Regional attachment and support for European integration

Abstract

The European Union has evolved into a system of multilevel governance. EU actions have facilitated the involvement of regional/subnational authorities in EU policy formulation and implementation, thus enabling a downward shift in competencies toward subnational authorities. In this article, I argue that regional attachment facilitates support for European integration. Specifically, I show that individuals holding inclusive regional attachment sentiments perceive EU policies that result in the empowerment of subnational authorities in a positive light. In the process, these individuals are more likely to be supportive of European integration. This article’s findings contribute to the literature on public opinion and European integration by exploring regionalist sources of support for European integration. Moreover, these findings contribute to the growing research on the emergence of a ‘Europe of the regions.’

Keywords

European integration public opinion regional attachment regional/subnational authorities regional policy subsidiarity

The evolution of European integration has resulted in shifts of competencies away from nation-states. An unexpected development has been the downward shift of competencies to subnational authorities of European Union (EU) member states. In what Hooghe and Marks (1996) call multilevel governance, subnational authorities have gained voice in EU affairs. Through policies such as subsidiarity, key decisions affecting EU citizens are being made at lower levels of authority.

The empowerment of subnational authorities through European integration has received some attention from Europeanists (Downs, 2002; Jolly, 2007). However, none of these studies have examined the role subnational attachment sentiments play in influencing public opinion toward European integration. European integration continues to have economic, social, and political consequences for citizens and shapes to what extent they support the integration process. Various studies have explored the role of public opinion in citizens’ support for European integration (Eichengreen and Dalton, 2007; Sánchez-Cuenca, 2000), although none have explicitly addressed the effect of regional attachment sentiments on opinions about European integration. Since individuals at the regional level more directly experience EU policies that economically and politically strengthen subnational authorities, what effect does regional attachment have on citizens’ support for European integration? Building on literature on the relationship between the EU and subnational authorities as well as public opinion research on support for European integration, I argue that individuals holding inclusive regional attachment sentiments are more likely to support European integration because the EU has facilitated regional autonomy and input in EU policies.

In previous studies on public opinion and support for European integration, the economic and sociopolitical consequences of the EU are shown to influence citizens’ attitudes toward European integration. Hooghe and Marks (2004: 415) note that ‘[c]itizens do indeed take into account the economic consequences of European integration, but conceptions of group membership appear to be more powerful.’ It should be expected therefore that, when EU policies politically and economically strengthen regions, individuals with strongly held inclusive subnational attachment sentiments would perceive the EU positively and be more supportive of the integration process.

To test this argument, I construct a dataset using Eurobarometer public opinion poll results and statistics from the European Commission and the World Bank. These data are observed at the individual and country levels: individuals are nested in member countries of the EU. I use a mixed-effects model estimation in order to account for variance at the two levels of observation (Steenbergen and Jones, 2002). This approach follows that of others studying public opinion and European integration who view citizens’ attitude toward European integration as being shaped by individual- and group-level factors (De Vries et al., 2010; Stockemer, 2012). The results from this estimation lend credence to the primary hypothesis of this article: controlling for individual- and country-level factors, individuals possessing a strong sense of inclusive regional attachment are more likely to be supportive of European integration.

As the process of European integration continues, questions on democratic accountability in EU policy-making abound. The perceived democratic deficit, whereby unelected bureaucrats in Brussels take key decisions affecting the member states’ political, economic, and social well-being, may be fueling Euroskepticism. Yet, policies and actions of the EU that encourage subnational participation in policy-making and implementation may be aiding in correcting this unintended consequence. Inclusive regionalists would view such EU policies as addressing their preferences for devolution and involvement in EU affairs while not deeming it a threat to their identity. The findings of this article point to this regionalist aspect of support for European integration and contribute to the growing research on the emergence of a ‘Europe of the regions.’

Regional authorities and the European Union

The European Union has become a system of multilevel governance characterized by multiple nodes of decision-making. ‘One of the most important consequences of European integration is the multiplication of extra-national channels for subnational political activity’ (Hooghe and Marks, 1996: 73). Research on the relationship between regional/subnational authorities and European integration has acknowledged this multilevel characteristic of the EU. Although initially conceived as involving an upward shift of competencies away from national governments to bureaucrats in Brussels, European integration, especially since the signing of the Maastricht Treaty, has inadvertently resulted in the shifting of some nation-state powers downwards to subnational authorities (Jeffrey, 2000). The Maastricht Treaty strengthened the role of subnational authorities by establishing channels for them to influence EU decisions. The treaty established the Committee of the Regions and articulated the principle of subsidiarity. Other EU initiatives, such as its regional policy, have also continued to empower subnational authorities.

The notion that the EU might be moving towards a ‘Europe of the regions’ became prevalent in the aftermath of the Maastricht Treaty that opened up various channels for regional authorities to influence EU policy-making. The EU’s economic integration has reduced the cost of regional autonomy as all EU members came to operate under the same market rules (Hooghe and Marks, 2008). Devolution is now being perceived as one means of enhancing the efficiency of policy implementation while heightening accountability and citizen participation in EU affairs (Brusis, 2010). The EU has facilitated devolution through its institutions and treaties such as the Maastricht Treaty (Mandrino, 2008; Tatham, 2008). These devolution processes have increased the influence of regions in EU policy-making and implementation. In certain instances, devolution has resulted in these regions bypassing their national governments and dealing directly with Brussels (Tatham, 2010).

The EU has opened up several institutional channels for regions to influence their own local affairs and EU policy formulation and implementation. These include the Committee of the Regions, direct links with the EU Commission, and the establishment of regional offices in Brussels (Hooghe and Marks, 1996). These institutional channels, especially the Committee of the Regions, were a means of addressing the democratic deficit perceived to widen the rift between citizens and EU’s unelected bureaucrats in Brussels (Roht-Arriaza, 1996).¹

Other aspects of the EU that have facilitated devolution and the empowerment of regions are provisions emanating from various EU treaties. Researchers on this subject have focused on subsidiarity. The principle of subsidiarity, as envisioned in the Maastricht Treaty, advocates for decision-making to be conducted at the lowest levels of authority in the EU (Treaty on the European Union, 1992). Subsidiarity has generated a wide range of research that has empirically demonstrated how regions have used it to seek a greater voice in EU affairs and has reignited calls for decentralization and regional autonomy (Scott et al., 1994). Subsidiarity ‘has provided one of the stimuli for regionalism, emboldening regional leaders to assert their territorial autonomy, to travel with the attitude and air of a head of state, to open quasi-embassies (“information offices”) in multiple countries, and to sign high profile agreements with other regions’ (Downs, 2002: 73).

