Jump to ContentJump to Main Navigation
The Europe of ElitesA Study into the Europeanness of Europe's Political and Economic Elites$

Heinrich Best, György Lengyel, and Luca Verzichelli

Print publication date: 2012

Print ISBN-13: 9780199602315

Published to Oxford Scholarship Online: May 2012

DOI: 10.1093/acprof:oso/9780199602315.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 01 December 2020

National elites’ preferences on the Europeanization of policy making

National elites’ preferences on the Europeanization of policy making

Chapter:
(p.67) 4 National elites’ preferences on the Europeanization of policy making
Source:
The Europe of Elites
Author(s):

José Real-Dato

Borbála Göncz

György Lengyel

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780199602315.003.0004

Abstract and Keywords

The chapter focuses on the preferences of national elites towards the Europeanization of specific policy areas as an expression of the multidimensional nature of the forces and conflicts behind the process of European integration. The analysis tests the most important hypotheses considered in the European integration literature, taking into account individual elite characteristics and country-specific contextual factors. Results confirm that explanatory patterns differ depending on the type of policy area. For example, preferences toward Europeanization are stronger in those areas dealing with transnational problems, and on whether opinions refer to the current situation or in the mid-term––elites in the later case being more favourable to Europeanization. Preferences also differ among types of elites, whereby economic elites are more pro-European, and according to country; elites of former state socialist and more Eurosceptic countries, such as Denmark and the United Kingdom, are less likely to approve of delegating national authority to the EU.

Keywords:   political elites, economic elites, Europeanization, EU policy making, European political space, European integration, European Union

4.1 Introduction

European integration can be described as a process of reallocation of policy-making competencies between member states and supranational institutions. Since the Treaty of Rome, and following subsequent treaties and reforms, decision-making processes in many policy areas have moved from the traditional state sovereignty to intergovernmental and supranational arenas (Börzel 2005; Wessels and Kielhorn 1999; Schmitter 1996). Nevertheless, research on citizens’ attitudes towards the Europeanization of specific policy areas has been less frequent than that concerning general support for European integration. Only recently has research on this topic been undertaken, mostly interested in explaining the factors determining public opinion towards the Europeanization of specific policy domains (i.e. Dalton and Eichenberg 1998; Vössing 2005; Eichenberg and Dalton 2007), the structure of the European political space according to specific policy preferences (Gabel and Anderson 2002), and on the role of the outputs of Europeanized policies in fostering a European identity (Kritzinger 2005).

There are fewer studies on elites’ preferences concerning the Europeanization of specific policy domains, mainly as a consequence of the scarcity of data.1 Some exceptions are Wessels and Kielhorn (1999), who analysed the (p.68) driving forces behind the Europeanization of national policy competencies, and the differences in preferences among policy areas and countries, and types of elites; and Hooghe (2003), who compared public and political elites’ preferences on policy Europeanization, and attempted to explain the variation and underpinning logic.

In this chapter, we follow the line of these previous works, believing that the study of specific policy preferences has the virtue of providing ‘a more fine-grained measure’ (Hooghe 2003: 283) of national elites’ views of the European integration process. The reasons behind national elites’ support of the transfer of former national (sub-national) powers to supranational institutions where their influence on decision making is lower or uncertain constitute a major puzzle when studying European integration. Besides this, it is useful to explore whether patterns of support differ according to policy area. These differences can be seen as an expression of the multidimensional nature of the forces and conflicts behind the process of European integration. In this respect, this chapter aims to contribute to the debate on the configuration of the European political space of contestation (Gabel and Anderson 2002; Marks and Steenbergen 2002; Hix 1999).

The elite survey of the IntUne project provides the necessary data with which to explore these questions. In addition, compared to previous studies, it adds new value in three aspects. First, it allows us to compare the views of political elites (national MPs) with those of economic elites, an aspect not dealt with by previous works, and where difference in preferences over Europeanization between the two types of elites would suggest the impact of political authority. Second, we are able to contrast attitudes in the short term with those in the long term in order to look for different time frames concerning the support of policy-making Europeanization. Finally, it enables us to explain patterns of support using not only individual but also country-level variables.

The rest of the chapter is structured as follows. First we offer a description of national elites’ preferences concerning the Europeanization of a number of policy areas covered by the IntUne elite survey. Then, in order to account for such preferences, we review the main theories explaining support of policy-making Europeanization and set a number of propositions, which are subsequently tested and discussed in the corresponding sections. The chapter ends with a brief summary of the main findings in our analysis.

(p.69) 4.2 Descriptive Analysis of Policy-Making Preferences

In this section we briefly examine our dependent variables. Two types of questions in the IntUne elite survey refer to preferences concerning the Europeanization of policy making; one relates to the preferred level of policy making, ranging from the sub-national level to the full allocation of decision-making authority to European Union institutions, with regard to a number of policy areas: namely, unemployment, immigration, environment, fighting crime, health care, and taxation. The second type of question examined relates to the agreement with the full long term (within ten years) Europeanization of taxation, social security, and foreign policy. Despite some limitations, the catalogue of variables covers the spectrum of activities performed by contemporary states—basic sovereignty and security functions (foreign policy, fighting crime, and immigration), economic regulation and distribution (taxation), welfare (health care, social security, and unemployment), and activities concerning post-materialist values (environment). Thus, the preferences of national elites in each of these policy areas may be used as a proxy for the preferences of national elites on each of those general fields of public activity.

National elites’ preferences on the Europeanization of policy making

Figure 4.1. Preferred level of government in policy areas (political and economic elites) (valid percentages)

Note: the valid number of cases (N) by variables are: Environment = 1988; Immigration = 1992; Fight against crime = 1990; Taxation = 1988; Fighting unemployment = 1987; Health care = 1989.

(p.70)

Table 4.1. Factor analysis of level of policy-making variables

All

Political elites

Economic elites

Factor 1

Factor 2

Factor 1

Factor 2

Factor 1

Factor 2

Fighting unemployment

0.030

0.448

0.170

0.431

0.126

0.487

Immigration policy

0.440

0.141

0.454

0.256

0.468

0.282

Environment policy

0.845

−0.129

0.782

0.112

0.782

0.097

Fight against crime

0.396

0.178

0.416

0.325

0.445

0.222

Health care policy

−0.057

0.572

0.087

0.482

0.150

0.566

Taxation

0.105

0.515

0.244

0.484

0.255

0.557

Explained variance

33.415

32.055

35.624

KMO (sig.)

0.734

0.734

0.718

Note: method of extraction: principal axis factoring; method of rotation: oblimin for the whole sample (factors are correlated); varimax for subsamples (factors are uncorrelated).

