‘We No Longer Love You, But We Don’t Want To Leave You’: The Eurozone Crisis and Popular Euroscepticism in Greece

Abstract This article analyses whether and how public opinion towards the European Union (EU) in Greece has changed in the context of the current Eurozone crisis. It provides the first detailed treatment of how the crisis has affected citizens’ views in a traditionally pro-European member state. It examines whether public opinion has become more Eurosceptic and which societal groups have changed their views and in what direction. It uses data from Eurobarometer surveys conducted before and during the current crisis. Unsurprisingly, the findings show that negative sentiment towards the EU has increased across all social groups in recent years. However, we find a paradox of a decline in general support for the EU and an increase in support for the Euro. In a country seen as traditionally pro-European, Greek public opinion has fallen out of love with the EU, but it clearly does not want to leave the Eurozone or renounce membership altogether.


Introduction
This paper analyses trends in public support for the European Union (EU) in Greece, the Eurozone member most severely affected by the current economic crisis. It provides the first detailed understanding of how citizens' attitudes towards the EU have evolved in response to the ongoing economic crisis and related developments in domestic politics. There has been recurrent speculation about Greece leaving the Eurozone. Domestically, the crisis has caused much political controversy, electoral volatility and civil strife, affecting political parties, voting behaviour and the institutions of governance. According to Eurobarometer data, prior to the crisis, the Greek public was among the most pro-integrationist in the EU. Subsequently, in 2007-11, Greece became the member state with the highest increase in negative attitudes towards EU membership, as noted by Serricchio, Tsakatika and Quaglia (2013). Despite this rather startling shift, there has been little analysis of how Greek public opinion may have changed in response to the crisis. Have particular societal groups changed their views or has there been a more general 'sea-change' in attitudes towards European integration within Greek society? Has discontent with the EU led to a rise in 'hard' Euroscepticism (Szczerbiak and Taggart 2000), implying rejection of the integration project and of national membership? These are questions of critical significance, not only for the future of Greek-EU relations and Greece's status as a long-standing pro-integration member state, but also because of their wider implications for the impact of economic crisis on support for European integration.
Using data from the bi-annual Eurobarometer surveys conducted before and during the crisis, the present article analyses general support for the EU and attitudes towards the single currency. The paper contributes to research into party-based and popular Euroscepticism and the role of the EU issue in the member states of southern Europe (Costa Lobo and Magalhã es 2011;Ruiz Jiménez and de Haro 2011;Llamazares and Gramacho 2007;Mavris 2004;Quaglia 2011;Verney 2011aVerney , 2011b and to recent analysis of public attitudes in the context of the Eurozone crisis (Serricchio et al. 2013). It also contributes to the wider public opinion literature on explaining general support for the EU (Gabel 1998;Hooghe andMarks 2004, 2005;McLaren 2006). Beyond this, the paper aims to deepen our understanding of the impact of the crisis on the EU member states and of how this may affect the future prospects of the integration project itself.
The paper is structured as follows. First, it sets out the wider context of Greece and the Eurozone crisis. Second, it examines how public opinion towards the EU has changed over time in the context of the crisis. Third, it looks more closely at whether and how the opinions of different social groups towards the EU have changed. Fourth, it sets out the multivariate analysis of Greek attitudes towards the EU using the most recently available survey data. Fifth, it presents the main findings from the multivariate analysis. Finally, it concludes the analysis.

