Accounting for confidence in institutions in Greece: institutional performance, social capital & EU public policy

Transcription

1 Accounting for confidence in institutions in Greece: institutional performance, social capital & EU public policy Christos J. Paraskevopoulos University of Thessaloniki, Egnatia 156, Thessaloniki, Greece Tel: Paper prepared for the ECPR Workshop on The Interrelationship between Institutional Performance and Political Support in Europe: Causes and Consequences Muenster, Germany, March 2010 DRAFT Please do not quote 1

2 Introduction Institutions are the basic pillars of representative democracy and society at large and, given their longevity, they are far more important factors for the functioning of democracy than several actors (politicians, political leaders or citizens) (Newton and Norris, 2000). Furthermore, in the complicated and impersonal modern world institutions are viewed as key components of political stability and integration, primarily through their function as information conveyors among actors. Therefore, confidence in major public institutions constitutes a prerequisite of a well-functioning democracy, while, conversely, erosion of this confidence indicates public disaffection with and poses a major threat to democracy. Moreover, although they are viewed by many as a far more crucial variable than interpersonal and social trust, institutions may play a key role in the creation of generalized/social trust and social capital at large, which is also considered a key concept for well-functioning liberal democracy and increased levels of performance in public policy at large. Indeed, relatively recent research has shown that the fairness and impartiality of formal state institutions and institutional performance at large constitute crucial variables affecting the level of social trust/social capital and consequently public policy outcomes. In this theoretical framework, Greece demonstrates low levels of institutional performance, especially in the implementation of crucial EU public policy areas, such as macro-economic adjustment, environmental policy, socio-economic cohesion and so on. Additionally, Greece is widely considered as a weak country in social capital endowments and strength of civil society (see Mouzelis, 1995; Paraskevopoulos, 2001a-c, 2005a,b, 2007), characterized by what Tsoukalas (1995) has identified as peculiar Greek individualism or free-riders in wonderland, high levels of corruption and relatively low trust in political institutions. In this respect, obviously, the interplay between corruption, low levels of institutional performance affecting public policy outcomes and low social capital may constitute a vicious circle that might indeed become an explosive mix for Greece s political system and public policy-making structure. Hence, Greece may also be considered a representative case study country of the wider regions of S/E and/or C/E Europe, which demonstrate similar characteristics. In light of the above, this paper, drawing primarily on the institutional theory of trust, concentrates on the following objectives: the first section presents a very brief discussion of the relevant theoretical dichotomy in the literature about confidence in public institutions between the socio-cultural and institutional approaches; the second section examines the 2

3 crucial interplay between institutional performance, social capital and public policy in Greece; the third section analyses and accounts for the specifics of confidence in public institutions in Greece; the fourth section presents a brief report on the implications of the level of institutional performance and social capital for the implementation of EU public policy in Greece; finally, the fifth section summarizes the main conclusions with regard to the interplay between institutional performance, confidence in institutions and public policy outcomes in Greece. Confidence in public institutions: culture vs performance The literature and research on confidence in institutions over the last two decades has been dominated by a fundamental and crucial theoretical dichotomy. On the one hand, the so called socio-cultural approach views institutional confidence as embedded in and generated by culturally and historically determined networks of civic engagement (i.e. associations, civic society organizations) and relevant experiences of early life socialization. On the other hand, the so called institutionalist approach emphasizes the role of citizens evaluations of the actual institutional performance in generating confidence in particular institutions (see Paraskevopoulos, 2010). Indeed, the role of participation in civil society and associative networks in the creation and generation of trust and norms of reciprocity is of crucial importance for the theory of social trust and social capital at large, and can be traced back to J.S Mill and Tocqueville (see Coleman, 1990). Yet, more recently, it has been operationalized and, indeed in a rather popular way, by Putnam (1993). Thus, this approach views institutional confidence and wellfunctioning democracy as an extension of social trust and civic/cooperative culture which is learned early in life through participation and involvement in voluntary activities (see Almond and Verba, 1963; Ostrom, 1990; Uslaner, 2002) and, therefore it assumes a strong correlation between social trust and confidence in public institutions. In that respect, it is also similar to Sabel s notion of studied trust, which is based on the hypothesis that 'trust is a constitutive -hence in principle extensive- feature of social life' (1993:140) and hence it can be learned through actors participation and involvement in civil society networks. Yet, in this theoretical framework, the creation and destruction of trust and social capital are marked by virtuous and vicious circles (Putnam, 1993:170), which has engendered criticisms focusing on historicism and path dependence logic, while several scholars have questioned 3

