Inequality,state ownership and the European Union: How economic context and economic ideology shape support for the European Union

Abstract

This article investigates the extent to which economic ideology affects people's support for European Union integration and how this is conditioned by economic context. We argue that people on the economic left who live in a country with conditions of high income inequality and little state ownership will support European integration, because more integration would move public policy in a left-wing direction. By contrast, people on the left who live in a country with conditions of low income inequality and widespread public ownership are likely to be eurosceptic, as further integration would result in a more right-wing public policy. We empirically confirm our hypotheses and discuss the implications for European Union democracy.

Keywords

Economic context economic ideology public support for European Union integration

Introduction

The formation of public opinion on European Union (EU) integration is likely to be influenced by the distinct processes of deepening and widening. The initial dominance of economic explanations in the opinion formation literature has, as a result of deepening integration and the pooling of sovereignty in areas other than economics (especially since the Maastricht Treaty), been challenged by an increased focus on explanations based on identity and culture (Eichenberg and Dalton, 2007; Gabel, 1998a, 1998b; Hooghe and Marks, 2005; McLaren, 2002). While the process of deepening may affect the relative importance of economic factors in shaping opinion, we suggest that the process of widening also has implications for how economic factors influence opinion formation. This is due to the increased economic heterogeneity resulting from enlargement. Specifically, we focus on the normative ideological aspects of economics and argue that the variation in the economic context of the 27 member states of the EU conditions how people's economic ideology drives opinions about EU integration.

Our point of departure regarding the context dependency of the ideological underpinnings of citizens' attitudes to where power should lie is illustrated in Figure 1. Public policy in a potentially highly powerful federal context (such as the USA or the EU) is characterised, in the example, as being slightly on the economic right. A moderately right-wing citizen would prefer power to stay at state level if they live in region A, an economically free market context (Figure 1(a)). An ideologically identically positioned citizen would prefer power to move to the federal level if they live in region B, an economically interventionist context (Figure 1(b)).

Figure 1.

How economic ideology affects preferences over regional/federal power dependent on the regional economic context. (a) Free market region A, (b) interventionist region B.

This line of thinking has informed a small amount of previous work on attitudes to regional integration in the context of the pre-2004 enlargement 15-member EU. Brinegar and Jolly (2005), Brinegar et al. (2004) and Ray (2004) found, using different measures and approaches, that economic context moderated the influence of economic beliefs on attitudes to EU integration. We contribute to this area of scholarship in three distinct ways. First we expand the context. While previous analyses only looked at the 15 member states of the mid-1990s, we examine the much increased contextual variation across the 27 member states of 2009. In a recent review of differences in economic context within the EU, Höpner and Schäfer (2012: 14) emphasise the increased economic contextual diversity in the EU caused by enlargement. They state that ‘EU member states' production and welfare regimes have become more heterogeneous with each round of enlargement’ and specifically highlight that ‘the EU-27 is notably more diverse than the old EU-15’ (Höpner and Schäfer, 2012: 12).

Second, rather than relying on respondent self-placement on a general ‘left–right’ scale (Brinegar and Jolly, 2005; Ray, 2004) or on measures of one particular aspect of economic ideology (Brinegar et al., 2004), we directly measure the two main underlying aspects of economic interventionist-liberal ideology. On the one hand, we have a measure of people's support for state intervention regarding inequality: should the government redistribute wealth? On the other hand, we have a measure of people's support for state intervention regarding ownership: should the government run major services and industries?

Third, and crucially, we operationalise context in a more effective manner than previous research. We directly measure objective differences between countries in terms of inequality and the state control of industry and services. This avoids grouping countries together into ‘types of capitalism’ blocs (Brinegar and Jolly, 2005; Brinegar et al., 2004), which assumes homogeneity but masks considerable cross country variation, or measuring country differences on only one economic dimension (Ray, 2004). Our objective country level measures directly match the measures of ideology at the individual level, and we are able to connect actual income inequality with preferences for redistribution and actual state intervention with preferences for public ownership.

