Diverse Europe. Public opinion on the European Union & Cultural diversity, economics and policy. European Outlook 4

Diverse Europe Public opinion on the European Union & Cultural diversity, economics and policy European Outlook 4 Annex to the State of the European Union 2007 Paul Dekker, Sjef Ederveen, Henri de Groot, Albert van der Horst, Arjan Lejour, Bas Straathof, Henk Vinken and Charlotte Wennekers 1

PREFACE Europe presents a palette of cultures. The growing expansion of the European Union only increases the richness of its hues. This diversity of cultures is a great achievement, and Europe has learned to bridge the differences in order to work together in union. The previous European Outlooks covered the social systems, migration and working hours in Europe. In comparing the EU member states, we repeatedly encountered differences in institutions, preferences and cultures. It is this cultural diversity that is the focus of this Outlook. Key questions in part B include the following: Where do the cultures of Europe differ? What does this mean for the economic achievements of the countries? What influence does cultural diversity have on cooperation within the EU? As in previous European Outlooks, before focusing on this year s main theme, part A looks at public opinion on the EU. Even after the Referendum of 1 June 2005, there is apparently still extensive support for European integration in the Netherlands. The percentage of people in favour of a European constitution is close to the European average in our country. However, support for further enlargements has clearly dropped in recent years. Qualitative research shows that there are strong opinions on the topic of enlargement, but Europe has little place in public opinion outside the realm of research. A broad discussion on Europe was not initiated in society, and there is little indication that Europe will be a major issue in the upcoming elections. This European Outlook, like its predecessors, is a co-production of the Netherlands Bureau for Economic Policy Analysis (CPB) and the Netherlands Social and Cultural Planning Office (SCP). The SCP is primarily responsible for part A (Paul Dekker and Charlotte Wennekers) and the first chapter of part B (Henk Vinken and Paul Dekker). The CPB bears primary responsibility for chapters 2, 3 and 4 (Sjef Ederveen, Henri de Groot, Albert van der Horst, Arjan Lejour and Bas Straathof). On behalf of ourselves and the authors, we would also like to thank the members of the interdepartmental sounding board group for their useful suggestions and the staff of the Netherlands Ministry of Foreign Affairs Forward Strategy Unit (DG European Cooperation) for the active encouragement they have provided in commissioning the publication and the support they have provided towards completing the publication. These words of thanks in no way detract from the fact that the responsibility for the content of the Outlook is borne by the CPB and the SCP. This survey does not therefore necessarily represent the views of the Dutch government. Prof. Dr. Paul Schnabel Director of the Social and Cultural Planning Office of the Netherlands (SCP) Prof. Dr. Coen Teulings Director of the Netherlands Bureau for Economic Policy Analysis (CPB) 2

TABLE OF CONTENTS p. Part A. Public opinion on Europe 4 Outline and important findings 4 A1 Countries compared 6 1.1 Involvement and support 6 1.2 Positive and negative aspects of the EU membership 8 1.3 Preferences for joint policy 10 1.4 Developments in the short term 11 Appendix tables to chapter A1 13 A2 The Netherlands 15 2.1 General attitudes 15 2.2 Differences in the population 18 2.3 Topical European themes 24 2.4 Qualitative analysis of opinion formation 27 2.5 Concluding observations 31 Appendix to chapter A2: NederlandinEuropa.nl and Eurobarometer 33 References part A 35 Part B. Diverse Europe: cultural diversity, economics and policy 37 Outline and important findings 37 B1 Cultural diversity 39 1.1 Five approaches to national cultures 40 1.2 The relationship between the approaches 51 1.3 Trust 54 1.4 Cultural characteristics and attitudes towards the EU 61 B2 Culture and economic development 64 2.1 Social capital and institutions 64 2.2 Economic growth 65 2.3 Background factors 69 2.4 Concluding remarks 71 B3 Culture and economic interaction 72 3.1 Cultural differences 72 3.2 Intensity of economic interactions 76 3.3 Concluding remarks 85 B4 Culture and economic policy 86 4.1 Domestic economic policy 86 4.2 Cross-border economic policy 89 4.3 Benefits and limitations of European policy 92 4.4 Concluding remarks 96 References part B 97 3

PART A. PUBLIC OPINION ON EUROPE Outline and important findings The referendum held on 1 June 2005 on the European Constitution 1 was the definitive breakthrough in public opinion on European policy, a field which had been primarily reserved in the preceding decades for political elites and parties who had a direct interest in the effects of such policy. Public discussion on the European Union and the integration of the Netherlands into the EU brought these topics to life for the first time for many people. The collective opinions formed in response to the referendum and its results surprised politicians and others. It was widely noted that too little had been done to involve the people in Europe and that it would be good to catch the public interest and increase general knowledge on Europe. This new realisation never led to an official public discussion on Europe, headed by the government and the parliament, and the spontaneous discussion in society has died down. That is probably inevitable, considering the many domestic and international issues competing for the people s attention. It is precisely at times of reduced public discussion that it becomes important to survey how opinions on Europe develop and how population groups differ in that respect. No matter how diverse the attitudes measured in the surveys are and how unfounded the opinions may often be, they do offer information about the involvement in Europe and the support for European policy. At the least, the opinion-based survey serves to put into perspective firm statements in the political arena and in the media about what the people think about Europe. In the previous European Outlook, which was published in the year of the referendum, we mentioned the possibility that public opinion on Europe might become more negative due to the discussions. Has that happened? Has public opinion in the Netherlands (and the other no-voter, France) become more negative than in countries where referendums yielded a positive result or were not held? Have the discussions about the constitution left their traces in the form of greater involvement and a different kind of discussions? The following two chapters offer material that may help answer these types of questions. In general terms, it can be said that the attitude towards the EU in the Netherlands after the referendum did not become more negative. However, the public debate did subside considerably. We will return to this aspect in the concluding remarks in chapter A2. As in last year s Outlook, this part first compares member states, then taking a more detailed look at the Netherlands. The first chapter on country comparisons covers preferences for enlargement and for joint European policy, which will be examined in part B in the context of cultural differences. In chapter A2 on the Netherlands, the analysis of changing moods from the previous Outlook is continued based on data from the Eurobarometers and from observations made in focus groups. The availability of the large-scale Netherlands in Europe internet survey made it possible this time to take a closer look at local differences as well. 1 It is actually the constitutional treaty, but this year, as in past Outlooks, we prefer to use common terms rather than official titles. In the country comparison, the Netherlands still has a relatively positive attitude towards EU membership. Very positive moods are seen in Ireland, Luxembourg and to a slightly lesser extent Spain; negative sounds are heard from Sweden, the United Kingdom and Austria. Across the line, the new 4

