Social trust and responses to political
and economic transformation in East-Central Europe*
Natalia Letki and Geoffrey Evans
Nuffield College
Oxford OX1 1NF
Nuffield College Politics Working Paper 2002-W9
*Earlier version of this manuscript was presented at the symposium
Trust As A Pre-Condition To Communication, Social Thinking And Social Practices During Democratic Transition In Post-Communist Europe,
British Academy, London 13-15 September 2001
Abstract
Social trust forms a major component of current conceptions of social capital and as such has been attributed a significant role in providing the social context for the emergence and maintenance of stable, liberal democratic polities and effective economies. Its role in these processes has in turn been generalized to post-communist societies in East Central Europe undergoing ‘dual transitions’ from authoritarian states with command economies to democratic free market societies. In this paper, however, we show that the relations between trust and perceptions of democratic functioning in Eastern Europe imply a rather different ‘top-down’ process, in which levels of trust reflect rather than influence the effectiveness of political and economic institutions. This calls into question the generalization of models developed in democratic societies to the post-communist context and provides the basis for an alternative understanding of the process of social capital formation.
Introduction: Social trust in East-Central Europe
In recent years the post-communist transformation has become a topic of major significance in political science and sociology. A substantial part of the debate on the transformation has been devoted to considerations of ‘civil society’. Civil society, defined generally as a sphere of social organisation occupying the space between the state and the market, has been acclaimed not only for its role in abolishing the communist systems, but also because “the concept of civil society may … be regarded as containing elements necessary, if not sufficient, for the development of the liberal-democratic basis of modern -- or ‘post-modern’ -- socio-economic life” (Miller 1992). A central focus of scholarly interest in developments in the region has thus concerned the benefits of active civic association and the development of non-governmental organisations for the emergence of liberal democracy. In this respect, academic and journalistic attention has been captured by Robert Putnam’s general theory of the link between civil society and workings of democracy. In his Making Democracy Work (1993) Putnam stressed the importance of ‘concerted action’ for the condition of democracy, but he focused not on the explicit principles of participatory democracy, but on their ‘by-products’, i.e. norms of generalised reciprocity and trust. In his most recent work Putnam insists that social trust is learned from participation in various types of associations. Social trust and associational membership form ‘social capital’, which is fundamental to the development of political institutions as well as economic activity (Putnam 1995a, b, 2000).
The novelty of the social capital approach in the version advocated by Putnam, Inglehart and others, results from supplementing networks and organisational activities (i.e. voluntary associations) with a system of norms (i.e. trust and reciprocity). Thus, the discussion about social trust in ECE forms a natural continuation of the debate about civil society in post-communist countries, but with more stress placed on a “moral resource” (Putnam 1993: 169) rather than on any type of activity. And although both networks and trust are the components of social capital, some authors emphasise the superiority of social trust over participatory behaviour: “civic networks may enhance social life, but this “social connectedness” … is distinct from – and secondary to – moral values.” (Uslaner 1999: 122).
The concept of social capital has gained such popularity not only because it refers to ever popular and reputable notions of civic virtues, tolerance and pluralism, but foremost because it offers an explanation for the flourishing of political and economic systems. As Jackman and Miller point out, the recent discussion of social capital treats social trust, norms and reciprocity as “exogenous factors in generating economic and governmental performance” (1998: 50). And thus, enthusiasts of cultural explanations put forward a list of reasons why institutional effectiveness hinges on norms and networks. Some of these refer to rational choice arguments of the positive effects of social trust, such as the reduction of transaction costs and increased predictability of cooperation, resulting from the belief that cooperation is more advantageous than individualism (Putnam 1993, Uslaner 1999). Thus Boix and Posner (1998) point to five dimensions of the positive influence of social capital on government: the relation between the articulation of citizens’ interests and expectations and the political elite’s responsiveness (verified later as a result of elections); the reduction of costs of policy and rule implementation; the transformation of citizens’ preferences from particularistic into collectivist; the increased effectiveness of bureaucracy (trust and reciprocity within agencies); and, leaders’ ability “to make necessary compromises without losing the support of their group members” (Boix and Posner 1998: 686-93). Links between cultures of trust and tolerance and the quality, durability and stability of democratic systems have been examined empirically and the results seem to confirm that social capital is of crucial importance for the development of democratic systems (Inglehart 1997, 1999).
Social trust and norms of reciprocity have also been considered to be essential for the development of civic engagement. Networks of civic participation in voluntary associations are, next to social trust, the other facet “of the same underlying factor – social capital” (Putnam 1995a: 73). A positive correlation between these two elements seems plausible: “generalised trust makes people more willing to take part in their communities and to endorse moral commitments” (Uslaner 1999: 123). Initially, this relationship was hypothesised to be mutual, but more recent research has stressed that “the connection is stronger from participation to interpersonal trust, rather than the reverse” (Brehm and Rahn 1997: 999, see also Newton 1999b, Putnam 1995b: 666). Moreover, social trust is also considered to be responsible for generating various forms of participation in politics, such as discussing politics, voting, supporting a political party or a candidate etc. “In short, people who trust others are all-round good citizens” (Putnam 2000: 137).