4 its underlying assumption of spill over effect from learning to trust other co-members in civic associations to trusting unknown people in more general contexts and public institutions at large (see Boix and Posner, 1996; Sabetti, 1996; Levi, 1996; Tarrow, 1996). The institutional literature s critique of the main assumptions of the socio-cultural approach is based on the fundamental assumption that institutional performance and not culture determine confidence in institutions and therefore it does not necessarily predict a close association between institutional confidence and social trust (see Newton and Norris, 2000; Mishler and Rose, 2001; Stolle and Hooghe, 2003; Espinal et. al., 2006). Yet, the institutional approach provides for a relationship, albeit an indirect one, between confidence in public institutions and social trust. In particular, the so called institutional theory of trust (Rothstein and Stolle, 2003; Rothstein, 2002, 2005) attributes a very important role to the perceptions of fairness and impartiality of public institutions on the part of citizens as a crucial variable/determinant affecting the creation of generalized trust and hence the building up of social capital. In a similar vein, Herreros (2004) points to the crucial role of formal social and political institutions -as providers of external solutions to dilemmas of collective action- in the creation of social capital in two important respects: first, a direct one, as guarantor of agreements, that is sanctioning agent; and second, an indirect one, as facilitator of increased participation in associations and hence of building social capital through the provision of relevant incentives. While the former function refers to the role of social and political institutions in the form of impartial state (street-level) bureaucracy, the latter refers to the universal welfare state or, in a broader perspective, to the role of institutions in enhancing generalized trust through absorbing or diminishing the risk associated with trusting other people (see Levi, 1996). Yet, this measure becomes relevant at the societal and cross-national and not at the individual level of analysis. In other words, while institutional performance and not culture determines trust in institutions at the individual level, at the societal/national level social capital may play a crucial role as determinant of institutional performance that leads to trust in institutions at the individual level (see Newton and Norris, 2000). This has also been vindicated by research in post-communist countries (see Mishler and Rose, 2001). Perceptions of corruption, such as those collected by Transparency International, are widely used as a very important proxy of institutional performance and subsequently of confidence in institutions at large (della Porta, 2000). Additionally, Delhey and Newton (2005) have found strong evidence that good government, evaluated on the basis of several rule-of-law, efficiency, political stability and accumulated freedom indices, has a positive impact on the 4

5 level of generalized trust. Finally, Cook et.al. (2005) have conceptualized in some detail how collective action and cooperation could be achieved without trust, but rather enabled through institutional design. Moreover, Uslaner (2008) has attempted to link corruption, and particularly high level corruption, with inequality. Whilst the institutional literature views corruption of all sorts (i.e. high but also petit corruption) as a symptom of the malfunctioning of institutions that has a negative impact on the level of inequality, Uslaner points to the opposite direction: namely that high inequality leads to high corruption and low trust and then to more inequality. In other words, he identifies the relationship between inequality, corruption and trust as a case of what he calls inequality trap (2008:23-57). Obviously, the theoretical dichotomy analyzed above is particularly relevant to countries where the conditions of social capital/social trust and confidence in institutions are considered as generally poor, such as the post-communist countries of Central-Eastern Europe (CEECs) (see Howard, 2003; Kornai and Rose-Ackerman, 2004) and/or Cohesion countries of South-Eastern Europe (see Paraskevopoulos, 2005a,b, 2007; Paraskevopoulos and Leonardi, 2004), among which Greece can be viewed as one of the most characteristic cases. The interplay between institutional performance, social capital and public policy in Greece Capturing the interplay between institutional performance, level of social capital and public policy outcomes is a crucial research task, especially in the case of Greece, given that, taking into account the limitations with regard to data availability (lack of historical/time series data), there seems to be a close interconnectedness between the three variables (i.e. extremely low level of institutional performance, social capital and public policy outcomes) which coincides with/in a period of intense efforts towards modernization/europeanization of the country, which, obviously, poses a crucial challenge to modernization theory itself. Thus, while, as it is well-known, measuring social capital is not an easy research task, in the case of Greece the lack of historical database constitutes an additional impediment. Thus research for this paper has relied on two main sources of data that became available relatively recently: the Special Eurobarometer 62.2 (223) which has been carried out by the European 5

6 Commission (DG-V) and released in 2005 (see Paraskevopoulos, 2005b); and the European Social Survey (ESS) (1 st and 2 nd waves) which were released in 2003 and 2006 respectively. Although both surveys provide rich databases for measuring social capital and are very similar in their findings, there are important differences in their sample size and coverage of specific variables that allow for differentiation in their analytical capacity. In that respect, the data from Eurobarometer is used mainly for macro-analytical comparative purposes, that is for comparing Greece vis-à-vis the other EU member states on the basis of a social capital index and regressions in relation to specific measures of institutional performance and public policy outcomes. The ESS, on the other hand, involves generally larger samples and covers a wider variety of variables, including communication, control, trust in institutions and other crucial variables, and therefore it provides for capturing the micro-foundations of social capital and confidence in institutions at the domestic level. In that respect, it is used for micro-analytical purposes, namely for the analysis of specificities related to the composition of social capital in Greece and its interconnectedness with other determinant and/or outcome variables. The social capital index involves the capturing of the nexus between the main aspects of social capital, namely structure and culture, through measures of associational membership and social trust 1. 1 Measures of associational membership normally include membership of at least one voluntary organization (vol.any), a number of multiple organizational memberships (vol.org) and a combined score (vol.act) of active membership, passive membership and no membership at all in any category of organization (Norris, 2001). The measurement of social trust, on the other hand, is based on the proportion responding most people can be trusted in each society. The construction of the social capital index on the basis of Eurobarometer 223 data, however, has been based on a combination between social trust and number of multiple organizational memberships (vol.org). This is because it was very difficult from the data provided to distinguish between active and passive memberships and therefore to end up with a cohesive combined score (vol.act). Nonetheless, we are confident that the measure of organizational memberships adequately captures the dynamism of associational membership across the EU member states. 6