Greece and the Eurozone Crisis: The Re-emergence of European Integration as a Contested Issue
The Greek crisis began in October 2009 with the admission by the newly elected Socialist government that Greece was running an unsustainably high budget deficit and public debt. The public revelations concerning the true state of the Greek economy and the manifest reluctance of the EU leadership to provide guarantees against the sovereign default of a Eurozone member triggered a crisis of confidence in the financial markets which rapidly developed into a sovereign debt crisis.
After Greece was definitively excluded from international financial markets in the Spring of 2010, the EU finally intervened, offering a joint EU/IMF loan in May 2010, followed by a second bailout accompanied by private sector sovereign-debt restructuring in March 2012, and then another minor debt 'haircut' in November 2012. The bailouts were granted in exchange for a programme of drastic deficit reduction, imposing severe austerity, and of deep structural reforms, including the reduction of the minimum wage and the rolling back of labour rights. The international creditors were unambiguous about the loss of Greek economic sovereignty. Implementation of the programme was monitored on a regular basis by the "troika", consisting of representatives of the European Commission (EC), the International Monetary Fund (IMF) and the European Central Bank (ECB), who regularly threatened to recommend the withholding of loan instalments if the Greek government did not meet its targets.
For the Greek public, the EU thus became inextricably linked with a fiscal adjustment programme and a loss of sovereignty. As the pain of the deepening crisis and the drastic reduction of state provision were increasingly felt across Greek society, the initial relief of avoiding default and a disorderly departure from the Eurozone and possibly even the EU was increasingly replaced by the feeling that the effects of these measures represented an unfair and indiscriminate punishment, particularly targeting vulnerable population groups such as the young unemployed and elderly pensioners. The EU role in the design and implementation of the Greek adjustment programme meant that disaffection with the latter seemed likely to produce a eurosceptic response.
Such a response could reflect a crisis of 'output legitimacy', given that the EU can no longer guarantee prosperity and growth for its citizens but instead has become involved in the delivery of harsh austerity. The crisis has also brought a crisis of 'input legitimacy'. In the era of the EU/IMF bailouts, citizens in Greece have become increasingly aware that they can no longer influence public policy through traditional forms of political participation, such as voting in national elections, because the main policy directions are laid out in the agreements with the international lenders and cannot be changed. This is in line with Peter Mair's (2011, 15) concluding remarks in his paper discussing the impact of the crisis on political representation in Ireland: that 'this signals the onset of failure of representation and democracy without choices' (see also Krastev 2002).
The realisation that the EU matters a lot could trigger a positive process whereby citizens demand to become more actively involved in debates on the future of the integration process, the content of EU policies, their ideological direction and their impact on questions of equity and solidarity within their societies. However, it could also trigger a negative response, possibly leading even to a popular rejection of European integration itself. In order to see what is actually happening in Greece, let us now turn to the data.

