Explaining Attitudes about European Integration

2 Explaining Attitudes about European Integration

2.1 Instrumental self-interest

Far beyond the other approaches, the economic considerations of citizens of the EU have been the most thoroughly examined. In Western Europe, attitudes regarding EU membership have been explained by personal economic situations (Gabel 1998a ,b ) and individuals’ perceptions of their own economic well-being and national economic performance (Eichenberg and Dalton 1993 , 2007 ; see also Gabel and Whitten 1997 ). Similarly, the broader classification of citizens into integration’s economic ‘winners’ and ‘losers’ has been argued to affect EU support in the West (including education and occupational groups, see Gabel 1998a ). Essentially, these explanations rest on the notion that support for EU membership comes from the implicit cost/benefit analysis of individuals’ likely economic benefit to be gained from integration.

Given the EU’s origin as an economic organization intending to bring economic efficiency and affluence among European states, support for the continuation of this project has often been understood in economic terms (Eichenberg and Dalton 1993 ; Anderson and Kaltenthaler 1996 ; Duch and Taylor 1997 ). However, which economic factors are the most relevant continued to be a matter of debate. Early work tended to cite national economic performance in the form of rates of growth, inflation and unemployment (Anderson and Kaltenthaler 1996 ), country net benefits from EU (Eichenberg and Dalton 1993 ; Anderson and Reichert 1996 ; Carrubba 1997 ; Gabel 1998a ), and more general ‘economic perceptions’ (Eichenberg and Dalton 1993 ) as key economic variables that influenced citizens’ perceptions about the EU.

Throughout the 1990’s, scholars attempted to discern the precise source of this evaluative proxy mechanism. Appealing to the outright ‘costs of non-Europe’, many found favorable national-level economic evaluations to be the source (Anderson and Kaltenthaler 1996 ; Eichenberg and Dalton 1993 ) and others addressed regional and sectoral economic variation (Smith and Wanke 1993 ).

A distinct facet of the sociotropic approach is the assessment of being a net beneficiary of the net transfers from the EU to the nation (Anderson and Reichert 1996 ; Gabel 1998a ; Carrubba 1997 , 2001 ; Gabel and Palmer 1995 ; Smith and Wanke 1993 ). While economic in nature, the underlying assumption is that further EU expansion implies continued market liberalization. As such, citizens of EU member states are able to capitalize on the availability of human and national-level resources according to their own socio-economic profiles. However, what became increasingly clear was perceptions about one’s own economic situation was often more powerful than ‘objective’ measures of occupation, class, or income.

Therefore, these sociotropic approaches gradually gave way to a more specific process of egocentric utilitarianism. Born out of the economic voting literature (Lewis-Beck 1988 ), this individual-level approach tried to explain why objective economic predictors (GDP, inflation rates) only weakly related to attitudes about the EU. For example, Gabel and Palmer (1995 ) examined how different economic benefits of integrative policy relate to individual-level differences in public support for integration. Not only did they find support for this economic explanation but also coherent and accompanying socio-economic status (SES) and social location effects in the form of education, age, income, and occupation (see below). In another effort, Gabel (1998b ) refuted the sociotropic economic argument of Eichenberg and Dalton (1993 ) demonstrating that declining unemployment and inflation, and rising GDP are associated with less support for integration. Furthermore, when Gabel and Whitten (1997 ) included both objective national-level indicators and subjective individual-level indicators, they found the latter to be a stronger predictor of EU support.

This fairly simple economic model of EU support was a precursor to the broader approach of understanding individuals’ cost/benefit analysis. More recent analyses now include explicit theorizing about these socio-economic perceptions but combine them with social location variables that lead to a more complex assessment of how economic assessments affect EU attitudes. The utilitarian cost-benefit approach states that as material gains within a country increase – particularly through the liberalization of trade within the EU – support for the EU will increase. Importantly, in contrast to simple macro-level economic indicators, this was particularly true of those who are positioned to take advantage or further integration, distinguished by specific SES and social location variables (Gabel and Palmer 1995 ; Gabel and Whitten 1997 ; Gabel 1998a ,b ). In short, this argument rests on the assumption that individuals in different socio-economic locations experience integration differently, some as winners and some as losers (Gabel 1998a ).⁸