In order to have a more parsimonious view of the preferred level of authority over policy making, we have recoded the variables and regrouped the original categories, which comprised different combinations of national, regional, and European levels of government, into three categories: (1) authority only at national or sub-national level, (2) shared authority between national/sub-national and EU levels, and (3) full authority at the supranational/EU level (Figure 4.1).2

Looking at the distribution of preferences along these categories, there seems to be an underlying pattern regarding levels of government preferences depending on policy area: national and sub-national levels are clearly preferred in health care, unemployment, and taxation. In contrast, participants clearly show a preference for the Europeanization of environment and immigration policies. Finally, despite the fact that 40.6 per cent prefers the fight against crime to remain under national and/or sub-national authority, the proportion of those preferring shared or full Europeanized control shows that there is a general feeling about the importance of the EU level of government in dealing with this policy area. These results are very much in line with Wessels and Kielhorn’s (1999: 177) hypothesis that the level of Europeanization of a policy area depends on the cross-border character of the problem it addresses—more globalized problems, such as environmental issues and immigration, are more acceptable as areas to be dealt with at the EU level. We examine this in greater detail in the following paragraphs.

Factor analysis (Table 4.1.) confirms this underlying pattern regarding levels of government preferences depending on policy area both on the whole sample and by type of elite. Although the factor solutions are not quite robust, (p.71) factor loadings reveal the same underlying structure. In the first factor, environment policy has the highest loadings, followed by immigration policy, and finally fighting crime. Policies with high loadings on the second factor are health care, taxation, and unemployment.

These patterns of policy-making preferences fit into Wessels and Kielhorn’s (1999) ‘globalization of problems’ hypothesis, and more specifically into the sub-hypothesis about the ‘cross-border’ (transnational) character of problems. Respondents favour Europeanization where a policy is related to a problem requiring cooperation between countries. This is clearly the case with environment and immigration, to which Wessels and Kielhorn also attribute a high cross-border nature (1999: 178). Regarding the fight against crime, recent developments in economic globalization, technologies, and the growing phenomenon of organized crime crossing the borders of member states (profiting from freedom of circulation within the EU) have clearly raised awareness of the need for a more coordinated, even common, approach.

Wessels and Kielhorn also expected that a convergence of problems between countries (what they term problem ‘load’) (1999: 177–8) would lead to a stronger desire for higher Europeanization, although to a lesser extent than in the case of cross-border problems. Our cross-sectional elite data only allows us to state that significant differences in terms of readiness for the Europeanization of problem solving still exist across countries. Furthermore, preferences for the Europeanization of tackling unemployment––one of the policy areas to which Wessels and Kielhorn attributed a high degree of convergence across countries (1999: 178)––are lower than those concerning taxation, a policy area which allegedly has a low ‘load’.

This general pattern of differential preferences between policies concerning transnational problems and the other policy areas is generally reproduced within the countries, although there are variations concerning the intensity of preferences according to policy area (Table 4.2). However, in all countries, the percentages of those who favour some kind of Europeanization (full or shared) in policy areas of ‘transnational’ nature are always higher than those in any of the other policy fields. The only exception is the strong support in France for the Europeanization of taxation, which is even higher than for fighting crime.

Along with these general patterns, there are differences among countries, as can be seen in Table 4.2. For instance, elites in two Baltic countries, Estonia and Lithuania, seem to be more zealous concerning national sovereignty in all policy areas, while French national elites tend to have a more pro-EU standing. Nevertheless, apart from these outstanding cases, there is much variation between countries according to policy area, which suggests that a more sophisticated analysis is needed to shed light on underlying patterns.

Concerning the type of elite, Figure 4.2 shows how economic elites are on the whole slightly more pro-European than political ones in all policy areas (p.72)

Table 4.2. Preferred level of government in policy areas (valid percentages by country)

Immigration

Environment

Fight crime

Unemployment

Health care

Taxation

N/S

S

EU

N/S

S

EU

N/S

S

EU

N/S

S

EU

N/S

S

EU

N/S

S

EU

Austria

28.1

16.7

55.3

16.7

20.2

63.2

35.7

19.1

45.2

50.9

14

35.1

85.2

8.7

6.1

56.1

14.9

28.9

Belgium

17.7

19.4

62.9

14.5

26.6

58.9

26.6

40.3

33.1

63.7

25

11.3

70.2

22.6

7.3

44.7

30.9

24.4

Bulgaria

28.2

56.5

15.3

22.8

66.7

10.6

35.4

57.5

7.1

62.9

33.1

4

54.4

40.8

4.8

60

37.6

2.4

Czech republic

27.9

13.9

58.2

31.1

16.4

52.5

54.5

19

26.4

72.7

10.7

16.5

85.2

6.6

8.2

73

9.8

17.2

Denmark

40

13

47

15

13

72

34

29

37

76.3

14

9.7

85

5

10

75.8

7.1

17.2

Estonia

47.3

15.2

37.5

60.4

17.1

22.5

75.9

13.4

10.7

83

8

8.9

93.8

4.5

1.8

86.6

9.8

3.6

France

5.9

47.1

47.1

7

45.2

47.8

23.3

52.6

24.1

35.3

52.1

12.6

53.9

34.8

11.3

19.3

48.7

31.9

Germany

26.3

13.6

60.2

15.3

14.4

70.3

42.4

19.5

38.1

78

11

11

83.1

7.6

9.3

60.2

12.7

27.1

Greece

17.5

60

22.5

14.2

63.3

22.5

36.1

54.6

9.2

39.3

52.5

8.2

50.4

43.7

5.9

44.9

43.2

11.9

Hungary

38

12.4

49.6

15.6

26.2

58.2

32.8

33.6

33.6

65.3

16.5

18.2

76.9

14

9.1

43

26.4

30.6

Italy

15.9

19.8

64.3

19

23.8

57.1

43.7

27.8

28.6

52.4

17.5

30.2

72.2

15.9

11.9

55.6

22.2

22.2

Lithunia

55.1

2.5

42.4

42.4

8.5

49.2

72

3.4

24.6

82.2

0

17.8

86.4

2.5

11

64.4

6.8

28.8

Poland

31.4

55.1

13.6

20

67.5

12.5

23.3

68.3

8.3

52.1

42

5.9

74.2

23.3

2.5

64.2

35

0.8

Portugal

13.3

16.7

70

10

18.3

71.7

32.5

26.7

40.8

47.5

15.8

36.7

74.2

10.8

15

43.3

15.8

40.8

Spain

25

24.3

50.7

26.4

31.1

42.6

36.6

37.9

25.5

65.5

23.6

10.8

76.2

19

4.8

52.7

24.3

23

Slovakia

35.6

11.9

52.5

24.8

24.8

50.4

45.3

22.2

32.5

87.3

5.9

6.8

88.9

5.1

6

88.1

5.9

5.9

United Kingdom

32.9

55.7

11.4

14.3

51.4

34.3

44.3

50

5.7

65.7

28.6

5.7

85.7

11.4

2.9

83.6

13.4

3

Total

28.2

26.2

45.7

21.9

31.2

46.9

40.6

33.7

25.7

63.2

21.9

14.9

75.8

16.6

7.6

58.7

22

19.3

Note: N/S = ‘Subnational and national level’; S = ‘Shared powers between subnational/European or national/European levels or the three of them’; EU = ‘European level’.

(p.73) .
National elites’ preferences on the Europeanization of policy making

Figure 4.2. Preferred level of government in policy areas by type of elite (valid percentages)

National elites’ preferences on the Europeanization of policy making

Figure 4.3. Preferences on the Europeanization of three policy areas in 10 years (valid percentages)

Note: the valid number of cases (N) by variables are: unified tax system = 1992; Common Social Security = 1995; single foreign policy = 1998.