Subsidiarity studies have also examined the role of regions in making and implementing specific EU policies such as the regional/cohesion policy and regulatory policies dealing with environmental and social issues (Baun and Marek, 2006; Borghetto and Franchino, 2010; Brusis, 2010; Tatham, 2012). Through subsidiarity, regions have come to perceive the EU ‘as an ally against the central state’ (Jolly, 2009: 4). Indeed, the involvement of subnational authorities in EU affairs such as its regional policy has encouraged regions to further seek a greater role in their affairs and a greater voice in EU policies (Brusis, 2010).

An extension of this research has focused on how EU-facilitated devolution is promoting regionalist parties to support European integration. Through subsidiarity, the EU has empowered subnational authorities and in the process strengthened the positions of regionalist parties by increasing their ‘credibility of demands for greater autonomy, ranging from independence to devolution to cultural rights, and therefore… [these regionalist parties’] credibility’ (Jolly, 2007: 111). According to Jolly (2007: 124), regionalist parties such as the Scottish National Party, favor ‘European integration because it creates a more favorable political opportunity structure for their subnational autonomy movements.’ These regionalist parties cue their adherents to be supportive of European integration because of the benefits they derive from the EU in terms of further autonomy from their nation-states (Jolly, 2009).²

The above review demonstrates that the EU has empowered subnational authorities to influence EU policies. However, a missing link in this research is the extent to which citizens holding strong attachments to these subnational levels support the process of European integration. Individuals form opinions on European integration through evaluating the effects of the integration process on their identities and economic well-being (Brinegar et al., 2004; Carey, 2002; Hooghe and Marks, 2005). Given that the EU empowers subnational authorities and given our knowledge of Euroskepticism, a fundamental puzzle emerges: can the EU’s support for regional authorities function as an indirect mechanism to improve support for European integration? I provide a theoretical basis for answering this question in the next section, building on research on public opinion and support for European integration.

Regional attachment and support for European integration

Research on public opinion and support for European integration has focused on the political, economic, and identity consequences that result from European integration and how these shape individuals’ attitudes toward the integration process. In the first category, European integration as an economic venture that has distributional effects influences individuals’ views of the integration process based on perceived economic gains or losses resulting from economic liberalization (Brinegar and Jolly, 2005; Mayda and Rodrik, 2005). However, European integration entailing social and political cooperation has resulted in a shift in national authority that has traditionally been tied to national identity and attachment. In this second category, the EU’s perceived encroachment into notions of national identity influences individuals’ opinions on European integration based on the intensity of their national attachment (Hooghe and Marks, 2004). Since the EU has also politically and economically strengthened the competencies of subnational territories, which has direct implications for individuals holding strong attachments to these territories, this section proposes an argument linking the intensity of regional attachment to support for European integration.

The findings of the relationship between national attachment and support for European integration are useful in mapping out the theory I propose. Nation-states form the primary political unit above which there is no other authority. Hooghe and Marks (2004: 416) elaborate: ‘Humans and their ancestors evolved an emotional capacity for intense group loyalty long before the development of rational faculties, and such loyalties can be extremely powerful in shaping views towards political objects.’ The locus of this loyalty is the nation-state (McLaren, 2002). As a process that involves the shift of traditionally national areas of influence to supranational institutions, European integration may be perceived as a threat to national identity among citizens of member states (Carey, 2002). In forming their opinion on European integration, individuals would be ‘concerned about problems related to the degradation of the nation-state’ by the process of European integration (McLaren, 2002: 554).

However, national identity is not always associated with less support for European integration. Diez Medrano’s (2003) research points to the important role national histories have in shaping support for or opposition to European integration. These national histories shape the extent to which national identity is inclusive or exclusive, leading Hooghe and Marks (2005) to argue that individuals with an exclusive sense of national attachment would find the EU to be more of a threat than those with an inclusive sense of national identity. Politicians in turn can tap into these notions of national attachment and influence citizens’ support for or opposition to European integration (Hooghe and Marks, 2005).

Regional attachment should also have an impact on support for European integration in a manner similar to national attachment. In the preceding section, I summarized previous research that finds that the EU, through various institutional and legal mechanisms, has encouraged regional autonomy and participation in EU affairs. Regionalist adherents, as discussed in Jolly’s (2007, 2009) research on regionalist parties, would welcome these acts toward autonomy that the EU facilitates. Political economists examining the effects of European integration on individuals have observed that those who benefit economically from the integration process would tend to be supportive of it (Brinegar et al., 2004). Extrapolating from this finding, individuals with inclusive regionalist sentiments should be more supportive of the integration process because the EU facilitates regional autonomy and participation in EU policies. Since the EU is giving individuals at the regional level and their direct representatives more say in their local affairs and a voice in EU policy-making, these inclusive regionalists should perceive European integration positively.

A distinction needs to be made between exclusive and inclusive regional attachment in this relationship between regional attachment and support for European integration. Since the EU facilitates regional autonomy and participation in EU policies within the nation-state framework, those with an inclusive regional attachment sentiment are the ones most likely to support European integration. On the other hand, those with an exclusive sense of regional attachment that restricts membership to those from a particular territory might not be as receptive to European integration because the EU does not go far enough in promoting regional autonomy and the integration process constitutes a threat to their exclusive identity.