This instrumental approach approximates individuals’ cost/benefit analysis associated with integration and in addition to advantageous economic positions, education, occupational skills, and proximity to borders impact individuals propensity to support continued EU expansion (Anderson and Reichert 1996 ; Gabel and Palmer 1995 ).⁹ Conceptualizing the EU as an international economic policy, Gabel (1998a ) used economic factors that explain support for the EU but in addition to shoring up further support for the utilitarian approach, he delineates sources of support across occupational differentiation in the support for EU membership. Higher income earners benefit from continued integration as it creates increased investment opportunities while lower income earners are subject to diminishing welfare brought about by increased capital liberalization. Correlated with this are individuals’ sectoral location and occupational differentiation, for example, younger Europeans are also more likely to benefit from continued integration as they are more cosmopolitan, mobile, and flexible (see also Inglehart and Rabier 1978 ). The approach tapped the distributional consequences of economic integration for individuals and demonstrates that both winners and losers can be identified and can be shown to differ in their support for the EU project (Gabel and Palmer 1995 ; Gabel 1998a ,b ); this explains not only differentiated support but also an individual-level process of ‘calculating’ the personal impact of further integration.

2.2 Social Location

In addition to the utilitarian approaches above, others have demonstrated at least perceptible differences across individual attribute variables. In its earliest form, post-material theory suggested that individuals’ incremental political sophistication created a value system of appreciation for the democratic organization (Inglehart 1970 ; Inglehart and Rabier 1978 ; Janssen 1991 ) and eventually extended to support for the EU (Inglehart 1990 , 1997 ). In the post-materialist language, this development was accorded to a more highly refined rejection of materialist considerations and a resultant capacity for the abstract consideration of the EU project. However, the process of sophistication was predicated on the notion that individuals’ cognitive mobilization increased political awareness which lessened the ‘threat’ of integration. Studies in this approach operationalized cognitive mobilization with high levels of political awareness and skills in political communication and were based on assumptions that higher levels of cognitive skills are necessary but needed to understand the highly abstract nature of the EU project and that more information about integration would cultivate support for the EU project. While post-materialism as a determinant was ultimately demonstrated to be at best a weak relationship (Anderson and Reichert 1996 ; Gabel 1998a ), the underlying implications of cognitive mobilization, increases in education and political interest and the ability to receive and manage incoming and increasingly accessible information, have resonated in a broader approach. As citizens developed more cosmopolitan outlooks, their apprehension to the EU project fell accordingly and has been supported by empirical evidence of the positive correlation between political involvement and knowledge and support for the EU integration (Janssen 1991 ).

This is but a part of the social basis from which political skills develop. Age, income, occupation, and political values are not merely controls in this analysis but rather contribute to individuals’ cognitive development and thus understanding of the EU project (Inglehart, Rabier, and Reif 1991 ). Education, age, gender, and the socio-economic status of individuals have consistently been found to be salient contributors to individuals’ support for the EU (Weßels 1995 ). Gabel (1998a ) pointed to several socio-demographic characteristics and political ideological preferences as sources of support for the EU, particularly center (vs. strong left/right) ideological positions (see also Anderson and Reichert 1996 ), that were meaningful in the instrumentalist approach discussed above.

However, beyond socio-economic status variables as components of a singular economic consideration of support, some have explicitly theorized these SES variables as separate processes. In particular, one study suggests a modest gender gap exists with women being less enthusiasm about EU integration (Nelson and Guth 2000 ). More significantly, Nelson, Guth, and Fraser (2001 ) have pursued religion as a determinant of support for the EU. While they find that Catholics are far stronger supporters of integration than Orthodox, Protestant, and ultimately the least supportive atheists and agnostics (see also Nelson and Guth 2003 ), their approach steers clear of the more provocative role of religion in the EU, namely the increasing Muslim population and conflict that has accompanied the more insular diasporas in the Netherlands, France, Germany, and Denmark (as recent examples), as well as the on-going debate over Turkey’s admittance.