member states that acceded in 2004 were more negative in the autumn of 2005 than the old EU-15. However, this latter group did express the most serious reserves where further enlargement of the EU was concerned. In assessing the favourable effects and threats of EU membership, the extent of the differences between individual countries is the most striking feature. For example, no less than 89% of the Irishmen and only 20% of the Germans see EU membership as having positive effects on employment. In France, 70% fear a loss of social welfare facilities, compared to 28% in Estonia. The follow-up analyses of the Netherlands still have not unearthed much general euroscepticism. An increasing distaste for further enlargements can be detected. If a list of countries is presented in surveys, Albania is rejected somewhat more often in the Netherlands than Turkey is, but Turkey takes centre stage in social discussions and occupies participants in focus groups more. Support for Turkey s accession grows proportionate to a more positive attitude towards the EU. Of the respondents in the Netherlands in Europe survey that consider Dutch membership in the EU to be a very bad thing, only 4% are in favour of Turkey s accession. However, among the respondents that consider Dutch membership a very good thing, 44% are in favour, but even here 43% are against it (and 13% don t know). In a list of opinions from this large-scale survey, the attitude towards Turkish membership is described in more nuanced terms. Dutch people of Turkish origin are more often in favour than others. This may in part explain the differences in support for Turkish membership between neighbourhoods in the biggest cities. 5

A1 COUNTRIES COMPARED Having 25 EU member states complicates country comparison studies: there are too many to look at individual countries and the differences between them, but too few to perform rigorous quantitative analyses in which countries are reduced to numbers. It is difficult to select a few characteristic countries. 1 We accordingly present data on all 25 member states again this year, even though it fringes on an overabundance of numbers and threatens to obscure a clear overview. The figures are presented in a number of large tables, first listing the fifteen member states from before 1 May 2004 (the old member states or EU-15 ) and then the ten new member states ( EU-10 ), more or less in order from north to south. To make it easier to compare the countries, the tables present the figures as a deviation from the country average (without taking into account any differences in population size). 1 If there were a small number of groups in bordering countries or countries in close proximity that shared extremely similar public opinions on the EU, then group averages or a limited number of countries as group representatives would suffice. Unfortunately for the purposes of this chapter but probably fortunately for Europe such a classification into groups is no longer possible. It is, for instance, possible to distinguish between clusters of countries that have a more or less EU-friendly population across the entire line (see CPB/SCP 2005: 10), but the clusters do not form homogeneous areas, either geographically or culturally; they change in composition; and, most importantly, they hold little predictive value for all sorts of opinions on important European issues. Opinions on enlargement, a constitution or joint policy differ only slightly in relation to the clusters that were identified. 2 One important side note is that we must assume that all translations of questions and statements in the surveys are correct and bear the same associations and emotional connotations. That was confirmed in general terms for the English and Dutch formulations, but not for the other languages. 3 In the first three cases, it is only necessary to choose between correct and incorrect, so many answers may be correct by coincidence. The connection between supposed insight and factual test of knowledge is positive, at the level both of countries and of individuals within the countries, but not very strong (for the countriesr=0.55 (n=25), for individuals within countries r= 0.12 0.29 (Netherlands 0.26);n = ± 1000 per country). It was noted on an earlier occasion that the Dutch are not inclined to admit ignorance on knowledge questions by choosing don t know (Lanting 2006). That was not the case here: in the Netherlands, 29% of the wrong answers were a don t know, compared to the average of 30% among the 25 countries (only 12% in Belgium; 40% and up in the United Kingdom, Spain, Estonia, Latvia, Lithuania and Cyprus). The table in section 1.1 looks at indicators for involvement in Europe and support for EU membership, for enlargement and for a European constitution. Section 1.2 discusses individual advantages of the EU and fears that membership evokes. Section 1.3 covers the preferences for joint European policy. In the closing section, we examine the four referendum countries from 2005 (Spain, France, the Netherlands and Luxembourg) and look at the developments in the short term. The chapter is primarily based on Eurobarometer 64.2 from autumn 2005, which offers measurements from after the referenda. The final section also uses earlier measurements and data from the report by the European Commission on Eurobarometer 65.0 from spring 2006 (EC 2006). 2 1.1 Involvement and support Table 1.1 shows the Netherlands, Sweden and Hungary as very involved in Europe, diametrically opposite the United Kingdom, Slovakia and Malta. People in the Netherlands also seem to understand relatively well how the European Union works, only surpassed in this respect by Luxembourg, Poland and Cyprus. These high scores for the Netherlands do not correspond to its position on a factual test of knowledge of Europe. In that respect, the Netherlands takes a middle position with an average of 38% correct answers to four questions, scoring considerably better than the United Kingdom (29%) and Italy (28%), but clearly less than Denmark, Luxembourg, Slovenia, Cyprus and Malta, where respondents answered half or more of the questions correctly on average. The following offers an indication of the knowledge that the various countries demonstrate. On average 52% of the people in the 25 member states know that the Union does not have fifteen members (Netherlands: 44%). On average 54% (Netherlands 46%) know that the European members of parliament are elected directly. On average 30% (Netherlands 17%) know that the last European elections were not held in June 2002. Finally, on average 19% (Netherlands 10%) list agriculture as the EU s most important budget expenditure. These low figures probably paint an overly optimistic picture of the actual knowledge. 3 Example from table 1.1 and the tables following: The average percentage of the population aged 15 and older in each of the 25 countries that considers EU membership to be a good thing was 50% in autumn 2005; that percentage varies from about 32 in Austria (50 18) to about 82 in Luxembourg (50+32) and is about 70 in the Netherlands (50+20). 6

