Original scientific paper	UDK 316.472.4:338.124.4(497.4)
Hajdeja Iglič
THE CRUMBLING OR STRENGTHENING OF SOCIAL CAPITAL? THE ECONOMIC CRISIS' IMPACT ON SOCIAL NETWORKS AND INTERPERSONAL TRUST IN SLOVENIA
ABSTRACT: The economic crisis is bringing uncertainty, change, worries and distress into the everyday lives ofpeople, which has an effect on the level and form of social capital. The literature offers two opposing arguments regarding the impact of shocks and material uncertain^-y on social capital. The first one emphasises the instrumental nature of interpersonal relations and claims that the increased need for mutual help and support that arises during crisis periods strengthens social networks. In contrast, the second argument brings forward the problem of a social networks overload, which threatens to break them up during crises. More importantly, interpersonal dependence can lead to the closing off of social responsibility and solidarity into narrow family networks, thus reducing social capital among social groups and in society in general. In this paper, we empirically confront these two arguments in the context of the economic crisis of2008 in Slovenia. We compare the characteristics of social networks and interpersonal trust before the crisis with those measured several years after it began.
KEYWORDS: social capital, social networks, trust, economic crisis, welfare state
Krhanje ali rast socialnega kapitala? Zaupanje in socialna omrežja v Sloveniji v času ekonomske krize
IZVLEČEK: Ekonomska kriza vnaša v vsakdanje življenje ljudi spremembe, negotovost in skrbi, kar vpliva na njihov socialni kapital. V literature zasledimo dva nasprotujoča si argumenta o vplivu kriznih dogodkov in materialne negotovosti na socialni kapital. Prvi poudarja instrumentalno naravo medosebnih odnosov in pomen osebne odvisnost za razvoj družbene odgovornosti. Trdi, da povečana potreba po socialni opori, ki se pojavi v času krize, krepi socialna omrežja. V skladu z drugim argumentom pa medosebna odvisnost vodi v zapiranje družbene odgovornosti in solidarnosti v ozka družinska in sorodstvena omrežja, s čimer se krha socialni kapital med družbenimi skupinami in v družbi na splo-
Hajdeja Iglič šno. Še (
razpada. Veljavnost obeh argumentov preverjamo s kvantitativno analizo podatkov o stanju socialnega kapitala v Sloveniji v času pred krizo in nekaj let po začetku krize.
KLJUČNE BESEDE: socialni kapital, družbena omrežja, zaupanje, ekonomska kriza, država blaginje
1. Introduction
The economic crisis and the anti-crisis policies in Europe, with their emphasis on austerity and reduced public spending, have sharply increased uncertainty, distress and change in the daily lives of people. Individuals and families have suddenly been faced with lower incomes, rising unemployment and limited access to public services and the welfare state's support systems. At the same time, most countries have seen a continued trend of rising social inequality, which points to the fact that the burden of the crisis has not been shouldered equally by all social classes.
The drastic changes in the economic sphere are causing changes in the social integration of individuals through their networks and interpersonal trust. There is, however, strong disagreement in the literature about the predicted direction of these changes. On one side, we find the argument that the weakening economic situation of individuals in the labour market and the contraction of the welfare state are leading to the crumbling of the social solidarity, triggering a weakening of social networks in public and private spheres. On the other side, some argue that the reversal of responsibility for material well-being from the state back to the individual and their families, civil society, and local communities is leading to increased interpersonal responsibility that is strengthening social networks in private and public spheres.
The contrasting views on the effect of economic and political changes on social integration extend beyond scientific debate among social scientists and enter into the field of political discourse about the welfare state's past achievements and its future development. Social integration is a value in itself and the contribution of social security systems to social welfare is also measured by the extent it stimulates or hinders the development of social networks and interpersonal trust. The welfare state's potential negative impact on social integration would provide an additional argument to critics of the modern welfare state, while a positive impact would support those who caution against the disintegrative logic of the neoliberal attack on the welfare state at a time of economic crisis.
Data are readily available for exploring the economic crisis' impact on social capital. Survey data were collected within the framework of international research projects such as the European Social Survey and the World Values Study. Based on an analysis of these data, we will chart the trends in social networks and trust during the second half of the 2000s in order to see whether there have been changes in social capital, what have been the patterns of those changes, and how deep have they reached into the social fabric. We will attempt to answer the central research question of whether the worsening of the economic situation is leading to the consolidation or disintegration of social networks and interpersonal trust in Slovenia.
