Allianz Summer Academy 2005

DEMOCRATIC GAP IN THE EU: Assessment, Causes and some Proposals

JACOPO ALLEGRINI CARLO CAPRA VALENTINA DOLFI ANDREA LOCATELLI NATALIA MILANESE

_________________________________________________ UNIVERSITÀ BOCCONI via Sarfatti, 25 - 20136 Milan (Italy)

Coordinator: STEFANO RIELA tel. +39 328 2913443 - e-mail: [email protected]

1. DEFINING DEMOCRACY The term “Democracy” is derived from two Greek words: demos (people) and kratos (rule). Thus democracy means “rule by the people”. The two major types of democracy are direct democracy and representative democracy. 1.1 Direct Democracy Direct Democracy means direct vote of citizens on all important issues. Each vote should be preceded by wide discussion and self-education of the citizens on the issue(s) to be decided. This would require unrestricted access of all citizens to information. Theoretically, direct democracy is the superior form of political rule; practically it can only work in a small group, while in a mixed society it would lead to ineffective management, unwanted inefficiency, and political instability. However, as the membership increases, people become less involved1 and once the membership reaches the size of a country, the participation still exists, but is limited to as low as it can go. 1.2 Representative Democracy Representative democracy is conceived as political process with “the people freely choosing representatives, those representatives debating and enacting policy and later standing for re-election, and administrators enforcing that policy”2. The basic claim to be “democratic” is dependent on the process itself, based on a cluster of institutions and rules which define the relationship between governed and governors in the determination of how representatives are chosen and how popular control is exercised over representatives. In a representative democracy both election (participation) and representation (indirect decision-making) are defining characteristic of the political system. Repeated elections serve as a key incentive in encouraging representatives to take account of the views of the represented 3; elections decide who will make decisions rather than deciding issues themselves.

1

Sartori, G. Democrazia: cosa è, Bur, 2000, p. 56-7 Murphy , W.F., Constitution, Constitutionalism and Democracy ,1993, in Judge D., Earnshaw D., The European Parliament, Palgrave MacMillan, 2003, p. 67 3 Judge D., Earnshaw D., The European Parliament, Palgrave MacMillan, 2003, p. 80 2

2

2. DEMOCRACY IN THE EUROPEAN UNION In the structure of the European Union (EU), the European Parliament (EP) constitutes the representativedemocratic element par excellence4, as it is the only body which is directly elected by the citizens of the member States (MSs) and represents the democratic expression of the political will of the peoples of the EU. Not surprisingly, it is also the most trusted (Fig. 1). Fig. 1 - Trust in European Institutions (April 2004) Question: For each of them please tell me if you tend to trust it or tend not to trust it? (EB61)

2.1 Democracy v Efficiency: the decision-making of the EU The EU has a complex set of decision-making rules5 involving mainly three Institutions: • • •

the Council of the European Union the European Commission the European Parliament

The Council of the EU is made up of MSs’ representatives, usually ministers, and is the key institution of the EU. It is the core legislator and, acting through national governments and bureaucracies, it has considerable executive power. It derives its legitimacy from the fact that it is constituted of democratically elected governments.6

4

Blondel J., Sinnot R. and Svesson P., People and Parliament in the European Union: Participation, Democracy and Legitimacy, Clarendon Press, 1998, p.10 5 Hix S., The Political System of the European Union, MacMillan, 1999, p. 3 6 See Baldwin et al., Nice Try: Should the Treaty of Nice be Ratified?, Monitoring European Integration 11, Centre for Economic Policy Research, 2001

3

Yet, despite this, the Council lacks accountability and does not score well on representation. The Council is the Institution perceived as the least trustworthy7 (Fig. 2). In elections, each government’s performance is judged mainly on the basis of national issues; European issues play a secondary role at best8. In addition, minorities’ views that are not taken up by governments are not heard in the Council. Finally, and little transparency of the Council voting system tends to aggravate this accountability deficit. Thus national governments are rarely brought to account by their constituents for decisions taken in the Council. Fig. 2 - Trust in the Council of the EU (April 2004) Question: Please tell me if you tend to trust it or tend not to trust it? (EB61)

Besides the Council, the European Commission is placed at the heart of the EU policy making process. An independent bureaucracy, the Commission is the guardian of the treaties. Its exclusive right to initiate legislation and broad regulatory powers make the Commission a powerful engine of integration. The Commission is also the EU executive, but in comparison with the executive of the typical nation-state, it has limited powers9. The mandates of Commissioners are given by the Council, which has the right to approve, or disapprove, every agreement negotiated by the Commission, leaving to the latter mainly powers of implementation10. The Commission enjoys a reasonable degree of trust (Fig. 3). Under current procedures, the Commission is not politically accountable. Its members are not elected, but appointed by the European Council, subject to the approval of the EP. The latter can fire the Commission, by casting a censure vote, but only in its entirety and subject to a two-thirds majority. However, making the Commission more politically accountable entails a trade-off. Although the Commission would thus obtain the legitimacy to carry out new tasks, it would become a less powerful engine of integration, a less vigilant guardian of the treaty. A Commission more politically accountable to

7

Berglof E., Eichengreen B., Roland G., Tabellini G., Wyplosz C., Built to Last: A Political Architecture for Europe, Monitoring European Integration 12, Centre for Economic Policy Research, 2003, p. 36 8 Ibidem, p. 37 9 The most important domains in which its executive powers are deployed are competition and external trade policy, but it has no executive powers over the second pillar (security and foreign policy) or the third pillar (justice and home affairs). 10 Ibidem, p. 39

4

national interests would have to worry about seeking consensus and not moving too far ahead of its constituents’desires; would therefore be less committed to integration.11 Fig. 3 - Trust in the European Commission (April 2004) Question: please tell me if you tend to trust it or tend not to trust it? (EB61)

2.2 The role of the European Parliament The role of the EP in the decision-making had varied across time (and Treaties) in accordance with changes to its inter-institutional relations with the Commission and the Council and the context in which it operates. The Commission used its right of initiative to propose legislation that the Council approved or rejected. Even when the Parliament was directly elected for the first time in 1979, its role in the legislative process was restricted to the so-called consultation procedure. The EP had only the right to express non-binding opinions. The powers of the Parliament were expanded under the Single European Act (1986) with the introduction of the cooperation procedure, involving two readings in Parliament, giving members ample opportunity to review and amend the Commission’s proposal and the Council’s preliminary position on it. The role of the Parliament has been particularly strengthened by the introduction of the co-decision procedure12, established by the Maastricht Treaty (1992) and revised by the Amsterdam Treaty (1997). The Commission has been weakened under the co-decision procedure as the Parliament is no longer required to secure the support of the Commission for its amendments. Co-decision now covers a great deal of EU legislation, such as employment, transport, education, culture, health, consumer protection, research, the environment, transparency... with the exception of EMU, agriculture, fisheries and fiscal harmonization. In some co-decision areas (citizenship, mobility of workers, tax treatment, self-employed, culture) unanimity is still required in the Council, making bargaining particularly difficult.

11

Ibidem, p.40 Firstly, it allows the EP to reject legislation at the second reading. Secondly, if the Council does not agree with an amended proposed by the Parliament at the second reading, the proposal goes to a conciliation committee composed of an equal number of members from each body. The co-decision procedure has noticeably increased the powers of the Parliament, as it gives effective bargaining power to the EP, as it can use its right of rejection to negotiate compromises with the Council.

12

5

After these reforms, the EP is a major player in the legislative process and it is a co-legislator in important policy areas with the Council13. Furthermore, being it the only directly-elected institution, the role played by the EP in legitimating the EU political system is fundamental. 2.2.1 Why the concern with electoral turnout? What is distinctive about legitimacy in a liberal democratic political systems is that democratic participation is largely conceived in terms of authorization of representatives through the electoral process14. From the outset, direct elections were seen as a means of enhancing the direct legitimacy of the representatives in EP. European elections were identified as the process through which the actions of Members of the EP (MEPs) would be authorised and through which they would be directly accountable to the people who elected them.15 The belief that the legitimacy of the entire EU is enhanced by the means of direct elections lies at the heart of the debate about the democratic deficit. There was an assumption of an automatic correlation between an increase in democracy (through the act of voting) and an increase in legitimacy. Not surprisingly, given this assumption, there has been an intense concern with the level of turnout throughout the history of direct elections.16 This was largely because members of supranational institutions (the EP and the Commission) feared that ‘a low turnout would undermine the legitimacy of the EP or that it might be interpreted as an expression of no confidence in the EC.’17 2.2.2 The EU citizens’ perceptions In the 1990s a key element of the ‘democratic deficit’ identified at the heart of the EC was ‘the weaknesses and deficiencies in inter-institutional relations and specifically to the idea of inadequate parliamentary influence over the Commission and Council [...] in particular the impact of EP elections on the composition of decision-making institutions was seen to be flawed’18. The emphasis upon giving the EP greater powers in the EU legislative process was therefore explicit in proposal to address the ‘deficit’.19 Nevertheless, as June 1999 and 2004 elections demonstrated, that potential remained unfulfilled. There appears to be only a tenuous connection between perceptions of the power of the EP and electoral turnout. Despite the EP’s enhanced powers, and despite recognition of this fact by knowledgeable voters (as Fig. 4 and Fig. 5 shows that the perceived importance of the EP role has increased), most voters still have little knowledge of the role or impact of the EP.20 Paradoxically, the EP does play a more influential policy role than many of its national counterparts; however, the potential to exploit this at election time is mitigated by voter ignorance and the inability of transnational parties to claim credit for what are essentially inter-institutional policy outputs rather than identifiable outputs of party government as such. Blondel et al.’s survey 21 found that the most striking characteristic of voters’ general attitudes to the EU was a sense of a lack of involvement rather than opposition. Correspondingly, voters’ attitudes to the EP itself were not generally grounded in political knowledge or direct experience of the Parliament. In this respect the basic problem for the EP seem to be that a substantial proportion of its electorate has no image of the Parliament, whether good or bad. Lack of 13