Table 1.1 Involvement and support, population aged 15 and older in 2005: country average and deviations involved a perceived insight b knowledge of the facts c good benefit e enlargement thing d f favour a constitution g favour the constitution h Country average 25 43 39 50 56 54 61 52 Finland FI 0 3 2 12 11 9 13 4 Sweden SE 19 3 3 11 24 6 17 9 Denmark DK 9 4 12 5 13 8 16 4 United Kingdom UK 13 8 10 17 19 11 15 10 Ireland IE 5 7 3 23 30 0 3 3 The Netherlands NL 18 9 1 20 5 6 0 5 Belgium BE 4 3 2 9 9 7 16 21 Luxembourg LU 6 14 18 32 19 23 8. Germany DE 2 3 8 3 10 18 13. Austria AT 3 0 4 18 21 25 13. Italy IT 9 5 11 0 8 1 9. Greece EL 8 1 4 3 11 20 7. France FR 1 4 1 4 5 22 6 6 Spain ES 2 14 7 16 13 2 1. Portugal PT 1 10 2 7 9 1 2 3 Estonia EE 8 3 5 9 0 2 12 2 Latvia LV 4 5 7 14 6 8 4. Lithuania LT 7 2 3 7 14 15 4. Poland PL 3 16 2 3 7 19 1 5 Czech Republic CZ 9 8 4 6 1 11 11 14 Slovakia SK 14 10 6 1 6 14 3. Hungary HU 26 2 5 11 15 13 14. Slovenia SI 5 7 12 9 17 13 10. Cyprus CY 1 13 10 7 1 3 1. Malta MT 10 8 13 7 1 20 13. a Percentage that tends to/totally agrees with the statement I feel very much involved in European affairs. In this case and later, don t know is counted toward the complementary categories, such as not involved, false and disagree. b Percentage that tends to agree rather than disagree with the statement I understand how the European Union works. c Percentage of correct answers in response to The European Union currently consists of fifteen member states (false), The members of the European Parliament are elected directly by the citizens of the European Union (true) and The last European elections took place in June 2002 (false), and choosing agriculture from seven options as the EU s most important budget expenditure. d Percentage that, generally speaking, think that their country s membership in the European Union is a good thing. e Percentage that, taking everything into account, say that their own country has on balance benefited from being a member of the EU. f Percentage that is in favour of enlarging the Union to include other countries in future years. g Percentage that is in favour of a constitution for the European Union. h Percentage that, according to what they know of it, tend to be or are totally in favour of the European Constitution. Source: Eurobarometer 64.2 (October-November 2005) Support for the EU membership of the respondents own country displays major differences, ranging from 32% of Austrians to 82% of Luxembourgers, and from 32% of Swedes to 86% of Irishmen. Both indicators for support ( good thing and benefit ) correlate, but not perfectly (r= 0.83 for the 25 countries) and there are striking differences in the degree to which a country deviates from the country averages. For example, Denmark and Greece score relatively high on benefit and the Netherlands and Luxembourg score relatively high on good thing. That may be a matter of the different translations and their emotional associations, but it may also have something to do with an actual difference in the link between seeing their own advantage and being in favour of the EU. According to Bruter (2005) the latter is the case. He argues that good thing and benefit were once the same thing in the six old member states and were also strongly correlated at an individual level, but that benefit now often scores lower than good thing and the individual link has decreased due to the development of a European identity. This has theoretically led to an increasing proportion of Eurobarometer respondents who keep 7