2. Economic uncertainty and social capital
Analysis of the social consequences of changes in the economic conditions of individuals lies at the centre of interest in two research fields. The first one explores the influence of different types of shocks on the lives of people in small local communities. A shock is understood as an unusual event that in a relatively short time can change relations in a community's social networks. Rather than considering unusual and sudden events, the second field focuses on long-term social processes wherein part of the responsibility for the welfare of individuals is transferred from the family and networks in local communities to state-run social security systems. The two fields of study approach the question of the impact of material uncertainty on social capital from two different vantage points: during shocks the material uncertainty of individuals and their families increases, while in the case of development of the welfare state material uncertainty declines.
2.1 Shocks and changes in social trust
Communities can face different types of shocks such as an economic crisis with related unemployment and reduced economic opportunities, the building or the closure of a nuclear power plant, natural disasters, the construction of highways and other infrastructural projects affecting the normal lives of people. Communities react to shocks in different ways. In some communities, a shock can lead to the strengthening of social capital followed by the restoration or even improvement of the quality of living in the community (Couch and Kroll-Smith 1994). This is because communities faced with collective threats come together, which results in the development of community networks and interpersonal help, cooperation among residents in the search for solutions and in their readiness for the joint mobilisation of resources. But shocks can also reduce social capital and the quality of life when they trigger large social divisions and conflicts in the community (Freudenberg and Jones 1991). How a community will react does not depend so much on the nature of the shock itself as on the how the shock is interpreted by residents in the process of managing the consequences (Besser, Recker et al. 2008). A shock has a destructive effect on the social capital in a community when the benefits, costs and risks related to the shock are not equally distributed, and when in the process of managing the consequences of the shock differences in the values and social status of the residents are strengthened.
Despite creating economic uncertainty at the level of the individual, an economic crisis can thus lead to higher or lower levels of social capital. However, it is most frequently expected to have a destructive impact on social integration. A substantial number of studies exploring the factors of generalised trust1 established that lower GDP, when that also means lower social welfare, hinders the development of trust, making
1. Generalised trust is one of the elements of social capital, along with social networks and norms of reciprocity (Putnam 2000). It is measured by the question: "Generally speaking, would you say that most people can be trusted, or that you can't be too careful in dealing with people?" Answers to this question express trust in anonymous others.
trust seen as a 'luxury' which cannot be afforded by everyone (Knack and Keefer 1997; Zak and Knack 2001; Beugelsdijk, de Groot et al. 2004; Algan and Cahuc 2010; Steijn and Lancee 2011). People with access to fewer resources have greater difficulty covering their losses following a potential abuse of their trust, which is why they are more cautious in placing their trust in others. Trust is also seen to go down with the rise of social inequality that accompanies economic crises. In these circumstances, social inequality implies taking advantage of and a lack of care for others, leading people to distrust others in general. While a higher degree of equality and material well-being enables an optimistic view of the world in which socialising with others creates life opportunities, in the situation of greater inequality and shrinking resources socialising with others is not seen as something that would produce opportunities (Knack and Zak 2002; Uslaner 2002; Rothstein and Uslaner 2005; Uslaner and Brown 2005; Berggren and Jordahl 2006; Leigh 2006; Bjornskov 2007; Gustavsson and Jordahl 2008).
In contrast to the above-mentioned studies, Searing (2013), when analysing the social consequences of the recession in the states of Latin America, found a positive effect of the economic recession on social capital. She showed that the longer the recession continued the more the social capital grew, although great shocks in terms of large drops in GDP slowed down the dynamics of social capital's increase. An important factor contributing to the rise of social capital during the economic crises was institutional trust which expressed a positive valuation of the ways in which the elites were striving to manage the crises situation. The quality of political institutions, particularly the degree to which the rule of law is developed, the non-corruption of public employees and the extent to which democracy is developed, encourage generalised trust as they, in the same way as social equality, testify to the trustworthiness of politicians, their adherence to universalistic norms and caring for others in general (Rothstein 2001; Iglič 2004; Delhey and Newton 2005). In the case of Latin America, however, institutional trust tended to be maintained through political populism (Searing 2013). But regardless of the source of institutional trust, if it recedes during a crisis, the impact of the crisis on social capital shifts towards the negative side.
Beside institutional trust, the dynamics of social capital during an economic crisis depend on so-called positive and negative loops acting as the mechanism for reproducing inequality in social capital (Bourdieu 1986; Lin 2000). When facing a shock, attempts to unequally distribute the costs of the shock and to widen the social divisions among the residents, which both tend to further reduce social capital, are more likely to occur in communities with a low initial level of social capital. Conversely, residents of communities with a high initial level of social capital tend to react to a shock by trying to evenly distribute its costs while also cooperating across the social divisions in the community. In this case, the experience of the shock will strengthen the community's social capital. In their study of national reactions to the 2008 crisis among OECD countries, Helliwell, Huang et al. (2014) found that a high level of social capital influenced the countries' ability to manage the crisis in a way that further strengthened their social capital, and led to greater satisfaction among the population.