Greece and The EU: The Changing Face of Public Opinion Over Time
Greece is one of the member states traditionally labelled as 'pro-European', and this is evident in relation to public opinion from the time series data presented in Figures 1 and 2. Trend data are shown for public opinion in Greece and for the EU as a whole (reflecting the number of member states at each point in time), based on two long-standing questions which have been asked regularly in Eurobarometer surveys. Notwithstanding some fluctuation of opinion, it is clear that before the crisis, there were generally high levels of support for Greece's EU membership, expressed as agreement with the statements that Greece has benefitted from membership and that membership has been 'a good thing'. When comparing the time series, it is also evident that there has been greater support for membership in Greece than across the EU as a whole.   Next, we turn to the period of crisis and examine the rise in negative sentiment. Figure 3 uses various indicators of public opinion, including the two Eurobarometer questions used in Figures 1 and 2, but this time focusing on the eurosceptic responses, as well as a further three questions designed to measure general attitudes towards the EU. Given that some of these questions have not featured in Eurobarometer surveys over an extended time period, we report data for the period 2003-2012 (or the most recently available data). For each indicator, we report the proportions holding negative views: Those who say Greece has not benefitted from EU membership. Those who say Greece's membership is a bad thing. Those who tend not to trust the EU. Those who have a very or fairly negative image of the EU. Those who think that the EU is going in the wrong direction.
The trend is very clear. During the first year after the collapse of Lehman Brothers, the general view in Greece was that the international crisis would not really affect the country, while it was felt that participation in the euro offered Greece protection, especially in comparison to the sharp decline suffered by the Swedish kroner. This is reflected in our negative indicators, which all show a small decline in Greek negative sentiment during this period. This is followed by a sharp jump in all cases between Autumn 2009 and Spring 2010 -the period identified earlier in this article as the real beginning of the crisis as far as Greek public opinion was concerned. Subsequently, all indicators show increasing levels of negative sentiment, climbing steadily.
Specifically, in a period of three years, on the three indicators measuring general attitudes towards the EU, the proportion of Greek respondents holding a negative image of the EU almost tripled (from 14% in November 2009 to 49% in November 2012), while those tending not to trust the EU and those believing the EU was going in the wrong direction more than Figure 3. Indicators of negative sentiment in Greece towards the EU Source: Eurobarometer surveys doubled (from 38 to 81% and from 31 to 74%, respectively). On the latter two indicators, the overwhelming majority of Eurobarometer's Greek sample now has a negative stance. The two indicators measuring attitudes towards national membership of the EU were somewhat controversially discontinued by Eurobarometer after the Spring 2011 survey. However, in the 18-month period separating the latter from the beginning of the crisis, those regarding EU membership as 'a bad thing' jumped from 13% to 33%. This made Greece in Spring 2011 the EU member state with the highest proportion of respondents giving a negative answer to this question, marginally ahead of the traditionally eurosceptic UK. Meanwhile, between November 2009 and May 2011, those Greek citizens who considered the country had not benefitted from membership doubled from 25 to 50%, the second highest proportion in the EU (this time behind the UK).
Does this mean that three decades after accession, the crisis had turned the Greek population into 'hard' eurosceptics (Szczerbiak and Taggart 2000)?
The data do not support such an interpretation. While all our indicators showed a sharp upward trend, they also display significant quantitative differences. In May 2011 (the last occasion on which it is possible to compare the two indicators), while almost three-quarters of the Greek sample agreed the EU was going in the wrong direction, only one-third concurred with the statement that EU membership was 'a bad thing'. It is also worth noting that on the membership question, while in May 2011 Greece had the highest level of negative responses in the EU (33%), the 'hard' Eurosceptics were still outnumbered by those who regarded EU membership as 'a good thing' (38%). This was not the case in the UK, where those with a positive evaluation of EU membership (26%) were significantly fewer than the 'hard' Eurosceptics (32%). What the Eurobarometer data suggest, therefore, is that while the Greek population has clearly lost its former enthusiasm for the EU, it has moved in a Euro-critical rather than a Euro-rejectionist direction.
Perhaps the most striking indication of this is provided by the issue of the single currency, one of the EU's flagship initiatives. During the economic crisis, the very real prospect of 'Grexit' -a probably disorderly Greek departure from the Eurozone -meant that this was the issue on which the country was facing an immediate 'in' or 'out' dilemma. Figure 4 charts longer-term attitudes towards the single currency, covering a period of more than two decades from 1990-2012. The data are based on questions whose wording has altered in successive Eurobarometer surveys to reflect the various stages of development of EMU and the single currency. The figure displays public support in Greece over time. Generally, levels of support in Greece have been high over time, though they have fluctuated, in particular falling after the introduction of the Euro. Support for the euro, though, has risen again since the beginning of the crisis, providing an interesting counterpoint to the direction of public opinion evident for the various indicators shown in Figure 3. Compared to the EU average (not shown here), the single currency was more popular in Greece during the 1990s, less popular once the euro was introduced but has once again become more popular since the beginning of the international economic crisis. In the most recent survey for which data are available (Autumn 2012), 65% of the Greek sample were in favour of the euro compared to 53% across the EU.
The examination of the longitudinal data thus suggests that the majority of Greek public opinion has lost confidence in the EU and disagrees with its current direction, but clearly wants to remain within the Eurozone: an attitude that could be described as 'We no longer love you, but we don't want to leave you'. The next section examines the changing dynamics of opinion across social groups. , 2003-2007 and 2007-2011 This section focuses on group attitudes for three Eurobarometer questions at three points in time, using data from surveys conducted in 2003, 2007 and 2011. The questions utilised in this section are the following: whether EU membership is a good thing, a bad thing or neither; whether the country has   benefitted or not from EU membership; and whether for or against a single currency. Table 1 shows the overall distribution of opinion in Greece for these three questions and Table 2 presents changes in opinion for 2003-2007 and 2007-2011 for societal groups classified by sex, age, education and occupational status. Table 2 reports the changes in negative sentiment:

Greece and the EU: Changes in Group Attitudes
Those who think that membership is a bad thing; Those who think their country has not benefitted from membership; and Those who are against the single currency.
Looking first at Table 1, we can see that the overall proportion who believe their country has not benefitted barely changed between 2003 and 2007 but then rose substantially between 2007 and 2011 (by 32.9 percentage points). Looking at the breakdown for the various social categories, we can see that, while there was mixed response from groups in the 2003-2007 period, there was a uniformly strong shift in negative sentiment in the latter period. All groups evince large percentage point increases in those who perceive that Greece has 'not benefitted' from membership. Looking at attitudes towards the single currency, there was a strong shift in a negative direction between 2003 and 2007, again occurring across all social and economic groups. The proportion against the single currency rose from 33.8% in 2003 to around half in 2007 (49.4%). This coincides with the introduction of the euro and can be explained by the latter's association with inflation. However, the opposite occurred between 2007 and 2011, with a substantial decline in negative sentiment to 20.6%. In both periods, the extent of the attitudinal change varies across social and economic groups, but the direction of change is always the same. Of the three indicators, attitudes towards the single currency therefore display the greatest contrast between the two periods captured in Table 2. Even during a period -2007-2011 -when the public mood was clearly growing dissatisfied with Greece's membership of the EU, support for one of the EU's flagship projects was considerably strengthened and, moreover, outpaced support across the EU as a whole.
Data from the most recently released EB survey (Autumn 2013) confirm strong support for the single currency (62.0% in Greece compared to 52.0% across the EU as a whole). But the negative sentiment towards the EU clearly remains: 54.0% have a negative image of the EU; 72.0% think it is going in the wrong direction; and 77.0% tend The Eurozone Crisis and Popular Euroscepticism 255 not to trust the EU (European Commission 2013). This suggests that the crisis seems to have convinced Greeks that the single currency offers a safe haven in the storm. Thus, in Greece, the crisis has clearly produced a crisis of confidence in the EU; however, this does not extend to rejection of European integration. This interpretation is further underlined by a question only recently included in EB surveys, asking whether member states would be better off outside the EU: in response, a majority in Greece disagreed (57.0%), with 38.0% agreeing and just 5.0% unsure (European Commission 2013).

Greece and the EU: Multivariate Analysis of Public Attitudes
This section introduces the data source and measurement of dependent and independent variables before presenting and discussing the findings from the model estimations.

Data Source and Dependent Variables
The data for this analysis come from three surveys.   Table 3 (indicating the change in percentage points between the two time points). For all analyses, the dependent variables are measured so that the Eurosceptic responses are scored as 1 and pro-EU or neutral responses are scored as 0: Those who say their country has not benefitted from membership are scored as 1 while those who say it has benefitted are scored 0.
Those who say membership is a bad thing are scored as 1 and those who respond a good thing or neither good nor bad are scored as 0.
Those who are against the single currency are scored as 1 and those who are in favour of it are scored as 0.
We use binary logistic regression analysis, suitable for dichotomous dependent variables. Missing data are excluded from the analysis.

Independent Variables
The independent variables used here are informed by the main theoretical approaches used in the wider literature on public opinion towards the EU. These approaches relate to economic interests, group identities and political cues (Hooghe andMarks 2004, 2005;McLaren 2006). Survey data permitting, we operationalise a range of independent variables which reflect insights from the economic interests and political cues approaches. Due to the lack of suitable measures for group identities or attachments being available in all EB surveys and since we aim to use equivalent model specifications for the multivariate analyses of the 2007 and 2011 surveys, we do not examine their impact in this paper. It has been well established that citizens' economic interests and material circumstances are important determinants of their views towards the integration process (Hooghe andMarks 2004, 2005;McLaren 2006). In particular, economic integration -and major projects such as the Single Market and EMU -has varying consequences for citizens in the member states, leading to winners and losers from a country's EU membership. According to existing studies of public attitudes towards the EU, those with higher levels of education or who are in higher-level occupations are more likely to benefit from economic integration (Gabel 1998;Hooghe 2007;McLaren 2006), including particular policies such as the single currency.
As well as 'objective' indicators of socio-economic location 'subjective' assessments of personal and national economic conditions have also been shown to be consequential for attitudes on EU-related issues (Anderson 1998;Gabel and Whitten 1997;Hooghe and Marks 2005). Positive appraisals of past or future economic performance or conditions are thought to underpin support for the EU and its policies. As de Vreese et al. observe: 'the "subjective utilitarian model" suggests that European integration is supported if perceptions and future evaluations of the economy are positive ' (2008, 513).
We use both objective and subjective indicators of socio-economic circumstances. The former is measured, first, by the age at which an individual finished their education (measured as a series of dummy variables: aged 15 and under; 16-19 years; 20 and over; still in education). It is measured, second, by occupational status (dummies for self-employed, in manual employment, in non-manual employment or not currently working -a student, retired, unemployed or a home person). We use retrospective assessment of the national economy as our subjective indicator of socio-economic circumstances. Higher scores represent more positive assessments.
While it may be a powerful determining factor, public opinion towards the EU is not only influenced by individuals' economic interests. Indeed earlier 'accounts of attitudes towards European integration resemble the old adage of "where you stand determines where you sit", and are almost exclusively concerned with economic utilitarianism' (McLaren 2006, 31). However, public opinion is also shaped by the domestic political context of member states. Increasingly, research into public opinion and the EU has found that citizens rely on cues and proxies rooted in domestic politics to form opinions about the integration process (Anderson 1998;Sá nchez-Cuenca 2000). Indeed, because of the complex and multifaceted nature of multilevel governance within the EU, 'one would expect domestic politics to shape public views on European integration' (Hooghe and Marks 2005, 425). While, unfortunately, measures of party support are not available in the surveys used here, we can employ measures of frequency of political discussion or level of interest in politics. We also examine trust in governing institutions -specifically, whether respondents tend to trust or tend not to trust the national parliament and government. Finally, in order to reduce the chances of omitted variable bias, all models employ socio-demographic controls for gender (scored as 1 if male and 0 if female) and age (a continuous variable). A summary of the independent variables is provided in the Appendix.