Following the initial recognition of social trust as a prerequisite for the efficiency of political institutions and liveliness of citizens’ political engagement, it has also been proclaimed as a “missing link” in economic development (Fine and Green 2000). The importance of trust for the emergence of “ethical habits and reciprocal moral obligations” (Fukuyama 1995: 9), decisive for functioning of industrial structure, has resulted in the identification of social capital as a - hypothesised, but not yet fully evaluated - positive influence on economic productivity. Thus, researchers focusing on the link between social capital and democracy supplemented their analytical framework with economic indicators. Their conclusions seem unanimous: “where trust and social networks flourish, individuals, firms, neighbours, and even nations prosper” (Putnam 2000: 319, see also Inglehart 1990, 1999).
Such benevolent effects of social capital imply that its level is likely to be particularly important for the future of countries undergoing the ‘dual transition’ from non-democratic, non-market regimes to market democracy in East-Central Europe. Levels of social capital may be expected to be crucial for the success of consolidation in democratic and market institutions and civic attitudes in the post-communist countries (c.f. Boix and Posner 1998, Inglehart 1997, 1999, Newton 1999b, Putnam 1993, 1995a, 1995b, 2000). Moreover, theorists of social capital, apart from putting forward claims about the link between trust and networks, and the development of liberal democracy, join the laments about the weakness of social networks and potentially disadvantageous levels of distrust in post-communist societies. The theory of social capital assumes that trust and social interconnectedness are the components of a democratic political culture, thus low levels of social trust seem to be an inevitable heritage of the 50 years Communist rule[1] (Inglehart 1999, Rose 1994). In this paper, we examine these assumptions in the light of extensive cross-national survey-based evidence that address questions of the significance of levels of interpersonal trust in the region more thoroughly than has hitherto been the case.
Data and methods of inquiry
The data used in this paper come from a survey conducted in 11 ECE countries in the midst of transformation, 1993-1994. They give a unique opportunity to test the role social capital plays in generating the emergence of political activism and positive responses to political and economic transformation at both the individual and aggregate levels. Assessment of both individual and country-level evidence is especially valuable in this context, as the aggregate-level analysis favoured by many authors offers apparently impressive evidence for the powerful effects of social capital, but is unable to examine individual-level mechanisms that might account for such ‘effects’.
Our analysis uses survey-based indices, as we are more interested in the relation between social trust and other attitudes, such as perceptions of politics and economic developments, than in the hypothesised influence of trust on official estimates of economic growth, voting turnout etc. The main focus of our analysis is on respondents’ evaluation of political and economic developments rather than on normative attitudes towards democracy and market economy. First, we introduce the measures of the main concepts examined in the analysis, while in the following section we examine the associations between these measures implied by social capital theory. This involves investigating the relationship between the stocks of social trust and various aspects of political and economic transformation in East-Central Europe at the aggregate level. Then, we treat trust as an individual-level resource and use it as an independent variable in several regression models explaining responses to political and economic transformation. Later, we merge these two pictures using multi-level models that allow us to account for the contextual effects while analysing individual-level data. As our findings diverge from expectations derived from social capital theory, in the final section of the paper we offer an alternative interpretation of the relationship between social trust and the quality of institutional arrangements.
Indicators
The first indicator refers to the concept of social trust. Social trust is one of the components of social capital and is most often defined as norms of generalised reciprocity and trust, learned from participation in networks of civic engagement (Putnam 1993, 2000, Uslaner 1999, Inglehart 1990). In survey-based research social trust is usually measured by one simple agree/disagree survey question about “trust in others” (Putnam 1993, 1995b, Inglehart 1997, 1999, Dekker et al. 1997, Torcal and Montero 1997, Stolle and Rochon 1997, Rose 1998, 1999, Rose et al. 1997, Mishler and Rose 1998). In a few cases this measure has been further developed to include faith in other people’s honesty and helpfulness (Brehm and Rahn 1997, Paxton 1999, Putnam 2000), although still measured in terms of ‘agree/disagree’ items. However, one question does not seem to be a sufficient measure of a multifaceted attitude, such as “a set of institutionalised expectations that other social actors will reciprocate co-operative overtures” (Boix and Posner 1998: 686). Single item measures have been criticised for their limitations in capturing multifaceted topics, such as values or beliefs (e.g. Evans et al. 1996, Heath et al. 1994). Therefore, to maximise the reliability of our measure of social trust and to reduce the influence of random errors we selected five items that express individuals’ beliefs about trust and norms of reciprocity[2]. Using a wider range of indicators is also necessary to more adequately capture developments in the social capital theory: the initial stress on trust in others has been moved to norms of reciprocity, as they imply interactions which are crucial for the quality of social life (Putnam 2000). Therefore, we used the following five items to construct the scale:
a) It is human nature to cooperate with other people.
b) Most people can be trusted.
c) If someone is in serious trouble, no one else cares about it.
d) If you are not always on your guard other people will take advantage of you.
e) A person cooperates with other people only when he or she sees it in his or her own interest.