Our findings demonstrate that variation in economic context plays a significant role in conditioning the relationship between economic ideological beliefs and attitudes to the integration process. People on the economic left (with pro-equality and pro-state ownership views) who live in a country with ‘rightist’ economic conditions (of high income inequality and little state ownership) support European integration because more integration would likely move public policy to the left. By contrast, people on the left who live in a country with ‘leftist’ economic conditions of low income inequality and widespread public ownership favour less integration because more integration would result in more right-wing public policy. Given variation in economic context, it is only by theorising and testing conditionality that we can make sense of how ideology underpins EU attitudes.

What determines support for the EU?

Early work on the individual level economic determinants of EU support focused on citizens' levels of human capital and found that highly skilled/educated people (who were well placed to benefit from the market opportunities that flowed from the EU integration process) were broadly supportive of the EU, while people with lower levels of education and fewer marketable skills (who occupied a more vulnerable economic position) were more sceptical about EU integration (Gabel, 1998a, 1998b; Gabel and Palmer, 1995). Some have argued that human capital effects may be conditional upon context. Specifically, the economic liberalisation associated with integration should mean further specialisation amongst countries depending on their levels of cheap capital or cheap labour. In wealthy states where unskilled workers are less important to production than is capital, it is those unskilled workers that lose out from integration (Brinegar and Jolly, 2005; Hooghe and Marks, 2005; Scheve, 2000).

Attitudes to EU integration in former Soviet states have also been studied using this instrumental ‘winners and losers’ model. People are characterised as either ‘winning’ or ‘losing’ from the transition from the communist command economy to the post-communist free market economy. Those who gained from free market reforms are assumed to favour EU integration because the EU is seen as a vehicle for guaranteeing, and furthering, the free market process. Those who did not benefit from the same reforms are likely to oppose integration and the furtherance of the free market that integration would be perceived to deliver. This model has been operationalised by a number of authors using people's subjective evaluations of the economy as a proxy for gaining or losing (Christin, 2005; Cichowski, 2000; Tucker et al., 2002). Garry and Tilley (2009a) sought to integrate contextual effects into the winners and losers model. Using pooled data from eastern and western states, they found that retrospective economic evaluations were in fact a greater predictor of EU attitudes in the East than in the West, consistent with this post-transition winners and losers model.

In addition to human capital and evaluations of economic performance, a further aspect of economics that may drive opinion on EU integration relates to citizens' normative or ideological beliefs, an area that some authors believe has been relatively neglected. For example, Rohrschneider and Whitefield (2006) bemoan the focus on the instrumental rather than the ideological in investigations of attitude formation in the former communist states (see also Loveless, 2010; Rohrschneider and Whitefield, 2004). They argue that because former Soviet states have so recently undergone a transition to the free market, a normative consensus on the merits of the free market does not exist. Given the strong ideological value-based divisions that exist on the question of free market versus interventionist economic management, it is this diversity of opinion that drives attitudes to the EU. Pro-free market views are likely to be strongly associated with pro-EU integration views in former communist states because pro-free market citizens see the EU as a vehicle to cement the transition to capitalism. The EU is seen as a means of furthering free market reforms and lessening the chances of a reversion to a more interventionist, and authoritarian, state. It is the specific post-transition context that leads to a strong free market pro-EU linkage:

citizens may apply their views about the domestic economy, which to a considerable degree represent socialist views, in appraising various aspects of the emerging international markets … if one rejects the idea of a market economy, one is not likely to endorse economic integration that is justified by a desire to strengthen market relationships. (Rohrschneider and Whitefield, 2006: 148)

Rohrschneider and Whitefield (2006: 147) therefore argue that ‘western’ instrumentality-based theories of attitude formation may be ill suited to the post-communist context as Eastern Europeans are ‘much more likely to make judgements about integration based on their underlying economic values than on expected material payoffs’.

This incorporation of economic values and economic context into modelling the determinants of attitudes to integration is also consistent with earlier work in the old 15 member states (Brinegar and Jolly, 2005; Brinegar et al., 2004; Ray, 2004). Ray (2004) focused on one contextual feature of member states – welfare provision – and one general ideological measure, taken from citizens' self-placement on a left–right scale. He found a positive relationship between being ‘right wing’ and being supportive of integration among citizens living in states with generous welfare provision, and a negative relationship between ‘right wing’ and EU support in less generous welfare states. Hence, right-wing citizens were in favour of European integration when they believed that integration would move the status quo in a right-wing direction; when right-wing citizens thought that integration would move the status quo in a left-wing direction they were opposed to integration.