Results and Discussion
The binary logistic regression results for the 2007 survey are shown in Table 4 and for the 2011/12 surveys in Table 5. In contrast to the earlier analysis of change in social group attitudes, this section reports two sets of cross-sectional analyses. The same model specification is used for the three dependent variables, to permit comparison of the effects of the independent variables across models and time points. First, looking at the results reported in Table 4 for the 2007 survey it is clear that gender has a consistent impact in all three models. Compared to women, men are less likely to think Greece does not benefit from membership or is a bad thing, and less likely to be against the single currency. Age is only significant in the third model (single currency), where older people are less likely to be against the Euro. In terms of the objective indicators of socio-economic circumstances,  the results are rather patchy, with both education and occupation having more effect in the model of attitudes towards the single currency. Here, those who left education aged 16-19 years or 20 and over, as well as those still in education, are less likely to be against the Euro compared to those who left education aged 15 or under. In relation to occupational status, those in manual employment are more likely to be against the Euro compared to those not currently in work. However, the occupational status dummies have no effects in the first two models (not benefit from membership and membership is a bad thing).
Turning to the attitudinal variables in Table 4, hese have collectively larger impact and the directions of the relationships are generally the same across the different specifications of the dependent variables. Higher life satisfaction has consistent effects: it is negatively related to thinking Greece has not benefitted from membership, believing membership is a bad thing and being against the single currency. Frequency of discussion of politics similarly has a consistent and negatively signed relationship with the three dependent variables. Greater discussion of politics leads to less likelihood of holding anti-EU views and being opposed to the Euro. Economic perceptions also have significant effects in all three models: positive economic appraisals of the national economy are negatively related to anti-EU attitudes and being opposed to the Euro. It appears then that the wider economic context is shaping the attitudes of Greek citizens towards both membership of the EU and their country's involvement in the Euro. Finally, it is clear that political attitudes rooted in the domestic context -which can act as 'cues' or 'proxies' are shaping public opinion on membership and the single currency. Specifically, those who tend to trust national political institutions -the government and parliament -are less likely to hold unfavourable assessments of EU membership and to be against the Euro. These findings confirm insights from the political cues approach.
Turning to the results for the Spring 2011 and Spring 2012 surveys reported in Table 5, we can see that there is a mixed performance for the socio-demographic and socio-economic variables. In particular, education has stronger effects in Table 5 than in Table 4 and is a more potent explanatory factor than is occupational status. Generally, those who left aged 20 years and over, or who are still in some form of education, are less likely to hold negative views of membership and the Euro compared to those who finished education at an earlier age. This is also the case for those who finished education aged 16-18 years, but only for opinion towards the single currency. The occupational status variables have no significant impact in the model of attitudes towards the Euro. In the models of general support for membership, manual workers are more likely to think that membership is a bad thing and those in non-manual employment are more likely to think that the country has not benefitted from being in the EU, but these are otherwise isolated effects as occupational status performs poorly overall. Age again has a significant effect on opinion towards the single currency in 2012, with older people less likely to be opposed.
As with the results from Table 4, all of the attitudinal variables -with the exception of subjective economic perceptions -have consistent and significant effects on general support for the EU. Evaluations of the national economy are only significant in the first model, relating to views as to whether Greece has benefitted from its membership. Again, positive retrospective assessments of the national economic situation are negatively related to unfavourable views of the EU. In contrast to Table 4, however, this variable has no significant impact in the other two models. Next, general life satisfaction again is related to less likelihood of holding negative views of the EU, as was evident in Table 4. With the exception of trust in government in the model assessing attitudes towards the single currency, the measures of trust in political institutions have significant effects in all models, again supporting the political 'cues' approach to explaining public support for European integration. Those who tend to trust either parliament or the government is less likely to hold unfavourable views of EU membership and the Euro.
Taken together, the findings for evaluations of national political institutions support one set of findings from the political 'cues' approach (Anderson 1998;Sá nchez-Cuenca 2000). Specifically, we find that positive attitudes towards the national parliament and government are less likely to be related to opposition to the EU and single currency, rather than the opposite relationship, whereby some previous studies have found that negative views of domestic institutions are related to greater support for the EU and its institutions (Sá nchez-Cuenca 2000). Also, there are similar effects for subjective economic perceptions across models, which are again in keeping with established findings from the wider literature. That is, positive appraisals of the national economy are less likely to be related to negative evaluations of the EU and the Euro (Gabel and Whitten 1997). Finally, in relation to objective economic interests, although the results are less impressive and consistent overall, when they are significant we find that those groups whom we would expect to be more secure in their economic circumstances -precisely those groups thought to have the human capital to take advantage of the opportunities offered by membership and key elements, such as the Single Market and its economic freedoms -are less likely to be unfavourable towards the EU and the Euro (Gabel 1998). The results here for individuals' economic circumstances and public attitudes towards the EU are interesting in the light of Roth et al.'s findings that, amongst the four member states at the centre of the crisis -Greece, Spain, Portugal and Ireland -there has been a more pronounced fall than across the European Area-12 -in public trust in the ECB, which has been driven by increasing unemployment rates (n.d.).