Judge D., Earnshaw D., op. Cit., p. 203 Beetham and Lord, Legitimacy and the European Union, Longman, 1998, p. 8 15 Herman and Lodge, The European Parliament and the European Community, Palgrave Macmillan, 1978, p.74 16 Judge D., Earnshaw D., op. Cit., p. 76 17 Smith Europe’s Elected Parliament, Sheffield Academic Press, 1999, p. 110 18 Lodge, The European Community and the Challenge of the Future, Pinter, 1993, p. 22 19 Judge D., Earnshaw D., op. Cit., p. 86 20 Judge D., Earnshaw D., op. Cit., p. 87 21 Blondel, Sinnot and Svesson, People and Parliament in the European Union: Partecipation, Democracy and Legitimacy, Clarendon Press, 1998, p. 238-244 14

6

interest, inadequate information and dissatisfaction with the EP electoral system may be the reasons for not voting. Fig 4 - Perceived importance of the EP role (April 1993) Question: How important would you say is the European Parliament in the life of the European Community nowadays? Would you say it is...? (EB39.0)

Fig 5 - Perceived importance of the EP role (June 2000) Question: How important would you say is the European Parliament in the life of the European Community nowadays? Would you say it is...? (EB53)

7

3. ELECTORAL GAP IN EU ELECTIONS The erosion of turnout evident in June 1999 EP elections has set widespread alarm bells ringing in Brussels bringing evidence that the public is becoming disenchanted and disengaged with the EU. The level of voting participation fell in fact from almost two-thirds (63%) in the first direct elections in 1979 to just under half (49.8%) of EU citizens in June 1999, and down again to just above 45% in 2004 in occasion of the last EP election (Fig. 6). Fig. 6 - EP elections turnout trend 22

Set in a broader context, the decline in the EP turnout is even more worrying because it proves contrary to trends elsewhere. Also within the EU, the trend in participation in national elections has generally declined only marginally23, but the same figure for EU voting is in general remarkably lower. The last low turnout cannot explained only by a downward trend in participation in MSs, since the average size of the democratic gap in both the EU-15 and in the ten new ones is around 30%. 3.1 Assessing the Electoral Gap In appraising the gap24: The EU-25 average lies at 45.7%, closer to the turnout rate of the EU-15 than to that of the 10 new Member States (NMS) due to the bigger size of the former group. For the first time, the citizens of the NMS were called to vote their MEPs, but despite the importance of this election for the new MS only the 26.9% decided to show up. We want to compare this result with the average in the turnout at the most recent general elections held in these countries. This figure is roughly 56%, 30% more than the results recorded for the EP election. The average turnout rate in the EU-15 was 49.4% at the 2004 EP elections, with just a slight fall from the 1999 figure (49.8%). Even if average participation in the vote is higher here than in NMS, yet the gap that separates that average from the most recent national elections is analogous: 29.5% 22

Source http://www.europarl.eu.int/press/Eurobarometer/pdf/en/PostEuropElections2004ReportEN.pdf In EU-15 only 3 countries out of 15 showed a consistent declining trend for voting participation comparing the three last elections organized in each country with the EU elections in 2004. They are Italy, the UK and Ireland (Annex 1), but they recorded a significant rise in their participation in EU PL elections (compared to 1999). 24 Data are taken from http://www.europarl.eu.int/press/Eurobarometer/pdf/en/PostEuropElections2004ReportEN.pdf 23

8

3.2 Electoral System and Practice According to a statistical analysis led by the IDEA Institute, compulsory voting, although not really enforceable but rather reinforcing the idea of voting as a duty, has a strongly positive correlation with EP turnouts (Annex 16). The countries that have experienced this rule in their constitution are Belgium, Luxembourg, Greece, Cyprus, the Netherlands and Italy. Having the date of EP elections together with national elections generally boosts turnout through contemporaneously setting more issues at stake; an additional element is the choice of Sundays to let busy people consider the practice of going to a polling station. The former matter has been followed by Luxembourg, Belgium and in part by UK, the latter, instead by the large majority except for Ireland, the Netherlands, Czech Republic and Malta; results, however, don’t seem very significant for these countries in terms of causality for lower turnouts. On the contrary, Poland, which held EP elections during a long weekend, registered very low participation rates. An other important point is the trust in politics and trust in government; indeed, more in detail, it has been verified that not only the more the governing party’s vote is down, the more turnout is down, too, but also citizens use the practice of European elections as a mid-term ballot to send a warning to the party they favour by abstaining25.

Our sample is made of six countries: France, Germany, Italy, Poland, Sweden and the UK. The participation to national elections in all six countries is around 80% with remarkable downward trends only in the case of France26. Italy, together with Belgium, Luxembourg and Malta, present the highest levels of participation with an average between 90 and 80%, notwithstanding no sanctions are imposed on nonvoters anymore. In contrast, we can clearly see from the figures that the drop in EP elections have been quite sharp, even if the lowest levels ever for EU-15 27 on average have been reached by UK28. A last point will be made on Sweden, whose analysis shows two distinct peculiarities with respect to all other nations: in 1999 it was the first country in the world to have more women ministers than men; on the other side of the spectrum within Europe we have Italy. 3.3 Information and democratic gap In order to assess whether there is a link between the level of information provided to potential voters in these countries and the level of turnout recorded there, we analyse data provided in the Flash Eurobarometer 162 Post Election Survey (July 2004). Question Q.6 asks whether all necessary information had been provided to allow for a sensible voting decision at the EP elections. Surprisingly, only a minority of the respondents declared to have perceived a lack of information (Annex 1). On the contrary, more than half of the interviewed citizens declares to have had all necessary information to make up their minds before going to the polls at the 2004 EP election. 39% of people interviewed acknowledged that they did not have all the information they needed to choose the candidate for whom they would vote. This result is far from negligible but it is still a minority. Yet, if we look with attention at the data concerning our 6 MSs, there seems to be a very strong correlation between the

25

Rose, R., Poltical trust,Tturnout and Governance Capital, Glasgow, University of Strathclyde, 2004 France records a 60% in the last parliamentary elections, a level which is relatively low for a Western European Country. However, Presidential elections for the same country show a 82% (annex 17), perhaps driven by the intense electoral challenge between the incumbent Chirac and the right wing candidate Le Pen. 27 Remember that NMS in the 2004 elections, except for Cyprus where voting is compulsory, have showed on average a 36% turnout 28 The UK has also experimented with introducing new technologies to the main purpose of increasing participation rates in the voting process. During the local elections in May 2002 pilot projects were run to try out new, innovative techniques such as remote electronic voting from any computer, touch-screen kiosks, voting by phone or text message, and voting at different public institutions such as libraries. Results by now, however, have been not that satisfactory. 26

9

level of democratic gap in those countries and the lack of information perceived by respondents in the same MS. There is a correlation between the level of Country demo gap no, rather does not democratic gap and the proportion of 25% 69% PL potential voters who answered “NO” to the 44% 54% SE question whether they had had enough 30% 42% FR information to go to vote: as expected, we 36% 38% DE can observe that there seems to be a very 21% 36% UK strong correlation b/w the level of 8% 27% IT democratic gap in the countries in our 27% 44% simple avg sample and the lack of information 29 perceived by respondents (see table on Correlation coefficient = 46.82% Covariance = 71.4% the right). According to the observations put forward by the Eurobarometer on the other 19 MSs, our result can be extended in general (Annex 1). At the top there are the two countries where the participation was particularly strong, Luxembourg and Malta (a country where media coverage of the elections is generally very strong), and, at the bottom of the list we find Poland, the Czech Republic and Slovakia, countries where – conversely – participation was very weak. Only the Slovenians counter this observation30. The Eurobarometer stresses also that data reveal the split between the UE-15 and the 10 new MSs: in the NMS a majority of respondents deem that they did not have sufficient information (54% versus just31 36% for the EU-15). This is a simplified analysis on for the 6 countries in the sample:

Country

PL SE FR DE UK IT simple avg

EP elections

Last national election

20.9% 37.8% 42.8% 43.0% 38.2% 73.1%

46.3% 81.3% 72.9% 79.1% 59.1% 81.5%

25.4% 43.5% 30.1% 36.1% 20.9% 8.4%

92.7% 158.8% 109.9% 131.8% 76.3% 30.7%

28.0% 44.0% 57.0% 60.0% 62.0% 71.0%

69.0% 54.0% 42.0% 38.0% 36.0% 27.0%

% of national NO over avg (*) 155.6% 121.8% 94.7% 85.7% 81.2% 60.9%

42.6%

70%

27.4%

100%

53.7%

44.3%

100%

Demo gap

Demo gap over avg (^)

Yes rather

No rather does not

=*/^

1.7 0.8 0.9 0.7 1.1 2.0 1.2

29

Data analysis in Annex 2 76% of them consider that they had all information needed to make their choice, but only 28% of them went to vote. 31 Just does not imply that this figure is low per se, on the contrary the fact that more than 1 third of respondents felt to be lacking adequate information at the ballot box is alarming. Yet the figure for the NMS is even worse. 30

10

Instead of the correlation coefficient a simpler measure we can look at is the ratio obtained dividing the figure of the “% of national NO over average” by that for the “demo gap over average demo gap”. This ratio stays close to 1, in particular the average value of these figures is 1.2. This offers intuitive evidence that the level of democratic gap is linked (and partially due) to lack of information in that country (see graph on the right).

100%

75%

50%

25%

0% 0%

25%

50%

75%

100%

11

4. ABSTENTION IN EP ELECTIONS From the previous analysis it is clear that a challenge of democracy is to avoid abstention and indifference towards EU matters: who absents the most and what are the possible solutions to this problem? Of course, we focus mainly on those who did go to national elections but undervalued the importance of voting for the EP, with particular attention to on the day of election abstentionists, which, in practical terms, should be the easiest to convince for the next electoral date. 4.1 Who lacks information the most? The young What emerges from the Eurobarometer survey32 is that younger voters are less informed than adults, who are on their turn less informed than elderly voters.

At the EU-25 level, the propensity to abstain from voting is higher when the voter is young: more than 2/3 of voters between 18 and 24 (67%, compared to 54.3% on the average) did not go to the polls. Conversely, a minority of people over 55 chose to abstain (41%) 33. Following a basic approach, we can define a simplified correlation as the ratio between “the ratio of non voters in an age group to the average for all groups” and “the ratio of those who lacked information in an age group to the average lack of information”. In this case, the average value of the obtained figures is not only near 1, but actually equal to 1,00 – an extremely strong result.

Group

age

Young

18-24

67%

119%

50%

123%

96%

young adult

25-39

63%

112%

43%

106%

105%

Elderly

40-54

54%

96%

37%

91%

105%

Elderly

>55

41%

73%

32%

79%

92%

56%

100%

41%

100%

100%

simple average

32 33

did not vote as a % (*) lacked info As a % (^)

Correlation = */^

Still making reference to question 6.a from the Flash EB 162 already mentioned above. See Annex 3

12

Yet we should see whether lack of information is the cause of abstentionism or whether it is a consequence of lack of interest on the part of the youngsters, in which case both lack of info and abstention from voting play the role of consequences of disinterest for the European subject matter. 4.1.1 Information is not all, but still… Even though the role played by lack of information is undeniable, it is interesting to note that nearly one abstentionist out of two declared that they had the information they needed to make a choice at the polls. Consequently, although a potential voter has sufficient information, this does not mean that it will go to the polls: reticence with regard to politics, indeed, can be so great that the level of information is not a sufficient criterion to prompt voters to go to the polls. 4.1.2 Vote, why not? … Criticism on politics & lack of information In the Flash Eurobarometer no. 162 - Post Election Survey, abstentionists were asked to give spontaneous answers to the question “What were the main reasons why you did not vote in the recent EP elections?” The weight of the general criticism of politics is much more significant than rejection or ignorance of European affairs in explaining the abstention in EP (Annex 5). Yet, even though the role played by lack of information does not represent the main cause of the observed democratic gap, still one fifth of the respondents agreed that its influence was heavy in deciding not to vote. 4.1.3 Perception of the level of knowledge about the EU: a black box?! There is in general a feeling among EU-25 citizens that their knowledge about the UE is fairly limited (Annex 6), but to have a better understanding of this issue we should look at the situation of each single MS, and we are not surprised to see that there are significant differences across countries (Annex 7). In particular criticism or ignorance of abstentionists with regard to the European question is high in the Scandinavian countries and in Slovakia. Sweden is also the country of our sample where the democratic gap came out to be largest, with a proportion just below 45%. Yet, if we look at the same figure for the other five countries, what comes out is an almost zero-correlation between the lack of information and no vote. 4.2 Lack of information & the tough ratification of the EU Constitution The point here is the following: do EU citizens possess the adequate level of knowledge and information about EU matters and institutions to be able to vote properly in a referendum such as that for the ratification of the EU Constitutional Treaty? Probably not. Indeed 1/3 of the EU population declares that they have not even heard about the proposed Constitution34, while just more than 10% of the respondents feel quite confident with the text, and says they have an overall knowledge of the contents. The average voter has at least heard about the European Constitution, but yet it seems to know very little about its contents. Naturally the situation is not homogenous in Europe, but it is distributed unevenly across MSs.

34

Source: Special Eurobarometer no. 214

13

Fig. 7 - Information about the EU Constitution

Fig. 7 shows inconsistent levels of knowledge within the EU. In particular, we can look at the 10 countries in which a referendum has been already called, or is due in the future (they are indicated by an arrow in Fig. 7). The level of knowledge of the Constitutional text was actually lower in these ten MSs35. A further aspect deserves being analysed, in as much as the different social groupings of EU citizens do not seem equal in terms of level of knowledge. The general results that come up are the following (Annex 9): - women are less informed than men, - young people are less informed than adults, and also the elderly are a bit less informed than adults, - the longer the education, the more respondents know about the Constitution, and - “High skilled” workers know more than “low skilled” workers. As a consequence of these observations we can understand that specific campaigns should be adopted for the different social categories, to address in the most appropriate way every citizens’ group. Finally, to appreciate whether there is a link between the rate of knowledge about the constitutional text and the level of approval, we can compute the Pearson coefficient between these dimensions: not surprisingly it indicates a positive correlation (31%). This means that the more citizens are informed, the more likely they are to support the draft36. 35

We observe particularly low averages for right answers in Ireland (33%), Latvia (37%) and the UK (39%). However, it is in those countries which have organized a referendum soonest that the situation appeared even more worrying: just 3 Months before the referendum in Spain, the average rate of correct answers was only 35% there; in Portugal, where the referendum date was initially set for Spring 2005, it was only 33%. What’s more, this observation applies to most of the countries in which a referendum will be organized, which may seem paradoxical. 36 Some European citizens thus prefer to oppose the Treaty as a precaution, not knowing its contents; others are opposed to it because they misunderstand the contents (Annex 13).

14

4.3 The Widespread Euroskepticism A profound lack of information arouses skepticism among EU citizens about the proper functioning of EU institutions and about their political ability to manage the complexity of the current political and economic situation. Turnout trends decline together with the level of satisfaction with the way democracy seems to be working at the Union’s level. To make some clarity about this hot issue, the Eurobarometer has tried to see whether citizens are more or less satisfied with the way democracy works in their home county and how it does at the EU level. Although 57% of respondents are satisfied with the way democracy works in their country, the level of satisfaction is 48% as regards the way democracy works in the EU, almost 20% less! However this result, recorded more than four months after the latest EP Elections, has increased by 5 points compared with last spring37 but historically it has followed a flat path around the 40% level (now moving around 45%, slightly better – yet below 50%).

37

See Annex 8

15

5. SOME ECONOMIC EXPLANATIONS FOR DECLINING TREND IN EP ELECTION TURNOUTS In this section the relationship between the EP election turnouts and some economic features on the EU MSs is analyzed. Firstly we try to show whether a correlation between the national macro-economic performance of the MS can be used as an explanatory variable of the decreasing interest for EP elections (Fig. 6). Thus we lead an empirical analysis about the role that the national economic context may play in shaping the public support for EU. The link with the decreasing trend concerning the EP election is provided by a Eurobarometer survey (2004), which clearly shows that the more the support towards the European project and institutions is, the more citizens will participate to the elections, thus elevating the turnout. Secondly, in order to analyse what is the EU influence on the economies of each MS, we focus on the main redistributive EU intervention: the Structural Funds. Then we relate the amount of these funds provided for each region to the political participation. Finally we lead a qualitative study of the feasible causes of the very low turnout of the NMSs at the EP elections in 2004. 5.1 National macroeconomic performance and democratic gap in Europe The foremost indicator we employ to assess the democratic gap at European level is the ever declining turnout level of the EP elections, with respect to the national elections one. In what follows, we relate the economic condition of the old MS to their public sustain for the European political building. Two clarifications about this choice are needed. First, we use the level of public support as a proxy for election turnouts since we will prove that they are closely related each other. Second we consider only the UE-15 because the period taken into consideration in this analysis is the decade 1990-1999, significantly before the accession of the new MSs 38. We follow the analysis developed by Nejes to assess whether national economic performance can be pointed out as a prime explanation of the declining support for EU institutions which a lot of members have experienced, particularly since the beginning of the 1990s. Thanks to a regression analysis39 (Annex 18), it is evident that in all the models considered (in each of them, different explanatory variables are taken into account) the national economic performance is highly significant in explaining support for Europe. Net transfers are also significant (even though their coefficient is far lower than the GDP one): the EU support increases when MS benefit from EU net transfers. The results of this analysis confirm what Van Kersbergen (2000) suggested: “as long as national economies perform, support for European Institutions (and thus political involvement) is secured”. However, the vice-versa has been occurring in the last years. Up to now we have related economic features to public support; the latter can be also taken as a proxy for electoral participation of EU citizens in EP elections. As a matter of fact the Post European Elections Survey 2004 results (Fig. 8) allow to clearly identify this relation: there is a considerable higher satisfaction with EU institutions, among those who voted at the last elections rather than among the abstensionists.