One of the limitations of Ray's findings is that the ‘left–right’ self-placement scale is a less than ideal measure of economic ideology as general ideological self-positioning may be a function of a range of influencing factors in addition to interventionist versus free market views (Inglehart and Klingemann, 1976; Knutsen, 1997). Brinegar et al. (2004) use a more sophisticated operationalisation of economic beliefs, generating a summed scale from a battery of questions relating to citizens' attitudes towards state provision of social welfare. Regarding contextual variation, Brinegar et al. divide the EU 15 into three groups on a 1–3 scale. Britain was allocated a score of 3 as the most right-wing (residual, liberal welfare state) context, Scandinavian countries (Sweden, Denmark and Finland) were allocated a score of 1 as the most left-wing (social democratic welfare state) context and all other member states were allocated a score of 2 (whether they were conservative Christian democratic welfare state contexts or other states (economic laggards)). This three-point context scale thus aimed to capture variation in ‘varieties of capitalism’ and was interacted with citizens' views on welfare in a model predicting attitudes to integration (as measured by a battery of items generating an ‘overall view of integration’ index). This interaction was statistically significant, suggesting a contextual conditioning effect. The limitation of this approach is that the contextual variation essentially differentiates Britain from Scandinavia, which is problematic when trying to capture variation in an EU of 27 member states. Moreover, by focusing on only one specific aspect of economic policy (welfare provision), Brinegar et al. do not directly measure people's underlying ideology about the role of government in society.

The Brinegar et al. analysis is replicated by Brinegar and Jolly (2005) using a slightly different model. Instead of measuring citizens' attitudes to welfare, like Ray (2004) they use citizens' self-placement on a left–right scale and also find a statistically significant interaction effect. Yet this measure of ideology is, in many ways, equally problematic. Not only do conceptions of what left and right are vary from person to person in idiosyncratic ways,¹ but crucially they vary from country to country. Building on the work of Inglehart and Klingemann (1976), Knutsen (1997) shows that both the partisan and the value-based component of left–right self-placement is affected by societal development and the fragmentation of the political system. Similarly, Piurko et al. (2011) find distinct relationships between different types of ideological values and left–right self-placement in particular country contexts (namely liberal versus traditional versus post-communist states).

These articles by Brinegar et al. (2004), Brinegar and Jolly (2005) and Ray (2004) provide the departure point for our analysis. We seek to build upon this work and provide as comprehensive a test as possible for the argument that economic context conditions the influence of economic ideology on EU support. In Table 1 we seek to clarify how our approach is distinct from these three articles. First, we almost double the amount of variation in economic context, analysing 27 rather than 15 member states (see the discussion regarding the increase in economic heterogeneity resulting from the post-2004 enlargement in Höpner and Schäfer, 2012).

Table 1.

Existing research compared to our suggested approach.

Authors	Dependent variable	Predictor variables	Contextual variables	Countries	Year
Ray (2004)	Multi-item scale	Left–right self-placement	Social protection spending as % of GDP	15	1996
Brineger et al. (2004)	Multi-item scale	Multi-item pro–anti-welfare spending scale	Type of welfare regime	15	1994
Brinegar and Jolly (2005)	Multi-item scale	Left–right self-placement	Type of welfare regime	15	1996
Suggested approach	Multi-item scale	Measure of attitudes to reducing income/wealth inequalities Measure of attitudes to state involvement in the economy	Levels of income inequality Levels of public sector employment	27	2009

