9 Citizen Support for Policy Integration RUSSELL). DALTON AND RICHARD C. EiCHENBERG ONE of the most remarkable features of modem European politics has been the process of European integration. Beginning with the Coal and Steel Community, the nation-states of Europe have gradually and relatively steadily moved toward ever-closer union. Moreover, in the last decade the rate of progress has accelerated, as the European Union expanded its policy responsibilities and membership, and institutionalized its presence in European politics. The initial process of European integration was dominated by the strategies and actions of political elites in line with the intergovernmental model described by Stone Sweet and Sandholtz (Chapter 1). Over time, however, this process has gradually moved toward a model of supranational governance that includes a range of other societal actors. Party leaders, parliamentary committees, transnational interest groups, and other political groups are now more involved in the process of building Europe. Our research specifically focuses on the role of public opinion as a societal actor in the integration process. Only ten years ago, unification research often ignored public opinion for understandable reasons. The integration process began as intergovernmental bargaining, and factors such as international power, elite preferences, or the actions of organized interests dominated the frameworks of neofunctionalists and realists alike. When researchers did consider the role of public opinion, at most they concluded that the public offered a "permissive consensus" that provided political leaders with considerable latitude in carrying out the European project (Lindberg and Scheingold 1970). Others bluntly dismissed the significance of public opinion (Haas 1958:17). Several factors have contributed to an increased awareness of the significance of public opinion in the integration process. In retrospect, it appears that scholars underestimated the role of public opinion in defining the national preferences that guided elite bargaining during the early decades of the integration process. Adenauer's efforts to build a European identity among the Germans, the public debate that accompanied the failed attempt to develop a We are grateful to Anthony Messina for a thorough reading of an earlier version of this chapter. Citizen Support for Policy Integration 251 European defense force, or the popular endorsement of the Common Market concept were affected by public preferences, or at least elite perceptions of these preferences. Even under an intergovernmental model, public opinion is important in defining national preferences.' In addition, as the integration process has evolved from intergovernmental bargaining toward "normal" politics, this expanded the role of public opinion in the process. EU policy is no longer a policy domain that is distant from the everyday life of Europeans. Just as in the formation and implementation of domestic policy, EU policy involves public debates about the political choices facing each member nation. For example, public opinion on integration policy contributed to Margaret Thatcher's political vulnerability toward the end of her administration. Similarly, public debates over currency union and public reactions to proposed cuts in government spending to meet EU targets are highly visible parts of French and German domestic politics. Public preferences inevitably condition the actions of interest groups, political parties, and elites toward EU policies. Furthermore, a changing framework for EU policymaking increases the potential ways in which public opinion can influence the integration process.2 Today, public opinion (and the positions of other national and transnational actors) is politically relevant in determining the activities of the EU politics of the member-states to a degree that violates a simple intergovernmental model or the integration process. — Public opinion also plays a role in moving the integration process along the continuum from intergovernmentalism to supranationalism. The public's pol- , icy preferences can influence which areas are most susceptible to further integration efforts. When there is permissive consensus or positive support, national governments are more able to endorse European action. When the publics of the member-states disagree, this is likely to retard further integration. Moreover, discussion of the "democracy deficit" within the EU necessarily creates pressures to move away from intergovernmental modes of decision-making and toward institutional arrangements that increase the input from the public and other societal actors. The recent history of the European Union illustrates that public opinion wields real influence. Public reactions to the reforms of the 1980s and 1990s— principally the Single European Act and the Maastricht accords—demonstrated that European citizens were neither "permissive" nor "consensual" in their appraisal of the dramatic relaunching of European integration. The Danish referendum of 1992 temporarily detracked the integration process, while tľie Irish, French, and Danish referendums of 1993 moved the process ahead. In 1 Indeed, neofunctionalist writings in the second wave of integration research readily accepted the roie of public opinion in the integration process (Haas 1971; Schmitter 1971). On the interrelationship of domestic preferences and intergovernmental bargaining see Moravcsik (1989). 2 The introduction of a directly elected parliament gives the public a direct jepresentation within the EU. In addition, the institutionalization of the integration process has increased the public's potential points of access into the process, such as contacting EU officials, to contacting national parliament officials, petitioning the ECJ, and other methods- 252 Russell ]. Dalton and Richard C. Eichenberg fact, one could argue that public opinion and the votes of citizens in the ratification of the SEA and Maastricht have been a profound influence on the recent process of integration. In the wake of the Danish and French referendums on Maastricht, sensitivity to public opinion became a hallmark of European integration.3 In summary, public opinion has grown from a relatively minor role in the integration process to a principal focus of political and scholarly attention. We acknowledge that public opinion has a broad and diffuse influence, and that the policy positions of specific societal actors—such as political parties and interest groups—may be more important in explaining the immediate course of the integration process (see Chapter 3). But public opinion provides the broad context for these policy debates, much as it does for the domestic politics of the member-states. Public support for the European Union can facilitate the process of further union, just as public skepticism toward the EU can slow the integration process. As the Union moves toward further reform, it seems no exaggeration to speak of a Citizens'Europe. This research examines the patterns of citizen support for the process of European integration. We focus our attention on support for policy integration in specific issue areas. That is, to what extent do Europeans believe that policy responsibility in specific areas such as health, environment, defense, and other fields should be transferred from national governments to the European Union. Like others in this research project, we are interested in explaining how the integration process moves along the continuum from state-centered action to supranational action—this time measured in public preferences (see Chapter 1). We believe that public opinion is one factor that can influence movement along this continuum. In addition, the variations in public support for policy integration provide a medium for testing theories of the integration process as measured in public sentiments. 