7 Table 1: Social Capital Index Social Trust Vol.Any Vol.Org Social Capital Index Most people Social Trust x Vol.Org Country N can be trusted Sweden (SE) Denmark (DK) Netherlands (NL) Finland (FI) Luxembourg (LU) United Kingdom (UK) Ireland (IE) Austria (AT) Belgium (BE) Germany (DE) Slovenia (SI) France (FR) Spain (ES) Estonia (EE) Malta (MT) Italy (IT) Cyprus (CY) Portugal (PT) Czech Republic (CZ) Hungary (HU) Slovakia (SK) Greece (EL) Latvia (LV) Lithuania (LT) Romania (RO) Bulgaria (BG) Poland (PL) EU EU NMS Total Source: EuroBarometer 62.2 (2004). Data weighted. Note: Explanation of variables. Social Trust: the proportion responding most people can be trusted. Vol. Any: the proportion of the adult population who say they belong to at least one category of voluntary organization. Vol.Org: the number of organizational sectors to which people belong. Thus, from macro-analytical perspective, namely in comparison with the other EU member states, Greece s position is depicted within a group of primarily Southern and CEE countries, demonstrating very low levels of social capital and capacities for collective action. Obviously, this is the outcome of Greece s weakness in both the social capital components, 7

8 namely social trust and associational membership, which, additionally, seems to be closely associated with weaknesses in other crucial from public policy s point of view determinant and/or outcome variables, such as social networks and access to and satisfaction from the quality of public services (see Paraskevopoulos, 2005b). The overall state of social capital in Greece and particularly that of associational membership, viewed from a macro-analytical perspective, raises serious doubts about the post-olympic games euphoria with regard to the development of the voluntary and NGOs sector, which was rooted in the increasing levels of volunteers participation in the games. This is particularly true, if one takes into account the additional factor of the low intensity of membership/participation (see Whiteley and Seyd, 2002), if any, in voluntary and civil society organizations, as becomes evident from the descriptive statistics (see Eurobarometer 62.2 data). Finally, this state of social capital points to the limitations of the role played by other, primarily socializing-related, variables, such as contacts with friends, colleagues etc., where Greece, along with other countries of Southern Europe, appears to be in a better position vis-à-vis the other EU member states (see Paraskevopoulos, 2005b). As for the specificities of social capital in Greece based on the ESS data (see Paraskevopoulos, 2007), the following points deserve reference. First, Greece demonstrates important similarities with other countries of Western and/or Southern Europe, with regard to the role of control variables, such as education and gender, as predictors of social trust, associational membership and social capital at large. The same applies to other determinant and/or outcome variables, such as trust in institutions (see Herreros, 2004). Second, TV viewing constitutes a very important variable/predictor of the level of social capital, since it affects -in varying ways- the level of associational membership and participation in the voluntary sector at large. Given the unregulated-problematic character of the TV industry and the comparatively high rates of TV viewing in Greece, there might be grounds for drawing comparisons between the Greek and the U.S. cases (see Putnam, 2000), with regard to the role of TV viewing in civic malaise and the erosion of social capital at large. Third, although the end of the 1980s is widely considered as a turning point, in terms of levels of political participation in the post-authoritarianism period (see Lyberaki and Paraskevopoulos, 2002; Sotiropoulos, 2004), and what one might expect would be a shift from increased levels of party membership and interest in party politics at large towards interest in and membership of civil society and voluntary sector organizations, there is no evidence to support such a hypothesis. Indeed, civil society appears to be particularly weak in Greece in terms of both the level of membership and intensity of participation (see ESS data). Finally, there 8

9 seems to be no evidence to support the hypothesis about some sort of informal strength of civil society in Greece (see Sotiropoulos, 2004) either, given that surveys are generally capable of capturing even informal patterns of participation. Yet, what would best describe the phenomena Sotiropoulos (2004) refers to, might be mainly ad hoc, spontaneous and reactionary attitudes, albeit without any organizational structure and hence without continuity, with the exception of some local and/or regional specificities, which undoubtedly exist (see Paraskevopoulos 1998, 2001a-d). Social capital and social networks Social networks are considered as a well-established predictor of social capital in previous research. Thus, as it might have been expected, it appears to be very highly (.683) and statistically significantly (at the 0.01 level) correlated with social capital. Undoubtedly, social networking constitutes a very important proxy/measure for identifying similarities and differences in social capital and capacity for collective action among EU member states. The measurement of the variable on the basis of the data from Eurobarometer 223 has relied on calculations of mean numbers of receiving and providing support to others in selected fields of life, such as personal care, borrow money, etc. What the correlation of figure 1 demonstrates is that the Scandinavian countries and the Netherlands -with very high levels of social networking and social capital- are depicted at the top-right corner of the graph, well above all the other EU countries. Countries of Western and/or continental Europe, i.e. France, UK, Germany, Austria, etc., because of their rather medium levels in social networking and social capital, are depicted around the central part of the graph. What is worth stressing, however, is the position of Southern and Eastern European counties. In particular, primarily countries of Southern (Italy, Greece, Portugal) and secondarily of Eastern Europe and the Balkans are depicted at the bottom-left corner of the graph, demonstrating very low levels in both social capital and social networking measurements. Obviously, this may be an indication of potential links with the existing variation in the types of welfare states and social policy styles at large. This, however, will be the subject of the next section. Figure 1: Social Capital and Social Networks 9