Sources of regional identity help to clarify this relationship between regional attachment and support for European integration. These sources influence positions on regional autonomy that the EU has inadvertently promoted. Cultural distinctiveness, including ethno-linguistic differences, fuels regionalist sentiments (Diez Medrano and Gutiérrez, 2001). Ethno-linguistic differences between the Flemish and Walloons can explain the strong regionalist sentiments one observes in Belgium (Cook, 2002: 138). Additionally, economic disparities can help generate a sense of regional attachment both for poorer regions feeling discriminated against and for richer regions identifying more with their affluence (Fitjar, 2010).³ Moreover, although not a direct source, political factors including regionalist parties, a history of political autonomy, and processes facilitating the institutionalization of regions can bring to the fore latent regionalist sentiments (Fitjar, 2010; Jolly, 2007). EU policies facilitating regional involvement in EU policy-making processes are examples of these political factors facilitating regional attachment (Fitjar, 2010; Paasi, 2009).

Individuals possessing regional attachment sentiments are also more receptive to calls for greater autonomy, for instance among the Scottish, Welsh, and Irish in the UK and Catalans, Basques, and Galicians in Spain (Keating, 1998; Painter, 2002). It could be argued that the strong regionalist sentiments of the Flemish and the Walloons may have influenced calls for greater regional autonomy after Belgium’s political post-election crisis in 2007. Figure 1 plots the means of inclusive regional attachment for respondents from the EU15 states against the regional authority index (RAI), a measure of the extent of regional autonomy. In this chart, inclusive regional attachment increases with RAI.

Figure 1.

Mean inclusive regional attachment and regional autonomy for the EU15 states.

As previously discussed in this article, the EU has evolved into a process that has encouraged and facilitated devolution and the empowerment of regional authorities closer to those with regionalist sentiments. Through institutions and legal mechanisms that enable regions to influence their local and EU affairs, regions have become more socially, politically, and economically empowered. This strengthening of regions not only encourages regional attachment; it should also enable those with regionalist sentiments to positively perceive the agents and processes that facilitate regional empowerment, in this case European integration.

Nevertheless, the EU’s actions of facilitating devolution to regions take place within the nation-state framework. The EU remains a supranational entity composed of nation-states as the main decision-makers. Its policies and directives are meant to facilitate social, political, and economic cooperation among the member states (Bache, 1999). Policies such as subsidiarity serve the purpose of encouraging regional participation in EU matters within this nation-state structure. These policies are not meant to threaten the existence of the EU’s member states (Van Kersbergen and Verbeek, 2004).

If the EU operates within this nation-state framework, then it is essential to draw a distinction between inclusive and exclusive regional attachment and how these two attachment types influence support for integration. Inclusive regionalists should positively perceive those actions of the EU that advance regional autonomy and be more supportive than exclusive regionalists of the integration process. Like inclusive nationalists, inclusive regionalists are more likely to accommodate multiple identities (De Vries and Edwards, 2009; Hooghe and Marks, 2005). Diez Medrano and Gutiérrez (2001) find that individuals with nested identities that can accommodate other forms of attachment are not necessarily against European integration. Because of their tendency to adapt their regional identity with other attachments, these inclusive regionalists are more likely to positively perceive EU actions on decentralization that are accommodative of their quest for more autonomy within their nation-state. Therefore:

H1: Individuals with a strong sense of inclusive regional attachment are more likely to support European integration.

Conversely, exclusive regional attachment does not lead to more support for European integration. Exclusive regionalists primarily identify with their regions and are less likely to accommodate other territorial attachments such as European identity that the process of integration inadvertently brings about. For these exclusivists, European integration might represent and facilitate their desire for regional autonomy. However, these exclusive regionalists might also view these actions of the EU as not going far enough and only serving to enmesh them in a Europe made up of nation-states. Hooghe and Marks (2005) rely on a similar causal mechanism when arguing and finding that exclusive nationalists feel that European integration threatens their identity and they are more likely to be Euroskeptics. From the perspective of these exclusive regionalists, European integration threatens their regional identity because it introduces another type of identity that their exclusiveness has difficulty accommodating, while also falling short of their demand for even greater autonomy. Thus:

H2: Individuals with a strong sense of exclusive regional attachment are less likely to support European integration.

Data and method

To test my argument, I analyze survey results from Eurobarometer 69.2 for the year 2008 (European Commission, 2008). Individuals form the units of analysis. Data for this empirical test are drawn from this opinion poll survey, the EU Commission, and the World Bank. A detailed discussion of the variables estimated and their operationalization is provided in the Web Appendix. The primary model is estimated using a mixed-effects multilevel logistic model to account for both individual- and country-level effects. Three sets of analyses are conducted: respondents from the EU15 states, respondents from all 27 EU states, and those from the United Kingdom. Since the EU15 states have been members of the EU for over 10 years, it should be expected that citizens in these states have witnessed the effects of various EU attributes that strengthened regions. Additionally, a separate examination of the UK is provided to highlight how regional attachment influences support for European integration in one of the EU’s most Euroskeptic states.

The dependent variable, EU benefit, is operationalized from the Eurobarometer question asking respondents to evaluate whether their country has benefitted or not benefitted from EU membership. The respondents’ answers to this question are recoded so that (1) indicates those who see the EU as having benefitted their country and (0) indicates those who do not perceive any benefits from the EU. Since this article focuses on the benefits that adherents of regional attachment perceive from EU membership, this ‘benefit’ question is appropriate for evaluating such opinions. This ‘benefit’ question is also one of the main questions researchers on Euroskepticism have relied on to evaluate public support for European integration (Carey, 2002; McLaren, 2002).

The two main independent variables of interest are drawn from respondents’ answers to questions evaluating their level of adherence to their regions and nations. Inclusive regional attachment is operationalized from respondents’ answer to the question asking them to what extent they are attached to the regions they inhabit. This question has four categories: ‘not at all’ (1), ‘not really’ (2), ‘somewhat’ (3), and ‘to a great extent’ (4). This question to measure inclusive regional attachment is similar to that used by Hooghe and Marks (2005) to evaluate inclusive national attachment. This regional attachment question is broad and does not specify what ‘region’ means. However, since it is asked alongside other territorial attachment questions, it is suitable for generally appraising how much individuals are attached to their regions without gauging its level of importance vis-à-vis other territorial attachments.

