for Official Publications of the European Union, 1987, p. 209-11). The Single European Act modified the original Treaty of Rome by replacing certain passages and adding new ones to define new or expanded competencies for the European Union. The categories were considered "expanded" if the Single European Act explicitly changed the competencies in a particular arena. A new category was formed when a whole new chapter was added to the original Treaty (see Treaties Establishing the European Community, p.215-422). The Treaty on European Union specifies the competencies of the European Union by replacing Article 3 of the Treaty of Rome with a new Article 3, which specifies 20 separate domains in which the European Union has precedence (European Community, Treaty on European Union, Luxembourg: Official Publication of the European Communities, 1992, p. 12). These include many of the domains already contained in the Treaty of Rome and the Single European Act. Wherever the competencies in the Treaty of Rome were substantially changed to increase the scope of the European Union (either by defining new responsibilities, changing the voting rules in the Council, or changing the procedures), the competence was counted as being "expanded".
Table 2 takes the original arenas specified by the Treaty of Rome and maps them onto the organization of the European Commission in 1970. One can observe that there is a one to one mapping between the titles and chapters of the Treaty of Rome and the Directorates General of the Commission in 1970. The structure of the Commission comes from the Directory of the Commission of the European Communities, a yearly publication of the Commission. Table 3 presents the titles of the Directorate Generals in 1970, 1980, 1987 (after the Single European Act) and 1993 (after the Treaty of European Union). These also come from the Directory of the Commission of the European Union.
Table 4 contains the topics considered at the meetings of the European Council of Ministers for 1970, 1980, 1987, and 1993. These are the categories reported by the Council. These appear in the yearly report of the European Council (European Council of Ministers, Annual Review of the Council's Work, Luxembourg: Official Publication of the European Communities, 1970, 1980, 1987, and 1993).
Information for EU legislation passed between 1958 and 1994 was dervied from the Directory of Community Legislation in Force (European Community 1995). The Directory provides bibliographic summaries of all legislation broken down by twenty analytical subdivisions defined by the EU. These subdivisions, listed in Table 11, are: general financial and institutional matters; customs union and free movement of goods; agriculture; fisheries; freedom of movement for workers and social policy; right of establishment and freedom to provide services; transport policy; competition policy; taxation; economic and monetary policy and free movement of capital; external relations; energy; industrial policy and internal market; regional policy and coordination of structural instruments; environment, consumers, and health protection; science, information, and culture; law relating to undertakings; common, foreign and security policy; cooperation in the fields of justice and home affairs; and People's Europe. As noted in the text, these subdivisions closely correspond to the policy domains of the Directorates General.
Our data include all legislation passed between 1958 and 1994 categorized by the EU as secondary legislation (CELEX documentary sector=3), including amendments. Our dataset contains a total of 15,643 legislative events, composed principally of decisions, directives, and regulations. Of the total pieces of legislation coded, 43.4% are classified by the EU as regulations, 30.6% as decisions, 16.7% as directives, and the remainder composed of other acts. Listings for legislation passed in 1995 have not been included in the dataset, since the Directory only documents legislation passed through December 1st of that year. Including 1995 legislation in our dataset would have led to an undercount (when comparing to other full year counts). The handful of legislation passed before 1957 (n=29) was also omitted, as it occurred before the Treaty of Rome and thus might not be properly considered as Community legislation for the purposes of our analysis here.
Philips (1994) provides what remains one of the most comprehensive records of EU pressure group activity available to date. The compendium lists major attributes (e.g., staff size, purpose, budget, DG contacts) for approximately 800 pressure groups active in lobbying the European Community as of the early 1990s. Data provided by Philips was compiled from questionnaires sent in early 1991 to several hundred organizations, and was supplemented from a number of other sources available to the contributors. The compilation includes organizations that range in size from post office boxes with no full time staff, on the one hand, to well-endowed international labor groups and trade unions with multiple offices (e.g., ETUC), on the other.
It is very likely that Philips (1994) is not an exhaustive inventory of organizations active in lobbying the European Commission. The contributors acknowledge that the recent growth in number of new voluntary organizations, combined with the absence of any central information source on lobbying activity in Brussels more generally, make the task of providing a comprehensive list difficult at best. Given these limitations and the methodology employed by the contributors, we can reasonably assume the organizations most likely to be missed by Philips (1994) are those that are either relatively small, were founded very recently, or that maintain such a low lobbying profile that they will not be familiar to an astute observer. To the degree that the list provided by Philips (1994) may be biased, then, it is likely to be skewed toward organizations that are larger, longer-established, and/or more prominent in discussions of EU policy. Nevertheless, for purposes of the current exercise, such a possible bias need not present a major problem.