10 SE DK Li near Regression Social Capital Index 1.50 NL 1.00 FI IT MT CY HUEL PT AT ESEE RO BGLT PL IE DE LU BE UK FR CZ SKLV SI Social Ne tworks (providing and rece iv ing support) Corruption and social capital The level of corruption, measured as score of the Corruption Perceptions Index (CPI), constitutes a crucial indirect proxy for confidence in public institutions and hence a powerful predictor of social capital, widely used in social capital research. This is the main reason why we introduced the measure in our analysis: to contemplate for the lack of measures relevant to trust in institutions in Eurobarometer 223. Indeed, the level of corruption appears to be extremely highly (.739) and statistically significantly (at the 0.01 level) correlated with social capital/social trust. In that respect, the level of corruption is viewed as a very important proxy for identifying similarities and differences in social capital and capacity for collective action among EU countries, as well as for investigating the role of institutions in this process. The measurement of the variable has been based on the CPI 2004 scores. Figure 2: Corruption Perceptions and Social Trust 10

11 % Trusting DK Li near Regression 0.60 NL SE FI 0.40 EE ES IE BE UK AT DE LU 0.20 RO PL HU IT BG CZ EL LT LV SK CY SIPT MTFR Corruption Perception (Transparency International 2004) Note: The CPI scores range from 10 (corruption-free) to 0 (totally corrupt) The picture emerging from the correlation in Figure 2 is similar in many respects to the previous correlations: that is the Scandinavian countries and the Netherlands -with very low levels of corruption and very high levels of social capital- depicted at the top-right corner of the graph, well above all the other EU countries, followed by countries of Western and/or continental Europe, i.e. UK, Germany, Ireland, Austria, etc., with relatively low levels of corruption and relatively high levels of social capital. Greece, along with the countries of South-Eastern Europe and the Balkans (Romania, Bulgaria) are depicted at the bottom-left corner of the graph, demonstrating very low levels of social capital and very high levels of corruption. What is striking, however, is the position of Poland at the far bottom-left corner, demonstrating the highest level of corruption and the lowest level of social capital. Social capital and access to public services Citizens satisfaction with regard to access to public services would normally be considered as an outcome measure for the identification of social capital in the sense of a crucial public policy outcome depended on the level of pre-existing institutional resources, such as social 11

12 % Trusting capital endowments linked to high levels of institutional performance. According to the institutional theory of trust, however, it might also be viewed as a determinant proxy related for instance to the specifics of the type of welfare state (i.e. the universal model) and/or institutional and social policy styles that facilitate the creation of generalized trust among citizens and building social capital. In any case, access to public services appears to be very highly (.603) and statistically significantly (at the 0.01 level) correlated with social capital/trust. This finding is extremely crucial, especially from comparative public policy s point of view, because it vindicates the link between satisfactory public policy outcomes and social capital. In that sense, access to public services constitutes a very important proxy for identifying similarities and differences in social capital and capacity for collaborative collective action among EU countries. The measurement of the variable on the basis of data from Eurobarometer 62.2 has relied on the mean (per cent) of people responding satisfied to the question about satisfaction with regard to the access to public transport, educational system, health system, social housing, training and child care. Figure 3: Access to Public Services and Social Trust DK Li near Regression 0.60 SE NL FI 0.40 EE ES IE UK DE LU AT BE 0.20 BG EL RO LV PT HU IT FR LT PL MT CY SK SI CZ Access to Public Services (%) As in the other correlations, the Scandinavian countries and the Netherlands are well above all the other EU member states, demonstrating high levels of citizens satisfaction 12

13 regarding access to public services and social capital. Increased variation among the other continental and Western European countries (i.e. France, UK, Germany, Belgium), an outcome of the significant differentiation in citizens satisfaction from access to public services and/or social capital across the EU, is the other main feature of the correlation. The candidate countries of the Balkans (Bulgaria, Romania), as well as Baltic states (Lithuania, Latvia) and Greece, are depicted at the bottom-left corner of the graph. This is particularly crucial for the case of Greece, given that it seems to vindicate results of previous research (Paraskevopoulos, 1998, 2001, 2004), pointing to the link between the extremely low levels of social capital, on the one hand, and generally low levels of institutional performance in most of the public policy areas in this country on the other. Social capital and satisfaction from public services As with citizens satisfaction from access to public services, satisfaction from the quality of public services is normally considered as an outcome measure for the identification of social capital, that is a crucial public policy outcome depended on the level of pre-existing institutional resources which in turn are linked to the level of institutional performance. In the light of the fundamental assumptions of the institutional theory of trust, however, it might also be viewed as an indirect, determinant measure related to specific type(s) of welfare state (i.e. the universal model) and/or institutional and social policy styles that are viewed as favourable for the creation of generalized trust and the building of social capital. In that sense, in a similar vein to access to public services, satisfaction from the quality of public services appears to be similarly highly (.601) and statistically significantly (at the 0.01 level) correlated with social capital. This is another important finding from comparative public policy s point of view, because it vindicates the link between public policy outcomes and social capital. Thus satisfaction from the quality of public services constitutes a very important proxy for identifying similarities and differences in social capital and capacity for collaborative collective action among EU countries. The measurement of the variable on the basis of data from Eurobarometer 622 has relied on the mean (per cent) of people responding satisfied to the question about satisfaction from the quality of public transport, educational system, health system, social housing, training and child care. 13