Exclusive regional attachment is operationalized differently, using the regional and national attachment questions in the Eurobarometer survey. For this type of attachment, I recode a binary variable that gauged whether an individual indicated a higher level of regional attachment than national attachment.⁴ Since the Eurobarometer survey does not have a question asking respondents to evaluate whether they feel more attached to their regions than to other territories, the operationalization employed here in capturing exclusive regional attachment is necessary and appropriate because it takes into account both national and regional attachment sentiments.

The models include several control variables at the individual and country levels used in previous studies on support for European integration. These include identity/cultural, utilitarian/economic, and political factors. Similarly to Hooghe and Marks (2005), I control for inclusive national attachment, which is an individual-level variable operationalized from the Eurobarometer question asking respondents to what extent they are attached to their particular nation-state. I also control for exclusive national attachment, which is a dichotomous variable indicating whether respondents expressed higher levels of national attachment than attachment to Europe.⁵

Additionally, I control for multiculturalism, a dichotomous variable operationalized from the Eurobarometer question asking respondents whether the EU means cultural diversity for them. Another cultural variable I control for is cultural identity fears, which is operationalized from the Eurobarometer question asking respondents whether the EU means cultural identity loss for them. Finally, at the country level I control for ethnic fractionalization, which evaluates to what extent a country is ethnically fractionalized (Alesina et al., 2003).⁶

Economic factors that influence support for European integration include personal economic expectations and national economic expectations. Individuals with better expectations for themselves and their country are more likely to support European integration (Hooghe and Marks, 2005). Personal economic expectations is an index based on three Eurobarometer questions that ask respondents about their expectations for the next 12 months of their lives, their financial situation, and their job situation (1 = worse; 2 = same; 3 = better). National economic expectations is an index operationalized from two Eurobarometer questions asking respondents about their expectations for the next 12 months of their nation’s economic and employment situation (1 = worse; 2 = same; 3 = better).

I also include two occupation variables: professional worker and manual worker. These are interaction terms of a dummy variable coded from the Eurobarometer question asking respondents about their occupation and the per capita gross national income (GNI) of a respondent’s country divided by 1000. Per capita GNI statistics are obtained from the World Bank. Professional/white-collar workers are expected to be more supportive of the EU, whereas manual/blue-collar workers are less likely to support the integration process (Hooghe and Marks, 2005).

At the country level, I control for net contributor states. These are member states that give more to the EU financially than they receive back. I rely on the 2008 EU Financial Report for this variable. Citizens of net contributor states are less likely to be supportive of the EU. I also control for cohesion funds, which is the percentage of the total funds each nation receives that goes toward cohesion policy projects. These statistics are also obtained from the 2008 EU Financial Report.⁷

For political factors, I control for ideology, which is a respondent’s self-placement of his/her ideology on a right–left scale. Additionally, I control for opinion leader, which is an index created from two Eurobarometer questions asking respondents about the frequency with which they discuss politics with family and friends and to what extent they influence the opinions of others. Opinion leader is recoded into four categories. At the country level, I control for the regional authority index, which measures how politically autonomous regions of a given state are on a 24-point scale (Hooghe et al., 2008).⁸ Finally, I include a set of demographic control variables that are standard in research on support for European integration. These include education levels attained, gender, and age of the respondents.

These variables are estimated using a multilevel mixed-effects logistic regression model because the dependent variable is dichotomous and the variables are observed at two levels of analysis. A multilevel mixed-effects model accounts for both within- and between-country effects (Steenbergen and Jones, 2002). A multilevel model is also suitable because both individual- and country-level factors have been shown to affect support for European integration.

Results

In this section, I present and discuss statistical results for the relationship between regional attachment and support for European integration. About 88 percent of the variance is explained at the individual level and roughly 12 percent is explained at the country level for the multilevel models presented. A multilevel estimation is thus appropriate to reduce the likelihood of type I error: without the use of a multilevel model, the standard errors would be too small (Steenbergen and Jones, 2002). Another methodological concern is whether the identity variables including regional and national attachments are endogenous to support for European integration. Quantitative studies that use individual attitudes from public opinion surveys have raised this concern (Gabel and Scheve, 2007; Kritzinger, 2003; Van Kersbergen, 2000). In acknowledging this issue, I also note that notions of attachment and identity plausibly predate opinions on European integration as discussed in Risse (2005). Additional tests reveal that the independent variables are not jointly associated with each other and there is no indication of heteroskedasticity. In the two sets of multilevel model estimates, most of the individual-level variables are statistically significant, whereas the country-level variables are not significant.

Table 1 presents estimates for the relationship between regional attachment and support for European integration for the EU15 states. Model 1 displays estimates for the control variables, Model 2 includes inclusive regional attachment and all control variables, Model 3 has estimates for exclusive regional attachment and all control variables, and Model 4 is the full model. These models reveal that inclusive regional attachment is positively associated with support for European integration; however, exclusive regional attachment has no statistically significant relationship with support for European integration. Other results including those for inclusive and exclusive national attachment, multiculturalism and cultural identity loss, and most of the individual-level economic and political control factors are also statistically significant and corroborate findings from previous studies. For instance, inclusive national attachment increases the probability of support for European integration, yet exclusive national attachment diminishes the likelihood of support for European integration. These findings give credence to those of Hooghe and Marks (2004, 2005) on the effect of the intensity of national attachment on support for European integration among citizens of the EU15 states.⁹

Table 1.