(p.74)

Table 4.3. Preferences on the Europeanization of three policy areas in 10 years by country (valid percentages)

 

Unified tax system for Europe

Common system of social security

Single foreign policy

Somewhat in favour

Strongly in favour

Somewhat in favour

Strongly in favour

Somewhat in favour

Strongly in favour

Austria

36

28.9

31.6

18.4

39.5

44.7

Belgium

38.5

27

39.7

19.8

37.9

57.3

Bulgaria

34.9

36.5

36.5

44.4

28

64.8

Czech Republic

28.1

9.9

29.5

10.7

43.4

31.1

Denmark

22.2

7.1

33

9

40

43

Estonia

30.9

2.7

50

14.5

41.5

50.9

France

23.8

53.3

41.2

37.8

17.4

74.4

Germany

30.2

32.8

29.1

15.4

36.8

50.4

Greece

27.9

50.8

32

57.4

16.1

80.6

Hungary

44.3

27

36.4

44.6

38.5

55.7

Italy

 

67.5

 

87.9

 

90.4

Lithuania

32.8

26.9

39.2

30

40.8

51.7

Poland

41.3

16.5

32.8

30.3

35.5

42.1

Portugal

50.8

17.8

57.1

26.1

40.8

49.2

Spain

49.3

25.7

61.5

23.6

36.5

56.1

Slovakia

26.1

8.4

31.7

18.3

33.6

50.4

United Kingdom

1.4

 

4.3

1.4

37.1

8.6

Total

31.5

26.9

35.4

29.9

32.8

54.5

(p.75) except for tackling unemployment. However, differences between the two elite groups are only statistically significant in the case of taxation, health care, and environmental policy making

With regard to the second type of question (those referring to the preferences of elites concerning the Europeanization of taxation, social security, and foreign policy within the next ten years), the majority of respondents favour Europeanization in all three policy areas (Figure 4.3). The Europeanization of foreign policy obtains the greatest support among European elites, while the level of consensus on the policy areas of taxation and social security are somewhat lower.

In all, the general pattern seen for short-term preferences in relation to the Europeanization of policy areas is repeated on the long term within countries, and variation between countries is also found (Table 4.3). Here, the case of the United Kingdom is particularly interesting, in that both types of national elites (p.76)

National elites’ preferences on the Europeanization of policy making

Figure 4.4. Preferences on the Europeanization of three policy areas in 10 years by type of elite (valid percentages)

(political and economic) are overwhelmingly against the full Europeanization of social security and taxation, and although somewhat less vehement in their rejection, they are still the elites most against a single foreign policy.

Concerning the type of elite––as with short-term preferences––economic elites are more in favour of the Europeanization within a ten-year term of foreign policy and taxation (Figure 4.4). In contrast, they show a less favourable attitude towards social security, where political elites have a more positive stance. Again, there are differences across countries and between elite groups across the three policy areas, but less so than for short-term preferences.

4.3 Theories Explaining National Elites’ Policy-Making Preferences

The literature on European integration has identified a number of drivers accounting for the process. First, there is the hypothesis that the Europeanization of policy making can be explained as a result of functional needs: policy making should be transferred to the European Union in the hope that it contributes to better problem solving or that the subsequent economies of scale produce more efficient results (Alesina, Angeloni, and Schuknecht 2001). This is the rationale behind the functionalist theory of European integration, and it is consistent with the globalization of problems perspective. Similarly, there is the argument that economic internationalization is a driver for European integration. International economic interdependence undermines national governments’ ability to control economic actors and transactions in their own territory (Schmidt 2002: 18), so that it is logical for national elites (mostly economic elites) to favour Europeanization ‘in order to have the same conditions of market participation with respect to the structure of regulation, incentives, and the like’ (Wessels and Kielhorn 1999: 184), and that this should be mainly in policies directly affecting a country’s economic performance.

An alternative driver to explain policy-making preferences––and one that may be considered to be closely related to the functional explanation––is the instrumental-evaluative argument taken from public opinion research (Gabel 1998; Gabel and Palmer 1995; Eichenberg and Dalton 1993), which is based on the reasoning that people support Europeanization if they can derive direct benefits from the EU. Concerning elites, it is plausible to think that they are more homogeneous as a group than the general public, and that their usually more affluent position makes direct material benefits less important. However, it is possible to apply a ‘sociotropic’ extension of this utilitarian argument in their case (Hooghe and Marks 2005)––elites may (p.77) evaluate Europeanization of policy making based on the effect they perceive it may have on their countries’ welfare, rather than on their own per se. This evaluative judgement may also be extended to the perception of the functioning of European institutions compared to national ones (Rohrschneider 2002; Sánchez-Cuenca 2000).

Political attitudes play an important role in shaping the degree of support for the European Union, and perhaps one of the most important attitudes with regard to EU integration is that of ideology. Certainly, the relationship between ideology and positions taken on EU integration has been widely discussed and is part of the debate about the configuration of the European political space (Marks and Steenbergen 2002; Gabel and Anderson 2002). According to Marks and Steenbergen, four basic standpoints have been distinguished regarding ideology—interpreted in the sense of left–right positioning, ideology is irrelevant to understand anti- or pro-integration positions (a position shared by realist, intergovernmentalist, and neo-functionalist theories on EU integration). In this way, elites’ preferences on Europeanization should not be influenced by their ideological positions. A second stance is that ideology and integration collapse in one dimension (Tsebelis and Garrett 2000), that is, the politization of the European integration debate implies that actors take the dimensions of national debate to the European level. In this case, variation in preferences on the Europeanization of policy areas would be explained by ideological positions. According to the third approach, the European space is formed by two basic unrelated dimensions (Hix and Lord 1997). Thus, domestic conflict over socio-economic issues cannot be extrapolated to the European level, where the dominant dimension is that of national sovereignty, ranging from independence to integration. The ideology dimension would express a cleavage between functional groups, while the national sovereignty dimension involves a cleavage between territorial groups (Marks and Steenbergen 2002: 884). Thus, ideology would not explain variation in preferences on the Europeanization of policy making. Finally, Hooghe and Marks (1999) propose a fourth vision of the European political space. They identify two dimensions: the first runs along the ideological spectrum (left–right), ranging from social democracy to market liberalism; and the second refers to European integration, ranging from nationalism to supranationalism. According to the authors, these dimensions are related, collapsing into a diagonal dimension with its extremes lying between regulated capitalism (encompassing centre-leftist and pro-supranational positions) and neo-liberalism (combining rightist views and an acceptance of European integration limited to minimal regulation––economic and monetary––that allows markets to work more efficiently).

(p.78) When dealing with political ideology it is also important to take into account the extremeness of ideological positions’ (Aspinwall 2002). Thus, centrist voters would tend to be more open to international interdependence and support European integration on an instrumentalist basis while extreme voters are more likely to reject the idea of the EU, but for different reasons: extreme left-wing voters see the EU as too much in favour of a liberal free market, while extreme right-wing voters reject the supranational character of the EU.