14 % Trusting Figure 4: Social Trust and Satisfaction from Public Services DK Li near Regression 0.60 SE NL FI 0.40 EE ES UK DE IE LU AT BE 0.20 BG EL RO LV LT HU IT PL PT SI FR MT CY SK CZ Satisfaction from Public Services (%) The correlation between satisfaction from public services and social capital in figure 6 is very similar to that of access to public services. The Scandinavian countries and the Netherlands are well above all the other EU member states, demonstrating high levels of citizens satisfaction from the quality of public services and social capital. Additionally, the variation among the other continental and Western European countries (i.e. France, UK, Germany, Belgium) is also similar, reflecting the significant differentiation in citizens satisfaction from the quality of public services and/or social capital across the EU. The Balkan (Bulgaria and Romania) and Baltic (Lithuania, Latvia) countries, along with Greece, are similarly depicted at the bottom-left corner of the graph, demonstrating very low levels of satisfaction from the quality of public services and social capital. In the case of Greece, this finding vindicates results of previous research, pointing to the link between the extremely low levels of social capital, on the one hand, and generally low levels of institutional performance in most of the public policy areas in this country, on the other. Overall, as figure 5 below with graphical representation of differences in social capital from a comparative (pan-european) perspective demonstrates, the level and type of social capital in Greece is very much similar to countries of Central-Eastern Europe, notably Poland and Hungary. This becomes evident from the fact that it is very highly correlated with 14

15 extremely low level of institutional performance (high level of corruption) on the one hand and public policy outcomes in very crucial policy areas on the other. Figure 5. Similarities and Differences in Social Capital across European Societies. A Graphical Representation of Multivariate Data in the Form of STARS (R). Source: the European Social Survey Round I (adapted from Wallace Claire et. al., CEC Report, 2005) 15

16 Accounting for confidence in public institutions in Greece: disentangling the paradox of normality Accounting for confidence in institutions in Greece is proved a rather complicated/difficult exercise. This is because, according to major surveys, Greece demonstrates rather normal, namely similar, characteristics to other EU countries in relation to confidence in institutions. In particular, the indicative table 2 below based on ESS data shows, Greece, along with other similar countries, demonstrates rather medium or above medium level of confidence in all the major political (national parliament), order (legal system, police) and international (EP, UN) institutions, and rather low level of trust in politicians, again along most of the other European countries. Additionally, along with other comparable countries (Ireland, Italy, Hungary etc.), Greece is characterized by well above average trust in the European Parliament, especially if this is compared with trust in domestic institutions. Additionally, there seem to be rather strong correlations at the aggregate level between confidence in key public institutions and social trust (figs. 6 and 7). Yet, at the individual level of analysis (tables 3, 4, 5) only confidence in the legal system seems to be significantly correlated with social trust. In particular, as the stepwise linear regression with dependent variable confidence in the legal system (table 3) shows, the latter is significantly correlated to social trust, trust in police, national and European Parliament, and education. Additionally, as the model of table 4 where the dependent variable is confidence in national parliament demonstrates, the most important variables/predictors of trust in national parliament are trust in the police, the legal system, the political parties, the EP and financial companies. However, it is worth stressing that political interest and church attendance are negative predictors, namely the less you attend church and the less you are interested in politics, the more you express confidence in the national parliament. Finally, as the model of table 5 where the dependent variable is confidence in political parties demonstrates, the most important variables/predictors of trust in political parties are again trust in the police, the national and European Parliament and financial companies. Yet, it has to be stressed that, strikingly enough, political interest and gender are negative predictors, namely women who are less interested in politics is more likely to express confidence in political parties. 16