Regional attachment and support for European integration (EU15 states)

Variables	Model 1	Model 2	Model 3	Model 4
Individual level
Inclusive regional attachment	0.083** (0.039)	0.099** (0.043)
Exclusive regional attachment	0.0003 (0.108)	−0.105 (0.123)
Inclusive national attachment	0.407*** (0.049)	0.381*** (0.051)	0.407*** (0.06)	0.343*** (0.069)
Exclusive national attachment	−0.846*** (0.06)	−0.842*** (0.06)	−0.846*** (0.06)	−0.841*** (0.06)
Multiculturalism	0.757*** (0.062)	0.757*** (0.062)	0.757*** (0.062)	0.757*** (0.062)
Cultural identity loss	−1.236*** (0.076)	−1.235*** (0.076)	−1.236*** (0.076)	−1.233*** (0.076)
Personal economic expectations	0.288*** (0.063)	0.281*** (0.063)	0.288*** (0.063)	0.281*** (0.063)
National economic expectations	0.41*** (0.048)	0.407*** (0.048)	0.41*** (0.048)	0.407*** (0.048)
Professional worker	0.006*** (0.002)	0.006*** (0.002)	0.006*** (0.002)	0.006*** (0.002)
Manual worker	−0.009*** (0.001)	−0.009*** (0.001)	−0.009*** (0.001)	−0.01*** (0.001)
Ideology	0.042*** (0.012)	0.041*** (0.012)	0.042*** (0.012)	−3.598*** (0.012)
Opinion leader	0.165*** (0.03)	0.167*** (0.03)	0.165*** (0.03)	0.167*** (0.03)
Education	0.15*** (0.03)	0.154*** (0.031)	0.15*** (0.03)	0.154*** (0.031)
Gender	−0.274*** (0.052)	−0.272*** (0.052)	−0.274*** (0.052)	−0.272*** (0.052)
Age	−0.01*** (0.002)	−0.01*** (0.002)	−0.01*** (0.002)	−0.01*** (0.002)
Country level
Ethnic fractionalization	1.523 (1.642)	1.525 (1.644)	1.523 (1.642)	1.528 (1.643)
Net contributor	−0.0001 (0.0001)	−0.0001 (0.0001)	−0.0001 (0.0001)	−0.0001 (0.0001)
Cohesion funds	0.003 (0.013)	0.003 (0.013)	0.003 (0.013)	0.003 (0.013)
RAI	−0.03 (0.028)	−0.03 (0.028)	−0.03 (0.028)	−0.03 (0.028)
Constant	−1.46** (0.611)	−1.633** (0.616)	−1.46** (0.626)	−1.534** (0.627)
Intra‐class correlation	0.138	0.139	0.138	0.139
Wald Chi-square	1014.79	1011.88	1014.79	1012.37
Log likelihood	−4679.230	−4661.965	−4679.231	−4661.600
N (countries)	9008 (15)	8976 (15)	9008 (15)	8976 (15)

Notes: Standard errors in parentheses. Statistical significance (two-tailed): **p ≤ .05; ***p ≤ .001.

The results for inclusive regional attachment support this article’s first hypothesis. Individuals with an inclusive sense of regional attachment are more likely to support European integration. Estimates in Models 2 and 4 show that the probability of supporting European integration increases with an increase in inclusive regional attachment. Substantively interpreting estimates from Model 4, the probability of supporting European integration increases from 0.66 for those not at all attached to their regions to 0.73 for those very much attached to their regions, holding other variables at their means. This positive relationship provides additional evidence for the effect that identity factors have on the likelihood of supporting European integration. Previous studies show that national identity influences attitudes toward European integration because the multicultural element of the integration process ‘erodes exclusionary norms of “us” and “them” that are deeply rooted in the creation of European nation states’ (Hooghe and Marks, 2005: 423). Since the integration process also entails empowering subnational authorities in line with the preferences of regionalists, inclusive regionalists would tend to support European integration.

The results for respondents from the EU15 states, however, do not lend support for this article’s second hypothesis: there is no statistically significant relationship between exclusive regional attachment and support for European integration. This finding is interesting because it shows that exclusive regionalists may be ambivalent toward European integration. Since they still benefit from the EU’s policies that facilitate regional autonomy, these exclusive regionalists may nonetheless be irresolute about whether to support the integration process owing to its potential to undermine their exclusive identity.

Table 2 provides estimates for all 27 EU member states. Similarly to the previous table, four models are provided that include the control variables (Model 5), the control variables with inclusive regional attachment (Model 6), the control variables with exclusive national attachment (Model 7), and the full model (Model 8). These estimates lend support to the first hypothesis: when considering all EU27 states, inclusive regional attachment has a positive impact on support for European integration. However, exclusive regional attachment has no statistically significant relationship with support for European integration.

Table 2.

Regional attachment and support for European integration (all member states)

Variables	Model 5	Model 6	Model 7	Model 8
Individual level
Inclusive regional attachment	0.089** (0.033)	0.093** (0.037)
Exclusive regional attachment	0.087 (0.094)	−0.024 (0.107)
Inclusive national attachment	0.432*** (0.042)	0.397*** (0.044)	0.459*** (0.051)	0.388*** (0.059)
Exclusive national attachment	−0.811*** (0.049)	−0.804*** (0.049)	−0.811*** (0.049)	−0.804*** (0.049)
Multiculturalism	0.68*** (0.053)	0.679*** (0.053)	0.681*** (0.053)	0.679*** (0.053)
Cultural identity loss	−1.18*** (0.062)	−1.176*** (0.062)	−1.18*** (0.062)	−1.176*** (0.062)
Personal economic expectations	0.363*** (0.051)	0.357*** (0.051)	0.363*** (0.051)	0.357*** (0.051)
National economic expectations	0.47*** (0.04)	0.468*** (0.04)	0.47*** (0.04)	0.468*** (0.04)
Professional worker	0.006*** (0.002)	0.006*** (0.002)	0.006*** (0.002)	0.006*** (0.002)
Manual worker	−0.01*** (0.001)	−0.01*** (0.001)	−0.01*** (0.001)	−0.01*** (0.001)
Ideology	0.056*** (0.01)	0.055*** (0.01)	0.056*** (0.01)	0.055*** (0.01)
Opinion leader	0.138*** (0.024)	0.14*** (0.024)	0.138*** (0.024)	0.139*** (0.024)
Education	0.16*** (0.026)	0.163*** (0.026)	0.161*** (0.026)	0.163*** (0.026)
Gender	−0.184*** (0.043)	−0.181*** (0.043)	−0.184*** (0.043)	−0.181*** (0.043)
Age	−0.008*** (0.001)	−0.008*** (0.001)	−0.008*** (0.001)	−0.008*** (0.001)
Country level
Ethnic fractionalization	0.331 (0.865)	0.301 (0.867)	0.328 (0.865)	0.301 (0.867)
Net contributor	−0.0001** (0.0001)	−0.0001** (0.0001)	−0.0001** (0.0001)	−0.0001** (0.0001)
Cohesion funds	−0.002 (0.008)	−0.003 (0.008)	−0.002 (0.008)	−0.003 (0.008)
RAI	−0.015 (0.018)	−0.016 (0.018)	−0.016 (0.018)	−0.016 (0.018)
Constant	−1.913*** (0.484)	−2.075*** (0.489)	−2.019*** (0.497)	−2.053*** (0.499)
Intra-class correlation	0.133	0.133	0.133	0.133
Wald Chi-square	1486.57	1479.57	1487.06	1479.61
Log likelihood	−6963.176	−6940.295	−6962.747	−6940.270
N (countries)	14001 (27)	13953 (27)	14001 (27)	13953 (27)