The shortcoming of socio-political or socio-economic approaches is that while we can assign individual-level attributes to varying levels of support in the national contexts, this ignores the supra-national character of European integration. Although, this fixation on a materialist explanation in the 1990’s eclipsed the rising number of ‘new politics’ issues, the economic approach makes the most intuitive sense as the greatest impacts from EU policies have been economic or directly affect economic considerations. However, this utilitarian approach is limited to the output based conception of representation, that is, limited to the explicit capacity of the EU to deliver the goods. While this instrumentalist approach has explained a great deal of support into the mid-1990’s, the EU has continued its expansion into non-economic policy areas (Franklin and Wlezien 1997 ) with consequences for how public support for integration is analyzed.

2.3 National vs. European Identities

One result is that national identities have increasingly become the focus of EU analyses. The classical conception of ‘Europe’ forces upon its members a question of what it means to be European. Member states such as the Netherlands, Germany, and Denmark have begun to confront these issues through specific policies of immigration and migration, national policies that are likely to become more prominent in EU discussions and politics at large. Therefore, these crises may have more to do with the Europeans themselves than the EU. Again returning to the legitimacy of the EU, it is the seeming absence of a European demos that limits citizens of thinking of themselves as Europeans, or as Schmitt and Thomassen put it, represented are “the people of Europe” not the “European People” (1999 : 256).

If legitimacy refers in part to the belief that the existing political order is right, then popular support for the EU is a matter of value congruency; that is, an affective recognition of the EU as necessary and representative of the collective will of the peoples of Europe. This of course is premised on the actual existence of a ‘collective will’ of a European demos, placing a heavy demand on the citizens of Europe to identify themselves as the legitimate origin of the EU and to define its scope and function (Cederman 2001 ). However, the difficulty of the demos question lies in addressing the continuing ambiguousness of representation at the EU level: to whom are EU decision makers accountable?

Some argue that the institutional uniqueness of the EU as a supra-national body, not a national one, weakens the congruence between the representatives and their constituents by its distance not only to the constituents themselves but in the daily affairs of individuals. Yet, over the past two decades, the EU has not only expanded in the number of members but also in breadth of its involvement at national-level and European publics responded to what they accurately perceived as an increase in both the number of policy areas for which the EU is partially or completely responsible (Schmitt 2005 : 654) as well as a substantial increase in the volume of legislation that the EU now produces (Franklin and Wlezien 1997 ). Not only have Europeans noticed the increase but have – in general – responded negatively to it. At the same time and sharing the same origin, the EU has been seen to be encroaching on state sovereignty.

Although attempts have been made to address this (the introduction of the three Pillars¹⁰ and the Principle of Subsidiarity), the expansion of the role of the EU in national affairs (e.g. through harmonization and EU enlargement) has catalyzed popular debate as to its necessity in doing so. Originally designed to aid the progress in areas related to implementing the complete economic package of the Single Market and accompanying the Single European Act, the Council of Ministers exchanged unanimity for majority voting. In the late 1980’s and early 1990’s, this undermined the ability of national parliaments to examine the actions taken by other countries’ representatives (i.e. national ministers acting in the Council of Ministers) and was perceived as a forfeiture of an uncomfortable degree of national sovereignty, particularly as there was no real trade off in parliamentary power.

Explicitly addressing the contest between the EU and states, the “subsidiarity principle” guides EU decisions to be taken as closely as possible to the citizen, that the EU does not take action (except on matters for which it alone is responsible) unless EU action is more effective than action taken at national, regional or local level.¹¹ Some have argued that it is through this principle that the EU has reinvigorated national sovereignty by highlighting national-level competencies and economic abilities (Moravcsik 1993 ); however, it is the subsidiarity dispute that has turned the scope of EU government into a salient issue in the larger debate on the support for the EU (de Winter and Swyngedouw 1999 : 47). Set against this backdrop, the term ‘European identity’ is not merely publics’ general and ambiguous feeling about ‘Europe’ but, more importantly, a constellation of attitudes regarding the role and nature of the EU and the strength of their attachments to state sovereignty.