17 Table 2: Trust in Institutions of Political Interest. Means at the National Level. Institutions Country National Parliament The Legal System The Police Politicians The European Parliament The United Nations Difference between European and national parliament Austria 5,1 6,1 6,4 3,5 4,2 4,5-0,9 Belgium 5,0 4,4 5,6 4,3 4,9 5,1-0,1 Czech Republic 3,6 3,8 5,0 3,2 4,7 5,3 1,0 Denmark 6,2 7,1 7,9 5,5 4,8 6,5-1,4 EU15 4,7 5,3 6,3 3,7 4,6 5,2-0,1 Finland 5,8 6,8 8,0 4,8 4,9 6,5-0,9 France 4,5 4,8 5,9 3,6 4,4 4,6-0,1 Germany 4,5 5,7 6,7 3,5 4,5 5,2 0,1 Greece 4,8 6,3 6,4 3,5 5,8 4,6 0,9 Hungary 5,0 5,1 4,9 3,9 5,7 6,0 0,7 Ireland 4,4 5,1 6,5 3,8 5,1 5,7 0,7 Israel 4,7 6,6 6,3 3,3 4,1 4,0-0,6 Italy 4,8 5,5 6,7 3,5 5,5 5,6 0,7 Luxembourg 5,7 6,2 6,4 4,8 5,0 5,1-0,7 Netherlands 5,2 5,4 5,8 4,9 4,7 5,4-0,5 Norway 5,7 6,3 7,0 4,6 4,7 6,8-1,0 Poland 3,5 3,7 5,0 2,7 4,8 5,6 1,3 Portugal 4,4 4,3 5,1 2,8 4,9 5,4 0,4 Slovenia 4,0 4,3 4,9 3,1 4,7 4,9 0,6 Spain 4,8 4,3 5,4 3,4 4,8 4,7 0,0 Sweden 5,9 6,1 6,8 4,7 4,0 6,6-1,9 Switzerland 5,8 6,2 6,8 4,9 4,8 5,4-1,0 United Kingdom 4,7 5,0 6,0 3,8 3,6 5,3-1,0 Total 4,6 5,1 6,1 3,6 4,6 5,3 0,0 Source: the European Social Survey Round I, Weighted data (dweight). 17

18 Social Trust Figure 6: Confidence in the Legal System 7,00 DK NOR FI SE 6,00 IE NL CH UK LU 5,00 ES BE EU15 DE AT 4,00 CZ PT SI FR HU IT PL EL 3,00 R Sq Linear = 0,401 3,00 4,00 5,00 Confidence in Legal System 6,00 7,00 Source: ESS 1 st wave ( ) 18

19 Social Trust Figure 7: Confidence in National Parliament 7,00 DK NOR FI SE 6,00 IE NL CH UK AT LU 5,00 4,00 CZ SI PT DE FR ES EU15 IT BE HU PL EL R Sq Linear = 0,641 3,00 3,50 4,00 4,50 5,00 5,50 Confidence in National Parliament 6,00 6,50 Source: ESS 1 st wave ( ) 19

20 Table 3. Stepwise linear regression, dependent variable: Confidence in legal system Variables Coefficients Social trust.077*** Political interest.027 Trust in police.441*** Trust in national parliament.210*** Trust in European parliament.180*** Trust in political parties.023 Left-right scale.006 Church attendance Trust in financial companies Gender.022 Education.049** Occupation status ***Significant at 99 percent ** Significant at 95 percent * Significant at 90 percent Source: ESS 2 nd wave ( ) Table 4. Stepwise linear regression, dependent variable: Confidence in national parliament Variables Coefficients Social trust.028 Political interest -.061** Trust in police.082*** Trust in legal system.193*** Trust in European parliament.167*** Trust in political parties.402*** Left-right scale.021 Church attendance -.044* Trust in financial companies.065*** Gender Education Occupation status.023 ***Significant at 99 percent ** Significant at 95 percent * Significant at 90 percent Source: ESS 2 nd wave ( ) 20

21 Table 5. Stepwise linear regression, dependent variable: Confidence in political parties Variables Coefficients Social trust.023 Political interest -.120*** Trust in police.082*** Trust in legal system.033 Trust in national parliament.430*** Trust in European parliament.259*** Left-right scale.005 Church attendance Trust in financial companies.062*** Gender -.040* Education Occupation status.003 ***Significant at 99 percent ** Significant at 95 percent * Significant at 90 percent Source: ESS 2 nd wave ( ) Overall, it needs to be stressed that there may be a paradox of normality in relation to the regressions at the individual level. This may be linked to the difficulty on the part of major surveys for capturing the obscured or overshadowed micro-foundations of low performance/low confidence in aspects/units of specific institutions and/or policy areas, such as specific units within the civil service (i.e. hospitals, local level-municipalities, urban planning authorities etc.), which are widely perceived as main nests of corruption. In sum, seems to vindicate the institutional rather than the socio-cultural approach to confidence in public institutions, in the sense that what seems to be taking place is actually a vicious circle between low institutional performance (high level of corruption) which leads to/feeds low level of social capital which in turn leads to poor public policy outcomes reinforcing/contributing again to low institutional performance. 21