38

The choice about the time period is not trivial as those were the years when several initiatives, in the fields of economic and political integration, were implemented to set up an authentic political and economic Union. As a matter of fact, both Maastricht and Amsterdam Treaties were agreed upon and signed in that decade. 39 Derived from Nejes (2004)

16

Fig. 8 - Eurobarometer, Post European Elections Survey 2004

Therefore, the more a citizen is in favour of the European integration (higher support), the more he will take part to EP elections: Euro-enthusiasts were are frequent among voters than among abstensionists 40. EU citizens are likely to actively join European political life when they are satisfied with the economic performance of their own countries. One might expect this conclusion to be counterintuitive since one should be motivated to address political institutions especially when his social and economic conditions are threatened. Instead the high level of abstention, particularly in the last three electoral appointments, gives the impression to be a sort of protest against the negative structural economic condition of the continent as a whole. This trend has been further stressed during the last five years when the increasing competition coming from HPAEs41 has provoked an impressive negative impact on a great part of the European economic activity42. In this perspective, the latest election turnouts are extremely consistent with our empirical findings for the decade 1990-1999. What is of concern for our study is that the identification of the European integration process as one of the main causes of the worsening economic conditions is likely to have been stimulated by national mass media and by national political parties43. The economic issue is used either by information channels or political leaders to lead an overspread social discontent towards the EU. Even though NMSs show a major economic dynamism with respect to a lot of the old MS44, are characterized by significant levels of withincountry economic inequality which can be a fruitful ground for disaffection to the European issues. 40

Post European Elections Survey 2004 High Performing Asian Economies, above all China and India 42 The textile sector is only one example of such shock 43 Especially those which constitute the opposition to the government leaders in order to grab the leadership 44 They show a significantly higher rate of growth than the old members 41

17

5.2 European intervention against across-countries inequalities and the European election turnout Is there any relationship between the European cohesion policy and the political participation in the countries beneficiary of the redistribution at European level? For the period 2000-2006, the structural funds constitute 86% of the total resources available to promote economic and social cohesion. The European Regional Development is mainly financed by the ERDF, which is the most important instrument to achieve the Objective 1, namely the support to the development and structural adjustment of lagging behind regions45. In order to apply for such funds an area has to show an average income lower than 75% of the EU. Fig. 9 gives a idea of what are the regions which represent the receivers of these economic supports. In what follows, we will concentrate on the six major falling behind regions within the old MSs. In particular in the EU-15, the six poorest macro-regions are: Greece, Ireland, Portugal, Spain, South Italy and East Germany. Fig. 9 - Regional Inequalities in EU-27, Regional per Capita GDP in PPS (NUTS2, 2001)

45 Here, region is defined as the administrative unity of the NUTS2 classification. In particular in the EU-15, the six poorest macroregions are: Greece, Ireland, Portugal, Spain, South Italy and East Germany

18

We will not consider the MSs since almost all these countries are eligible for the funds and the structural policy will change in 200746. One would expect the democratic deficit to be far less large in those regions which receive the structural funds, also because it has been proven that in a lot of these areas (except South Italy) this financial endowment has promoted an economic convergence towards the European average values (Fig. 10 and 11). Inhabitants of those zones should be more interested and willing to take part to EU political life, since they have gained so much from the Union, in economic terms. They should regard at the vote as a way to support this supranational project which has clearly benefited to them. Fig. 10 - The EU Growth Performance in the 1990s

Fig. 11 - Convergence in six EU lagging behind macro-regions

46

“Phasing in” and “phasing out” policies

19

To assess whether this is the case in the EU we have performed two regressions(Annex 21). In the first one the democratic gap47 is related to the percentage of each county’s population who is concerned by the structural fund48. This value is used to account for the extension, in terms of potential voters, to which these funds are addressed. In the second regression we set a relationship between the democratic gap and the ratio of the funds received by each country over its GDP level. This is employed to consider the economic impact of the funds at the national level, taking into consideration the size of the national economy. Therefore no significant relationship between the little participation of the EU population to the EP elections and the amount of financial subsidies they receive can be claimed. The explanatory power of these regressions is very low as confirmed by the very low level of R2. This study leads to the conclusion that the level of financial support from EU institutions is not a feasible explanation of the decreasing turnout at the European elections. The previous results suggest that redistributive policies by themselves are not capable of encouraging a higher participation to EP elections. This result may be explained by a home-bias of the citizens towards national governments and parties. In this perspective, population is likely to judge initiatives, such as the promotion of the regional development, as a merit of the local government, capable of getting from Brussels the maximum gain they can. In other words citizens may be prone to see local improvement projects as a prime virtue of the government and this would heighten the probability they will take part to the national elections49. Therefore EU remains solely an economic opportunity to be exploited and the role of the national politicians is to the take the utmost advantages for their country. Then, the economic benefit doesn’t translate into a political concern for EU. Furthermore the lack of information among the population who mostly benefited from the structural policy can be pointed out as another explanation of these findings.

A brief overview of the situation in the New Member States Given the fact that new MS have taken part to the EP elections for the first time in 2004 and that the structural funds’ provisions specifically addressed to these countries will change appreciably in 2007, we propose some hypotheses to explain the substantial lower rate of participation in such countries, 40,3%, with respect to the old members’ one which amounted at 52,8% in 2004 elections. This low percentage is even more stressed out when looking at the difference with the latest national elections, whose turnout was 69%. The vast majority of these countries conceived, and still conceive, the accession to the EU mostly (if not exclusively) as the opportunity of entering the Single European market and attracting foreign investments (due to the significant low labour costs) and structural funds (coming from Brussels). Moreover the European policy, promoting an harmonious development through the goal of the economic convergence, is benefiting and will certainly benefit the Eastern Europe countries. That’s why it may appear shocking to ascertain such a little interest for the EP elections. The reason for that may lie on the detachment of the political elites, which have prompted the process of integration to the EU, and the vast majority of the population which still thinks at Europe essentially as economic chance for the country. In other words, those countries’ citizens do not share the ideal of the EU, but they perceive the enormous financial advantages they can take from the accession.

47

Measured as the difference of turnout at EP elections of 2004 and the last national elections turnout. Even though the structural funds are targeted to single regions we take into consideration the national level because it considerably simplifies our analysis, since data on turnout are easier to derive for countries than for regions. This choice doesn’t alter our results. 49 This is consistent with the considerable discrepancy between national and European elections shown above 48

20

6. DEMOCRATIC GAP AND EUROPEAN POLITICAL SYSTEM In order to analyse the relationship between the democratic gap and the European political system, we firstly analyze the cohesion and the competition of EP, taking into consideration voting data from the plenary session of the EP from 1979 to 2004. Our goal is to define which the dominant trends are. Then, by using the Eurobarometer surveys, we look for the most important feature in the mandate which Europeans accord to MEPs. Finally we discuss benefits and drawbacks of these tendencies. 6.1 The EU party system EU politics have two main dimensions. One regards the traditional left-right conflict the parties have in their national parliament, and the other one concerns the pro/anti-Europe attitude about the speed of the integration process and the amount of sovereignty which should be given to EU institutions. The EP party groups, which are ideally responsible for these two aspects, are trans-national party federations which have developed as “extra-parliamentary” organizations since 199250. Their political mission is threefold: 1) Set the EP agenda; 2) Secure the placement of own MEPs into the EP offices; 3) Pursue political office and achieve policy goals. In order to guarantee a fair competition between parties, the following rules of the game have to be respected: - Political parties should compete in EP elections about European issues and for the European political office; - Voters should choose on the basis of rival policy platform or candidates; - Cohesive electoral parties should translate their policy goals, claimed in the manifesto, into the legislative political agenda at the EU level. Then, if competition works properly, which happens when actors of the same party behave consistently, the change of electoral choices should produce a shift in the political agenda, for example from market deregulation to social policy.