1. Public Opinion and Policy Integration Even if European integration is described as a general political process, it involves separate decisions to take common action on discrete policies. Progress comes not from a broad movement on all fronts, but by specific initiatives in specific areas. Integration is a process by which the EU gradually accumulates policy responsibilities. However, most prior opinion research has focused on generalized public 3 Further evidence of the relevance of public opinion comes from the European Union itself. The European Commission has displayed its concern for public opinion by its efforts to monitor opinions in the member-states. In addition to the long series of biannual Eurobarometer surveys of European public opinion, the Commission recently instituted a new series of periodic "flash" surveys to gauge opinion in the wake of important EU leform initiatives and a new monthly monitoring poll to keep the Commission abreast of public reactions. Citizen Support for Policy Integration 253 support for the process of European integration (Eichenberg and Dalton 1993; Anderson and Kaltenthaler 1995; Gabel and Palmer 1995; Gabel 1998). Scholars have devoted less attention to the specific policy preferences of Europeans and their support for European action in specific policy domains.4 Yet many theories of European integration stress the policy-specific nature of this process. The recent debates over the terms of the Single European Act or the Maastricht agreement reinforce the point that specific policy choices are being debated, and the relative speed of the integration process is a function of the specific policies being discussed. Therefore, our analyses focus on public opinions toward specific policy domains. In what areas do Europeans feel that policymaking should be the responsibility of national governments, and in what areas should this be the responsibility of the European Union? More important, what factors determine these policy preferences? Although opinion research has not studied policy integration in detail, existing integration theory provides a fertile starting point for generating our hypotheses. The following sections review this literature. 1.1. Cross-policy variation The question of how policy integration develops is directly linked to some of the central theoretical questions in integration research. The work of early functional theorists focused on the sequence of broadening policy responsibility for a supranational body that results from initial integration efforts (Haas 1958; Mitrany 1966). Partially because certain problems require it (Mitrany 1966), and partially because of the problem of overcoming national loyalties (Haas 1958), the neofunctionalist strategy begins integration in sectors of scientific, technical, or economic interdependence. Once in place, these initial steps lead to increased interdependence, which may create pressure for further integrative steps, a process labeled "ramification" by Mitrany and "spillover" by Haas. Of course, there has been much debate about the theoretical and political IgE? merit of the functional strategy (e.g. Dougherty and Pfaltzgraff 1990: 433-42). *^ What interests us here, however, are the theoretical implications for public support for policy integration in various domains. If public sentiments follow the neofunctionalist logic, then a direct implication of functionalism is that: -PjiWčacĚepfai^^moliey integration shoulďbe greaterrrtnarfow areas'of scientific or-techmccii integration5 fe For example, European publics may see mutual agreements on occupational *^ safety standards as a technical matter and easily accept EU competency on 4 Sinnott (1995) is one of the few exceptions; he presents European trends in support for policy integration, 5 More precisely, we might expect that potential public resistance to integration will be lower in these areas, enabling elites to use them as an initial basis of integration. Then success in these areas will generate public awareness and support for the integration process. 254 Russell). Dalton and Richard C. Eichenberg this issue. However, issues that directly touch on employment practices or employment security may evoke different popular responses. In short, policy integration starts with "narrow" policy concerns and then broadens to more fundamental policy domains. Functionalist theory generates additional hypotheses about the process of policy integration. The impact of interdependence is crucial for both Mitrany and Haas because it creates two types of effect.6 First, it increases the mutual sensitivity of societies, and thus increases the need for policy coordination. Second, and perhaps more important, it increases the interest of the parties in a relationship of mutual gain. Haas's later work emphasized the successful provision of "welfare" as a factor contributing to integration (1984). This notion is nicely summarized by Nye: "Neofunctionalists prefer a strategy of increasing policy interactions and assume that identities and loyalties will gradually follow interests and expectations in clustering around (and supporting) institutions associated with policy integration" (1971: 44, emphasis ours). Interests, of course, must be seen as the successful provision of gains. This version of neofunctionalist theory thus presumes that existing and potential international cooperation drive the integration process, a theme similar to Stone Sweet and Caporaso's research in this volume (Chapter 4). European action develops in areas where the potential benefits from international cooperation are greatest If the public perceives Europe in these terms, then we might expect that p^hc^pp^^p^^j.ntegrsation should be greater for thosi a$j^fta$«gjí^ level or which have dear potential benefits-frorrnnternationakcoordmaiibn' * In contrast to the neofunctionalist approach, Stanley Hoffmann (1966; also see Moravcsik 