22 From Social Networks to Policy Networks: implications for EU public policy in Greece Confidence in public institutions and social capital are considered key concepts in the academic debate on the impact of Europeanization and/or modernization processes on domestic policy and institutional change. This is primarily because of their role in facilitating the learning process among actors within policy networks/policy communities, which is viewed as a fundamental precondition for domestic policy change (see Paraskevopoulos, 1998, 2001a-d, 2004, 2005a, 2006). In Greece, the problematic state, namely the lack, of social capital has been identified as a key factor/variable linked to the blocking of the reform and policy-making processes at large in several policy areas from the economy (Lyberaki and Tsakalotos, 2002) to the environmental and regional policies (Paraskevopoulos, 1998, 2001a-d, 2004; Paraskevopoulos et.al., 2006). This is attributed to the fact that the family and kinship-based, essentially bonding (that is unsocial ), type of social capital, similar in many respects to the amoral familism identified by Banfield (1958) in Southern Italy, that had underpinned the spectacular economic growth in the 1960s and 1970s with the support of the patronizing state (see Lyberaki and Tsakalotos, 2002), has gradually been replaced by another sort of bonding -that is unsocial- capital which is based on relatively small and strongly-tied interest groups. Given that the creation of these groups has been based on a variety of selective incentives and that rent-seeking behaviour is their dominant feature, they impose important limitations to the exit and voice options of actors, thus inhibiting learning and policy change in several policy areas. Indeed, there is evidence to suggest that the state of social capital in Greece constitutes a key variable affecting the reform process in public policy in two important respects. On the one hand, the low level of social trust and trust in institutions is linked to the absence of weak ties among the actors within social and/or policy networks that would facilitate the crucial for the learning process- diffusion of information and knowledge (see Granovetter, 1973). Hence social and policy networks are characterized by the predominance of small, strongly-tied and rent-seeking interest groups which block the reform process. On the other hand, simultaneously, the low level of social capital/social trust crucially affects the level of other actors participation in policy networks, and hence inhibits the creation of advocacy coalitions and/or issue-specific policy networks, such as epistemic communities and independent think-tanks, that would facilitate the policy learning and reform processes in public policy at large (see Sabatier, 1999). In this way, the lack of social 22

23 capital may actually play a key role in undermining crucial reforms in public policy-making. Thus, while the particular type of bonding -unsocial- capital is considered as having facilitated a spectacular and of particular type economic growth during the 1960s and 1970s based primarily on family enterprises and a patronizing state, it seems that it constitutes an impediment to the policy reform and adaptation processes in public policy in contemporary circumstances, characterized by high transaction costs. In that respect, given that the EU level of governance provides as the only option for actors exit and voice functions, this section presents evidence on the state of the reform process essentially the response to pressures of Europeanization- in two crucial policy areas for Greece, namely environmental and regional. The empirical evidence derives from a recently completed comparative research study between Cohesion and CEE countries whose main focus was on policy structures, namely policy networks, as well as on outcomes of policy implementation (Paraskevopoulos, et.al. 2006). Environmental Policy Environmental policy in general and waste management in particular is a policy area which perfectly reflects the weaknesses of the institutional and policy-making structures related to the low level of social capital. In particular, rent-seeking and anomic behaviour, both consequences of lack of social capital and cooperative culture, facilitated partly by institutional deficiencies, such as the lack of a land registry system (see Paraskevopoulos, 2001a), has led to crucial environmental problems: namely, the degradation of the built environment in almost all urban areas; the increased levels of air pollution, mainly because of the huge expansion in the use of private cars; and land and water pollution, because of failures in the waste management policy (Paraskevopoulos et.al., 2006). Additionally, the absence of independent think-tanks and subsequently of advocacy coalitions and/or issue networks from the policy-making process, accompanied by the rather weak involvement of civil society actors, has led to the predominant role of central state actors in the policy process. Thus, the main institution responsible for the formulation and implementation of environmental policy at the national level is the Ministry for the Environment, Spatial Planning and Public Works. However, given that policy-making is shared with other - sectoral- Ministries, such as the Ministry of Agriculture, the Ministry of Culture and the Ministry of Commercial Marine, and there is lack of effective co-ordination mechanisms, the 23

24 Greek environment ministry has been characterized as a "weak" one, if compared to ministries of other EU member states, such as the Danish and British ones (see Paraskevopoulos et.al., 2006). Hence, the command-and-control type of regulation, which involves limited discretion and flexibility for the administrative actors, is considered as the main characteristic of environmental policy in Greece. In that respect, although even the formulation of environmental policy as a coherent and distinct policy area in the second half of the 1980s was strongly influenced by the need for harmonisation of national legislation with EU rules, Greece is considered as belonging to the so called latecomers or laggards group of EU member states. Yet, the gradual identification of the fundamental weaknesses of the command-and-control, namely the statist, system of environmental policy-making has led recently to some steps towards increasing flexibility and efficiency in monitoring and inspection procedures, such as the delegation of certain inspection, observation and accreditation functions to some sort independent bodies (experts) or voluntary organizations- NGOs (see Paraskevopoulos et.al., 2006). In the area of waste management, in particular, the lack of an integrated management/co-ordination strategy that would link the national, regional and local levels of government on the one hand, and the intense conflict among social and institutional actors at the regional and local levels about the location of the disposal or recycling areas on the other, the well-known NIMBY syndrome, create conditions of a mis-regulated policy-making environment, especially at the local level. Additionally, the institutional infrastructure at the national (central state) level includes the EU directives which are enforced by insufficient trans-ministerial decisions and the law on the protection of the environment, which has a rather declarative character, resulting in a lack of effectiveness of the policy-making process. Thus, although Greece appears to have a rather good record of adopting EU legislation (in fact all the relevant Council Directives -75/442, 91/156, 94/62- have been transposed), there are serious delays in the process (4-6 years), accompanied by lack of thorough examination of the conditions and needs at the national level (e.g. through research and production of reports). It is indicative that there was a significant delay (7 years) in the incorporation of the Packaging Waste Directive (94/62) into the Greek legislation and there are still several steps that have to be taken for its enforcement (e.g. setting up of an organisation for alternative waste management schemes). With regard to policy-making structures (policy network) as the case of the region of Attica, which was selected for fieldwork -Social Network Analysis (SNA)- research, reveals, 24