6.2 Cohesion: a balance of incentives The MEPs’ behaviour is strongly influenced by three variables: first of all their personal values and ideological beliefs, secondly the national party whereby candidates are elected, and the EP party group in which they take part. All these factors are taken into account to state which is the dominant one. 6.2.1 Personal Values MEPs have personal beliefs, single policy preferences. The ideal location of a party about an issue can be though as a mean of the respective positions of its MEPs. The latter, with the exception of those sitting at far left/right of the most lefty/righty party, are ideally situated between two party groups. If MEPs vote purely following personal beliefs, with no sanctions nor incentives, many members may find themselves closer to the ideal position of other EP parties. Hence, the closer their policy preferences to the mean position of their EP group are, the less likely they will situate on the opposite side of a cutting point.

50

Art. 191 Maastricht

21

Fig.12 shows that there is a considerable overlap between ideological self-placement of PES and EPP members only regarding the traditional left/right dimension. Within the latter there is room for a fair “supra-national” competition. Instead, since they have similar policy views, the two parties have no incentives to compete about European integration. Fig. 12 - Kernel density Plot of ideological self-placement of PES and EPP survey respondent Where would you place yourself on the left-right spectrum?

(1=Left 10= Right) Where would you place yourself on the question of the European integration?

(1=Anti 10=Pro) 6.2.2 People respond to incentives, MEPs too Beside their personal policy preferences, MEPs face two kinds of incentives directly provided by national parties and EP groups. National parties control the candidate selection in EP elections, and they have a tough power on candidate re-election. This control is possible through the central national party’s committee, which manages the list of candidates51, and through the party’s regional organs52. In case of conflict between the national and the EP group views, MEPs are expected to vote according to the national party because their potential re-election depends on it.

51 52

France, Belgium, Denmark and Spain. Italy, United Kingdom, Germany and Ireland

22

EP groups impose sanctions on the MEPs who defects from the EP group position and provide incentives for those who adopt the voting instructions given by the groups’ leaders. However not all the EP party groups are able to supply the same incentives. The PES and EPP, the main groups which actually run together the Parliament, have a lot to offer to the MEPs who behave coherently since they control the nomination process of the key offices in the Parliament, in the Committee chair and in the organization of the legislative agenda. The bigger the party group is, the more it is cohesive, and the more likely it will influence policy outcomes and it will attract votes (Annex 19), and individual preferences and national parties alone cannot explain the changes in group cohesion (Hix, Noury and Roland,2005).

6.3 Competition: Coalition building The three main coalitions are more likely to be built in the EP involving the three main party groups: ELDR, PES, EPP and the analysis in Annex 20 outline a break-even point during the EP3: in the fourth and fifth parliaments the PES voted more similarly to the ELDR rather than to the EPP. In a democratic system the probability of a coalition establishment between two parties is directly linked to their ideological distance. The closer their ideological positions are, the more frequently a coalition will be set up. If this is not the case the system is collusive. Those are the reasons why PES and ELDR have become ideologically closer than the PES and the EPP53. The break-even points in the ideological distance and in the coalition patterns coincide (Fig. 13). Fig. 13 - Ideological distance between the principal parties from 1979 to 2001

Given the specific institutional structure of the EU, the EP is the only Institution where debates and votes concern the traditional left/right dimension. The institutional context of the EP offers MEPs with similar 53

The coalition between PES and PPE is a singular one from the national point of view, it is composed by parties that in the national arena collaborate with the most extreme movement in order to run a “coalition” government.

23

political preferences considerable motivations to behave cohesively. The increasing level of party cohesion suggests that, despite the increasing ideological fragmentation, the system is slowly evolving towards party organization based on trans-national party groupings. In this manner EP party groups are able to influence policy outcomes. 6.4 MEPs’ mandate and legitimacy The EP party system can work as a genuine trans-national political system and EP party groups can be regarded as instruments to strengthen democracy in the EU. However this may remain an unfulfilled opportunity. For “party democracy” to work, the EU parties must be able to gain legitimacy from European elections. The second-order argument, the lack of a real euro-debate, suggests that the votes at European level are completely distorted by national politics. The winning parties of the national elections may use electoral appointment in Europe as a way to obtain a public approval for their national operate. This consequently leads to a lower turnout and a higher “quasi-switching54” (Fig. 14). Voters want to punish or sustain the government in charge, instead of ‘vote with the heart’ and follow the last political choice. Hence the government parties and the opposition parties are likely to lose vote, small parties and protest group to gain. Fig. 14 - Voters who would have vote in a different way if it were a national election 50 45 40 35 %

30

89 94

25 20 15 10 5

K U

Be lg D ium en m ar Fr k a G nce er m an G y re ec Ire e la nd Th Lux Ital y e e N mb et u he rg rla n Po ds rtu ga l Sp ai n

0

In every parliamentary system the electoral debate is dominated by the left/right opposition and the mandate given by citizens is built upon an ideological choice. Within European institutions, the views about the sense and the scope of the EU matter too. EP parties can’t ignore it since they have to discuss and to decide about “supra-national issue”55. In order to test whether the perception of EU influences the political inclinations of Europeans, we relate the citizens’ vision and perception of Europe to their political choice. If supranational dimension 54 55

People who would vote different than if a national election would be held at the same time. Enlargement, competences…

24

matters, the percentage of votes to Euro-enthusiastic parties should be higher in those countries which show positive attitude and a considerable trust towards the European institutions. There is no correlation between electorate’s feelings about Europe and their votes to parties showing different attitudes with regards to the European integration (Annex 22). The result claims that European citizens vote only according to their left/right ideology, thus referring merely to the national parties. 6.5 The national filter The traditional ideological filter of each single national system turned down the chance for voters to perceive EP groups and MEPs as a set of “trans-national” actors. Candidates are not voted for the opinions claimed about supranational issue but simply because they belong to political groups which share the ideological position of the electorate. And the trust of national parties influences not only the political choices but also the citizens’ decision to enter the democratic process and to vote. Although it doesn’t play a significant role inside the “pool-box”56, trans-nationality is a building block of the EP and it is particularly remarkable after the election, both in the formation of political groups and in the legislative procedures. On one hand Europeans may not consider Europeanism as an essential variable to be accounted for when expressing their political preferences. On the other hand, during electoral campaign57, candidates and national parties discuss mainly about national topics, focusing on the ideology. Nevertheless, the elected candidates face incentives from EP party groups and they have to take position with respect to the agenda’s issues concerning the integration or the enlargement of European sovereignty. Yet, if these issues were approached only according to the Europeanism element many party groups might split, allowing new potential coalitions58 to grow up.

6.6 Filling the Gap For a genuine ‘party democracy’ to work in EU, Europeans elections must become a European contest rather than a national59 one. We will discuss some of the several policies that should be pursued such a context. 6.6.1 Uniform electoral procedure Electoral campaigns have always focused on national topics, thus being only a test for the popularity of the government in charge. To develop a pan-European campaign, a uniform electoral procedure is essential. 25 national set of rules avoid the chance to develop political issues, such as economic or immigration policies, into a trans-national dimension. However this is enough by itself, since national parties would still lead the elections on a national basis. 6.6.2 EP party groups European elections will be ‘European’ only if the role and the power of national parties in the elections is replaced by the influence of the European ones; special right and responsibility in the European elections should be given to them by means of a new body of EU ‘party law’60. With the intention to break the national 56

The increasing powers given to the EP should have stimulated citizens; this has not happened as a result of a shift of competences which has not been able to achieve people. 57 This seems to have become a general trend. 58 The extreme party groups may vote cohesively, while GRE and EPP may divide themselves. 59 Hix S., ‘Political parties in the European Union’ 1996 60 Based on art. 138A of the EC Treaty

25

party control, EP elections should be held limiting the control of national parties in the candidate selection. Then, subsidies from the EU budget for their EP election campaign should be granted to national parties if they respect two conditions: they should agree to cooperate with their European groups in the candidate selection and they should adopt only the European party manifestos in their campaign61. Therefore giving the complete control to EP parties may reach the opposite effect; there’s opportunity for some negotiation as a part of candidate may be chosen by regional rather than central organs and a minority chosen by EP party leaderships on European-wide ‘top-up’ lists. 6.6.3 Elections of the President of the Commission In a presidential system, differently from a parliamentary one, the election of the head of the executive is independent from the legislature. In this case the choice of the Commission President would be disconnected from the EP but it would bring some benefits for the legitimacy of the European parties. There would be a separate candidate for each European parties, the President once elected would be accountable for the manifesto of the supporting parties; at last it would be difficult for the national parties to monopolize the election debate with the performance of national governments. Thus voters would be able to choose between rival agendas for EU-level action.