This question has been approached in two ways.¹² The first includes a version of national identity that is important to individuals’ choice to support or approve the EU. This more threatening form of national identification resonates from perceived cultural threats and exhibits hostility toward other cultures; that is, feelings of national attachment and perceptions emerge out of threats to the nation state (Kritzinger 2003 ). Carey (2002 ) has demonstrated that not only is the intensity of feeling toward one’s own country or the level of attachment to the nation important but so is the fear of other identities and cultures (see also Deflem and Pampel 1996 )¹³ or more specifically a fear of immigration (de Vreese and Boomgaarden 2005 ). Hooghe and Marks examine the role of ethnicity in Western Europe and find qualified support for the influence of ethnic conflict on support for the EU (2005 ; see also Luedtke 2005 ; de Master and Le Roy 2000 on ‘attitudes toward minorities’; and de Vreese and Boomgaarden 2005 for ‘anti-immigration sentiments’). This is not unprecedented as earlier works argued that Euroscepticism was born out of identity politics such that the nation-state is the appropriate reference point for identity and the continuing EU project undermines this (Taggart 1998 ).

Another approach to this topic pursues the competing self-identification of Europeans as nationals or Europeans. Scheuer (1999 ) presents a set of criteria of this form of nationalism arguing that in place of nationalism is the identification of Europeans as citizens of Europe and the implicit pride in being European. That is, rather than an appeal to the soil and toil of ‘nationalism’, it is born out of a sense belonging in a political community in which mutual trust and the inclusion of new members is an aspiration (the ‘we’ feeling). A version of this is the loss of national identity to the EU itself (Christin and Trechsel 2002 ) such that integration is a symbolic threat to national sovereignty.

In considering the first branch and finding the latter, McLaren (2002 ) examined the perceived threats posed by other cultures (or ‘hostility toward other cultures’) and the support for EU; however, her examination concludes that the threat is not exclusively third country nationals but also other EU member states and their encroachment through continued EU expansion. However, McLaren ’s later work (2004 ) suggests that while there is rather widespread fear of a loss of national identity and culture, it is not central to the resistance of citizens toward continued EU enlargement.

Clearly, the literature on European identity has proceeded on the idea that national and European identification are competitors in a zero sum game. This has a significant impact on the thinking about support for the EU. But as Bruter suggests, not only is a European identity necessary to combat the deficit of legitimacy in the EU but also that national and European identities do not have to be competing but rather can be expected to be positively correlated (2003 : 1154). Further arguing that little attention has been paid to the conceptual basis of identity, Bruter (2003 ) demonstrates that the above conceptualization of European identity is fundamentally problematic, limiting scholars’ ability to get at what national and European identities actually mean for the legitimization of the EU (see also Bruter 2005 ; Habermas 1992 ). He bifurcates European identity into civic and cultural components and suggests that the broader – that is, cultural – European identity is driven more substantially by the shared symbols.¹⁴ Therefore, this twofold definition would suggest that what is often meant – and empirically captured in the common operationalization of European identity – is the civic component of identity of the feeling that respondents are citizens of a European political system.

2.4 Institutions and Institutional Performance

Almost from the beginning of the EU project, it has been the popular perception that EU institutions were largely insulated from direct public access. As such, the process has suffered from a “representation deficit” or an insufficiency in its ability to accurately gauge and act in accordance with the will of the governed. In addition, even at the functional level of providing a level playing ground, the institutions have failed to generate equitable distributions of power within the EU itself and among its member states.¹⁵ Whether symptomatic or deterministic, there has been a steady decline in participation of European Parliamentary elections since 1979. This deceasing participation roughly indicates a behavioral manifestation of political interest, engagement, and enthusiasm and has been linked to the empirical legitimacy of the EU (Blondel, Sinnott, and Svensson 1998 ). In contrast, the link between declining turnout in European Parliamentary elections and declining legitimacy of the EU has instead been argued to be a strict function of structural factors (Franklin 2001 ; see also van der Eijk and Franklin 1996 ).