25 the policy process is dominated by two main groups of actors. The first group comprises the most central actors, which are primarily central state (i.e., Ministry for the Environment and Public Works-YPEHODE) and subnational actors, such as local government associations (i.e., ESDKNA TEDKNA), as well as, organizations of interest intermediation (i.e., Technical Chamber of Greece-TEE). The second -peripheral- group comprises primarily policy and/or issue-specific professional associations, civil society organizations and some small private research companies, such as the Hellenic Recovery and Recycling Association- HERRA, the Ecological Recycling Association-ERA and the Hellenic Solid Waste Management Association-HSWMA). Overall, the structure of the policy network in waste management arguably reflects the pathologies of the pre-existing institutional structure in Greece mentioned above, namely statism, centralized and weak administrative structure, problematic state-society relations, and weakness of civil society, accompanied by some reluctant indications/early signs of institutional innovation such as, some involvement of civil society, experts/professionals in the policy-making process (Paraskevopoulos, et.al., 2006). As far as the implementation of environment/waste management-related EU programmes and initiatives is concerned, the evidence from the implementation of the -CSF II- ROP in Attica is indicative. In particular, the Sub-programme 1, corresponding to the strategic objective of improving the environment and quality of life consistently demonstrated the lowest level of both absorption rate and level of physical completion among all the sub-programmes/strategic objectives of the ROP, while it should be stressed that the general picture emerging from the evaluation regarding the implementation of other subprogrammes of the ROP was rather positive. Additionally, and most importantly, the evaluation provides evidence about the significant underperformance (low level of accomplishment) of the specific measure (M.1.2) related to the waste management tasks of the ROP, which is attributed to the so called social reactions factor, namely to reactionary attitudes towards the location of landfills (the NIMBY syndrome in the field) (see Paraskevopoulos et.al., 2006), actually another typical symptom of the predominance of bonding (namely anti-social) capital in Greece. As for the actual accomplishment of policy objectives, the main waste management-related problem of the Attica region is supposed to be two-fold: on the one hand, while the bulk (95%) of the estimated 3,500 tonnes of solid waste is still disposed of to the sanitary landfill located in the Municipality of Ano Liosia, this landfill has already been saturated; and on the other hand, the remainder is disposed at several small and uncontrolled and illegal landfills and dumpsites. As far as recycling is 25

26 concerned, only an estimated 8% of the total municipal solid waste was directed into material recycling with a goal of increasing the amount of recycled waste to 25% by the end of 2006, which however, has yet to be achieved (see Paraskevopoulos et.al., 2006). Regional Policy Although regional policy is a policy area mostly affected by the Europeanization process almost since Greece s accession into the EC/EU, the weaknesses of the domestic institutional infrastructure in general and the low level of social capital in particular are evident in both the level of non-state actors (i.e., civil society, expertise etc.) involvement in policy-making (policy network) and in the actual outcomes/accomplishments of policy implementation. Thus, even though the interaction between EU pressures and domestic institutional structures has led to a series of important reforms of the domestic policy-making structure over the last twenty five years or so, involving the strengthening of administrative capacity and increased participation of non-state actors, such as private sector and expertise, in policy-making, this has taken place primarily at the central state level, while the overall response of the domestic institutional structure to the challenges of Europeanization has been rather poor (see Paraskevopoulos, 1998, 2001a-d, 2005a; Paraskevopoulos et.al., 2006). With regard to policy structures (policy network) the Southern Aegean Islands region, which was selected for fieldwork -Social Network Analysis- research, may be seen as reflecting the reality of policy-making structure, given that it used to be considered as a rather font-runner region, in terms of institutional capacity, and has been studied on a longitudinal basis, given that the first wave of SNA research was conducted in the region to evaluate the implementation of the Integrated Mediterranean Programmes (IMPs) and the first Community Support Framework-CSF (see Paraskevopoulos, 1998, 2001). Yet, the structure of the policy network demonstrates rather limited improvement, as far as institution-building and actors participation is concerned (see Paraskevopoulos, et. al., 2006). In particular, the policy network is characterized by the predominant role of state actors, that is central state (Ministry of Economy) or regional (Regional Secretariat and ROP-Managing Authority), as well as by a rather upgraded role of the sub-regional development agencies, the chambers of commerce and industry and the local associations of municipalities and communes, mainly due to their ex-officio participation in the ROP-Monitoring Committee. Nonetheless, the participation of civil society actors and NGOs is minimal (see Paraskevopoulos, et. al., 2006). Overall, the main lessons to be drawn with regard to the policy structures policy network- in 26