61

The adoption of a manifesto and the selection of candidates should take place at either regional or local level according to national and European parties. However, where candidates’ selection is decentralized, party cohesion breaks down since parliamentary parties have no instruments to punish legislative defection. In Italy, before the election, broad alliances are formed between parties that choose their candidates in the constituencies assigned and the same alliances, after the election, divide into a series of parliamentary parties. Hix S., ‘Parliamentary Behavior with Two Principals: Preferences, Parties, and Voting in the European Parliament’, American Journal of Political Science

26

Bibliography

Altomonte, C. and M. Nava, Economics and Policies of an Enlarged Europe,2005, forthcoming Baldwin R., Berglof E., Giavazzi F. and Widgren M., Nice Try: Should the Traty of Nice be Ratified?, Monitoring European Integration 11, Centre for Economic Policy Research, 2001 Beetham D.and Lord C., Legitimacy and the Europan Union, Longman, 1998 Berglof E., Eichengreen B., Roland G., Tabellini G., Wyplosz C., Built to Last: A Political Architecture for Europe, Monitoring European Integration 12, Centre for Economic Policy Research, 2003 Blondel J., Sinnot R. And Svesson P., People and Parliament in the European Union: Participation, Democracy and Legitimacy, Clarendon Press, 1998 Copeland G. W. And Patterson S.C., Parliaments in the Modern World: Changing Institutions, University of Michigan Press, 1994 Eichenberg, R. and R. Dalton, Europeans and the European Community: The Dynamics of Public Support for European Integration, International Organization 47, 1993, 507-34 Eurobarometer 62, Published May 2005 Eurobarometer interactive search system: europa.eu.int/comm/public_opinion/index_en.htm European Commission “Allocation of 2000 EU Operating Expenditure by Member State”, September 2001 European Parliament “Report On The Implementation Of The European Union’s Information And Communication Strategy”, A6-0111/2005 Flash Eurobarometer 162, Post European Elections 2004 Survey Flash Eurobarometer 168, The European Constitution, Post-Referendum Survey In Spain Grunfeld, L. and U. Sverdrup, Economic Voting and European Integration, 2004 Herman V. And Lodge J., The European Parliament and the European Community, Palgrave Macmillan, 1978 Hirst P., Representative Democracyand Its Limits, Polity Press, 1990 Hix S., A Supranational Party System and the Legitimacy of the European Union, The International Spectator, 2003(1), 2003, 49-59 Hix S., Amie Kreppel and Abdul Noury, The Party System in the European Parliament: Collusive or Competitive?, Journal of Common Market Studies 41(2), 2003, 309-331 Hix S., Legislative Behaviour and Party Competition in European Parliament: An Application of Nominate to the EU, Journal of Common Market Studies 39(4), 2001, 663-688 Hix S., Lord C., Political Parties in the European Union, MacMillan, 1997 Hix S., Noury A. and Roland G., Power to the parties: cohesion and competitiveness in the European Parliament, 1979-2001, Cambridge University Press, 2005 Hix S., Parliamentary Behavior with Two Principals: Preferences, Parties, and Voting in the European Parliament, American Journal of Political Science 46(3), 2002, 688-698 Hix S., The Political System of the European Union, MacMillan, 1999 IDEA, Voter turnout in Western Europe, A Regional Report, Stockholm, 2004 (1-93) Judge D., Earnshaw D., The European Parliament, Palgrave MacMillan, 2003 Kreppel A., The European Parliament and Supernational Party System, Cambridge University Press, 2002 Kreppel A., The European Parliament and the Supranational Party System: A Study in Institutional Development, Cambridge University Press, 2002

27

Lodge J.,The European Community and the Challenge of the Future, Pinter, 1993 Mattila M., Fiscal Redistribution in the European Union and the Enlargement, Department of Political Science, University of Helsinki, 2002 Member State. European Commission, Budget Directorate, September 2001 Murphy W.F., Constitution, Constitutionalism and Democracy, 1993, in Judge D., Earnshaw D., The European Parliament, Palgrave MacMillan, 2003 Nejes C., All abroad? Explaining Public Support for European Integration in a Post Maastricht Era, 2004, paper prepared for the panel “Public Opinion and EU Enlargement” at 2nd Pan- European Conference in Bologna, Italy, 2-26th of June; Noury A., Ideology, Nationality and Euro-Parliamentarians, European Union Politics 3(1), 2002, 33-58; OJ C 47 R/400-406; Pippa N., Blaming The Messenger? Political Communications And Turnout In EU Elections, Harvard University, 1999; Report”, Oxford University Press, 2004; Pippa N.,'Women's Power at the Ballot Box', in Voter Turnout since 1945: A Global Report (Stockholm: International IDEA, 2002) Rose R., Europe expands, turount falls: the significance of the 2004 European Parliament election, International Institute for Democracy and electoral assistance, 2004; Rose R., Political trust, Turnout and Governance Capital, Glasgow, (University of Strathclyde, 2004) Rieu A., G. Duprat (edited by), European Democratic Culture, The Open University, 1995; Sapir A., P. Aghion, G. Bertola, M. Hellwig, J. Pisani-Ferry, D. Rosati, J. Viñals, H. Wallace,M. Buti, M. Nava, P.M. Smith, An Agenda for a Growing Europe: The Sapir Sartori G., Democrazia: cosa è, Bur, 2000; Sartori G., The Theory of Democracy Revisited, Chatham House, 1987; Scully R., Democracy, Legitimacy and the European Parliament, 2000, in Judge D., Earnshaw D., The European Parliament, Palgrave MacMillan, 2003; Smith D., S. Wright (edited by), Whose Europe? The turn towards democracy, Blackwell Publishers, 1999; Smith J., Europe’s Elected Parliament, Sheffield Academic Press, 1999; Special Eurobarometer 214, The Future Constitutional Treaty; van Kersbergen K., “Political Allegiance and European Integration”, European Journal of Political Research 37, 2000, 1-17;

28

Annex 1

29

Annex 2

Country PL SE FR DE UK IT simple avg

national demo no, rather =% of national gap over avg does not NO over avg EP 2004 last gen el demo gap yes, rather =H2/E2 0,2090 0,4630 0,2540 0,9270 0,2800 0,6900 1,5564 1,6789 0,3780 0,8130 0,4350 1,5876 0,4400 0,5400 1,2180 0,7672 0,4280 0,7290 0,3010 1,0985 0,5700 0,4200 0,9474 0,8624 0,4300 0,7910 0,3610 1,3175 0,6000 0,3800 0,8571 0,6506 0,3820 0,5910 0,2090 0,7628 0,6200 0,3600 0,8120 1,0646 0,7310 0,8150 0,0840 0,3066 0,7100 0,2700 0,6090 1,9866 0,4263 0,7003 0,2740 1,0000 0,5367 0,4433 1,0000 1,1684

Country PL SE FR DE UK IT simple avg

no, rather does demo gap not 25% 69% 44% 54% 30% 42% 36% 38% 21% 36% 8% 27% 27% 44%

Correlation 0,468154983

covariance 0,00714

30

Country PL SE FR DE UK IT

national demo =% of national gap over avg NO over avg 0,927 1,556 1,588 1,218 1,099 0,947 1,318 0,857 0,763 0,812 0,307 0,609

Correlation 0,468154983

covariance 0,058778333

SUMMARY OUTPUT Regression Statistics Multiple R 0,379731123 R Square 0,144195726 Adjusted R Square -0,055804274 Standard Error 0,413352049 Observations 6 df 1 5 6

SS MS 0,143942198 0,143942198 0,854299583 0,170859917 0,998241782

F 0,842457384

Significance F 0,410630902

0 0,967143296

Standard Error t Stat #N/D #N/D 0,161282542 5,996577702

P-value #N/D 0,001850858

Lower 95% #N/D 0,552554

Regression Residual Total Coefficients Intercept X Variable 1

Upper 95% Lower 95,0% Upper 95,0% #N/D #N/D #N/D 1,381732592 0,552554 1,381732592

RESIDUAL OUTPUT Observation 1 2 3 4 5 6

Predicted Y 1,5052531 1,178024165 0,916241017 0,828979968 0,785349443 0,589012083

Residuals -0,578245801 0,409567076 0,182299129 0,48853828 -0,022575721 -0,282442739

31

Annex 3

Source: Flash Eurobarometer 162 “Post Election Survey”

Annex 4

Annex 5 Flash Eurobarometer no. 162 - Post Election Survey Abstentionists were asked to give spontaneous answers to the question: “What were the main reasons why you did NOT vote in the recent EP elections?”

32

Annex 6

Annex 7

33

Annex 8

34

Annex 9

35

Annex 10

36

Annex 11

Annex 12

37

Annex 13

38

Annex 14

It can be observed that the feeling of closeness to some political party increases with age, from 8% of respondents below the age of 24, to 14% for adults and just above 20% for elderly voters. The share is not in fact insignificant, but more effective channels of communications can probably be found. .

However a more thorough analysis of the results of the question shows that there is a positive correlation between commitment to political parties and going to the polls: we can in fact observe that the more the people feel close to a party, the more they go to the polls. So this lead us to conclude that abstention is above all related to a low degree of politisation or even distrust with regard to politics of certain potential voters.