Reliance on a national-level ideal type of institutional arrangement (and subsequent performance) continues to limit the ability to develop a working framework to accurately define political representation at the EU level (Coultrap 1999 ). Models of party government are premised on three broad assumptions: all major decisions are made by elected officials; policy proposals are formulated and policy decisions are made within parties which then act cohesively to enact them; and elected officials are recruited and held collectively accountable through party. Two examples illustrate the shortcomings of these national-level theories. The inter-governmental model of the EU blurs the path of representation at each political transmission from each nations’ electorate to national sets of parties to variously arrayed national parliaments and finally to – what at best can be – refracted representation in the European Council and Council of Ministers (Newman 1996 ). In contrast, a federal model of European political representation suggests a European electorate with European parties that weakly represent their interests in European Parliament and government, again making the tenuous assumption of a coherent European-wide constituency.

In the former, the pathway linking representatives to their constituents is convoluted. In the latter, as EP elections are considered ‘second order’, citizens’ participation (aside from the scattered mandatory or ‘same day’ elections practices) is much lower given that ‘much less is at stake’. Therefore, one facet of popular perceptions of empirical legitimacy is the perception that the EU is a representative institution that poorly reflects the collective voice of Europeans.

Evaluations of institutional performance include both the input component, that the EU government is both selected by popular sanction and institutions are sufficiently democratic in their process, and the output component, which is the ability of the EU to deliver on policy and enforcement. However, as Schmitt and Thomassen (1999 : 14) note, “…there is not a single undisputed normative theory of political representation”.¹⁶ While the ability of the EU to function in supra-national capacity has grown, there remains the perception that an “…effective system of political representation is …missing” (Schmitt and Thomassen 1999 : 3). Further muddied by the EU’s institutional distinctiveness, there remains a lingering dispute over criteria for political representation at the EU level.

Although scholars have debated the validity and sufficiency of democratic practices, constituents of the EU have cited a general un-responsiveness and eroding democratic practices as the bases of what has come to be known as the democratic deficit (Scharpf 1997 ). The democratic deficit cites the failure of EP elections to accurately translate elections into the distribution of power and is a twofold issue. First, the only elected body of the EU is not the most powerful institution. Second, beyond popular elections, the processes by which policy proscriptions are reached fail to meet an adequately democratic standard.¹⁷ For the former concern of the weak linkage between peoples and the EP, the deficit not only indicates the weak EP but also the nature of the EP elections.¹⁸ These ‘second order’ elections fail to accurately link EP representative with their constituents (and their agendas) effectively handicapping their role in the policy making process at the EU level (Reif and Schmitt 1980 ; Hix and Marsh 2007 ). The EU’s internal institutional structure further dispenses policy responsibilities across these under-representative groups: a weak European Parliament, an un-elected European Commission, and that both the Council of Ministers (or the Council of the European Union) and European Council.

Scharpf (1999 ) has re-conceptualized the dual notion of legitimacy as input/output legitimacy (see above). Not only should institutions be ‘democratic enough’ but also produce policies that are congruent with publics’ own preferences (see also Rohrschneider 2002 ). As scholars have noted, these are not properly controlled or accountable to national institutions or constituencies (Newman 1996 : 173). Although, in contrast to the intensity of this debate, others have argued that the democratic deficit is largely an academic exercise as the EU is institutionally ‘democratic enough’ (Schmitt and Thomassen 1999 ; see also Majone 1998 ; Moravcsik 2002 ) or as Crombez (2003 : 101) argues: “The institutional set up of the EU does not lead to policies that are fundamentally undemocratic and that the composition of its institutions is not inherently less democratic than that of the US political institutions.”