1991) argued that the integration process is shaped by national interests. Integration proceeds most swiftly when it does not involve matters of essential national interests. In Hoffmann's terms, integration begins with "low-politics" issues such as the technical or scientific examples of neofunctionalist theory. In these cases, national elites (and the general public) are more willing to grant decision-making authority to a supranational body over which they will have less control. When significant accomplishments have been achieved in these areas, the integration process may move to other policy domains. Hoffmann argued that "high-politics" issues of national security or national identity would be the most resistant to policy integration. Thus, efforts to establish a customs union or a common foreign-aid program might gain broad support, but attempts to develop a single European Army would meet with popular (and elite) resistance. Hoffmann's definition of high and low politics issues was somewhat fuzzy; the former included such issues as national security, 6 This approach is also compatible with Karl Deutsch's work on intersocietal transaction patterns and the development of "community" {Deutsch ft al. 1957: 58). Deutsch saw public values and the integration process as interrelated: as societal transactions and interdependence increased, presumably values would change or. perhaps converge. For Deutsch, then, the development of support for "community" was rooted in structural factors, especially the cross-border flow of goods and other international interactions. Citizen Support for Policy Integration 25 5 - control of the domestic economy, and general rights of national sovereignty. Low politics issues encompass such matters as welfare policies, tariff policies, and other lower priority policies. If we accept this broad distinction, then: public support for policy integration should be greater for "low-politics" issues than for "high-politics" issues. An even more utilitarian argument assumes that the integration process is based on national calculations of specific policy gains from unification. The expansion of the EU's policy authority results from relationships of interdependence and calculations of gain derived from the integration process. Elites may start the ball rolling, but eventually they have to deliver the goods. Prior research generally supports this utilitarian assumption of public opinion. Richard Eichenberg and Russell Dalton (1993) show that Europeans evaluate the policy performance of the EU in two senses. First, gains from intra-European trade are strongly related to overall public support for the European Community. Second, there is now ample evidence that Europeans evaluate the EU on the basis of economic performance, that is, based on such factors as unemployment, economic growth, and inflation (Eichenberg and Dalton 1993; Gabel 1995; Gabe] and Palmer 1995). Further, Matthew Gabel's recent work (1998) demonstrates that Europeans are quite sophisticated in their evaluation of the costs and benefits of integration. Drawing skillfully on the theory of gains from trade in the customs union context, Gabel shows that Europeans seem well aware of the implications of trade liberalization given their particular combination of education and compensation, that is to say, their skills and (relative) wages. Although Europe as a whole has experienced huge gains in prosperity from trade, Gabel shows that individual Europeans are aware that there will be winners and losers in the process, and they adjust their support for integration accordingly This leads to another prediction about public opinion: publicsup-$ port fozpfiJl^,integřatióh\mll'jeflect of European action. 1.2. Cross-national patterns The preceding discussion has focused on possible differences in policy integration across specific policy domains; but, of course, another source of variation involves differences across nations. The debates on the Single European Act .. .clearly demonstrated that nations varied in the priority they attached to various policy domains and their willingness to support European action in specific domains. For instance. Southern Europeans advocated a social charter and greater EU efforts on social policy, while the Danes and the Dutch wanted the EU to assume greater responsibility for environmental policy. Even the specifics of trade and economic policy varied across nations; while some nations strongly endorse monetary union, others are openly skeptical about the idea. In addition to indicating the general evaluation of a policy area, utilitarian theories may explain why nations differ in their support for policy integration. For instance, Matt Gabel's research on the class bases of support for the EU 256 Russell). Dalton and Richard C. Eichenberg suggests that utility calculations may vary across nations (Gabel 1997). However, it is not entirely clear how to translate the general utilitarian logic into hypotheses linking costs and benefits in specific policy areas. We will develop the utilitarian hypotheses in more detail below when we examine the levels of policy integration in specific domains, but the general logic is clear: national differences in support for policy integration may be a function of specific national costs and benefits. Rather than a narrow utilitarian logic, cross-national comparisons of attitudes toward European integration may emphasize the importance of national traditions, culture, and values as an explanation. An early work in this genre was Karl Deutsch's Political Community and the North Atlantic Area (1957). Although Deutsch placed heavy emphasis on the impact of intersocietal communication theory, he also emphasized the "mutual compatibility of major values" as a variable crucial to the integration process The strength of national identity, or nationalistic senüments, has been a major factor m explaining national support for European Union fShepard 1975, Inglehart 1977). Even if phrased in more modest terms, national traditions and experiences with,tlie European Union can create a national response to European integration that fs apparent across specific pohcy domain? For example, Gabel and Palmer (1995) show that World War 11 experiences still condition overall levels of national support for the EU; Eichenberg and Dalton (1993) find that "national traditions" exert a significant impact on support for the EU even when a multitude of other factors are included in a multivariate model. Still, it is unclear whether such broad national characteristics hold promise for explaining national levels of policy integration across domains. While this literature suggests which factors should be associated with the decline of national identity and support for the general process of European integration, support for integration of specific policy areas actually varies greatly across societies (as we will see below), highlighting the question of what factors differentiate policy areas from one another. Nonetheless, this research does provide a possible cross-national hypothesis: public support for policy integration ßhouldbg gzeatet\amongthqse societies with thehighest level ofsupport for the general profect of European integratipn. By examining the levels of public support for policy integration in several policy domains and across the member-states of the EU, we can both assess the prospects for further expansion of the EU's policy authority, and use this evidence to test prior theorizing about the nature of the integration process. 