39

Annex 15

It is striking that the country of our sample with the highest democratic gap, namely Sweden, is that of the 6 in which the highest percentage of respondents declared to have been aware of a non-party campaign/advertisement encouraging people to vote in the EP elections. Annex 16 Electoral System and Practice (International Institute for Democracy and Electoral Assistance, 2004) Compulsory voting Holding national and EP elections on the same day National politics: trust in parties National politics: trust in government National politics: vote for biggest governing party Benefits of EU membership Duration of EU membership Over-representation in EU Parliament * significance at < .01 ** significance at < .05

correlation .68 ** .45* .62** .54** -.44* Not significant .46* .57**

40

Annex 17 ITALY Electoral System for the national Parliament (630 seats): Mixed Member Proportional Bicameral Legislature, Parliamentary Elections only, Status in EU: founder Year Total Vote Registration Vote/Reg VAP Vote/VAP Invalid 1946 24,947,187 28,005,449 89.1% 28,346,220 88.0% 7.7% 1948 26,854,203 29,117,554 92.2% 28,794,780 93.3% 2.2% 1953 28,410,851 30,267,080 93.9% 31,041,400 91.5% 4.3% 1958 30,399,708 32,436,022 93.7% 32,367,060 93.9% 2.8% 1963 31,766,058 34,201,660 92.9% 33,328,680 95.3% 3.2% 1968 33,003,249 35,566,681 92.8% 35,449,700 93.1% 3.6% 1972 34,524,106 37,049,654 93.2% 36,454,700 94.7% 3.2% 1976 37,741,404 40,423,131 93.4% 39,547,710 95.4% 2.7% 1979 38,112,228 42,181,664 90.4% 41,093,160 92.7% 3.9% 1983 39,114,321 43,936,534 89.0% 42,627,000 91.8% 5.7% 1987 40,599,490 45,689,829 88.9% 43,008,750 94.4% 4.9% 1992 41,479,764 47,435,964 87.4% 44,918,610 92.3% 5.4% 1994 41,461,260 48,135,041 86.1% 45,641,100 90.8% 5.9% 1996 40,496,438 48,846,238 82.9% 46,363,590 87.3% 7.8% 2001 40,195,500 49,358,947 81.4% 47,332,575 84.9% 7.2%

FH n/a n/a n/a n/a n/a n/a 3 3 4 3 2 3 3 3 n/a

Pop. Size 44,994,000 45,706,000 47,756,000 49,041,000 50,498,000 52,910,000 54,410,000 55,701,000 56,292,000 56,836,000 57,345,000 56,859,000 57,049,000 57,239,000 57,684,294

UNITED KINGDOM Electoral System for the national Parliament (lower House, 659 seats): First Past The Post Bicameral Legislature, Parliamentary Elections only, Status in EU: access in 1973 Year Total Vote Registration Vote/Reg VAP Vote/VAP Invalid 1945 24,117,191 33,240,391 72.6% 34,427,400 70.1% n/a 1950 28,771,124 34,412,255 83.6% 35,261,100 81.6% n/a 1951 28,596,594 34,919,331 81.9% 35,148,400 81.4% n/a 1955 26,759,729 34,852,179 76.8% 35,327,310 75.7% n/a 1959 27,862,652 35,397,304 78.7% 35,968,320 77.5% n/a 1964 27,698,221 35,894,054 77.2% 36,864,840 75.1% 0.1% 1966 27,314,646 35,957,245 76.0% 37,026,000 73.8% 0.2% 1970 28,386,145 39,342,013 72.2% 39,895,200 71.2% 0.1% 1974 31,382,414 39,753,863 78.9% 40,298,400 77.9% 0.1% 1974 29,226,810 40,072,970 72.9% 40,298,400 72.5% 0.1% 1979 31,233,208 41,095,490 76.0% 41,608,720 75.1% 0.4% 1983 30,722,241 42,192,999 72.8% 42,846,520 71.7% 0.2% 1987 32,566,523 43,180,573 75.4% 43,326,080 75.2% 0.1% 1992 33,653,800 43,240,084 77.8% 44,658,460 75.4% 0.1% 1997 31,289,097 43,784,559 71.5% 45,093,510 69.4% n/a 2001 26,365,192 44,403,238 59.4% 45,804,132 57.6% n/a

FH n/a n/a n/a n/a n/a n/a n/a n/a 2 2 2 2 2 3 3 n/a

Pop. Size 49,182,000 50,373,000 50,202,000 51,199,000 52,128,000 54,213,000 54,450,000 55,410,000 55,970,000 55,970,000 56,228,000 56,377,000 57,008,000 57,998,000 58,563,000 59,434,645

POLAND Electoral System: List, Proportional Representation Parliamentary and Presidential Elections, Status in EU: access in 2004 Year Total Vote Registration Vote/Reg VAP Vote/VAP Invalid FH 1989 16,994,732 27,362,313 62.1% 26,572,700 64.0% n/a 7 1991 11,886,984 27,516,166 43.2% 26,771,500 44.4% 5.6% 4 1993 14,415,586 27,677,302 52.1% 27,723,600 52.0% 4.3% 4 1997 13,616,378 28,409,054 47.9% 27,901,720 48.8% 3.9% 3 2001 13,559,412 29,364,455 46.2% 28,469,123 47.6% 4.0% n/a

Pop. Size 37,961,000 38,245,000 38,505,000 38,752,000 n/a

41

Poland - Presidential Elections Year Total Vote Registration Vote/Reg VAP Vote/VAP Invalid FH Pop. Size 1990 14,703,775 27,535,159 53.4% 26,683,300 55.1% 2.3% 4 38,119,000 1995 19,146,496 28,062,205 68.2% 27,790,560 68.9% 2.0% 3 38,598,000 2000 17,798,791 29,122,304 61.1% 28,443,524 62.6% 1.1% 3 38,742,748 SWEDEN Electoral System: List, Proportional Representation Unicameral Legislature, Parliamentary Elections only, Status in EU: Access in 1995 Year Total Vote Registration Vote/Reg VAP Vote/VAP 1948 3,895,161 4,707,783 82.7% 4,749,270 82.0% 1952 3,801,284 4,805,216 79.1% 4,916,250 77.3% 1956 3,902,114 4,902,114 79.6% 5,047,350 77.3% 1958 3,864,963 4,992,421 77.4% 5,112,210 75.6% 1960 4,271,610 4,972,177 85.9% 5,161,200 82.8% 1964 4,273,595 5,095,850 83.9% 5,286,090 80.8% 1968 4,861,901 5,445,333 89.3% 5,616,100 86.6% 1970 4,984,207 5,645,804 88.3% 5,708,400 87.3% 1973 5,168,996 5,690,333 90.8% 6,023,600 85.8% 1976 5,457,043 5,947,077 91.8% 6,166,500 88.5% 1979 5,480,126 6,040,461 90.7% 6,303,440 86.9% 1982 5,606,603 6,130,993 91.4% 6,327,000 88.6% 1985 5,615,242 6,249,445 89.9% 6,513,000 86.2% 1988 5,441,050 6,330,023 86.0% 6,580,860 82.7% 1991 5,562,920 6,413,172 86.7% 6,721,260 82.8% 1994 5,725,246 6,496,365 88.1% 6,848,400 83.6% 1998 5,374,588 6,603,129 81.4% 6,915,438 77.7% 2002 5,385,430 6,722,152 80.1% 0 n/a

Invalid 0.4% 0.5% 0.6% 0.5% 0.4% 0.7% 0.7% 0.2% 0.2% 0.4% 0.6% 0.9% 0.9% 1.2% 1.7% 1.5% 2.1% 1.5%

FRANCE Electoral System for the national Parliament (lower House, 574 seats): Two-Round System Bicameral Legislature, Parliamentary and Presidential Elections, Status in EU: founder Year Total Vote Registration Vote/Reg VAP Vote/VAP Invalid 1945 19,657,603 24,622,862 79.8% 26,588,000 73.9% 2.4% 1946 20,215,200 24,696,949 81.9% 27,416,240 73.7% 1.7% 1951 19,670,655 24,530,523 80.2% 28,721,840 68.5% 2.8% 1956 22,138,046 26,772,255 82.7% 29,813,240 74.3% 2.9% 1958 21,026,543 27,244,992 77.2% 29,560,740 71.1% 2.5% 1962 18,918,154 27,540,358 68.7% 31,018,680 61.0% 3.1% 1967 22,910,839 28,242,549 81.1% 32,207,500 71.1% 2.3% 1968 22,500,524 28,178,087 79.9% 32,441,500 69.4% 1.7% 1973 24,299,210 29,883,738 81.3% 34,405,800 70.6% 2.3% 1978 24,658,645 34,424,388 71.6% 38,892,210 63.4% 1.9% 1981 25,182,623 35,536,041 70.9% 39,395,180 63.9% 1.4% 1986 28,736,080 36,614,738 78.5% 41,104,040 69.9% 4.3% 1988 24,472,329 36,977,321 66.2% 42,088,500 58.1% 2.0% 1993 26,860,177 38,968,660 68.9% 43,826,920 61.3% 5.3% 1997 26,649,818 39,215,743 68.0% 44,521,902 59.9% 4.9% 2002 22,186,165 36,783,746 60.3% 0 n/a 4.4%