favouring further European integration while believing that it does not benefit them and their country anymore. This is in contrast to an earlier situation when support and perceived benefits were much more strongly correlated (Bruter 2005: 174). The figures in Table 1.1 offer some support for this idea of progressing Europeanisation. In the original six, the later nine and the latest ten member states, the difference between the average deviations for good thing and benefit are, respectively, 10 (+8-2), 0 (0 0) and -4 (-5-1). However, the link between the two views at an individual level does not show the expected decrease: 0.53 on average for the original six, 0.63 for the later nine and 0.54 for the latest ten. The people of the oldest and newest member states seem on average to be equally calculating in their assessment of the EU membership, less so than the people of the countries in the middle group. 1 General support for expanding the Union by adding new countries varies from 29% in Austria to 74% in Greece and Malta. In the following chapter on the Netherlands, this European Outlook takes a closer look at the expansions and Table 1 in the appendix to this chapter accordingly offers more information on this topic. This shows that most countries distinguish roughly between three groups of possible accession candidates: the popular Western European countries of Switzerland, Norway and Iceland, the unpopular candidates Albania and Turkey, and a number of Central and Eastern European countries between these two groups. It should be noted that a comparison of the 25 countries reveals no connection between the levels of support for enlargement and support for EU membership of the respondents own country. 2 1 Analyses for the entire EU of the answers to a Eurobarometer question regarding the degree of involvement with Europe have shown that seeing benefits for the respondents own country is by far the strongest predictor of the operationalised European identification. Dieser Befund deutet darauf hin, dass die subjektive Verbundenheit der Bürger mit Europa eher über den verstand, d.h. ökonomisch-rational, gesteuert ist, als dass sie den Europäern eine Herzensangelegenheit wäre (Noll and Scheuer 2006: 5). See also Scheuer (2005) for an in-depth analysis of the link between many measurements in the Eurobarometer of involvement in, evaluations of and emotions towards the EU. This analysis shows that support for European integration generally cannot be traced back to considerations of usefulness and self-interest. 2 The correlations between the two measurements of support for enlargement (in table 1.1 and the summary measure in appendix table 1; lowest measurement 0.68 correlated) and the two measurements of support for EU membership fluctuate around 0. However, the links at an individual level are positive, although not always equally strong: those who consider EU membership for their own country to be a good thing or a benefit often favour further enlargement. 3 Unfortunately, this survey on the various countries did not look at voting behaviour in referenda, so it is not possible to check how many no-voters changed their opinion after the fact or are underrepresented among the respondents. See also section 1.3 and section 2.4. The last two columns of table 1.1 are about the European constitution. On average, 61% of the citizens in the 25 member states are in favour of a constitution. That is less than one year earlier, but slightly more than six months before or after that point (see also table 1.4). The Dutch percentage corresponds to the country average, which is substantially more than in the Scandinavian member states, the United Kingdom, Austria, Estonia and the Czech Republic. The last column lists data for a more limited number of countries, reporting on the support for the constitution presented to the people in the 2005 referenda; the Netherlands (and its fellow no-voter, France) shows a markedly positive score. The majority of the respondents tended to support or totally supported the constitution. 3 In Sweden, the United Kingdom and the Czech Republic, a majority is nowhere to be seen. It is therefore highly improbable that France and the Netherlands would have been the only troublemakers if referenda had been used more widely in the ratification of the constitutional treaty. 1.2 Positive and negative aspects of the EU membership Support for EU membership of the respondents own country was covered in table 1.1 in the columns on good thing and benefit. But what benefits do people see in membership, and what dangers do they think are associated with it? The following two tables cover the perceived positive effects and feared dangers in four areas. Table 1.2 shows that the Dutch are on average positive about the effects of EU membership on national security, standard of living and employment. They are more positive about the consequences for exports than the people in other countries. There are some very major differences between the member states in how benefits are perceived, from 35% in Lithuania to 87% in Ireland that see a positive effect on the standard of living, and 20% in Germany to 79% in Ireland that observed a positive effect on employment. 8

Table 1.2 Perception of positive effects of the EU membership for the respondents own country in four areas,a population aged 15 and over in 2005: country average and deviations security exports standard of living employment Country average 70 65 56 40 Finland 1 12 2 2 Sweden 12 4 3 2 Denmark 9 22 21 12 United Kingdom 26 14 4 2 Ireland 3 18 31 39 The Netherlands 4 18 3 2 Belgium 4 14 5 8 Luxembourg 4 16 17 1 Germany 4 13 10 20 Austria 23 3 12 13 Italy 3 5 0 1 Greece 8 9 5 4 France 5 8 14 18 Spain 6 8 9 12 Portugal 14 20 7 13 Estonia 6 5 0 15 Latvia 5 7 21 9 Lithuania 5 3 6 22 Poland 7 4 3 1 Czech Republic 9 1 8 8 Slovakia 5 8 5 14 Hungary 1 17 14 8 Slovenia 11 2 4 5 Cyprus 10 13 8 9 Malta 3 19 3 14 a People disagree about the advantages or disadvantages of [OUR COUNTRY] belonging to the European Union. I am going to read out some points and, for each one, I would like you to tell me if [OUR COUNTRY] being in the European Union has a very good, fairly good, bad or very bad effect? The percentage of very good and fairly good is listed for a selection of 4 out of 7 points. Source: Eurobarometer 64.2 (October-November 2005) 1 The statistical strength of the eight characteristics together varies, from a paltry Nagelkerke pseudo R2 of 0.26 and 0.29 in Ireland to a respectable 0.52 and 0.74 in Greece. Table 1.3 shows the Netherlands back in a middle position, but now with a clearly increased fear of a loss of social facilities (58% compared to the average of 49%). Major differences can be seen again: 28% in Estonia versus 70% in France that fear the loss of social facilities, 29% in Poland versus 66% in the UK that fear a loss of national identity and culture, and 46% in Lithuania versus 89% in France that are afraid of the transfer of jobs to other member states. How are the perceptions of the positive effects (table 1.2) and the fears (table 1.3) related to the general support for membership (table 1.1)? The support in a country appears to be most closely related to the extent to which the people see a positive effect on the standard of living and slightly less closely related to the lack of fear that the respondents own country will have to pay more and more, and to the expected positive effect on employment. There is no relation to the level of fear for the loss of social facilities or cultural identity, or to the transfer of jobs within the EU. Looking at the individual background of support for membership, the effect on the standard of living is seen in nearly all the countries to be the most important of the eight perceived effects and fears. The fear of job transfers is almost never important. The statistical effects of the other aspects range between those two and differ among the countries. 1 9