Second, in this previous work there is a clear tension between operationalising economic ideology with a specific focus on one aspect of ideology, for example welfare provision (Brinegar et al., 2004), or aiming to measure economic ideology in a general way by using left–right self-placement. Here, we seek a general measure of economic ideology that captures variation in citizens' beliefs stemming from a leftist preference for government interventionism to a rightist preference for a more free market approach. However, we eschew the use of the left–right self-placement scale given the difficulties of interpreting its content, especially across context (Knutsen, 1997). Instead, we seek to directly measure the two main substantive areas of economic interventionist versus free market beliefs. The first relates to equality and welfare provision, whilst the second relates to ownership of services and industry. Building upon earlier work by Rokeach (1973), Evans et al. (1996: 97–98) emphasise the distinct nature of the economic equality dimension and the economic ownership dimension. The need to keep the market and inequality economic ideology dimensions separate in the context of investigating attitudes to the EU is also emphasized by Loveless (2010). Measuring these substantive areas distinctly provides valuable analytical leverage regarding context conditionality as we can operationalise economic context in two distinct ways that map directly onto these two aspects of economic ideology, hence allowing two different tests of the general argument that economic context conditions the impact of economic ideology on EU attitudes.

The suggested causal mechanism at play is the same for both aspects of economic ideology. Specifically, we suggest – following the logic of Brinegar et al. (2004), Brinegar and Jolly (2005) and Ray (2004) – that a highly integrated EU would be characterised as having a single economic policy that would be uniformly implemented across the EU. By definition, this would alter the status quo in each member state. Depending on which member state they live in, citizens would have reasonable expectations as to how this potential EU economic policy would be different from (and thus move) their existing member state position. Citizens in very free market member states could reasonably expect that an impending integrated EU would implement a common EU policy that would be more interventionist that the previous status quo. Conversely, citizens in more interventionist states could reasonably expect a potential EU regime to be more free market than the previous status quo. In short, a potential common EU economic policy would likely be approximately at the centre point of the current member state positions.²

Below we set out our twin hypotheses. The first relates citizen views of equality with actual levels of income equality in society. The second relates citizen views of state intervention with actual levels of state intervention.

H1: Pro-economic equality views in relatively economically equal states lead to an anti-EU integration position (and pro-economic equality views in relatively unequal states lead to a pro-EU integration position).

H2: Pro-state ownership views in relatively interventionist states lead to an anti-EU integration position (and pro-state ownership views in relatively free market states lead to a pro-EU integration position).

Data and measurements

We use data from the 2009 round of the European election study.³ Our dependent variable is support for the EU. There are a number of different ways to operationalise this. Different authors have used the image of the EU (Christin, 2005), measures of whether EU membership is seen as a good or a bad thing (Anderson, 1998; Carey, 2002; Eichenberg and Dalton, 1993, 2007), attitudes towards further integration (Garry and Tilley, 2009b; Inglehart, 1970), measures that combine the desired and perceived speed of integration (Brinegar and Jolly, 2005; Brinegar et al., 2004) and also composite measures that include many of these aspects of liking or loathing for the EU (Gabel, 1998a; Gabel and Palmer, 1995; Garry and Tilley, 2009a; Hooghe and Marks, 2005; McLaren, 2002; Ray, 2004). Similarly to these last analyses we use a composite measure, and hence avoid reliance upon a single item operationalisation. The dependent variable is a factor score from a principal components factor analysis of several different items that tap support for the EU. The precise questions are:

Generally speaking, do you think that [Britain's] membership of the European Union is a good thing, a bad thing, or neither good nor bad?

Some say European unification should be pushed further. Others say it already has gone too far. What is your opinion? Please indicate your views using a scale from 0 to 10, where 0 means unification ‘has already gone too far’ and 10 means it ‘should be pushed further’. What number on this scale best describes your position?

You trust the institutions of the European Union? Do you ‘strongly agree’, ‘agree’, ‘neither agree nor disagree’, ‘disagree’ or ‘strongly disagree’?

How much confidence do you have that decisions made by the European Union will be in the interest of [Britain]? A great deal of confidence, a fair amount, not very much or no confidence at all?

These four items were recoded so that higher scores indicated a pro-EU position and low scores an anti-EU position. We treat them as interval level variables. Principal components factor analysis using the correlation matrix produces only one factor with an eigenvalue above 1 (the alpha score for the four items when standardised is 0.69), and we treat this factor as a generalised measure of support for the EU. The exact way in which we measure EU support makes little difference to the results presented here, so we have chosen this relatively simple measure that includes different types of support.⁴

To measure citizens' economic ideology we use two questions, one measuring opinion on the redistribution of income and wealth and the other tapping views on state intervention in industry and service provision (both with Likert response scales).