2. Patterns of Support for Policy Integration We are primarily interested in explaining public support for policy integration in areas that were involved in the SEA/Maastricht agreements or that may represent the next steps of the integration process. On the one hand, public Citizen Support for Policy Integration 257 acceptance of EU action in specific policy domains marks areas that are most susceptible to further integration, or at least domains where the "permissive consensus" is most developed. On the other hand, by understanding what influences the public to shift their preferences along the national-supranationaj dimension, we can better understand the theoretical basis for policy integration. Our empirical base is the rich series of Eurobarometer surveys that the Commission of the European Community has conducted (Reif and Inglehart 1991). Since the 1970s the EU has asked Europeans about their willingness to see various policy areas handled at the national level or at the European level (Daltoa 1978). Then, beginning in the mid-1980s the EU began a new series using a different survey question: Some people believe that certain areas of policy should be decided by the (National) government, while other areas of policy should be decided jointly within the European Community. Which of the following areas of policy do you think should be decided by the (National) government, and which should be decided jointly within the European Community? Our analyses focus on the percentage of the public who respond that particular policy areas should be "decided jointly" within the EU. 2.1. Variation across policy issues Table 9.1 shows the percentage support for policy integration for the fullest set of items over the 1989-97 period. We present the overall responses for the Europe of the twelve (EU12), that is, for those twelve states who were members of the Community during the full period of these surveys.7 (Statistics for 1985-9 are shown in italics because question wording diverged from subsequent surveys and thus are not fully comparable.) Focusing first on the overall level of support for policy integration, the Community finds itself in a "half empty/ half full" situation. Across all policy areas, support for policy integration is about 50 percent. In more recent surveys, which include additional policy areas, support has occasionally dropped to below that level, standing at 49 percent in 1993 and 1994. However, these data ■ also show that support varies widely across issue areas. Taking the extremes, - only 35 percent of Europeans favored integrating "personal data" policy (in 1993), but almost 80 percent have at times favored integration in "cooperation with developing countries." Fig. 9.1 ranks policy domains according to the overall level of support for EU decision-making. An "above average group" contains six policy areas: developing countries, scientific research, foreign policy, environmental protection, 7 Percentages are based on the Community-wide sample, weighted to reflect national populations 15 years of age and over. 258 Russell ]. Dalton and Richard C. Eichenberg ai a u BS o -=. o c J= 5 J ä III! •S * * Ill c ~ o "a _9 llílllf E O ■*~* .*-o. .y o 11 Ť Ä «.Ü « - Ľ .2 Q- E 3 Ol OJ Q. |l 3 oi Iis-läi c E Citizen Support for Policy Integration 259 Protection of personal data Education Health and social security Codeterminatioii/worker participation Cultural policy Basic rules for broadcasting (radio/tv) Worker health and safety Rates of value added tax industrial policy Security and defense Fight unemployment Currency Political asylum regulations Immigration policy Protection of the environment Foreign policy toward non-EC countries Scientific and technological research Cooperation with developing countries □ Avg: 1989-1991 O Avg: 1992-1995 l^a^iaiaBfa^aDlSfefc^-i sáW?^'ťľ asa«!Sg.y-Jg:«.] a&á^^Mfej-vĚ-^iir I ^^iSÍ^^/^f^^íffĚ^i^éSJÍÍiS&^iš^rííí ■'■'W-J;! ^^.^.^^^■■y.:,. i-w^i^3^ďšá'T^S-g^''»',y* Wlř='' ftavIS^SPBiEyTSIr-' ■' [v^iifai^j^Sofea^aaj iSft*fré!ňičiisw%Ŕsi-A ^S^T 25 -J— 45 35 45 55 65 75 % 5upport for policy integration Fíg. 9.1. Support for policy integration (EU12) asylum regulations, and immigration policy. Where comparisons to earlier periods are possible (Dalton 1978), there was broad support for European action on foreign policy and environmental policy long before the initiatives of the SEA and Maastricht. This is a fairly diverse set of policy areas, but they are characterized by at least one of the following traits: they.deäl with the external.•'■envii&^irfént ^foreign policy, developing countries, immigration issues);,with classic interdependence issues^environment, immigration); and 260 Russell J. Dalton and Richard C- Eichenberg thěydo not have substantial immediäte implications for individual standards of The "average" group basically contains two sorts of policies: defense and ecd nQmic management- The ranking of defense is understandable in terms of the "high-low" politics distinction discussed earlier—defense is, after all, the basic function of protecting national territory—but it is significant that this aspect of external relations is apparently considered separate from the broader task of foreign policy, which is listed in the first group. The remaining items in this "average" group essentially deal with economic management and market management: support for a common currency, industrial policy, VAT rates, and unemployment policy.' Similarly, Dalton (1978) found that in the 1970s there was broad public acceptance of EU competence on economic issues of fighting employment and inflation. Thus, the broad economic role of the EU is generally accepted by its citizens, even if monetary and employment policies still pnmanly remain the responsibility of national governments Finally, the, ^^»^PBS^lSS^aPRÍS^ftffÄPt the European Union policy* makmg'responsibihty on two types of policy issues individual ngrfts and stand-^ ards of lining (worker health; codetermination; social security; education), and policies dealing with national culture—what we earlier referred to as "identity issues" (education, privacy, cultural policy). Similar patterns have been observed for the 1970s (Dalton 1978). Do these statistics speak to the theoretical issues