2.2 Social networks in the welfare state context
Another research area relevant to our study is the examination of changes in social capital in the context of the welfare state's increased role in the second half of the 20th century. In this case, economic uncertainty was reduced. The central question here is whether solidarity organised by the state is supplanting the solidarity created within family networks, local communities and civil society organizations.
According to the communitarian argument, the development of the welfare state caused personal responsibility to be replaced by social responsibility which put an end to the intermediary organizational and communal social structures. When social obligations become public, citizens refer their needs for help to the state instead of to their co-citizens, which makes social bonds and networks lose their meaning, and isolated individuals, who no longer need each other, lose their ability for self-organisation, moral sense of responsibility for the community in which they live, and capacity to place trust in each other (De Swann 1988; Wolfe 1989; Etzioni 1995; Fukuyama 2000; Putnam 2000). The interdependence of individuals is considered to be the most important glue binding social networks together, which presumes an expressly instrumental understanding of social relations.
Not only according to the communitarian argument, but also to the authors of the 'third way' the modern welfare state should be adjusted so as to again be able to revitalise the potential of civil society (Giddens 1998); they see solutions in the so-called enabling welfare state aiming to assist individuals and communities in developing self-help mechanisms. Others, however, are sceptical of this argument, claiming that in the past few decades the supporting functions of the family and local community networks have eroded to the extent that they are no longer able to take on the tasks of providing welfare to their members (LeGrand 1997; Taylor-Gooby 1998). Also addressing this issue, a study on social policy in Portugal finds that while the Mediterranean type of welfare state is based on families being responsible for the care of their members, this does not necessarily mean that primary networks actually perform this task (Wall, Aboim et al. 2001). The absence of public systems for the provision of social security does not automatically imply that, in the case of economic uncertainty, the care will in fact be shouldered by social networks.
The opposing, liberal argument calls attention to the fact that, rather than being supplanted by the welfare state, intermediary structures and their development are even stimulated by it. The most frequently stated example is Scandinavian states where development of the third sector is stimulated by the welfare state by the latter's transfer to the former of the responsibility and financial resources to carry out social welfare activities (Kuhnle and Alestalo 2000; Rothstein 2001). Moreover, the welfare state is considered to have a positive influence on the individual's involvement and active participation in civil society associations. People who do not live in economic uncertainty, and have more time and money to develop their social capital by participating in various formal and informal networks, also show a high degree of trust in other people, take on responsibility for others and act in solidaristic rather than opportunistic
ways (Skocpol 1996; Salamon and Sokolowski 2003; Gelissen, Oorschot et al. 2012). Therefore, instead of limiting individuals the state enables their involvement in social networks by ensuring their right to social security. Social relations are assumed to have an intrinsic or expressive value, rather than simply an instrumental one.
Empirical analyses indeed show a positive statistical correlation between different indicators of social capital and the welfare state, operationalised either by the share of GDP earmarked for social security or by the type of welfare regime (universalistic, liberal, conservative, Mediterranean, post-socialist). In countries in which a large part of GDP is set aside for social security, and in the universalistic welfare regimes, people more often become members of voluntary associations, dedicate more time to voluntary work and show greater trust in other people and in institutions (Gaskin and Smith 1995; Costa and Kahn 2001; Hooghe and Stolle 2003; Van Oorschot and Arts 2005; Kaariainen and Lehtonene 2006; EU 2007; Larsen 2007; Wallace and Pichler 2007; Kaasa and Parts 2008). In fact, in Europe, the lowest levels of social capital are found in the Mediterranean and post-socialist welfare regimes (Oorschot, Arts et al. 2005; Kääriäinen and Lehtonen 2006) with low levels of social spending.
Rather than assessing a joint level of social capital, some authors suggest differentiating between bonding and bridging social capital. The first one measures individuals' integration into networks of strong bonds (family and close friends), while the second one measures their connecting into multiple networks of weaker bonds (co-workers, neighbours, members of organisations). A high level of bonding social capital is found in the Mediterranean welfare system, with an equally high level also present in Slovenia and the Czech Republic, although these are classified within the post-socialist type of regime. In contrast, the level of bridging social capital is highest in the universalistic welfare regime, followed by the liberal and conservative regimes (Kääriäinen and Lehtonen 2006).