Table 1.3 Fears about the EU,a population aged 15 and over in 2005: country average and deviations our country paying more and more to the European Union the loss of social benefits thelossofnational identity and culture the transfer of jobs to other member states which have lower production costs Country average 63 49 41 71 Finland 17 1 5 12 Sweden 7 5 11 8 Denmark 12 2 0 4 United Kingdom 8 4 25 2 Ireland 12 10 9 6 The Netherlands 3 9 2 4 Belgium 8 12 1 15 Luxembourg 14 9 2 8 Germany 11 15 3 13 Austria 7 13 2 0 Italy 4 10 4 5 Greece 4 4 5 14 France 7 21 1 18 Spain 8 4 7 15 Portugal 0 16 5 9 Estonia 1 21 4 17 Latvia 7 8 5 10 Lithuania 13 14 4 25 Poland 1 8 12 13 Czech Republic 3 11 4 3 Slovakia 2 3 2 8 Hungary 3 7 11 4 Slovenia 6 15 1 1 Cyprus 1 8 8 1 Malta 12 7 8 4 a Some people may have fears about the building of Europe, the European Union. Here is a list of things, which some people say they are afraid of. For each one, please tell me if you personally, are currently afraid of it, or not? The percentage of currently afraid of is listed for a selection of 4 out of 10 topics. Source: Eurobarometer 64.2 (October-November 2005) 1.3 Preferences for joint policy 1 In literal terms, this does not necessarily entail policy set by EU authorities; by choosing the option of joint decisions, the respondent could be indicating a preference for agreements between two or more EU countries without involving Brussels. However, this seems unlikely. An extensive table of preferences for European policy has been enclosed in the appendix to this chapter. Respondents to the Eurobarometer were asked for a number of areas whether they thought that decisions should be made by the [national] government or made jointly within the European Union. 1 In autumn 2005, a list of sixteen topics was presented. In appendix table 2, they are arranged in order according to the extent to which country-average European policy is preferred. The preference is strongest for fighting terrorism and weakest for pensions. In the table, as in the entire section, these figures for the countries do not take into account the size of the individual populations. As measurements in 2002, 2003 and 2004 showed (CPB/SCP 2003: 35, 2004: 16, 2005: 43), the citizens of the member states generally appear to be strong proponents of joint policy on clearly international topics and on dealing with specific large problems, and are more reserved about joint policy where traditional tasks of the social welfare state are concerned (see also Binnema 2005). Looking at the last row of appendix table 2, the new member states are more interested in joint policy (on average 60% want it) than the old member states (51%). Some extremes do present themselves, Finland in particular. Only 17% of Finns are interested in joint policy on defence and foreign affairs (80% in Estonia want it; 67% on average) and only 18% want joint immigration policy (73% in Poland; 58% on average). The Netherlands displayed a very average score across the board, but clearly deviated in specific areas. For example, the 10

Netherlands often agrees with Hungary in favour of joint environmental policy and often agrees with Cyprus and Malta in favour of joint support for weak regions. 1.4 Developments in the short term This year s Outlook discusses developments in the recent past (autumn 2004 spring 2006), focusing on the four countries that organised a referendum on the constitution in 2005. Three of the four countries were covered last year: Spain (referendum held on 20 February 2005, 42% voter turnout, of which 77% were in favour), France (29 May, 70% turnout and 45% in favour), and the Netherlands (1 June, 63% turnout and 38% in favour). Luxembourg (10 July, 87% turnout and 57% in favour; Dekker and Wennekers 2005) now joins that list. To what extent do the developments in public opinion differ between the two countries that rejected the constitution and the two countries that accepted it, and the rest of the EU? Table 1.4 offers the figures for four indicators from table 1.1. Table 1.4 Developments in the short term a autumn 2004 spring 2005 autumn 2005 spring 2006 difference between measurements before and after mid-2005 generally speaking, considers it a good thing that the own country isamemberoftheeu taking everything into account, says that the own country has on balance benefited from being a member of the EU in favour of further enlargement of the Union to include new members in future years in favour ofa constitution for the European Union The Netherlands 75 77 70 74 4 Luxembourg 85 80 82 72 5 France 56 52 46 49 6 Spain 72 66 66 72 0 average other 21 54 51 47 52 3 The Netherlands 59 67 61 63 1 Luxembourg 73 72 75 71 + 1 France 54 53 51 50 3 Spain 70 69 69 71 + 1 average other 21 56 57 54 58 1 The Netherlands 50 45 48 43 2 Luxembourg 38 33 31 27 6 France 39 32 31 31 4 Spain 67 56 56 55 6 average other 21 57 58 56 51 4 The Netherlands 73 53 62 59 3 Luxembourg 77 63 69 64 4 France 70 61 67 62 1 Spain 72 63 62 63 5 average other 21 65 58 60 57 3 a The percentages, including rejection of the statement or the answer don t know, add up to 100%. Source: Eurobarometer 62.0 (2004), 63.4 and 64.2 (2005), and the report on Eurobarometer 64 (EC 2006a) The table displays developments, some of which may initially seem to be predictable (a drop in 2005 in a good thing in the Netherlands and France, stability in Luxembourg and Spain), and others surprising (a huge drop in support for a constitution in the Netherlands prior to the referendum, followed by some recovery). If the fluctuations between measurements are taken out of consideration, looking in the last column at the different changes in public opinion between the two measurements before the referendum and the two measurements after, then no conflicting 11

development can be identified across the line between France and the Netherlands on the one hand and Spain and Luxembourg on the other. In comparing the countries, we found nothing to support our speculation in the previous Outlook (CPB/SCP 2005: 38) that the Dutch people might become more negative as a result of the discussions surrounding the referendum on 1 June 2005. The following chapter will take a closer look at the developments in the Netherlands. 12