Income and wealth should be redistributed towards ordinary people. Do you ‘strongly agree’, ‘agree’, ‘neither agree nor disagree’, ‘disagree’ or ‘strongly disagree’?

Major public services and industries ought to be in state ownership. Do you ‘strongly agree’, ‘agree’, ‘neither agree nor disagree’, ‘disagree’ or ‘strongly disagree’?

It is important to realise that these items are only correlated at 0.2 and can thus be clearly distinguished from one another.⁵ This supports our theoretical departure point that economic ideology consists of at least two distinct, but related, components: one relating to the government's role in producing income equality in society and one relating to the role of government as an owner of industry and services.

In terms of macro-economic context we also identify two facets of a country's economic situation. The first relates to income inequality, which we measure using the Gini coefficient for 2008 and 2009. Since the survey was conducted in the middle of 2009, we average the values across 2008 and 2009. The results we present here are not affected by picking one (or other) of the two years. The average Gini coefficient varies from 0.23 (in Slovenia) to 0.38 (in Latvia) with a mean of 0.30 (with a higher Gini score implying greater income inequality). The second contextual variable relates to state involvement in the economy, which we measure using government spending on wages as a proportion of Gross Domestic Product (GDP) within a country. This avoids conflating redistribution via the tax and welfare system with state intervention in the economy as our measure directly captures the proportion of the workforce tied to state employment.⁶ Again, we average the 2008 and 2009 figures. The lowest score for this measure is in Slovakia (0.07), the highest in Denmark (0.18) and the mean is 0.12. Full details of these measures for every country are included in the online appendix (Table A2).

Results

In Table 2 we report the results of three multi-level linear regression analyses. In each model the dependent variable is support for European integration as measured by the factor score discussed earlier. In model 1 a range of control variables liable to be related to both EU support and economic ideology are included, as well as the two economic context variables. The results relating to the control variables are in the expected direction (consistent with previous research): a higher social class, higher level of education, positive economic perceptions and a non-exclusive national identity predict support for the EU (Gabel, 1998b; Hooghe and Marks, 2005). Economic context on its own is not really a useful predictor, however. Greater equality and greater state involvement in the economy tend to be associated with greater support for the EU, but both effects are weak and neither is distinguishable from zero at conventional levels of statistical significance.

Table 2.

Multi-level linear regression models predicting support for the EU.

	Model 1		Model 2		Model 3
	B	SE	B	SE	B	SE
No occupation (reference)	–
Upper salariat class	0.02	0.02	0.02	0.02	0.02	0.02
Lower salariat class	0.01	0.03	0.01	0.03	0.01	0.03
Routine non-manual class	−0.02	0.02	−0.03	0.02	−0.03	0.02
Petit bourgeois class	−0.07*	0.03	−0.07**	0.03	−0.08**	0.03
Skilled working class	−0.06*	0.03	−0.07*	0.03	−0.07**	0.03
Unskilled working class	−0.10**	0.02	−0.09**	0.02	−0.09**	0.02
Male	−0.02	0.01	−0.02	0.01	−0.02	0.01
Aged 18–30 (reference)	‐		‐		‐
Aged 31–40	−0.08**	0.02	−0.08**	0.02	−0.08**	0.02
Aged 41–50	−0.08**	0.02	−0.08**	0.02	−0.08**	0.02
Aged 51–60	−0.07**	0.02	−0.06**	0.02	−0.06**	0.02
Aged 61–70	−0.01	0.02	−0.00	0.02	−0.00	0.02
Aged 70+	0.09**	0.02	0.10**	0.02	0.10**	0.02
Foreign parents	−0.03	0.02	−0.03	0.02	−0.03	0.02
Years of education post 15	0.03**	0.00	0.03**	0.00	0.03**	0.00
Economic perceptions (1–3)	0.14**	0.01	0.14**	0.01	0.14**	0.01
Exclusive national identity	−0.58**	0.01	−0.57**	0.01	−0.57**	0.01
Gini coefficient 08–09	0.31	1.41	0.43	1.44	1.76	1.48
% of GDP spent on govt wages 08–09	−2.31	2.25	−2.36	2.30	−4.17	2.40
Against redistribution (1–5)			0.02**	0.01	0.17**	0.04
Against public ownership (1–5)			0.04**	0.01	−0.03	0.02
Gini*redistribution					−0.54**	0.14
Govt wages*public ownership					0.59**	0.20

Intercept	0.37	0.50	0.20	0.51	0.02	0.53

* p<.05 **p<.01.