Conclusion
For the majority of Greek citizens, the EU was always positively associated with democracy, economic prosperity and a move away from the inherent weaknesses of national politics. The handling of the Eurozone crisis has changed this. The empirical analyses reported in this paper shed important new light on the response of Greek public opinion in the context of the current crisis. In particular, the detailed single-country focus (the 'microscopic' approach) complements broader cross-national research (the 'telescopic' approach) on public opinion towards the EU in the context of the crisis, which also finds changes in political support across member states (Braun and Tausendpfund n.d.). The traditional picture of Greece as a 'pro-European' member state, a characteristic it shared with other southern European countries, has changed during the period of economic crisis. The longitudinal evidence examined here show that negative evaluations of the EU increased sharply between 2007 and 2011 in contrast to the steady high levels of pro-European sentiment during the previous two decades. Our breakdown of opinion by social groups shows that this shift in negative opinion is a 'sea change' affecting society as a whole, with all social groups sharing in it.
The fact that public opinion in Greece has turned more Eurosceptic is hardly surprising. What is surprising though is that the decline in general EU support is accompanied by increase in support for the euro. Support for the single currency has increased between 2007 and 2011 and is now higher in Greece than the EU as a whole. This is also confirmed by the most recently released EB survey (Autumn 2013). What can account for the coexistence of these contradictory opinions? We argue that this picture is closely linked to the wider debate about the change from a 'permissive consensus' to a 'constraining dissensus; and the fact that European integration has become more politicised in the member states (Hooghe and Marks, 2009). Increase in public disenchantment with the EU is often seen as indication of a rejection of European integration and an obstacle to further integration (Tosun et al. n.d.). We argue that the picture in Greece supports a different interpretation to that of 'constraining dissensus'. Instead of Greek citizens wanting less Europe, their critical attitudes represent a move away from the norm of consensus governance and depoliticisation of the content of EU policies to one of conflict and politicisation. The majority of Greek people seem to adopt a pragmatic approach where they are aware that being in the Euro and the EU is the realistic alternative despite the pain of austerity policies domestically. However, they are also signalling their discontent with the lack of effective solutions from Europe in terms of outputs.
Future research should investigate the links between the positions of political parties and the opinions of their supporters. This is particularly important given the changing political and electoral landscape in Greeceincluding the increasing prominence of minor parties with Eurosceptic platforms -within the context of the unfolding Eurozone crisis. Such research should be accompanied by an analysis of how the EU issue in the context of the crisis has changed in Greece in terms of its salience for and the positioning of, the mainstream and non-mainstream parties, as Hutter and Kerscher have shown for France (n.d.) where there has been greater politicisation of the European integration issue. Retrospective evaluations of the national economy. Scale scored from 1 to 4. Higher scores represent more positive evaluations. Trust in national government Dummy variable scored 1 = tend to trust and 0 = tend not to trust.

Trust in national parliament
Dummy variable scored 1 = tend to trust and 0 = tend not to trust.