raised above? In both a positive and negative sense, they clearly do. First, there does not appear to be a "high-low" politics distinction in judging policy integration. To the extent that "high" politics meant pursuing the national interest in relation to the external environment, these surveys suggest that Europeans have long adopted the view that national interests are best pursued in the European context. National defense remains somewhat problematic in this regard,10 but there is broad support for a joint EU position on issues of Europe's foreign relations with the world. In addition, Dalton's (1978) early study of policy integration found that Europeans had generally accepted EU competence on a variety of foreign policy issues (relations with the superpowers, control of multinationals, and defense). European elites may once have emphasized the difficulties of coordinating foreign and defense policies among the European states (Rabier and Inglehart 1981), but these views have apparently changed (European Commission 1996). European publics see this as a natural area of joint action. a Immigration perhaps deserves further discussion. We treat it as an interdependence issue, because the experience of the 1990s suggested to Europeans that it was to some extent a collective problem that no state could solve alone. Of course, immigration may have budgetary and welfare (employment) implications, although its ranking here suggests that Europeans see it as a problem concerning the external environment. 9 VAT taxation might be considered an exception, but in the EU context it has long been discussed as an aspect of market liberalization rather than fiscal policy. 10 e.g. EiiTObaTOmeter 44 (spring 1996) found that 44% of the citizens in the twelve favored national action on defense issues, and 52% favored EU action; thus a majority do support a European alternative. Furthermore, a 1996 survey of European elites found very high levels of support for European action on defense and foreign policy in general (CEC 1996). Citizen Support for Poiicy Integration 261 "High-politics" issues no longer retain a special status among Europeans, if they ever did. In the early stages of the Community, overcoming traditional national sovereignty was clearly near the top of the psychological and political agenda for elites. As time has passed, however, it became clear that separate national positions on many international issues were less likely to succeed than a joint position. Beginning in the mid-1970s, moreover, a number of issues convinced European leaders—and apparently their constituents—that European interests were being neglected (principally by the senior alliance partner). Serious European efforts to construct joint positions began. Beginning most earnestly with the "Nixon shocks" of 1971, there ensued disharmony on the Middle East, energy policy, defense spending, nuclear weapons in Europe, the Soviets in Afghanistan, and the Iranian hostage crisis. Throughout this period, Europe attempted to increase its room for maneuver by laying the groundwork for a common foreign policy (Ginsberg 1989). Interestingly, when the Community first began surveying citizens on policy integration, it phrased the "foreign policy" item in the following terms: "make [our] presence felt in discussions with the superpowers." In short, as early as the 1970s, for Europeans, "high politics" meant aggregating their influence to balance the superpowers, and that could only be done on an EU level (Eichenberg 1997). If the "high-low" politics distinction seems to have outworn whatever usefulness it once had, other aspects of integration theory find more support in these data. Clearly, there is something to the neofunctionalist logic and strategy. Problems of interdependence and technical complexity attract high support, either because the logic of interdependence seems to demand it (environment, immigration) or because the issue is distant from the citizen's concerns (scientific research). Similarly, past research has shown that market liberalization conditioned overall support for the Community, presumably because of the prosperity to be gained from trade. We see moderately high support for the management of that market (currency regulations, VAT).11 Finally, support for policy integration decreases when the issue turns from the management of the market to two sets of issues: the definition of the national culture (education, cultural policy, privacy) and the distribution of national welfare, including questions of income distribution (social security) and workplace rights (codetermination, health and safety). Education spans both categories; and European elites also see education as primarily a national policy domain (CEC 1996). These latter patterns confirm the enduring utility of the early emphasis on national values, culture, and identity, as well as Dalton's (1978) emphasis on the personal salience of public policy as a factor conditioning support for policy integration. In summary, public support for policy integration suggests that the EU faces three distinct sets of political problems as it moves beyond the market emphasis of the SEA to the policy emphases of Maastricht. There is a broad basis of 1' In 1996 European elites rated currency as the top policy domain for European action) CEC 1996). This is an apparent change from earlier elite opinions (Rabier and Inglehart 1981). 262 Russell j. Dalton and Richard C. Eichenberg support for European action on issues of interdependence and relations with the outside world. There is moderate support, although not overwhelming, for the policy mechanisms that "regulate" the internal market. However, there is little support—indeed there is outright resistance—to any interference in the national state's traditional role as the final arbiter on questions of the national culture and the distribution of national welfare. 