Source: The 2009 round of the European Election Study (EES). Individual N = 20,891, country N = 27.

In model 2, the two economic ideology variables are included. Both opposition to redistribution and antipathy to public ownership predict support for the EU. That is, in aggregate across the EU, those on the left are less favourable to the EU than those on the right. Both effects are statistically significant and have a non-trivial magnitude. Since the dependent variable is a factor score, it has a mean of zero and a standard deviation of 1. We would predict that someone who gave the two most right-wing answers to the two ideology items would score nearly ¼ of a point more, i.e. ¼ of a standard deviation more, on that EU support factor score than someone who gave the two most left-wing answers. This effect is consistent with previous research characterising the EU as a right of centre project (for example Gabel, 1998a, 1998b; Rohrschneider and Whitefield, 2006; Scharpf, 2010).

Our main theoretical interest is in the interaction between economic context and the two economic ideology variables, however. Appropriate interactions are included in model 3 to test our hypotheses.⁷ We find that support for income redistribution is very strongly associated with opposition to EU integration in the most equal states, but that this effect reverses in the least equal states where it is those on the right that oppose EU integration. We find a very similar, though not identical, pattern in relation to public ownership. Here, in all countries opposition to public ownership is positively associated with support for integration, but this effect is essentially zero in those countries with the least involved state and is much stronger in countries with widespread state intervention.

The interaction effects from model 3 are illustrated in Figure 2. Here we show the predicted effects of moving from the left to the right for the two different ideology measures in different economic contexts. Figure 2(a) does this for redistribution and inequality. The black dots on this graph thus represent the difference between someone that is strongly in favour of redistribution and someone that is strongly against redistribution in two different contexts. The lines around the dots are 95% confidence intervals. The first context is one of high income equality (a Gini score of 0.23) and the second context is one of high income inequality (a Gini score of 0.38). Here we can clearly see that the context really does matter for how ideology affects support for the EU project. In countries that are relatively equal, it is people on the right that are cheerleaders for the EU, but in countries that are relatively unequal it is those on the left that like the EU.⁸ In other words, a person's ideology matters, but the country that they find themselves in affects how it matters.

Figure 2.

The effects of moving from a left ideological stance to a right ideological stance as a predictor of support for EU integration, conditional upon economic context. (a) Redistribution, (b) public ownership. Note: These figures show the predicted changes in support for the EU for (a) someone that moves from a leftist position on redistribution to a rightist position on redistribution in a country with a relatively equal income distribution (Gini score of 0.23) and a similar change for someone in a country with a relatively unequal income distribution (Gini score of 0.38), and (b) someone that moves from a leftist position on public ownership to a rightist position on public ownership in a country with relatively high state employment (18% of GDP expended on government wages) and a similar change for someone in a country with low state employment (7% of GDP given over to government wages).

Figure 2(b) tells a similar story with regard to public ownership. Here the contexts are of high state employment (18% of GDP expended on government wages) and low state employment (7% of GDP given over to government wages). In countries with high state involvement in the economy, people on the right of the public ownership debate are considerably more supportive of the EU project than people on the left. Interestingly, we do not see the reverse of this in countries with low levels of state employees. Rather, in those countries there is simply a much smaller difference between those who support public ownership and those who do not.⁹

Putting both of these effects together shows their impact. According to our results, in a country with very low levels of inequality (Gini of 0.23) and very high levels of state economic intervention (18% of GDP spent on government wages), people on the right are considerably more likely to support EU integration. In this context, model 3 predicts that someone who is opposed to public ownership and redistribution will score over half a standard deviation more on the EU support scale compared to someone who is supportive of public ownership and redistribution. By contrast, in a country with very high levels of inequality (Gini of 0.38) and very low levels of state intervention (7% of GDP spent on government wages) the effect of ideology on attitudes towards the EU will be relatively minimal; in fact, moving from the left to the right on both issues will actually reduce support for the EU overall.