Along with ideology, identity also arises as a factor to be taken into account in explaining attitudes towards European integration (Hooghe and Marks 2005, 2008; Risse 2006; Herrmann, Risse, and Brewer 2004; Carey 2002; Diez-Medrano and Gutiérrez 2001). Identity issues have gained strength as a defining axis of political conflict, forming a ‘new politics’ dimension along with other issues (Marks, Hooghe, Nelson, and Edwards 2006). To be more precise, the post-functionalist view claims that identification with territorial communities (national or sub-national) plays a decisive role in issues where economic implications are unclear and where there are strong communal implications (Hooghe and Marks 2008: 13). However, it is not the degree of territorial attachment that is important, but the exclusiveness of the attachment, i.e. the extent to which national or sub-national identities are viewed to be incompatible with European identity (Hooghe and Marks 2005). In this respect, exclusive attachment to national or sub-national territorial levels would be a factor fostering positions against Europeanization. Nevertheless, the proponents of this idea argue that elites’ preferences are not greatly influenced by identification (Hooghe and Marks 2008).

Finally, in explaining elites’ preferences on Europeanization it is also necessary to consider the effect of country-specific institutional and socio-economic configurations, and their mediating influence on other explanatory variables (Brinegar and Jolly 2005; Hooghe and Marks 2005). For example, Vössing (2005) shows how public opinion on EU policy making depends significantly on nationality, behind which are a number of institutional elements that may influence policy positions. For example, Wessels and Kielhorn (1999) have argued that political elites in corporatist countries are not likely to favour Europeanization in policies directly affecting organized interests, because such a process would undermine their position in internal policy making. Besides, and related to the above-mentioned globalization hypothesis, these authors test the influence of a country’s degree of economic internationalization on the level of national elites’ support for policy-making Europeanization. (p.79)

Table 4.4. Theoretical propositions and variables in analysis

Theoretical arguments

Propositions

Variables

Control variables: Gender (male) and age (centred to the overall mean)a

Difference between type of elites

Factors explaining political elites’ support to Europeanization of policy making differ from those of economic elites

Evaluative-pragmatic (outputs)

A positive evaluation of the effect of the European Union in one’s country has a positive influence over preferences on Europeanization of policy making

Country has benefited from EU membership (dummy)

Evaluative-pragmatic (institutions)

Elites favour Europeanization of policy making to a greater extent if they trust EU institutions highly

Trust on EU institutions (Parliament, Commission, and Council of Ministers) (average 0–10 scale)

Ideology (functionalist)

Left–right positions have no significant impact on preferences on Europeanization of policy making

Left (dummy) ideology scale 0 ≤ x 〈 4;

Right (dummy) ideology scale 6 〈 x ≤ 10;

Centre (Ideology scale 4 ≤ x ≤ 6) as category of referenceb

Ideology (Tsebelis)

Left–right positions have a significant effect on preferences on Europeanization of policy making in all policy areas

Idem

Ideology (Hix and Lord)

Left–right positions have no significant effect in those policy areas which are more directly related to the country’s sovereignty functions

Idem

Ideology (Hooghe and Marks)

Neoliberal/regulatory capitalism explains preferences on Europeanization of policy making Left–right positions are secondary

The EU main goal should be to promote: economic competitiveness (dummy) (alternative answer: better social security and both)

Ideology (extremeness)

Individuals with more extremist positions are more against policy-making Europeanization than those with moderate positions

Absolute distance from ideological centre (5)

Identity

Individuals who identify exclusively with national or sub‐national levels are less favourable to Europeanization of policy making in areas with unclear economic implication and/or strong communal implications

Exclusive identification with country (dummy)

Exclusive identification with regionc

Trust in national institutions (country level)

Individuals’ preferences on Europeanization of policy making are negatively related to the country’s level of trust in national political institutions

Country average of economic elites’ average trust on national institutions (national parliament and government, and regional or local government) (0–10 scale)

Evaluative-instrumentalist: economic benefits (country level)

Individuals in countries that are net beneficiaries in the EU will have more favourable positions towards Europeanization of policy making

Operating Budgetary Balance (OBB) as a percentage of GNI in 2007d

Functionalism: economic globalization (country level)

Individuals in countries with a high degree of international economic dependency will have more favourable positions towards Europeanization of policy making

Index of trade integration of goods and services as a percentage of GDPe

Corporatism (country level)

Individuals in countries with strong and centralized economic interest organizations are less favourable to Europeanization in areas directly related to welfare or economic policy

Centralization and coordination of union wage bargaining indexf

Year of EU membership:

–1973–1981

–1986–1995

–2004–2007

(founding members as category of reference)

Notes: a This variable has been centred to the sample mean in order to offer a clearer interpretation.

b We prefer to measure ideology in three categories ‘left’, ‘centre’, and ‘right’ (the latter as a category of reference) rather than using the 0–10 scale measurement, as we do not assume that ideology exerts a linear influence on the explained variables (that is, an individual is not more or less pro-EU whether his/her ideological self-positioning is 1 or 3.5). Besides, it allows us to better assess the specific behaviour within these ideological groups. However, due to the asymmetric distribution of cases (most of them concentrated on the centre-left of the ideological spectrum) we also included the pure ideological scale as a contrast. Results did not show any difference.

c These variables have been calculated as follows: [Attachment to territorial level (0–3)––Attachment to Europe (0–3)]. After obtaining results, a dummy variable was constructed at any territorial level, with 1 = results 〉 0 and 0 = rest of results.

d Operative Budgetary Balance (OBB) is defined as the difference between allocated operating expenditure––i.e. excluding administration––to EU member states and their own resources payments, excluding traditional own resources (European Commission 2008: 80), see 〈Shttp://ec.europa.eu/budget/library/publications/fin_reports/fin_report_07_en.pdf〉 (accessed 22 April 2010).

e Average of imports and exports of the items, goods, and services of the balance of payments divided by GDP. If the index increases over time it means that the country/zone is becoming more integrated within the international economy. Source: Eurostat http://epp.eurostat.ec.europa.eu/tgm/table.do?tab%20=%20table&init%20=%201&plugin%20=%201&language%20=%20en&pcode%20=%20tsier120〉 (accessed 9 November 2009).

f Source: AIAS (2009). For details on the elaboration of the index, see Visser (2009).

(p.80) From the different theoretical arguments set out in this section, we have determined a number of propositions displayed in Table 4.4, where variables used in the analysis are shown. Also, a number of control variables have been included for gender, age, and the year the respondent’s country joined the European Community/European Union.

(p.81) 4.4 Models and Results

In this section we test the previous propositions. We have run separate models for political and economic elites in order to see whether the factors explaining preferences on Europeanization of policy making differ along type of elites. Besides, differences between countries found in descriptive analysis suggest that variations at this higher level of aggregation may influence lower level (individual) positions. Statistically, an adequate treatment of such kind of data requires a multilevel approach (Snijders and Bosker 1999). Previous studies applying this modelling technique (Hooghe and Marks 2005; Brinegar and Jolly 2005; Steenbergen and Jones 2002) have shown that the multilevel character of attitudes towards the EU should not be ignored––higher levels of analysis, typically country-level and party-level data accounted for 14–20 per cent of the variance in EU support in those models.