2.2. Variations across nations Citizen support for policy integration varies among the EU member-states as well as across policy domains. For example, generalized sentiments toward European integration may extend across policies; nations that favor policy integration in one sector may also favor integration in others. Conversely, there are strong reasons to expect substantial national differences in the patterns of policy integration. Some nations may see benefits from integration in some areas, but neutral or negative consequences from integration in other sectors. Indeed, the process of integration has often involved balancing these contrasting national priorities to strike a European-wide bargain. Table 9.2 presents the national patterns in the percentage preferring Community action in each of the eleven core policy areas (also see Dalton and Eichenberg 1993). Comparing across nations, the greatest net support for European-level action is found among the six founding nations, with most policy areas showing a plurality of support for EU action. It is noteworthy that the Germans fall noticeably below the other founding states. Germans are less enthusiastic than their neighbors about the seemingly consensual issues for European action—science and technology, foreign aid, and environmental protection—and are hesitant to yield national autonomy on matters such as currency reform and VAT rates. While Germans and their government voice strong support for the European ideal, they are now more cautious in actually transferring authority to Brussels. A hesitancy toward policy integration is displayed among all six "new" states. The Danes and British express less support than average for policy integration in nineteen of the twenty-two areas presented in the table. Support for European action is equally restrained in Greece, Spain, and Portugal. Taken in aggregate, support for policy integration across a range of policy areas appears to be an extension of overall support for the European project. Indeed, it would be surprising if this did not occur, since those publics which endorse the EU overall should generally endorse greater policy responsibility for the Union. However, these aggregate patterns often coexist with substantial and different patterns at the national level. Such variations may be a function of national values or utilitarian calculations. 264 Russell J. Dalton and Richard C. Eichenberg 3. Case Studies of Policy Integration In our understanding the construction of the European project comes from developing public support where it is lacking, and building compromises between the differing policy priorities of societal actors and their national governments. In some areas the public may accept European action, facilitating policy integration in these areas. For instance, the public in the six EC founder states lean toward EC decision-making on foreign policy and security issues, and this support has increased in the post-Cold War period. The publics in the remaining nations prefer national policymaking on security and defense issues.12 In other areas, the opinions of the public lag behind policymakers; in several member-states a plurality of the public still favors national action on monetary and currency union. This section explores support for policy integration in three specific sectors: environmental policy, security and defense policy, and monetary union. These focused analyses will enable us to examine how national preferences differ and how these differences vary across time. Our goal is to determine how domestic policy preferences are formed and how they interact with policy change at the EU level. 3.1. Environmental policy Our first case study deals with public support for EU or national decisionmaking on environmental policy. We selected this area for several reasons. Environmental policy is one of the most visible new issues of European politics, and thus enables us to examine the formation of preferences in a new area that has not previously been integrated into the policymaking process of either the EU or national governments. Environmental policy also represents an area where the international dimension of the problem is obvious, but where industry and consumers may bear real economic costs through stricter environmental policy. Thus it is an area that contains a diverse mix of factors that might encourage or discourage a shift toward supranational policymaking. We also know that environmental policy is an area where EU responsibility has changed as a result of the SEA and Maastricht accords—thus we can track the interaction of public opinion and policy change. Environmental policy has a long, albeit sometimes surreptitious, policy history within the European Union (Sbragia 1993«; Vogel 1993). The European Community formed before environmental protection was a salient political issue. Thus, the Treaty of Rome makes no mention of environmental policy, 13 However, when given the choice between NATO or the EU as a forum for international security, the public in Mediterranean Europe favors the EU." Citizen Support for Policy Integration 265 and the early Community possessed no formal authority on environmental matters. Environmental issues began to gain the attention of policymakers in the early 1970s, as public pressure and international meetings brought this issue onto the political agenda. At the 1972 Paris Summit the heads of government called for the EC to develop an environmental program. The first EC environmental action plan followed in 1973; it assessed the state of the European environment and suggested a series of potential policy reforms.. A second action plan was published in 1977 and a third in 1983. In 1981 the Community formed DG XI, which was responsible for environmental issues. During this period the issue received some attention from the EC institutions. However, economic recessions in the 1970s and 1980s focused EC attention on economic issues. In addition, the EC lacked formal legal authority to act on environmental matters, except through consensual regulations and directives. The Community issued a number of directives on environmental issues over this time, but these were fairly modest proposals because of the unanimity requirement. As late as the 1980s, therefore, environmental policy remained primarily a policy domain of national governments. Although the environment remained off the EU's formal list of responsibilities, there was mounting public and interest group pressure during the 1980s for the EU to become more active on environmental matters. European-wide environmental problems, such as acid rain or periodic environmental crises, reminded Europeans that environmental problems spread across national borders. This was also a period of heightened public interest and activism on environmental issues (Dalton 1994). The public and environmental groups pressed Brussels to become involved in environmental policymaking, a role that was encouraged by DG XL The Single European Act transformed the EU's role on environmental issues. A revision of Art. 100 gave the Commission the legal right to set environmental policy. Additional changes to Art. 130 set the standards for environmental protection and mandated that environmental protection requirements be considered in other Community policy areas. Beyond the legal standing the EU gained by these provisions, these reforms also changed how environmental policy would be decided in the future. Instead of unanimous consent, as was the rule prior to 1987, Community action could be taken by a qualified majority when market-related issues were involved. The Maastricht Accords then extended the EU's environmental areas subject to a qualified majority and increased the European Parliament's role on environmental matters. Thus a single polluter nation no longer can block environmental protection policies at the EU level. A