Appendix tables to chapter A1 Appendix table A1.1. Support for accessions in the 25 member states of the European Union, 2005, (unweighted) averages of the national percentages of the population aged 15 and older and country deviation from this average in percentage points a country average average support per country for twelve accessions and deviation from the average in % points FI SE DK UK IE NL BE LU DE AT IT EL FR ES PT EE LV LT PL CZ SK HU SI CY MT average support for new accessions 54 55 70 56 45 51 57 55 43 44 33 41 64 47 50 49 55 52 57 62 56 60 54 67 59 62 deviations from average support: Switzerland 28 39 20 31 22 22 34 32 37 40 39 31 24 29 16 22 28 25 23 19 33 30 29 26 26 15 Norway 27 39 21 37 23 22 37 32 39 40 38 26 22 33 16 20 28 23 22 19 32 28 24 24 27 13 Iceland 19 38 21 33 16 12 27 25 33 27 23 20 16 24 7 6 22 18 16 10 24 10 11 20 17 11 Croatia 1 10 0 5 4 0 1 0 7 3 17 2 6 4 2 4 6 9 1 8 20 19 22 8 6 3 Bulgaria 1 2 0 3 1 1 5 3 4 4 12 5 0 2 0 0 8 10 11 3 8 9 3 2 5 2 Romania 4 7 2 6 3 7 6 8 4 12 16 5 3 6 3 4 2 4 2 8 14 6 1 5 10 0 Ukraine 6 16 11 15 6 7 18 11 12 14 14 4 5 9 6 5 1 5 10 3 15 5 10 1 6 1 Bosnia and Herzegovina 8 14 4 10 9 3 10 10 11 11 9 10 3 9 4 7 12 13 9 7 12 7 15 6 8 3 Macedonia 9 12 5 9 7 5 8 6 14 9 11 12 18 8 3 5 7 11 7 5 10 7 16 4 11 2 Serbia and Montenegro 10 17 6 14 11 8 14 10 13 14 13 14 10 10 6 8 15 14 12 7 10 11 14 6 0 4 Albania 18 18 13 15 12 8 20 22 19 22 20 15 27 14 5 6 18 18 20 14 30 29 23 20 24 7 Turkey 22 24 22 23 8 18 16 19 24 23 22 14 44 26 10 9 28 21 31 20 25 32 13 18 43 23 a For each of the following countries, would you be in favour or against it becoming part of the European Union in the future? The countries are listed in random order. Answer categories: in favour/against/don t know. The basic questionnaire refers to membership in the future. Example: In Finland, on average 55% are in favour of EU membership for the twelve countries listed; 94% (55+39) of Finland favours accession for Switzerland and Norway; 31% (55 24) favour accession for Turkey. In contrast to appendix table A1.2, the deviations per column total 0. Source: Eurobarometer 64.2 (October-November 2005); weighted results 13

Appendix table A1.2 Preferences for European policy in the 25 member states of the European Union, 2005, (unweighted) averages of the national percentages of the population aged 15 and older and country deviation from this average in percentage points a country average deviation from the country average in % points FI SE DK UK IE NL BE LU DE AT IT EL FR ES PT EE LV LT PL CZ SK HU SI CY MT fighting terrorism 84 0 4 4 20 5 5 4 3 0 9 9 10 2 10 4 8 6 5 6 5 8 9 3 1 4 scientific and technological research 78 16 6 5 14 4 1 2 11 17 13 4 2 2 1 1 6 5 4 4 4 8 7 6 17 6 support for regions facing economic difficulties 68 12 2 5 9 11 16 2 16 13 11 17 1 22 3 5 3 10 8 3 17 3 7 7 16 16 protecting the environment 67 18 1 4 6 8 14 3 1 2 9 6 4 6 7 2 1 3 2 7 2 5 13 9 11 0 defence and foreign affairs 67 50 29 10 25 4-8 12 13 3 8 6 7 9 6 1 13 10 7 2 9 18 14 3 4 13 competition policy 64 15 2 0 14 5 5 3 5 7 11 0 2 5 6 2 1 2 2 2 9 0 11 8 18 4 fighting crime 64 5 19 1 30 22 5 3 4 7 1 2 2 10 5 6 10 13 5 11 11 9 15 2 2 0 energy policy 64 25 10 3 13 7-3 2 11 6 17 4 8 9 9 9 5 11 3 7 15 2 6 5 22 10 immigration 58 40 27 11 18 1-4 8 2 1 16 10 4 12 15 4 9 8 12 15 1 9 10 8 6 14 agricultural and fishing policy 56 38 10 8 18 2 12 9 5 2 10 1 2 0 5 1 1 6 10 5 6 3 5 6 20 1 consumer protection 51 17 12 9 11 6-3 1 6 2 14 2 11 1 8 2 6 8 8 9 1 0 8 6 18 3 fighting unemployment 44 22 12 11 17 8 9 1 5 12 1 8 14 6 4 6 7 10 12 6 2 5 13 3 14 8 health and social welfare 34 25 23 20 9 5 11 4 6 14 10 9 18 8 7 4 16 22 21 4 1 5 10 1 31 1 the education system 32 20 14 7 16 10 14 4 13 7 1 11 13 4 6 2 1 12 11 3 12 3 9 9 17 6 taxation 27 13 12 16 10 2 6 1 15 6 2 9 14 2 10 4 1 16 18 0 3 8 14 2 9 4 pensions 25 19 20 19 8 2 15 6 15 14 8 9 20 5 11 10 10 18 15 6 1 0 8 0 17 1 average preference for European policy 55 20 12 7 15 2 1 2 0 5 9 2 5 1 5 2 3 10 9 4 1 2 10 2 14 6 For each of the following areas, do you think that decisions should be made by the [national] government, or made jointly within the European Union? 16 policy areas and topics were presented in random order; all They are included here in order of decreasing support for joint policy. Don t know answers were not included. a Example: On average 84% of the population in the member states thinks that terrorism should be dealt with in a European context; in Finland that figure is 84% (84+0), 64% (84 20) in the United Kingdom and 93% (84+9) in Hungary. In contrast to appendix table A1.2, the deviations per row total 0. The 15 old and 10 new member states have been listed in order from north to south as much as possible: Finland (FI), Sweden (SE), Denmark (DK), United Kingdom (UK), Ireland (IE), The Netherlands (NL), Belgium (BE), Luxembourg (LU), Germany (DE), Austria (AT), Italy (IT), Greece (EL), France (FR), Spain (ES) and Portugal (PT)/Estonia (EE), Latvia (LV), Lithuania (LT), Poland (PL), Czech Republic (CZ), Slovakia (SK), Hungary (HU), Slovenia (SI), Cyprus (CY) and Malta (MT). Source: Eurobarometer 64.2 (October-November 2005); weighted results 14