4.4.1 Short-Term Policy-Making Preferences: Transnational Policy Areas

Tables 4.5 and 4.6 show the results of the multilevel multinomial logistic models for short-term policy-making preferences by type of elite. Here, we will address how they confirm the different theoretical arguments and propositions considered above. Concerning what have been called ‘transnational’ policy issues (environment, immigration, and fighting crime), different patterns explaining the focus of attitudes emerge.

With respect to political elites’ individual characteristics, instrumental‐evaluative arguments (trust in EU institutions) are relevant for explaining preferences for full Europeanization against keeping control over these policy areas at the national/sub-national level. However, the significance of these arguments disappears when considering members of political elites with more moderate preferences (sharing powers between EU and national/regional authorities), where ideology factors are more important. Nevertheless, patterns differ according to policy area. Left–right positions are significant in the case of environmental and fighting crime policies, with leftist positions more prone to support an intermediate Europeanization of these areas. The neoliberal/regulatory dimension, on the other hand, is significant for explaining preferences for control over policies related to immigration and fighting crime. Members of national political elites who think the EU’s main goal should be exclusively to promote economic competitiveness are less likely to delegate authority to the EU level in these areas. Finally, political elites’ preferences concerning environment policy are also significantly influenced by their exclusive identification with their regions–– (p.82)

Table 4.5. Preferences on policy-making Europeanization: ‘transnational’ issues

Environment

Immigration

Fighting crime

Political elites

Economic elites

Political elites

Economic elites

Political elites

Economic elites

Level 1 units

1124

534

1126

533

1127

532

Level 2 units

17

17

17

17

17

17

Condition number

2458.873

2179.483

2165.223

1881.994

2542.094

1904.377

Log likelihood

−1042.320

−481.976

−1062.450

−501.058

−1106.416

−540.180

Cat: Shared

Coef.

 

Coef.

 

Coef.

 

Coef.

 

Coef.

 

Coef.

 

Age

−0.008

 

−0.011

 

−0.001

 

−0.014

 

0.019*

−0.010

 

Male

−0.098

 

−0.947***

−0.326

 

−0.456

 

−0.101

 

0.053

 

Benefited EU

0.720

 

0.971

 

0.550

 

0.034

 

0.412

 

−0.118

 

Trust EU institutions

0.125

 

−0.041

 

0.118

 

0.010

 

0.057

 

−0.063

 

Left

0.553*

−0.398

 

0.164

 

0.021

 

0.711*

0.332

 

Right

−0.580*

0.300

 

−0.335

 

−0.031

 

0.186

 

0.499

 

Extremism

0.054

 

−0.259

 

0.015

 

−0.172

 

−0.122

 

−0.256

 

EU more competitive

−0.292

 

−0.289

 

−0.415*

−0.369

 

−0.351*

−0.424

 

Exclusive national

−0.138

 

0.181

 

−0.125

 

−0.070

 

−0.059

 

−0.274

 

Exclusive regional

−0.584*

0.239

 

−0.432

 

−0.085

 

−0.337

 

−0.146

 

Trust national institutions

−1.044***

−0.688

 

−0.884***

−0.484*

−0.676*

−0.179

 

OBB 2007 per

−0.304

 

−0.211

 

−0.322

 

−0.385

 

−0.307

 

−0.235

 

Trade integration

−0.014

 

0.003*

−0.042

 

0.004

 

−0.005

 

0.002

 

Bargaining centralized

0.081

 

−4.100

 

0.855

 

−6.462***

−0.783

 

−1.694

 

EU member 73–81

0.760

 

0.928

 

0.097

 

0.658

 

0.634

 

0.095

 

EU member 86–95

0.425

 

−0.729

 

0.038

 

0.369

 

0.565

 

0.042

 

EU member 04–07

−0.581

 

−1.098

 

0.291

 

−1.095

 

−0.237

 

−0.202

 

Constant

5.575***

6.900

 

5.361***

6.534***

3.540

 

2.686

 

Cat: EU level

Coef.

 

Coef.

 

Coef.

 

Coef.

 

Coef.

 

Coef.

 

Age

−0.012

 

−0.009*

0.001

 

−0.019

 

0.034***

−0.002

 

Male

0.462*

−0.665

 

0.045

 

−0.275

 

−0.070

 

1.411***

Benefited EU

0.255

 

1.813

 

0.058

 

0.933

 

−0.115

 

−0.010

 

Trust EU institutions

0.133*

0.056

 

0.160***

0.095

 

0.123*

−0.009

 

Left

0.370

 

−1.282***

−0.047

 

−0.110

 

0.628

 

0.819

 

Right

−0.485

 

−0.823

 

−0.297

 

−0.653

 

0.390

 

0.159

 

Extremism

0.071

 

0.303

 

0.085

 

0.178

 

−0.070

 

−0.054

 

EU more competitive

0.150

 

0.256

 

−0.075

 

−0.112

 

−0.140

 

−0.171

 

Exclusive national

−0.011

 

−0.072

 

−0.208

 

−0.313

 

0.168

 

−0.046

 

Exclusive regional

−0.653*

0.145

 

−0.344

 

0.161

 

−0.342

 

−0.787**

Trust national institutions

−0.159

 

−0.733***

−0.074

 

−0.485**

0.210

 

−0.162

 

OBB 2007 per

−0.097

 

−0.049

 

0.143

 

−0.335

 

0.105

 

−0.232

 

Trade integration

0.012

 

0.037*

0.017*

0.033*

0.007

 

0.004

 

Bargaining centralized

−0.169

 

−2.600

 

−0.900

 

−5.100***

0.761

 

4.465

 

EU member 73–81

−0.064

 

1.405*

−1.235**

−0.023

 

−0.936

 

−0.028

 

EU member 86–95

0.071

 

0.254

 

−0.285

 

1.449***

0.035

 

−0.203

 

EU member 04–07

−1.598*

−2.623***

−1.935***

−2.277***

−0.832

 

0.209

 

Constant

0.649

 

3.754

 

0.658

 

3.864*

−2.565

 

−2.345

 

Level 2 variance (standardized estimate)

0.109 (0.078)

0.000 (0.000)

0.129 (0.074)

0.000 (0.000)

0.235 (0.116)

0.112 (0.077)

Note: multilevel multinomial logistic regression. Category of reference: ‘National/subnational level’. Robust standard errors.

* p ≤ 0.05;

** p ≤ 0.01;

*** p ≤ 0.005.

(p.83) (p.84)

Table 4.6. Preferences on policy-making Europeanization: ‘non-transnational’ issues

Fighting unemployment

Taxation

Health care

Political elites

Economic elites

Political elites

Economic elites

Political elites

Level 1 units

1125

 

532

 

1117

 

534

 

1126

 

Level 2 units

17

 

17

 

17

 

17

 

17

 

Condition number

1872.626

 

1129.145

 

2295.274

 

1353.347

 

1591.295

 

Log likelihood

−907.232

 

−447.690

 

−902.204

 

−511.828

 

−639.035

 

Cat: Shared

Coef.