All of these items have an agree-disagree format. Respondents almost unanimously agree or strongly agree with the statement about human nature being cooperative. Moreover, although only around 50% of respondents declare that they trust other people, this figure closely resembles the reported results for the most ‘civic’ nations – Great Britain and the United States (Hall 1999, Putnam 1995a, 1995b, Inglehart 1999, Newton 1999a). In the case of the remaining three items, however, ‘negative’ (i.e. ‘strongly agree’ and ‘agree’) answers outweigh the ones expressing trust and faith in cooperation. Thus, although 42% of respondents would expect help and support from other people, almost 80% are afraid that others may take advantage of them, and over 70% do not believe in unselfish cooperation. Whether or not these results can be considered ‘negative’ or ‘positive’ in tone is difficult to assess, but their most important feature is that they provide us with a range of responses across items that differ in terms of their general levels of agreement and disagreement and thus allow the construction of normally distributed multiple-indicator measure of the concept of social trust.
The second indicator we use is an index of political satisfaction. Although political and economic performance can be measured by means of ‘objective’ indicators, such as Freedom House rankings and per capita GNP, we use measures taken from our surveys that more directly index citizens’ experiences of transition[3]. Seven items have been selected to measure respondents’ opinions about these aspects of the workings of the political system in their country.[4] All of them are 5-point agree-disagree scales (see Appendix B for details). The seven items refer to the implementation of democracy, general appraisals of the government’s actions and respondents’ sense of political efficacy. Only slightly above 20% of respondents evaluate the way democracy is being implemented in their countries positively, while 46% hold a negative opinion. Two of three items referring to the government create a more optimistic picture: 34% think that the majority of the society benefits from the government’s actions and 46% believe in the government’s effectiveness. However, as much as 56% of respondents disagree with the statement that the government’s actions reflect ‘the popular will’. Moreover, the three items referring to respondents’ political efficacy indicate their sense of ‘political powerlessness’. As many as 65% of the respondents do not believe in the most basic feature of democratic system - that citizens can influence the election of the government - while 72% think they cannot influence government’s actions and 73% agree that officials are not concerned about ‘ordinary’ people’s opinion. In general, these figures indicate that citizens in the new ECE democracies had a negative opinion of their political systems in the mid-1990s. This applies not only to feelings of efficacy, but also to the democratic functioning of their political systems.
As mentioned above, however, there is one more element involved in the ‘social capital puzzle’: economic performance. To test the predictive power of the social capital concept thoroughly we also examine the relationship between social trust and a measure of satisfaction with economic performance. Again, five 5-point agree-disagree items are chosen as indicators of satisfaction/disaffection with economic transformation in East-Central European countries: they refer to the evaluation of market economy as well as observed and anticipated change in the economic circumstances on individual and national level. Less then 30% of respondents replied ‘positively’, i.e. expressing satisfaction or positive opinion about any of the five aspects of the economic situation. In the case of questions about experience of micro- and macro-level changes of economic circumstances, 66% and 76%, respectively, reported experiencing a fall of living standards. At the same time, 45% respondents were not expecting any future change in their economic situation and 43% did not predict such change at the country level. Overall, evaluations of economic situation and living standards in ECE in the mid-1990s were predictably rather low.
Table 1 below presents the average scores of particular countries. Countries are ranked on the basis of their score on the social trust index: this allows us to observe whether the pattern of distribution of social trust among countries is likely to be explained by any of the characteristics invoked by social capital theorists (such as the stage of democratisation, economic advancement, or cultural characteristics).
Table 1. about here.
As the table shows, in terms of stocks of social trust, three clusters of countries are visible; Hungary, Czech Republic and Slovak Republic have average social trust scores of 2.76 - 2.78, while Bulgaria, Lithuania and Poland have scores lying between 2.89 and 2.91, and a third looser cluster of Ukraine, Russia, Belarus and Estonia scores between 3.07 and 3.18. While the first cluster seems intuitively predictable based on the countries’ central European geographical location and economic characteristics, the second one is much more diverse in character. Even more counterintuitive is the fact that Ukraine, Russia and Belarus have social trust scores higher than those of any of the countries that are further advanced in the transition towards market democracy. It is also interesting that Romania, despite its high levels of support for democracy and the free market at that time (c.f. Evans and Whitefield 1995), has clearly the lowest score on the index of social trust. In general, with the exception of Romania and Estonia, countries with superior institutional performance seem to have systematically lower social trust scores[5].
Table 1 shows also that the governmental evaluation index scores are low (i.e. significantly below 3) across all countries. Country scores do not vary greatly, but there are substantial and statistically significant differences between the highest scorers (the Czech Republic and Romania) and the lowest (Ukraine and Russia). Levels of economic evaluation are also low: no country’s average score reaches the mid-point of the scale, i.e. 3. The general pattern of economic satisfaction scores is similar to what would be expected on the basis of more ‘objective’ measures of economic development, e.g. more economically ‘advanced’ countries tend to score higher than less advanced countries.