Notes: Standard errors in parentheses. Statistical significance (two-tailed): **p ≤ .05; ***p ≤ .001.

The results in Table 2 offer additional support to findings from previous studies. For all EU27 states, personal and national economic expectations, an individual’s type of occupation, ideology, political discussion, education, age, and gender affect levels of support for European integration. Close to the interest of this article, identity factors including both types of national attachment, multiculturalism, and cultural identity loss also give credence to previous arguments showing that these factors have an impact on support for European integration.¹⁰

Inclusive regional attachment increases the probability of support for European integration for all 27 member states (Models 6 and 8). Similar to the findings for the EU15 states, these results suggest that, although other identity, economic, and political factors influence support for European integration, inclusive regional attachment should also be considered.

However, the estimates in Table 2 do not support the second hypothesis. There is no statistically significant relationship between exclusive regional attachment and support for European integration for respondents from all EU states. As explained for respondents from the EU15 states, this insignificant relationship may be indicative of the ambivalence of exclusive regionalists toward European integration, which might satisfy their need for autonomy yet be perceived as a threat to their regional identity.

The results presented in Tables 1 and 2 provide empirical support for the argument I propose in this article. However, a closer examination of the United Kingdom offers further evidence that inclusive regional attachment does have a positive impact on support for European integration. The UK is interesting because it has long been singled out as one of the most Euroskeptic states in the EU (Spiering, 2004). Its opt-outs from the single currency and the Schengen zone attest to the UK’s noncommittal stance on European integration. Despite its Euroskeptic characteristics, regions within the UK have been noted to be supportive of European integration. An illustrative example, which I summarized previously in this article, is Scotland, where the Scottish National Party has been able to link its case for greater autonomy from London with support for European integration among its adherents (Jolly, 2007). Others examining British Euroskepticism have also observed that a more thorough investigation of support for European integration in the UK needs to consider the multicultural characteristics that distinguish the positions of the Irish, Scottish, and Welsh on European integration (Carey, 2002; Haesly, 2001).

Table 3 presents logistic regression estimates for the relationship between regional attachment and support for European integration among respondents from the UK using a subset of the data previously estimated in this article. Four models are presented that include the control variables (Model 9), inclusive regional attachment (Model 10), exclusive regional attachment (Model 11), and the full model (Model 12). The estimates for inclusive regional attachment provide substantively and statistically more significant evidence that in the UK regional attachment influences support for European integration. Interestingly, inclusive national attachment is not statistically significant in the full model. Moreover, most of the identity factors such as exclusive national attachment, multiculturalism, and cultural identity loss are statistically significant and in the expected direction, whereas economic and political factors do not seem to attain significance levels.

Table 3.

Regional attachment and support for European integration (UK only)

Variables	Model 9	Model 10	Model 11	Model 12
Inclusive regional attachment	0.256** (0.033)	0.366** (0.15)
Exclusive regional attachment	−0.248 (0.348)	−0.679 (0.434)
Inclusive national attachment	0.5** (0.166)	0.473** (0.17)	0.43** (0.192)	0.252 (0.219)
Exclusive national attachment	−1.471*** (0.294)	−1.513*** (0.299)	−1.478*** (0.295)	−1.521*** (0.3)
Multiculturalism	1.243*** (0.268)	1.161*** (0.27)	1.26*** (0.27)	1.19*** (0.272)
Cultural identity loss	−1.617*** (0.26)	−1.569*** (0.261)	−1.62*** (0.26)	−1.555*** (0.261)
Personal economic expectations	0.173 (0.239)	0.158 (0.241)	0.193 (0.241)	0.2 (0.244)
National economic expectations	0.727*** (0.211)	0.678*** (0213)	0.717*** (0.212)	0.651** (0.214)
Professional worker	0.009 (0.006)	0.011 (0.007)	0.009 (0.006)	0.011 (0.007)
Manual worker	−0.006 (0.005)	−0.006 (0.006)	−0.007 (0.005)	−0.005 (0.006)
Ideology	−0.231*** (0.057)	−0.226*** (0.058)	−0.232*** (0.057)	−0.226*** (0.058)
Opinion leader	0.268*** (0.108)	0.226** (0.109)	0.274** (0.108)	0.23** (0.11)
Education	0.122 (0.136)	0.15 (0.137)	0.123 (0.136)	0.143 (0.138)
Gender	−0.184*** (0.043)	0.038 (0.206)	0.063 (0.204)	0.046 (0.207)
Age	−0.01 (0.006)	−0.011 (0.006)	−0.01 (0.006)	−0.011 (0.006)
Constant	−1.439 (1.063)	−1.995 (1.11)	−1.191 (1.118)	−1.555 (1.144)
LR Chi‐square	238.36	236.21	238.87	238.66
Log likelihood	−299.212	−293.41	−298.959	−292.186
Pseudo R²	.285	.287	.286	.290
N	604	594	604	594

Notes: Standard errors in parentheses. Statistical significance (two-tailed): **p ≤ .05; ***p ≤ .001.