A2 THE NETHERLANDS This chapter looks at public opinion in the Netherlands. Developments in some general attitudes towards the EU and how they are connected are examined first (section 2.1). The chapter then looks at with differences between population groups, this year primarily looking at regional and local differences (section 2.2), continuing with several topical issues based on survey data (section 2.3) and the findings from qualitative research (section 2.4). It closes with concluding observations on this and the previous chapter. The data on the Netherlands in this chapter is derived from the Eurobarometers, the Netherlands in Europe internet survey from spring 2006 and the associated focus groups. The internet survey contains answers to questionnaires from people who went to the website on their own initiative, often in response to advertisements calling the people to share their opinions on the EU. The Netherlands in Europe website attracted 128,059 visitors; 97,452 people filled out the questionnaire in its entirety (Anker 2006: 5). However, we have some reservations in using this internet survey for statements on what the Dutch population as a whole thinks, because we suspect that people who have internet access, are interested in surveys and are involved in Europe are overrepresented. We have corrected for that problem as much as possible by including a weighting factor based on social demographics and one attitude about the EU in a survey in which a more effective check was possible on how representative the sampling was. However, some distortion due to self-selection remains probable, which is why we use the data primarily to compare population groups and to study the links between opinions. (The appendix to this chapter offers further explanation for the use of the internet survey.) 2.1 General attitudes Developments in support for Europe, 2001 2006 How has general support for the EU in the Netherlands developed in the recent past? As an introduction, figure 2.1 offers a graphic representation of the development of three indicators from table 1.1 and table 1.3. The assessments that Dutch membership is a good thing and that the Netherlands benefits from EU membership run parallel. In autumn 2003 and spring 2004, support was at its lowest point, only to climb up again. In spring 2005, both good thing and benefit were at their highest point in five years. After the referendum in June, both forms of support dropped, but both climbed again in spring 2006. A different development can be seen in the support for further enlargement of the Union: a steady decrease with an interim downward spike in spring 2004, when ten new member states acceded to the Union on 1 May. Support recovered in autumn 2004, only to drop again. Over a five-year period, support for further enlargement has decreased by 15 percentage points (from 58% to 43%). In table 1.4 in the previous chapter, we identified a decrease in other member states as well. Sections 2.3 and 2.4 return to attitudes towards further enlargement. 15

Figure 2.1 Developments in support for EU membership and for enlargement, 2001 2006 80 75 70 65 60 55 50 45 40 autumn 2001 spring 2002 autumn 2002 spring 2003 autumn 2003 spring 2004 autumn 2004 spring 2005 autumn 2005 spring 2006 Good thing Benefit For enlargement Good thing: Generally speaking, do you think that the Netherlands membership of the European Union is a good thing? Benefit: Taking everything into account, would you say that the Netherlands has on balance benefited from being a member of the European Union? Enlargement: Are you in favour of enlarging the Union to include new members in future years? Source: Eurobarometer 56.2, 57.1, 58.1, 59.1, 60.1, 61.0, 62.0, 63.4, 64.2 and report (EC 2006a) on 65; weighted results The speed of integration Since 1986, the average desired speed of European integration (how fast do people want European integration to take place?) has been higher than the average perceived speed of integration (how fast do people think European integration?), as figure 2.2 shows. Over time, however, perception and desire are converging. Between autumn 2004 and autumn 2005, the perceived speed of integration remained the same, but the desired speed dropped significantly. In that year, the percentage of people who felt that integration was taking place too fast rose from 20% to 26% (not shown on graph). 16

Figure 2.2 Perceived and desired speed of European integration,a 1986 2005 75 70 65 60 55 50 45 40 35 30 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 perceived speed desired speed a The perceived speed and desired speed of the process of building Europe were measured on a seven-point scale, visualised as seven little men ranging from standing still to running as fast as possible. Source: Eurobarometer 26.0, 28.0, 34.0, 37.0, 39.1, 40, 41.0, 42, 43.1, 44.1, 46.0, 48, 52.0, 53.0, 54.1, 55.0, 56.2, 58.1, 60.1, 61.0, 62.0, 63.4 and 64.2, weighted results Knowledge and involvement Table 2.1 displays developments in how the Dutch rate their own knowledge of and insight into the EU. In the period from 2001 2004, the percentage of respondents that said they knew quite a lot or a lot about the EU fluctuated between 31% and 41%. In spring 2005, this figure suddenly jumped to 50%. It would seem obvious to assume that the referendum played a part in this: people were suddenly confronted with European issues more directly and felt that they knew more about the EU. This (supposed) learning effect subsided quickly after the referendum. In autumn 2005, the percentage of people that gave themselves a satisfactory mark was back at 35%, which is more in line with earlier developments. Table 2.1 Knowledge of and insight into the EU, 2001 2005, in percentages of the population aged 15 and over autumn 2001 spring 2002 autumn 2002 spring 2003 autumn 2003 spring 2004 autumn 2004 spring 2005 autumn 2005 rates personal knowledge of EU as sufficient/good 34 32 38 34 31 38 41 50 35 states an understanding of how the EU works... 43. 45 51. 53 reports familiarity with at least seven out of nine EU institutions 30 27 30 20 27 26 20. 21 Source: Eurobarometer 56.2, 57.1, 58.1, 59.1, 60.1, 61.0, 62.0, 63.4 and 64.2; weighted results 17