FH n/a n/a n/a n/a n/a n/a n/a n/a 2 2 2 2 2 2 2 2 2 n/a

FH n/a n/a n/a n/a n/a n/a n/a n/a 3 3 3 3 3 3 3 n/a

Pop. Size 6,883,000 7,125,000 7,315,000 7,409,000 7,480,000 7,661,000 7,910,000 8,040,000 8,140,000 8,222,000 8,294,000 8,325,000 8,350,000 8,437,000 8,617,000 8,780,000 8,780,000 n/a

Pop. Size 39,100,000 40,318,000 42,238,000 43,843,000 44,789,000 46,998,000 49,550,000 49,910,000 52,130,000 53,277,000 53,966,000 55,546,000 56,118,000 57,667,000 58,581,450 n/a

42

France - Presidential Elections Year Total Vote Registration Vote/Reg VAP Vote/VAP Invalid FH Pop. Size 1965 23,744,400 28,200,000 84.2% 31,694,000 74.9% 0.7% n/a 48,760,000 1969 22,200,000 28,800,000 77.1% 32,708,000 67.9% 1.0% n/a 50,320,000 1974 25,100,000 29,800,000 84.2% 34,656,600 72.4% 9.0% 3 52,510,000 1981 30,350,568 36,398,762 83.4% 39,395,480 77.0% 2.5% 3 53,966,000 1988 32,164,400 38,200,000 84.2% 42,088,500 76.4% n/a 3 56,118,000 1995 31,852,695 39,976,944 79.7% 44,080,760 72.3% 6.0% 3 58,007,000 2002 32,832,295 41,191,169 79.7% 0 n/a 5.4% n/a n/a GERMANY Electoral System for national Parliament (at least 600 seats, currently 603). Mixed Member Proportional Bicameral Legislature, Parliamentary Elections only, Status in EU: founder Year Total Vote Registration Vote/Reg VAP Vote/VAP Invalid FH Pop. Size 1949 24,495,614 31,207,620 78.5% 32,035,480 76.5% 3.1% n/a 47,111,000 1953 28,479,550 33,120,940 86.0% 35,346,630 80.6% 3.3% n/a 51,227,000 1957 31,072,894 35,400,923 87.8% 35,484,630 87.6% 3.8% n/a 51,427,000 1961 32,849,624 37,440,715 87.7% 37,820,300 86.9% 4.0% n/a 54,029,000 1965 33,416,207 38,510,395 86.8% 41,328,000 80.9% 2.4% n/a 59,040,000 1969 33,523,064 38,677,235 86.7% 41,979,600 79.9% 1.7% n/a 60,840,000 1972 37,761,589 41,446,302 91.1% 42,552,300 88.7% 0.8% 2 61,670,000 1976 38,165,753 42,058,015 90.7% 45,519,620 83.8% 0.9% 2 61,513,000 1980 38,292,176 43,231,741 88.6% 46,786,360 81.8% 0.9% 3 61,561,000 1983 39,279,529 44,088,935 89.1% 48,522,590 81.0% 0.9% 3 61,421,000 1987 38,225,294 45,327,982 84.3% 50,954,310 75.0% 0.9% 3 64,499,127 1990 46,995,915 60,436,560 77.8% 64,285,650 73.1% 1.1% 3 79,365,000 1994 47,737,999 60,452,009 79.0% 65,942,100 72.4% 1.3% 3 81,410,000 1998 49,947,087 60,762,751 82.2% 66,313,874 75.3% 1.6% 3 82,172,259 2002 48,582,761 61,432,868 79.1% 0 n/a 1.2 n/a n/a

Definitions: FH: Freedom House from which we use the ratings calculated as an indicator of the levels of freedom in a country's political system. There have been added together the "political rights" and "civil liberties" ratings to yield a number from 2 to 14, where 2 indicates the highest possible level of rights and freedoms and 14 the lowest. Invalid: The number of invalid votes, as reported by each country. Population: The total population as described under Voting Age Population. Registration: The number of registered voters. Total vote: The total number of votes cast in the relevant election. Total vote includes valid and invalid votes, as well as blank votes in cases where these are separated from invalid votes. Vote/Registration: The number of votes divided by the number of names on the voters' register, in percentage terms. Vote/VAP: The number of votes divided by the Voting Age Population, in percentage terms. Voting Age: The lowest age at which the right to vote is obtained in the relevant country. Voting Age Population: It includes all citizens above the legal voting age.

43

Annex 18 The decreasing affection brings about a diminishing public affection to the EP elections. Actually Nejes analyzes the relationship between public support and a series of variables referring to the national cultural and political identity and to some highly significant political events, which occurred in that period, within MSs. Nevertheless, we focus on the three economic variables used in the model, namely the percentage of change in GDP (derived from the OECD dataset), net EU transfers as percentage of national GDP (Mattila, 2002) and a dummy variable defining the country’s access to the EMU. The first one is entirely determined by the national economic condition, while the other two account for the impact of the European integration on the national situation through fiscal and monetary policies. Country differences in change in support European integration between 1991-999

Source: Nejes, C. (2004) Furthermore, support for European project is measured through an index of three survey questions taken from the Eurobarometer Trend file: Generally speaking do you think (your country’s) membership in the Union is a good thing, a bad thing or neither good or bad? Taking everything into consideration, would you say that your country has on balance benefited of not from being a member of the EU? In general are you for or against the efforts being made to unify Western Europe?

44

Annex 18 - Determinants of Support for European Integration

Source: Nejes, C. (2004)

45

Annex 19 Fig. 1 shows the Agreement Index62 of the party groups based on the roll-call votes63 in each parliament. It is clear that parties do behave cohesively in order to be influential in the EP while cohesion is not the consequence of a shift of power to the EP. Fig. 1 - Index of Agreement64 EP Group

EP1

EP2

EP3

EP4

EP5

(79-84)

(84-89)

(89-94)

((94-99)

(99-01)

PES

75,7

86,9

90

90,1

90,4

EPP

89,9

93,4

90,7

89,8

85,9

ELDR LEFT

84,9 81,2

84,9 87,1

84,7 86,1

86,1 80,4

90,9 75,6

GAUL

80

84,2

84,9

78,8

71,7

81,3

85

91,3

90,6

91,8

89,2

93,2

87,8 67,3 90,7

53,5

GRN 89,4

CON RIGHT

83,4 87,2

ANTI REG IND

77,6

NA

74

79,4

80,5

63,4

65,1

886

2146

2732

3739

2124

No of RCVs

63,6

In Fig.2 the small party groups65 are correlated with the respective political strength, measured as % of seats in the EP. The high correlation coefficient gives evidence that the smaller the party groups is, the lower the cohesion become, as the European incentives relatives to nationals turn out to be minor. The small parties tend to join the biggest groups.

Index of Agreement

Fig. 2 - Small party groups and cohesion 100 95 90 85 80 75 70 65 60 55 50

R2 = 0,8122

0

2

4

6

8

10

Party group strengths (% seats)

62

AI = Max ( Y, N, A ) –1/2 ( Y+N+A) –Max (Y+N+A) / (Y+N+A), where “Y” is the total number of “yes” votes of the group “i” on an issue, “N” is the number of “no” votes and “A” is the number of abstainers. 63 Roll-call votes are called by EP parties or a large national delegation to show their position to the public on a issue, to embarrass another party or to keep a check on their own members behaviour. 64 Hix, Roury, Roland 2005 65 Alll except EPP and PES

46

Annex 20 Proportion of times majority of different parties voted the same way

Coalition building trends

0,9

PES-EPP

Proportion of Times

0,86 0,82

EPP-ELDR

0,78

PES-ELDR

0,74 0,7 0,66 0,62 0,58 0,54 0,5 EP1

EP2

EP3

EP4

EP5

47

Annex 21

45 40 35 30 25 20 15 10 5 0 0

1

2

R2 = 0,0505

3

4

5

Structural fund/GDP

Euro-Gap and Structural Funds y = -0,5855x + 50,882

120

R2 = 0,0964

100 % population

Euro Gap

Euro-Gap and Population Involved in Structural Policy Programs

80 60 40 20 0 0

10

20

30

40

50

Euro Gap

48

Annex 22 Scatter diagram, Trust in the EU Parliament and Votes to “Pro-Europe” parties66

% Trust in EP

90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

y = -0,0579x + 0,5447 2

R = 0,0013

0%

20%

40%

60%

80%

100%

% vote to "pro-Europe" parties

Scatter diagram, Positive perception of the EU membership and votes to “Pro-Europe” parties

Positive perception of EU membership

120% 100%

y = 0,2306x + 0,4646 2

R = 0,048

80% 60% 40% 20% 0% 0%

66

20%

40% 60% 80% % of votes to "Pro-Euro" parties

100%

the relative low level of R2 means that variables are not correlated.

49