In the UK, those more attached to their respective region inclusively are more likely to support European integration. For estimates in Model 10, the probability of supporting European integration increases from 0.30 for those ‘not at all’ attached to their regions to 0.51 for those attached to their regions ‘to a great extent’, holding other variables at their means. In Model 12, the probability of supporting European integration increases from 0.26 for individuals not at all attached to their regions to 0.52 for those very much attached to their regions, holding other variables at their means. Figure 2, which plots these predicted probabilities with 95 percent confidence intervals, graphically presents this positive influence of inclusive regional attachment on support for European integration. An interesting observation from Figure 2 is the large increase in predicted probabilities of support for European integration from 0.29 for those ‘not really’ attached (category 2) to 0.53 for those ‘somewhat’ attached (category 3) to their region. These findings suggest that, among UK respondents, support for European integration is more likely to increase with inclusive regional attachment, perhaps owing to EU actions that support further devolution in the UK, which gives more voice to the Irish, the Scottish, and the Welsh.

Figure 2.

The effect of inclusive regional attachment on predicted probabilities of support for European integration in the UK.

On the other hand, exclusive regional attachment does not seem to have any statistically significant relationship with support for European integration among UK respondents. As already discussed for EU15 and EU27 respondents, exclusive regionalists in the UK may be hesitant to support the integration process despite its policies that facilitate regional autonomy and empowerment.

Conclusion

Although it has received only paltry attention in public opinion literature, regional attachment helps to explain support for European integration among EU citizens. European integration does not just involve national governments; the EU has facilitated the input of subnational authorities in formulating and implementing its policies. In this article, I found that individuals with strong inclusive regional attachment sentiments, who are usually supportive of regional autonomy, perceive these acts of the EU that strengthen their regions in a positive light. Support for European integration therefore also hinges on the level of inclusive regional attachment among EU citizens.

These findings, however, fail to provide empirical support for this article’s second hypothesis. In all sets of results, exclusive regional attachment does not appear to have any statistically significant relationship with support for European integration. This finding appears puzzling; yet one could potentially argue that exclusive regionalists are ambivalent toward European integration because they may perceive the benefits of the EU in strengthening their regions but remain wary of the potential of the integration process to dampen their exclusivity. This finding also calls for a better measure of exclusive regional attachment that future research should address.

In a national setting, the regionalist sentiments investigated in this article would be interpreted as a threat to national cohesion. The case of Belgium attests to the precarious position in which a state with strong regional identity sentiments finds itself. Within the EU context, however, inclusive regionalist sentiments may not be as threatening. Because of its favorable policies toward regions, the EU may be benefitting from a situation that at first glance appears to threaten the nation-state system in Europe.

These findings, moreover, point to one of the mechanisms of addressing the perceived democratic deficit in the EU. Although supranational bureaucrats in Brussels lack a degree of democratic accountability similar to that enjoyed by national government elites, the EU continues to facilitate the involvement of authorities below national governments that might otherwise feel far removed from the integration process and its resulting implications. Through various EU institutional and legal mechanisms, regions that might have perceived European integration to be yet another level of distant bureaucracy have gained a seat at the EU and are able to exercise an influence that might not have been possible through national channels.

Finally, the findings I report suggest that, although the EU remains intergovernmental in its decision-making, the interests of regions are also gaining some traction. EU actions are tapping into regionalist sentiments by supporting policies of which individuals possessing such sentiments approve. In this regard, the integration process finds support from individuals with strong inclusive regionalist sentiments. Instead of perceiving European integration as a threat, inclusive regionalists, because of EU policies and actions, view the EU as an ally in their quest to gain more autonomy. In effect, the European Union may be taking the initial steps toward evolving into a supranational entity comprised of regions.

Footnotes

Acknowledgements

I would like to thank Maurits van der Veen, Ryan Bakker, Seth Jolly, Shane Singh, Johannes Karreth, Ann Kryzanek, Jonathan Polk, Szymon Stojek, Gerald Schneider, and EUP’s three anonymous reviewers for their helpful comments, feedback, and suggestions on improving previous draft versions of this article. The Web Appendix of this article includes the data and do files, description of variables, descriptive statistics, and additional figures.

Notes

References

Alesina

Devleeschauwer

Easterly

(2003) Fractionalization. Journal of Economic Growth 8: 155–194.

Bache

(1999) The extended gatekeeper: Central government and the implementation of EC regional policy in the UK. Journal of European Public Policy 6(1): 28–45.

Baun

Marek

(2006) Regional policy and decentralization in the Czech Republic. Regional and Federal Studies 16(4): 409–428.

Borghetto

Franchino

(2010) The role of subnational authorities in the implementation of EU directives. Journal of European Public Policy 17(6): 759–780.

Brinegar

Jolly

(2005) Location, location, location: National contextual factors and public support for European integration. European Union Politics 6(2): 155–180.

Brinegar

Jolly

Kitschelt

(2004) Varieties of capitalism and political divides over European integration. In: Marks

Steenbergen

(eds) European Integration and Political Conflict, Cambridge: Cambridge University Press, pp. 62–89.

Brusis

(2010) European Union incentives and regional interest representation in Central and East European countries. Acta Politica 45(1/2): 70–89.

Carey

(2002) Undivided loyalties: Is national identity an obstacle to European integration? European Union Politics 3(4): 387–413.

Cook

(2002) Belgium: A History, New York: Peter Lang Publishers.

10.

De Vries

Edwards

(2009) Taking Europe to its extremes: Extremist parties and public Euroscepticism. Party Politics 15(1): 5–28.

11.

De Vries

Edwards

Tillman

(2010) Clarity of responsibility beyond the pocketbook: How political institutions condition EU issue voting. Comparative Political Studies 44(3): 339–363.

12.

Diez Medrano

(2003) Framing Europe: Attitudes to European Integration in Germany, Spain, and the United Kingdom, Princeton, NJ: Princeton University Press.

13.

Diez Medrano

Gutiérrez

(2001) Nested identities: National and European identity in Spain. Ethnic and Racial Studies 24(5): 753–778.