Discussion

Increasing the power of a federation (regional authority) increases the likelihood of its associated public policy position being implemented. This is likely to change the policy position in any given sub-element of the federation. The direction of the change will depend upon the status quo position relative to the federation's position. Citizens may take into account this likely shift in public policy, and specifically the direction of this shift, when formulating their views on whether more federal/EU power, at the expense of regional/national power, is a good thing. We demonstrate here that national context conditions the manner in which citizens' ideological preferences drive views on EU power. Our analysis confirms and extends previous research. Brinegar and Jolly (2005), Brinegar et al. (2004) and Ray (2004) found that the influence of citizens' ideological views on their opinions about the EU was conditional on the type of economic context the citizen lived in. Our analysis provides a more comprehensive treatment than previous work by expanding the context considerably, improving the measures of ideology, improving the measures of context and more effectively linking context and ideology.

There are a number of implications of our findings. First, the conditioning effect that we find is likely to become more important as the enlargement process continues. Highlighting the increased economic context differences that are a consequence of enlargement, Höpner and Schäfer (2012: 14) point out that ‘if the Balkan states and possibly Turkey or Ukraine entered the EU, disparities would further increase’. This would likely lead to greater heterogeneity in the kind of economic context variables that we focus on, economic inequality and state ownership, and consequently significant conditioning effects on the impact of economic ideology on EU views. Second, while our analysis has focused on a snapshot of the conditional influence of economic ideology on EU attitudes, changes over time in economic contexts could be important in addition to variation between countries. This dynamic interpretation is consistent with work by Evans and Butt (2007) who, in the case of Britain, track the changes over time in the relationship between economic attitudes and EU attitudes. They find that in the early 1980s economic left-wing views are related to an anti-EU position, but this relationship disappeared as the UK moved in a free market direction over the course of the 1980s and 1990s. It is also possible that the ongoing financial/Eurozone crisis may result in new kinds of variation across economic context. Heterogeneity regarding the type of financial relationship each member state has with the EU (for instance bailout states, bailout exiting states and non-bailout states) may, in addition to the inequality and ownership dimensions we focus on, play an important conditioning role.

Finally, our findings have important implications for European democracy and specifically the alleged EU ‘democratic deficit’. If one advocates enhancing EU democracy via strong pan-European political parties that offer voters clear agreed programmes/manifestos at European Parliament election time (see Hobolt and Tilley, 2014 for a critical discussion of this), our results are problematic. Our analysis highlights the fact that parties with common economic policies across the EU may not have common policies on EU integration. Economic context heterogeneity militates against a common shared position emerging on EU integration and this is likely to prevent parties from different countries forming alliances that would allow them to ‘represent not merely fragmented national interests, but also coalitions of interests that cut across national boundaries’ (Hobolt and Tilley, 2014: 153). One might infer from this that direct, rather than representative, democracy may have more potential in terms of facilitating a flourishing EU democratic space.¹⁰ Either way, what remains clear is that while economic ideology is an obvious source of communality between voters and parties in different countries, those same ideological stances will continue to generate differences between ostensibly ideologically similar parties and voters on the issue of EU integration as long as there are differences between countries in their approach to the free market.

Footnotes

Notes

References

Anderson

(1998) When in doubt, use proxies: Attitudes toward domestic politics and support for European integration. Comparative Political Studies 31: 569–601.

Brinegar

Jolly

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

Brinegar

Jolly

Kitschelt

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

Steenbergen

(eds) European Integration and Political Conflict, New York: Cambridge University Press, pp. 62–92.

Carey

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

Christin

(2005) Economic and political basis of attitudes towards the EU in central and east European countries in the 1990s. European Union Politics 6: 29–57.