 

Coef.

 

Coef.

 

Coef.

 

Coef.

 

Age

−0.003

 

−0.010

 

−0.005

 

0.000

 

−0.007

 

Male

−0.105

 

−0.134

 

0.103

 

0.235

 

−0.409*

Benefited EU

0.268

 

0.538

 

0.792

 

0.179

 

−0.305

 

Trust EU institutions

0.014

 

0.058

 

0.090

 

0.128

 

0.084

 

Left

0.850***

1.420***

0.641*

0.473

 

0.373

 

Right

−0.228

 

0.548

 

−0.132

 

0.721

 

0.029

 

Extremism

−0.134

 

−0.322

 

−0.092

 

−0.379

 

−0.036

 

EU more competitive

−0.829***

−0.409

 

−0.632*

−0.427

 

−0.673***

Exclusive national

−0.190

 

−0.340

 

−0.051

 

−0.241

 

0.236

 

Exclusive regional

−0.025

 

0.070

 

−0.445

 

0.605

 

−0.486

 

Trust national institutions

−0.427

 

−0.083

 

−0.903*

−0.056

 

−1.000***

OBB 2007 per

−0.035

 

0.225

 

0.254

 

0.257

 

0.111

 

Trade integration

−0.027

 

−0.005

 

0.023

 

−0.020

 

0.005

 

Bargaining centralized

0.449

 

−3.405

 

−4.629

 

−1.761

 

−2.208

 

EU member 73–81

−0.026

 

−0.319

 

−1.558

 

−0.452

 

−0.707

 

EU member_86–95

−0.131

 

−0.513

 

0.358

 

−0.756

 

0.115

 

EU member 04–07

0.066

 

−1.009

 

−3.140***

−0.330

 

−1.909*

Constant

2.272

 

0.897

 

4.667*

0.675

 

5.225***

Cat: EU level

Coef.

 

Coef.

 

Coef.

 

Coef.

 

Coef.

 

Age

−0.012

 

−0.012

 

−0.014

 

−0.005

 

−0.008

 

Male

0.059

 

0.134

 

0.544***

0.687

 

−0.022

 

Benefited EU

−0.621

 

0.479

 

0.206

 

0.806

 

−1.073*

Trust EU institutions

0.045

 

0.113

 

0.078

 

0.201*

−0.123

 

Left

0.537

 

1.014

 

0.681*

−0.111

 

−0.955

 

Right

−0.518

 

−0.444

 

−0.223

 

0.086

 

−0.880

 

Extremism

−0.041

 

0.048

 

−0.009

 

−0.097

 

0.050

 

EU more competitive

−0.229

 

−0.322

 

−0.407*

0.065

 

−0.249

 

Exclusive national

−0.251

 

−0.155

 

−0.146

 

−0.048

 

0.250

 

Exclusive regional

0.170

 

−0.233

 

−0.250

 

0.178

 

−0.790*

Trust national institutions

−0.045

 

−0.308

 

0.144

 

−0.070

 

0.058

 

OBB 2007 per

0.324

 

0.258

 

0.576***

0.516

 

0.698***

Trade integration

−0.004

 

−0.001

 

0.023

 

0.010

 

0.020

 

Bargaining centralized

0.264

 

1.540

 

−2.188

 

−3.508

 

−0.303

 

EU member 73–81

−1.804**

−0.250

 

−2.091***

−0.938

 

−1.514*

EU member 86–95

0.012

 

−0.005

 

−0.072

 

0.133

 

−0.779

 

EU member 04–07

−1.284*

−0.774

 

−2.824***

−2.281*

−2.148***

Constant

−0.297

 

−0.981

 

−2.278

 

−1.122

 

−1.011

 

Level 2 variance (standardized estimate)

0.210 (0.077)

0.000 (0.000)

0.304 (0.094)

0.030 (0.072)

0.227 (0.128)

Note: multilevel multinomial logistic regression. Category of reference: ‘National/subnational level’. Robust standard errors.

* p ≤ 0.05;

** p ≤ 0.01;

*** p ≤ 0.005.

(p.85) (p.86) exclusive identities would reinforce attitudes towards keeping environmental issues in national hands.

Concerning the influence of contextual (country) variables on political elites’ attitudes, there is a significant negative effect in the case of interviewees from former state socialist countries concerning the full Europeanization of environment and immigration policies, and for the United Kingdom, Denmark, and Greece in the case of immigration policy. These models also provide further evidence for the globalization hypothesis, as political elites in countries with a greater level of international embeddedness are more favourable to the full Europeanization of national immigration policy. Another contextual variable––a country’s average level of trust in national political institutions––is significant, but only for intermediate Europeanization preferences. Here, results confirm the instrumentalist proposition, in that political elites in countries with a lower average level of trust in national political institutions are more favourable towards some kind of Europeanization in these transnational policy areas.

For economic elites, preferring shared authority over national/regional control in transnational policy issues has more to do with national contexts than with individual level variables. The only individual level variable having a significant effect is that of gender on environmental policy, where women are more favourable towards Europeanization. Individual level characteristics, however, are important in explaining strong preferences for full Europeanization (transferring all powers to the EU) in transnational policy issues. More specifically, ideology has a significant effect in the case of environmental policy––members of the economic elite who position themselves on the political left are less likely to favour full Europeanization; while exclusive regional identity seems to foster a preference to keep the fight against crime at national or sub-national level.

Some contextual variables also have significant effects. Regarding preferences on shared powers over policy making, there is evidence to support functionalist-pragmatic arguments. Thus, for environmental policy, economic elites in countries with higher trade integration are more likely to favour shared competencies between national/regional levels and the EU. In the case of immigration policy, in countries with higher trust in national political institutions, economic elites are less prone to support intermediate Europeanization. However, in this policy area, the higher the degree of corporatism (centralization and coordination of union wage bargaining) the higher the preference for keeping control over it in national or sub-national hands. For those members of the national economic elite preferring a full transfer of powers to the EU, country-level variables are significant for environment and immigration policy. In both cases, a country’s average level of trust in national institutions has a negative, significant effect, while the degree of trade (p.87) integration of the country increases the likelihood for preferring full Europeanization against exclusive national/regional authority. Top business leaders in countries joining the European Economic Community or the EU in 1986–1995 (Spain, Portugal, and Austria), however, are more likely to prefer complete Europeanization of immigration policy compared to those in founding member states. In contrast, economic elites in former state socialist countries are significantly less likely to delegate environmental and immigration policy to the EU level.