We are concerned here with transnational pressure groups whose resources, institutional access, and lobbying efforts are significant enough to allow participation in the formation and evolution of policy domains in the EU. These are the groups for whom representation in Philips (1994) is likely to be most complete Groups whose existence is mostly symbolic (e.g., no office or staff), or that have not been around long enough to muster a lobbying presence in Brussels, are not likely to be participants in a policy domain. Our use of Philips (1994) book is therefore consonant with a focus on organizations that are active participants in policy-making at the EU level.
For purposes of data collection, we coded all organizations listed in the main body of Philips (1994) that provided information about contacts with one or more Directorate Generals in the Commission. For each organization, information was recorded on all DGs associated with the group's activity as well as the organization's founding date (when available). A total of 245 organizations were coded, providing a total count of 761 "points of contact" between pressure groups and Directorate Generals. Most organizations listed one, two, or three DG contacts in the Commission; a handful listed six or more.
For purposes of comparing pressure group and legislative activity across policy domains, we recoded information on Directorate General lobbying activity to correspond with analytical classifications of legislation provided by the Commission. As we note in the text, there is a close correspondence between many of the DGs and the analytical categories used by the Commission to classify legislation. Table A1 provides a listing of DG classifications and the analytical domain categories onto which they were mapped.
As can be seen in Table A1, the majority of DGs map directly to a legislative domain. DGs that did not correspond in name to an analytical heading were mapped to the category closest in scope and content: Development was mapped to External Relations; Financial Institutions and Company Law was mapped to Law Relating to Undertakings; Credit and Investments was mapped to Economic and Monetary Policy and Free Movement of Capital; and Enterprise Policy and Small Businesses was mapped to Right of Establishment and Freedom to Provide Services. It should also be noted that two legislative categories were subsumed into other existing domains in order to more accurately correspond to the DG headings. Common, Foreign, and Security Policy was subsumed into External Relations; and, Taxation was subsumed under Customs Union and Free Movement of Goods.
Footnotes
[1]1. We think these categories are somewhat artificial and they overlap, but they are a good approximation to capture the major policy domains of the EU.
[2]2. The European Parliament committee structure has a similar set of domains defined. The annual report entitled General Report of the Activities of the European Communities contains a listing of the standing committees of the Parliament which parallel those of the Council and Commission. The Economic and Social Committee has a different organizational structure that is focussed on dividing into firms, workers, and other organized groups. But, it too has a committee structure that parallels the policy domains.
[3]3. The appendix contains a discussion of the data source for this table and considers the question of the representativeness of this group of organizations.
[4]4. Anderson and Eliassen (1993) assert that as many as 2000 pressure groups exist in Brussels currently. Our examination of different lists of these groups led us to conclude that many of these groups are very small (1 or 2 person offices) and that the major function of most of them is to provide a listening post for European wide organizations. The largest pressure groups, not surprisingly, tend to be older, bigger, and much smaller in number.
[5]5. We have already argued that the Commission was organized according to policy domains specified by the Treaties. It follows that groups interested in certain policies would locate the relevant people in the Commission and the Council in order to express their views, thus being participants in the policy domains. See the appendix for details on how organizations were coded into policy domains.
[6]6. We exclude the domains of Common/Foreign/Security policy, Justice/Home Affairs, and People's Europe because only 6 pieces of legislation were passed from 1992-94.
[7]7. This ratio has several obvious problems. First, it depends on the amount of legislation in a domain. Domains with lots of pressure groups with little legislation could in fact be dominated by intergovernmentalism. That is, states are not allowing legislation to be passed. Second, it does not adjust either for the importance of the legislation or the pressure groups. So, for instance, a small number of pressure groups might be very powerful and have a large legislative effect. The measure, however, has some face validity in two ways. The measure is bimodal in distribution implying that either domains have a high or low ratio. It is also the case that the domains that appear intergovernmental (taxation, financial/institutional matters, etc...) on the ratio are plausible candidates for remaining under the direct control of governments).