14.

Downs

(2002) Regionalism in the European Union: Key concepts and project overview. Journal of European Integration 24(3): 171–177.

15.

Eichengreen

Dalton

(2007) Post-Maastricht blues: The transformation of citizen support for European integration, 1973–2004. Acta Politica 42: 128–152.

16.

European Commission (2008) Eurobarometer 69: 2. The Europeans and Globalisation. Brussels: European Commission. Available at: http://ec.europa.eu/public_opinion/archives/eb/eb69/eb69_globalisation_en.pdf (accessed 27 September 2012).

17.

Fitjar RD (2008) Modelling regional identities. PSA British and Territorial Politics Specialist Group Conference, University of Edinburgh, 10–11 January.

18.

Fitjar

(2010) Explaining variation in sub-state regional identities in Western Europe. European Journal of Political Research 49: 522–544.

19.

Gabel

Scheve

(2007) Mixed messages: Part dissent and mass opinion on European integration. European Union Politics 8(1): 37–59.

20.

Garrett

Rodden

(2003) Globalization and fiscal decentralization? In: Kahler

Lake

(eds) Globalization and Governance, Princeton, NJ: Princeton University Press.

21.

Haesly

(2001) Euroskeptics, Europhiles and instrumental Europeans: European attachment in Scotland and Wales. European Union Politics 2(1): 81–102.

22.

Hepburn

(2008) The rise and fall of a ‘Europe of the regions’. Regional and Federal Studies 18(5): 537–555.

23.

Hooghe

(2004) Belgium: Hallowing the center. In: Amoretti

Bermeo

(eds) Federalism, Unitarism, and Territorial Cleavages, Baltimore, MD: Johns Hopkins University Press.

24.

Hooghe

Marks

(1996) ‘Europe with the regions’: Channels of regional representation in the European Union. Publius: The Journal of Federalism 26(1): 73–91.

25.

Hooghe

Marks

(2004) Does identity or economic rationality drive public opinion on European integration? PS: Political Science and Politics 37: 415–420.

26.

Hooghe

Marks

(2005) Calculation, community and cues: Public opinion on European integration. European Union Politics 6(4): 429–443.

27.

Hooghe

Marks

(2008) European Union? West European Politics 31(1–2): 108–129.

28.

Hooghe

Marks

Schakel

(2008) Regional authority in 42 democracies, 1950–2006. A measure and five hypotheses. Regional and Federal Studies 18(2–3): 111–302.

29.

Jeffrey

(2000) Sub-national mobilization and European integration: Does it make any difference? Journal of Common Market Studies 38(1): 1–23.

30.

Jolly

(2007) The Europhile fringe? Regionalist party support for European integration. European Union Politics 8(1): 119–130.

31.

Jolly SK (2009) Strange bedfellows: Public support for the EU among regionalists. Unpublished draft.

32.

Keating

(1998) The New Regionalism in Western Europe: Territorial Restructuring and Political Change, Cheltenham: Edward Elgar.

33.

Kritzinger

(2003) The influence of the nation-state on individual support for the European Union. European Union Politics 4(2): 219–242.

34.

McLaren

(2002) Public support for the European Union: Cost/benefit analysis or perceived cultural threat? Journal of Politics 64: 551–566.

35.

Mandrino

(2008) The Lisbon Treaty and the new powers of regions. European Journal of Law Reform 10(4): 515–534.

36.

Mayda

Rodrik

(2005) Why are some people (and countries) more protectionist than others? European Economic Review 49: 1393–1430.

37.

Paasi

(2009) The resurgence of the ‘regions’ and ‘regional identity’: Theoretical perspectives and empirical observations on regional dynamics in Europe. Review of International Studies 35: 121–146.

38.

Painter

(2002) Multilevel citizenship, identity and regions in contemporary Europe. In: Anderson

(ed.) Transnational Democracy: Political Spaces and Border Crossings, London: Routledge, pp. 93–110.

39.

Risse

(2005) Neofunctionalism, European identity, and the puzzles of European integration. Journal of European Public Policy 12(2): 291–309.

40.

Roht-Arriaza

(1996) The Committee on the Regions and the role of regional governments in the European Union. Hastings International and Comparative Law Review 20: 413–466.

41.

Sánchez-Cuenca

(2000) The political basis of support for European integration. European Union Politics 1(2): 147–171.

42.

Scott

Peterson

Millar

(1994) Subsidiarity: A ‘Europe of the regions’ v. the British constitution? Journal of Common Market Studies 32(1): 47–67.

43.

Spiering

(2004) British Euroscepticism. European Studies 20(1): 127–149.

44.

Steenbergen

Jones

(2002) Modeling multilevel data structures. American Journal of Political Science 46(1): 218–237.

45.

Stockemer

(2012) Citizens’ support for the European Union and participation in European Parliament elections. European Union Politics 13(1): 26–46.

46.

Tatham

(2008) Going solo: Direct regional representation in the European Union. Regional and Federal Studies 18(5): 493–515.

47.

Tatham

(2010) ‘With or without you’? Revisiting territorial state-bypassing in EU interest representation. Journal of European Public Policy 17(1): 76–99.

48.

Tatham

(2012) You do what you have to do? Salience and territorial interest representation in EU environmental affairs. European Union Politics 13(3): 434–450.

49.

Treaty on European Union (1992) Official Journal C 191, 29 July. Available at: http://eur-lex.europa.eu/en/treaties/dat/11992M/htm/11992M.html (accessed 28 September 2012).

50.

Van Kersbergen

(2000) Political allegiance and European integration. European Journal of Political Research 37: 1–17.

51.

Van Kersbergen

Verbeek

(2004) Subsidiarity as a principle of governance in the European Union. Comparative European Politics 2: 142–162.

52.

Verdier

Breen

(2001) Europeanization and globalization: Politics against markets in the European Union. Comparative Political Studies 34(3): 227–262.

53.

Verdonck

Deschouwer

(2003) Patterns and principles of fiscal federalism in Belgium. Regional and Federal Studies 13(4): 91–110.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.19 MB