4.4.2 Short-Term Policy-Making Preferences: Non-Transnational Policy Areas

In the case of policy areas dealing with problems without a direct cross-border nature (fighting unemployment, taxation, and health care; see Table 4.6) again, patterns between political and economic elites differ. For political elites, in the case of preferences for shared powers, apart from the significant negative effect of being male on health care policy, ideology is the only significant explanatory factor at the individual level. For the three policy areas, preferences for an intermediate Europeanization are significantly related to the dimension of ‘neo-liberalism–regulation’. Being a member of political elites and believing that the EU should promote economic competitiveness means that supporting the Europeanization of the two policies usually classified as ‘social’ (fighting unemployment and health care), and the intervention of EU decision making into national/regional taxation policy, is less likely. However, this ideological axis does not overrule the traditional ‘left–right’ dimension in the case of unemployment and taxation policies, where MPs with leftist self-positioning are more likely to support shared powers. Both ‘competitive/regulative’ and ‘left–right’ variables are also significant in the part of the model related to full Europeanization preferences in case of taxation policy. For the other policy areas, only in the case of health care we find two individual variables with a significant effect––exclusive regional identity and evaluation of the benefits of the EU for the respondent’s country. In the latter case, the direction of the effect is not as expected theoretically in that the likelihood of preferring full Europeanization of health care policy is lower for those members of the political elite who evaluate the EU positively. It is also interesting that no individual level variable accounts for this group’s preferences to transfer all the powers to the EU level in the case of policies to fight unemployment.

With respect to country level variables, a country’s average level of trust in national institutions again has a negative, significant effect on preferences in taxation and health care issues, where the preference is for intermediate Europeanization. Furthermore, political elites in former state socialist countries are also less likely to favour mixed powers in these areas. In the (p.88) case of those showing a preference for full Europeanization, in the issues of taxation and health care we find for the first time a significant effect of objective economic benefits received from the EU (measured through the OBB), which runs in the expected direction––more benefits would support a greater desire for Europeanization. Besides, MPs in former state socialist countries and those entering the EEC between 1973 and 1981 are less likely to support full EU Europeanization in these areas against keeping authority for policy making exclusively in national/regional hands than founding members.

For economic elites, only models for preferences regarding fighting unemployment and taxation could be computed. In the case of fighting unemployment, only ideology (self-positioning on the left) has a significant, positive effect in explaining preferences on shared powers as opposed to exclusive power for country/regional government. Trust in European institutions at the individual level has a significant effect on preferences on taxation issues––more trust would mean more support for exclusive EU-level authority in this matter. At the country level, being a top business leader in a former state socialist country has a negative effect on a preference for full Europeanization in taxation, while political elites have also a significant preference against transferring decision making to the EU level in health care and social security.

4.4.3 Long-Term Policy-Making Preferences

Table 4.7 includes the results of the multilevel logistic regression models for long-term (ten years) preferences on the complete Europeanization of taxation, social security, and foreign policy.3 Out of these areas, only taxation can be compared with short-term preferences––taking into account the differences in operationalization of variables, while social security only allows an indirect comparison with social policies included in the previous models.

In the case of taxation, ideology continues to be a key explanatory factor of political elites’ long-term preferences. Those with self-positioning on the right and who consider the EU’s main goal to be the promotion of economic competitiveness are less likely to favour the Europeanization of this policy area in the long term. Individual evaluation of EU institutions and the benefits they offer to respondents’ countries also have a significant positive effect. The same patterns concerning individual level variables are again found for the social (p.89) security model, where both the traditional and the neo-liberal–regulation axes are also at work in the same expected direction; the instrumental-evaluative judgements also play a significant role in this model. Institutional and performance evaluation are also significant factors accounting for preferences on the establishment of a single foreign policy. In addition, this is the only policy area where exclusive national identity has a significant effect––in the expected direction: members of the political elite who identify exclusively with their country are also less likely to support full Europeanization in this policy area.

Concerning country-level variables, it appears from models in Table 4.7 that political elites in countries that are net beneficiaries of the EU are most likely to favour Europeanization in the long term for all three policy areas. Also, a higher degree of trade integration (that is, a higher exposure to external economic forces) has a positive significant effect on political elites’ preferences on the Europeanization of foreign policy. Furthermore, just as in the case of short-term policy preferences, political elites in former state socialist countries are systematically less likely to favour Europeanization of the three policy areas. The same applies in the case of taxation and foreign policy for countries joining the EEC between 1973 and 1981.

With respect to economic elites’ long term policy-making preferences, evaluative-pragmatic factors have a positive significant effect in all policy areas. This is most true for trust in EU institutions, while subjective evaluation of benefits is significant only in the case of social security. However, ideology has a significant effect on the long-term policy-making preferences of this area. Here, we find both traditional left–right and neo-liberal–regulation dimensions working in the expected directions. Finally, regarding country-level variables, members of the economic elite in countries with positive OBB are significantly more likely to support the Europeanization of social security and foreign policy within the next ten years. As in other models, however, respondents in former state socialist countries are less prone to support Europeanization in taxation and social security.

4.5 Discussion

As we have shown, the factors behind national elites’ policy preferences present a complex picture, with no uniform pattern along policy areas or type of elite being evident. In this section, however, we will summarize our findings and try to relate them to previous theoretical suggestions.

The first thing that appears is that both pragmatic and ideological factors are key to understanding political (and, sometimes economic) elites’ preferences. Ideology is present as a key factor in explaining political elites’ preferences on shared authority between national/sub-national and European levels in the (p.90)

Table 4.7. Preferences on Europeanization of policy areas in 10 years

Taxation

Social security

Foreign policy

Political elites

Economic elites

Political elites

Economic elites

Political elites

Economic elites

Level 1 units

1111

537

1123

1123

1128

535

Level 2 units

17

17

17

17

17

17

Condition number

849.224

859.947

838.537

838.537

1606.738

911.582

Log likelihood

−627.054

−321.856

−549.591

−549.591

−376.708

−146.534

Cat: In favour

Coef.

Coef.

Coef.

Coef.

Coef.

Coef.

Age

0.020*

−0.006

0.023*

0.023*

0.006

−0.005

Male

0.229*

0.522

−0.032

−0.032

0.246

−0.090

Benefited EU

0.916*

0.986

0.840*

0.840*

1.844***

−0.088

Trust EU institutions

0.138***

0.184*

0.143*

0.143*

0.162*

0.278***

Left

0.209

0.628

0.570*

0.570*

0.509

−0.321

Right

−0.785*

−0.253

−0.113

−0.113

0.151

0.033

Extremism

0.110

−0.004

−0.097

−0.097

−0.079

−0.249

EU more competitive

−0.612***

−0.366

−1.114***

−1.114***

−0.161

0.186

Exclusive national

−0.171

−0.230

−0.099

−0.099

−0.488*

−0.109

Exclusive regional

−0.208

0.049

−0.160

−0.160

−0.212

−0.197

Trust national institutions

−0.161

−0.010

−0.078

−0.078

0.234*

0.156

OBB 2007 per

0.636*

0.543

0.912*

0.912*

0.926***

0.401*

Trade integration

−0.013

−0.003

−0.001

−0.001

0.024**

0.015

Bargaining centralized

1.973

−0.103

−1.953

−1.953

−1.695

−1.160

EU member 73–81

−2.643***

−1.709

−2.456

−2.456

−0.978**

−0.346

EU member 86–95

−0.844

−0.892

−0.336

−0.336

−0.536

0.159

EU member 04–07

−1.196*

−1.836*

−1.715**

−1.715***

−2.220***

−0.714

Constant

−0.148

−0.543

1.490

1.490

−1.560

0.519

Level 2 variance (standardized deviation)

0.227 (0.115)

0.401 (0.310)

0.408 (0.292)

0.408 (0.292)

0.000 (0.000)

0.000 (0.000)

Note: Multilevel logistic regression. Category of reference: ‘Not in favour’. Robust standard errors.