series of decisions by the European Court of Justice also expanded the EU's ability to promote strict environmental regulations (Vogel 1993). By the 1990s, the EU had become a significant arena for the making of environmental policy in Europe. Much if not most of the policy responsibility still lies with the national governments. However, the EU is increasingly 266 Russell ]. Dalton and Richard C. Eichenberg involved in setting environmental Standards, arbitrating between conflicts involving economic interests and environmental protection, and establishing European-wide environmental policies. This is an area where there has been a considerable shift horn intergovernmental decision-making in the 1970s to supranational decision-making in the 1990s. Thus, it is interesting to examine the relationship between public preferences and the activities of the Community. Did the public follow elite cues, as the intergovemmentalists would suggest, or is there evidence that public opinion existed separately from the EU's own actions? Because the Eurobarometers have changed their question wordings over time, we cannot exactly track the trends in public. opinion over the past three decades. Yet the evidence from different questions presents a single picture. When questions on environmental policy were first asked in the 1970s, there was already surprisingly high public interest in these issues. In the 1973 European Community Survey, a large majority in each member-state (except Ireland) already felt that environmental protection was an important or very important issue. Subsequent surveys showed that public interest in environmental matters remained widespread over the next decade, despite the economic recessions Europe experienced at the same time (Dalton 1994: ch. 3). In the late 1980s and early 1990s, public interest increased even further. The first EU public opinion surveys on this issue also found that a clear majority of Europeans favored EU action on environmental issues in the early 1970s. As early as 1974 and 1975, two-thirds of the European public favored Community-level decision-making on policies to fight pollution and protect the environment. Even after the economic recessions of the next several years, surveys in 1983 and 1984 found even higher levels of support for European action.1* Table 9.3 summarizes public support for EU decision-making on environmental matters over the last decade. Two findings are apparent from these data and earlier research. First, public support for European action preceded EU responsibility for this policy area. Since the early 1970s Europeans have generally favored EU policymaking, recognizing the international dimension of this issue and the need for common action on setting environmental standards. The European public created a "permissive consensus" for action on this issue —a consensus that took the "EU more than a decade to translate into formal authority. Furthermore, the development of formal EU authority on this issue has apparently little impact on public sentiments. Second, national differences in support for European action are fairly consistent and fairly stable over time. Southern Europeans, such as the Greeks, Spaniards, and Portuguese, are least supportive of EU action on environmental !3 Sinnott (1995) notes that the wording changes in Eurobarometer 6 and 10 appeared to lessen support for European action on all issues, and attributes this to a methodological artifact of the question wording. For this reason we will not discuss these two surveys. Citizen Support for Policy Integration fc 0\ (N Ol ÍN 00 CI ' ^ ifl V "O CO N- r^ í^ ^j vo N f in •& rn T !Sj v N IN ä 1 11 c í! íl Z o r; Mflfl 268 Russell). Dalton and Richard C. Eichenberg matters. The Irish also display significant reservations. In contrast, support is greatest in the Netherlands, which has a very active environmental movement and some of the strongest national environmental regulations in Europe. This cross-national pattern tends to undercut a narrow self-interest explanation of national opinions. If we assume that EU standards would reflect an "average" of present national environmental policies, then EU policymaking would raise environmental standards for most Southern European nations and might produce lower environmental standards in nations with strict legislation. However, support for EU action is greatest in nations with already strong environmental legislation, and weak in nations with less developed environmental protections {Spain, Greece, Portugal, and Ireland). Rather than national self-interest, it appears that support for EU action on environmental matters minors general public support for action on this policy. Where publics favor strong environmental regulation, as in the Netherlands, the expansion of EU policy responsibilities can increase environmental standards across the continent. Where public support for environmental reform is weak, as in Southern Europe, then an expansion of government activism (even by a European government) is not supported. In other words, in this area policy integration at the EU level appears to be an extension of national policy activism by other means. 3.2. A common foreign and security policy There is something of a paradox in the fact that Europe has found it most difficult to achieve progress on the integration of foreign and defense policy. To be sure, it is precisely in these areas that "high-politics" questions of national interest, national tradition, sovereignty, and territorial security are raised. Nonetheless, it is also the case that European unity and European institutions were conceived to serve the pursuit of a broad range of external political and security interests. Most obviously, since 1957 Europe has been a customs union, and the institutions of the Community have become thoroughly engaged in the diplomatic representation of foreign trade and other economic interests abroad. Second, the very creation of the European Community was strongly motivated by foreign and security policy considerations. With the failure of the European Defense Community in 1954, the creation of integrated European institutions offered the only option that would bridge the American-supported desire to rearm West Germany and the European (especially French) fears that this step understandably provoked. Like American participation in NATO, the new institutions of European unity served the dual purpose of deterring the Soviets through unity, while simultaneously integrating and containing the newly rehabilitated West Germany. A united Europe was seen as a balance—or lever—in an external environment characterized by bipolar dominance. As DePorte has put it (1986: 222-3): Citizen Support for Policy Integration 269 Finally, on the international level, [the Schuman plan] promised not only to strengthen Western Europe in face of the Russian threat but also—though this