The percentage of people that say they understand how the EU works rose as early as autumn 2004. There is little difference between autumn 2004 and autumn 2005. Familiarity with EU institutions has in fact declined in the past four years: from 30% in 2001 to 21% in 2005. All in all, the self-assessments do not paint a rosy picture of the average level of knowledge on the EU. Chapter A1 showed that the average Dutch person does in fact demonstrate a fairly mediocre level of information, but is still in the middle range in a European context. Are increasing involvement and knowledge accompanied by more support for the EU? Table 2.2 shows that that is indeed the case for involvement: 81% of the people who rate themselves as very involved also consider Dutch EU membership to be a good thing. Of those who see themselves as totally uninvolved, this figure is only 28%. However, there are also involved opponents: 11% of those who are closely involved view Dutch membership as a bad thing to a very bad thing. This group is not large. We see a different picture for knowledge about. Precisely among those people who consider themselves very well informed, support for Dutch membership is relatively low: 63% consider membership a good thing, 24% consider membership a bad thing. The conviction that Dutch membership is a bad thing only occurs more often among people who say they know (almost) nothing about it (35%). Support for EU membership is highest among people who feel that they know quite a lot or only a bit. There seems to be a curvilinear connection: as knowledge about Europe increases, support initially increases, only to decrease again. 1 Table 2.2 Assessment of EU membership based on involvement and knowledge in various groups (in %) groups (percentage of respondents in parentheses) membership is a good thing membership is a bad thing do you feel involved in European affairs? how well informed do you feel you are regarding issues that are relevant in Europe and that could influence the Netherlands? very involved (7) 81 11 involved (35) 82 7 neutral (27) 67 9 not involved (16) 52 20 completely uninvolved (13) 28 48 very well informed (2) 63 24 well informed (18) 73 13 a bit informed (42) 72 10 not informed (25) 61 16 completely uninformed (11) 41 35 Source: NederlandinEuropa.nl (spring 2006), weighted results 2.2 Differences in the population 1 There is a linear connection to support (r=0.22), but that is weaker than the relationships between support and involvement (0.44). 2 Research by Aarts and Van der Kolk (2005) shows that political parties with the exception of the small left-wing and right-wing parties are in fact more positive about the euro, the speed of integration and the enlargement than their supporters. According to Aarts and Van der Kolk, parties are inclined to underestimate the differences between themselves and their adherents. This year we are focusing special attention on socio-economic position and living environment. Accordingly, table 2.3 lists the background variables of education, profession, ethnic origin, level of urbanisation, social status of the neighbourhood and political affiliation. The first European Outlook (CPB/SCP 2003) showed that the level of education was an important predictor for attitudes towards the EU. Two years ago, the European Outlook spotlighted political party preference in combination with attitudes towards the EU (CPB/SCP 2004). One of the conclusions at that time was that voters often see political parties as being more in favour of the European Union than the voters themselves. 2 A great deal has happened in the meantime. Most of the major political parties 18

campaigned in favour of the European constitution, while most voters voted against the constitution. Where 85 per cent of the members of the Lower House were proponents of the European constitution, only 38.5 per cent of the voters who showed up proved to share the same opinion. (Aarts and Van der Kolk 2005: 45) After the referendum, there was a great outcry to increase politicalisation in the area of Europe (see e.g. AIV 2005; Nicolaï 2005; TK 2006a; Van den Berg, not yet published; Hollander, not yet published). Political parties were accused of doing too little to differentiate and shape public opinion on European themes. In response, most of the political parties considered the matter of Europe and announced their visions on the topic (AIV 2005: 15). However, this primarily led to discussions within political parties, rather than discussions between political parties. In this light, it is interesting to take another look at the relationship between party preference and attitudes towards Europe. To what extent is there still a discrepancy between what the voters think and what the parties think? The indicators in table 2.3 are knowledge, involvement, support for EU membership and support for Turkey s accession. The first three correspond to the attitudes reported in section 2.1, while the views on Turkey look forward to section 2.3. This concerns agreement with Turkish membership if Turkey is mentioned in a series of possible accession candidates. The support measured using this method (22%) is lower than when conditions for accession are also mentioned. 37% of the respondents in the Netherlands in Europe survey agree with the proposition Turkey may accede to the EU once it meets all the requirements (see table 2.6). Table 2.3 shows that level of education and party preference are particularly important predictors for attitudes towards Europe. There is a major difference between people who attended higher or lower education: people with higher education are much more often positive about Dutch EU membership (+37% compared to people with lower education) and much more often label themselves as involved (+27%). People with higher education are also more often in favour of Turkey s accession, but the differences are smaller here (+17%). On knowledge, people who attended higher education scored only 7% higher than people with lower education. Other groups that achieved notable scores include people in highly urbanised areas (above average positive and involved), people from neighbourhoods that have a high social status (idem), self-employed entrepreneurs (above average involvement), school pupils (above average positive about Dutch membership and Turkey s accession) and respondents of Turkish origin. This last group is largely in favour of Turkey s accession to the EU. 1 We will return to this group in section 2.3. We first look at differences based on political preference, urbanisation and the status of the neighbourhood. 1 An additional warning regarding the representative sampling is appropriate here. The 583 respondents of Turkish origin included a relatively high proportion of young people and highly educated people. This signifies more a vanguard of opinion leaders than a sound representation of the silent majority of the population of Turkish origin. 19