* p ≤ 0.05;

** p ≤ 0.01;

*** p ≤ 0.005.

(p.91) short term, as well as their long-term preferences for taxation and social security. In this respect, results do not show a clear dominance of a particular view of the ideological space. In immigration and health care, the ‘Hooghe-Marksian’ view dominates, while in environment the traditional ‘left–right’ axis is the important dimension. However, in most of the areas both dimensions are significant simultaneously. Explaining why these patterns appear would require an in-depth investigation into the particular characteristics of each policy area, and this is beyond the scope of the present chapter.

Pragmatic-evaluative arguments seem to hold for political elites in the case of strong pro-Europeanization preferences (transnational policy areas and health care policy). Trust in EU institutions, and lack of trust in national institutions, are driving forces of preferences concerning full Europeanization of transnational policy areas. It is also interesting that, in these cases, ideology is not significant, which allows us to conclude that strong Europeanization preferences in the short term are less dependent on the political-ideological debate. In the long term, however, pragmatic-evaluative and ideological arguments are both confirmed in the case of political elites’ preferences over taxation and social security.

Such a mix does not occur in the case of long-term preferences over foreign policy. Related to the discussion of the role of ideology in defining the political space, power over foreign policy is seen as a major state sovereignty function. In this way, and given that it is the only policy area where ideology is not significant at all in explaining political elites’ preferences (as opposed to exclusive national identity, which is), it could be taken to confirm the Hix-Lord’s intergovernmentalist hypothesis.

In contrast to political elites’ preferences, those of economic elites are much less influenced by ideological considerations. For this group, ideology is only significant in explaining short-term preferences for full Europeanization concerning environmental policy, shared Europeanization in unemployment policy, and long-term preferences concerning social security. In turn, pragmatic-evaluative arguments are only confirmed for long-term preferences. In general, the evident explanatory paucity of individual level variables in the models for economic elites, particularly in the short term, clearly raises the question of whether there are other variables, not included in our analysis, which account for variation in preferences.

Our analysis also shows the non-significance of ideological extremism, the low importance of exclusive territorial identity––with the exception of environment, health care, and foreign policy. Extremism’s lack of impact may be related to the fact that we investigated only elite groups. However, results do not suggest that extremist elite members are unable to influence policy preferences of the public.

(p.92) Concerning contextual (country-level) factors, the significant effect in some models of a country’s average level of trust in national political institutions is of particular interest. This partially confirms the Sánchez-Cuenca (2000) hypothesis on the inverse relationship between levels of trust in national and European institutions. Economic globalization arguments also make sense in the case of economic elites where policy issues have a clear transnational nature (environment and immigration), also confirming partially Wessels and Kielhorn’s (1999) cross-border hypothesis (see above). In contrast, these authors’ arguments about the possible role of the internal structure of interest intermediation are solely confirmed for economic elites and in case of immigration policy. Some models reflect a significant effect of the objective benefit countries receive from their membership in the EU. Regarding short-term preferences, this variable is more important for economic elites, while in the long term it also has a significant effect on political elites’ preferences. Finally, our analyses also confirm that the contextual variables of former state socialist states and, to a lesser extent, of those who joined the ECC between 1973 and 1981 (particularly the United Kingdom and Denmark) have negative influence over the supranational policy preferences of these countries’ elites.

4.6 Conclusion

The analysis in the previous pages has shown the complexity inherent to national elites’ preferences on the Europeanization of policy making. Nevertheless, some broad patterns emerge within such complexity. First, preferences differ depending on whether they are projected in the short or long term. In the long term, national elites tend to be more pro-European––considered as a whole, the absolute majority support the idea of single foreign policy, while a relative majority also support the unification of taxation and the social security system. In the short term, however, the picture is slightly different. Thus, those who prefer taxation to be dealt with at the European level (even if authority on this area is shared with national/regional institutions) are a minority. The same occurs with unemployment and health care policies. In this respect, we find the second general pattern: positions in the short term referring to what have been called ‘transnational’ policy issues (namely environment, immigration, and crime) are more pro-European than in the rest of the policy areas. These short-term patterns are in accordance with previous research findings concerning the globalization of social problems and with a functional approach, which takes into account their cross-border character.

A third general pattern is that economic elites are more open to Europeanization than their political counterparts in most policy domains––fighting unemployment and social security are the exceptions.

(p.93) Finally, the three previous general patterns vary along countries and country groups. Thus, in general terms, former state socialist elites––similarly to the Eurosceptic elites of the United Kingdom and Denmark––are less likely to approve of delegating national authority to the EU.

This chapter has also aimed to shed light on the reasons for these regularities. Statistical models show no uniform explanatory pattern along time span, type of policy issue, or type of elite. However, some interesting facts have been detected. Results indicate that political elites’ views are affected by ideological explanatory factors in all policy areas but foreign policy, which seems to adjust to a more pragmatic, intergovernmental logic. However, we have not found a predominance of a particular view of the ideological space in the remaining policy areas. Besides, ideology is mostly important when referring to mild (shared) positions over Europeanization. We have also found room for a significant impact of pragmatic-evaluative factors in political elites’ attitudes. Individual trust in European institutions and/or the average level of trust in national institutions are significant explanatory factors of strong preferences concerning Europeanization of transnational policy issues.

With respect to economic elites, it is more difficult to find significant variables explaining preferences at the individual level, which suggests that further research is needed to identify other factors that could account for this group’s specific preferences concerning the Europeanization of policy making.

Finally, explanatory models have also shown the importance of contextual/country-level factors. Along with others, they clarified––in line with the descriptive findings––the significant negative impact of former state socialist countries on the support for supranational policy solutions.

Notes:

(1) The survey among members of national parliaments conducted in 1996 in eleven EU countries within the umbrella of the Political Representation in Europe research project included a question of whether a number of policy areas (seventeen) should be decided at the national or at the European level (‘European Study of Members of Parliament 1996’ Core questionnaire 〈http://www.wzb.eu/~wessels/Downloads/Quest&Codebooks/CORE-MNP1.pdf〉 (accessed 18 December 2011)). Similarly, the survey conducted by EOS Gallup Europe also in 1996 on behalf of the European Commission among several groups of national elites (‘top decision makers’) in the fifteen member states (Spence 1997) included a question on fourteen policy areas, where respondents had to choose on a ten-point scale whether each policy area should be dealt with ‘exclusively at the national or regional level’ or ‘exclusively at the European level’.

(2) We have also considered as missing values the ‘none of them’ responses.

(3) For the dependent variable, it would be more appropriate to use ordinal logistic regression. However, our data did not adjust to the parallel regression assumption, so for reasons of simplicity we opted to dichotomize the dependent variable. Thus, the ‘In favour’ category includes former ‘Strongly in favour’ and ‘Somewhat in favour’ categories, while the category ‘Not in favour’ refers to the rest of the responses.