was less talked about—to strengthen it vis-á-vis its indispensable but overpowering American ally. These Frenchmen, like many West Europeans, did not doubt that they had to rely on American strength for their security. But they did not want to rely entirely on the Americans to manage relations with the East which involved no less than war and peace, that is, the lives of the European nations. At the least a more united Europe could better influence American decisions affecting its vital interests; at the best it could break through th« rigidity and risks of bipolar Europe by becoming strong enough to cease to be a stake of the superpowers in cold of hot war. However, this latter purpose of European foreign policy soon fell victim to the Cold War context of a Europe allied to the USA. Increasingly, foreign and security policy were defined as a stark choice between the Atlantic and European options, and the latter could not be pursued without risking the former. Although European integration was originally a child of the Cold War, in the ensuing years integration of foreign policy could not be seen as competing with the Atlantic connection. The very real threats and tensions of the Cold War (not to mention pressure from Americans) helps to explain why the choice was framed this way, but perhaps it also resulted from the fact that the most energetic proponent of a European voice was General de Gaulle, who complicated matters even further by making it clear that Europe should be led by France, and who in any case was a strong critic of supranational solutions (Grosser 1982; DePorte 1986). Yet the French position always contained a contradiction. Given the Soviet Union's geographic advantage, its (perceived) conventional dominance, and its large nuclear arsenal, European security required a nuclear deterrent. However, the first and most emphatic principle of French security policy under DeGauIíe was the insistence on independence, most particularly in the nuclear realm. In the absence of French willingness to share its deterrent, let alone integrate its defense forces, the remaining European states chose to privilege the Atlantic connection, a result that certainly slowed the evolution of common European positions in foreign and security policy. There matters rested until the end of the 1960s, when the European challenge to US foreign policy began to grow. Beginning with the Vietnam War and the negative economic consequences for Europeans, there ensued a number of transatlantic quarrels, especially on policy in the Middle East, energy, nuclear weapons, and the coordination of economic policy. Not surprisingly, therefore, the "launching" of formal efforts to institutionalize European cooperation in the foreign policy field began in earnest in the early 1970s in the form of European Political Cooperation (EPC), a form of intergovernmental coordination of foreign policy that nonetheless produced some important European positions on foreign policy, most notably the 1980 Venice declaration on the rights of the Palestinians. Perhaps equally important, the institutionalization of EPC led to biannual meetings of Foreign Ministers to discuss foreign policy specifically, and it also created a dense web of administrative links among Foreign Ministries 270 Russelt J. Dalton and Richard C. Eichenberg and between Brussels and the ministries (Ginsberg 1989; Wood and Yesilada 1996; Smith, Chapter 11). Foreign policy received little attention in the Single European Act because it was largely a market reform; but the Maastricht Treaty made official what Europeans had long aspired to: that the European Union "shall define and implement a common foreign and security policy" (see Chapter 11). The provisions of the Treaty on Common Foreign and Security policy are quite complex, but substantively they include the aspiration to an integrated defense and security policy in addition to the actual adoption of a common foreign policy, as for example in Art. J.4: "the common foreign and security policy shall include all questions related to the security of the Union, including the eventual framing of a common defense policy, which might in time lead to a common defense." The Treaty also commits the member-states to close consultation, collaborative actions in international bodies and conferences, and it specifies the West European Union as the "defense arm" of the Union. However, the most important aspect of the Treaty's foreign policy provisions is that intergovernmental rather than supranational decision-making is specified. In both areas, the Council must first decide by unanimous vote that "joint action" should be undertaken; subsequent decisions in the pursuit of these actions are to be taken by qualified majority vote. In summary, the language and indeed the symbolism of the Treaty pushes the Union quite a bit forward in the integration of foreign policy, but supranational integration of authority remains rather limited. Does this situation reflect the views of European citizens? To the extent that the Treaty continues the European tradition of ambivalence in the integration of foreign policy, it closely parallels citizen views. In the 1970s and 1980s, for example, responses to Eurobarometer surveys as well as US government polls showed that Europeans supported common European foreign policies. For example, in 1974 and 1975, almost 70 percent agreed that the Community should decide jointly to "make [our] presence felt in discussions with the US and the USSR," and in 1976, 55 percent of Europeans agreed that the EC should decide jointly to "defend [our] interests against the superpowers" (Dalton 1978). In addition, there was an increase from the 1970s to the 1980s in the percentage of Europeans who wanted Europe to conduct security policy "together" rather than "separately" (Eichenberg 1989: 14). However, throughout this same period, a substantial plurality of Europeans continued to believe that European security was best pursued within the NATO Alliance when survey questions provided that alternative (Eichenberg 1989:118-58). Thus, although Europeans responded favorably to the idea of a common foreign or security policy;'when phrased as an alternative to NATO, they turned distinctly less favorable. The same is true of more recent polls. Table 9.4 summarizes the Eurobaro-meter policy integration questions on foreign and defense policy. It includes three separate questions on EU-level decision-making: cooperation with developing countries; foreign policy toward non-EC countries; and security and Citizen Support for Policy Integration 271 i- *o o\ o ■— rvm\co\coTo iNmtNOmriiiiimmoji'i ul M N Ol T T U"] T T vOCO-q-TTOi— r-VO I í ! lllJ-šälo Ep> ~' -i- ir, x -. P P 3 E c -° S = c F >, c b E S " 5 S u ,«; d — u- CO O j t 272 Russell). Dalton and Richard C. Eichenberg r— O ■— [N0v0^r^>r^^0 u-1 vo tfi un >r xr T o•-u^'-,*, ^ S E £ ■£ q: d q o m s