Do national-level variables play a more significant role in shaping popular perceptions of the EU? Many authors have answered that not only are they important, they obscure the individual-level variation and may be thought to be developed in the national level context (Deflem and Pampel 1996 ; Kritzinger 2003 ). Several works have presented evidence that popular perceptions of the EU are conditioned by national institutional factors (Anderson 1998 ; Martinotti and Stefanizzi 1995 ; Norris 1999 ; Sánchez-Cuenca 2000 ; Rohrschneider 2002 ). The most fertile branch of this argument has been that individuals’ evaluation of the EU depends on the nation-state performance. Specifically, early work in this area demonstrated that support for integration depended on the legitimacy and efficiency of the nation-state (Anderson 1998 ; Janssen 1991 ).¹⁹ While satisfaction with the EU performance has been understood as a function of satisfaction with incumbent government (Franklin, van der Eijk, and Marsh 1995 ; Ray 2003a ) and positive evaluations of national government (Franklin, van der Eijk, and Marsh 1995 ), the argument has resulted in a discussion over the use of national governments as proxies for explicit supra-national performance-based assessments (Anderson 1998 ; Franklin, Marsh, and Wlezien 1994b ). Support for national governments and their parties has been understood as a heuristic through which citizens could make proxy assessments of the performance of the EU.

More recent works suggest that this institutional proxy argument does have limits. Karp, Banducci, and Bowler (2003 ) argue that political sophisticates can distinguish between national and EU institutions and that they are assessed on their own criteria. Sánchez-Cuenca (2000 ) demonstrated that in some cases, the proxy argument works as a contrasting lens producing an inverse perception between national and supra-national institutions. Rather than see both national governments and the EU as a singular, or overtly similar, set of political institutions, Sánchez-Cuenca argued that lower national institutional evaluation was correlated with support for the EU through not only the lower opportunity costs of transferring sovereignty to the EU but also the appeal to a less corrupt – as were many national governments were perceived as being – version of democratic governance. Rohrschneider (2002 ) refines this line of reasoning by arguing that individuals’ perceptions of how well governments are representing their interests are somewhat shaped by the performance of arbitrating institutions – specifically bureaucracies and judiciaries – as citizens’ interactions are likely to be with these institutions and through this experiential institutional exposure, they generate their orientation to the EU. As these work show, increasing amounts of the work on understanding the role of institutions in shaping individuals’ support and perceptions about the EU are being done in the literature associated with the democratic deficit.

While Karp and Bowler (2006 ) find that ‘hesitant’ Europeans’ attitudes, in contrast to the overt supporters and resisters of integration, are driven by economic and EU performance factors, two contributors to this debate have more explicitly linked the democratic deficit to the proxy argument from above. Both Rohrschneider (2002 ) and Sánchez-Cuenca (2000 ) argue that evaluations of EU institutions emerge from individuals’ assessments of the quality of national institutions; however, they do so in ways that contradict other works on proxy evaluations. Sánchez-Cuenca (2000 ) argument is somewhat different from the core of popular perceptions of the EU as a system of government than Rohrschneider’s. In the former, the dependent variable is related to the speed of integration (the Dynamometer) and is explained by variations in individuals’ perceptions of national level corruption and social protection. That is, rather than evaluate the EU as an extension of national level institutions, they are the bases for comparison. In the latter, Rohrschneider more directly links citizens’ perception of the lack of representation at the EU level and demonstrates that their support for the EU is reduced independent of economic perceptions. More provocatively for the case against the proxy argument, this is particularly true in countries with well-functioning institutions suggesting a mediated comparison between the function, or quality, of democratic institutions at the different levels.

At the core, the democratic deficit is founded on the idea that it is difficult for Europeans to care about a Union whose identity was for so long nebulous or at least limited, but which over time appears to increasingly impinge upon every aspect of their existence. The essence of liberal democracy rests on many foundations including the idea that government is designed to respond to its constituents, whether in the form of policy output or regular re-construction (Miller and Listhaug 1990 ). If people believe they are being fairly and accurately represented in government, their support for that government is likely to increase or be high and when this is not the case, little support is expected (Kornberg and Clarke 1994 ; Pitkin 1967 ). However, while the effect of feeling represented is necessary, it is not sufficient for widespread support. In sum, the democratic deficit represents both the substantive and procedural components (Dahl 1989 ; see Rohrschneider 2002 ), such that the substantive component is the output of government that, for the most part, the majority is able to get the government to do what it wants and the procedural component is related to representation such that the institutions are inherently ‘fair’.