Cichowski

(2000) Western dreams, eastern realities: Support for the European Union in central and eastern Europe. Comparative Political Studies 33: 1243–1278.

Eichenberg

Dalton

(1993) Europeans and the European community: The dynamics of public support for European integration. International Organisation 47: 507–534.

Eichenberg

Dalton

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

Evans

Butt

(2007) Explaining change in British public opinion on the EU: Top down or bottom up? Acta Politica 42: 173–190.

10.

Evans

Heath

Lalljee

(1996) Measuring left-right and liberal-authoritarian values in the British electorate. British Journal of Sociology 47: 93–112.

11.

Gabel

(1998a) Public support for European integration: An empirical test of five theories. Journal of Politics 60: 333–354.

12.

Gabel

(1998b) Interests and Integration: Market Liberalization, Public Opinion and the European Union, Ann Arbor: University of Michigan Press.

13.

Gabel

Palmer

(1995) Understanding variation in public support for European integration. European Journal of Political Research 27: 3–19.

14.

Garry

(2013) Direct democracy and regional integration: Citizens' perceptions of treaty implications and the Irish reversal on Lisbon. European Journal of Political Research 52: 94–118. .

15.

Garry

Tilley

(2009a) Attitudes towards European integration: Investigating east-west heterogeneity. Journal of European Integration 31: 537–549.

16.

Garry

Tilley

(2009b) The macro economic factors conditioning the impact of identity on attitudes towards the EU. European Union Politics 10: 361–379. .

17.

Hobolt

Tilley

(2014) Blaming Europe? Responsibility Without Accountability in the European Union, Oxford: Oxford University Press.

18.

Hooghe

Marks

(2005) Calculation, community, and cues: Public opinion on European integration. European Union Politics 6: 419–443.

19.

Höpner

Schäfer

(2012) Integration among unequals: How the heterogeneity of European varieties of capitalism shapes the social and democratic potential of the EU. Max Planck Institute for the Study of Societies Discussion Paper 12/5. http://www.mpifg.de/pu/mpifg_dp/dp12-5.pdf .

20.

Inglehart

(1970) Cognitive mobilisation and European identity. Comparative Politics 3: 45–70.

21.

Inglehart

Klingemann

(1976) Partisan identification, ideological preference and the left-right dimension among western mass publics. In: Budge

Crewe

(eds) Party Identification and Beyond, London: Wiley, pp. 243–276.

22.

Knutsen

(1997) The partisan and the value-based components of left-right self-placement: A comparative study. International Political Science Review 18: 191–225.

23.

Loveless

(2010) Agreeing in principle: Utilitarianism and economic values as support for the European Union in central and eastern Europe. Journal of Common Market Studies 48: 1083–1106.

24.

McLaren

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

25.

Piurko

Schwartz

Davidov

(2011) Basic personal values and the meaning of left-right political orientations in 20 countries. Political Psychology 32: 537–561.

26.

Ray

(2004) Don't rock the boat: Expectations, fears, and opposition to EU level policy making. In: Marks

Steenbergen

(eds) European Integration and Political Conflict, New York: Cambridge University Press, pp. 62–92.

27.

Rohrschneider

Whitefield

(2004) Support for foreign ownership and integration in Eastern Europe: Economic interests, ideological commitments, and democratic contexts. Comparative Political Studies 37: 313–339.

28.

Rohrschneider

Whitefield

(2006) Political parties, public opinion and European integration in post-communist countries. European Union Politics 7: 141–160.

29.

Rokeach

(1973) The Nature of Human Values, New York: Free Press.

30.

Scharpf

(2010) The asymmetry of European integration, or why the EU cannot be a “social market economy”. Socio-Economic Review 8: 211–250.

31.

Scheve

(2000) Comparative context and public preferences over regional economic integration. In:Paper presented at the annual meeting of the American Political Science Association,, Washington, DC, 31 August–3 September Available at: http://www.yale-university.com/leitner/resources/docs/2000-06.pdf.

32.

Tucker

Pacek

Berinsky

(2002) Transitional winners and losers: Attitudes toward EU membership in post-communist countries. American Journal of Political Science 46: 557–571.

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.

4.55 MB

0.00 MB

0.02 MB