While social networks or social capital represent the possibility of access to social support, the developed welfare state affects the need for social support. Contrary to the expectations of communitarian authors, the bonding social capital that prevails in countries with weaker welfare states, does not guarantee a high degree of social support, even though residents of these countries show a greater need for social support than those in countries with a stronger welfare state. Social support is weaker in countries with the Mediterranean and post-socialist welfare regimes, including Slovenia (Kääriäinen and Lehtonen 2006).2 Closed family networks tend to hinder the availability of social support, since it is wider social networks which could offer various kinds of support. Therefore, formal systems of social security developing under the protection of the welfare state do not prevent but, conversely, stimulate the development of larger social networks and bridging social capital which is an important source of various types of social support.3
2.	Slovenia displays a lack of social support mainly in situations where people need someone to turn to for help while experiencing psychological problems or bigger financial difficulties (Iglič 2004).
3.	The relation between the welfare state and development of the bridging type of social capital is not necessarily unambiguous, although for the case of the USA both Uslaner (2002) and
A study carried out in Slovenia on the impact of the political transition on social networks showed that in the first decade after the transition the increase in uncertainty as a result of the great institutional and economic changes led to the contraction of social networks (Hlebec, Filipovic Hrast et al. 2010). In their study of support networks, Mandič and Hlebec (2005) also find that during the 1990s community networks were far less responsive to the newly emerging risks and problems than personal networks of social support. We expect a similar effect to take place in the case of the economic crisis happening two decades later.
Social capital has been shown to be affected by economic uncertainty and contraction of the welfare state (Gelissen, Oorschot et al. 2012) through a reduction of the available resources and decreased generalised trust, both of which have led to a lower degree of involvement in associational and communal networks. The economic crisis is increasing mutual dependence among individuals, and requires higher, more frequent and unpredictable investments in terms of help, which is why support networks are closing in, forming bounded solidarities of reciprocity and decreasing general social responsibility. Low social responsibility is very vividly expressed in the cultural phenomena of "amoral familialism" (Banfield 1958) where a high degree of solidarity within the family circle is correlated with a lack of wider solidarity and lower level of readiness to offer help to individuals and social groups who are not part of the family network. This process is in stark contrast with the expectations of the communitarian view presented at the beginning of this section, which predicted that in the conditions of greater interpersonal dependence and need for help, social networks would enlarge and provide social support to a wide circle of acquaintances.
2.3 Research Model and Data
Our central research question is how economic crises influences social networks and interpersonal trust due to its deepening of the economic uncertainty facing citizens. Data used in the analysis were collected as part of two international projects, namely the European Social Survey and the World Value Survey (the ESS and WVS, respectively). In both cases, social capital levels will be compared in the last survey conducted prior to the crisis (ESS 2008 and WVS 2005-2009) and in the first survey carried out after the crisis started (ESS 2010 and WVS 2010-2014). Our study explores the short-term effects of the crisis, bearing in mind that the period of just a few years is not long enough to reveal the contours of any long-term social changes.
Patulny (2004) show that the fall in trust occurred following the contraction of the welfare state and the rise in social inequality. Some consider this a reciprocal causal relation with the welfare state and social capital strengthening each other: not only does the welfare state enable the development of bridging social capital, but the latter influences the formation of policies that strengthen the welfare state (Henriksen and Bundesen 2004). According to a third belief, however, the relationship is spurious, with the welfare state and bridging social capital both resulting from the same cause, such as the level of economic development (Inglehart 2003) or the democratic tradition of the state (Halman 2003).
Based on the arguments presented above regarding the impact of the economic shock and the development of the welfare state on social capital, a model of four possible scenarios was prepared (Figure 1). Arguments 1a and 1b speak about the changes in bridging social capital, while arguments 2a and 2b refer to bonding social capital. The left side of the model corresponds to the 'pessimistic' and the right one to the 'optimistic' scenario of the transformation of social networks during a crisis.
Figure 1: Impact of an economic crisis on social capital
Argument 1a
Economic crisis
BRIDGING SOCIAL CAPITAL Generalised trust
Trust in different social groups

Argument 2a
Trust in people we know personally
Argument 1b
Contacts with family and close friends BONDING SOCIAL CAPITAL
Economic crisis
Argument 2b
Argument 1a. In the first argument it is contended that during a crisis, when prosperity is falling and inequality is rising, generalised trust in people as an indicator of bridging social capital is also falling. The drop in generalised trust influences the decrease in trust at lower levels of social system, that is trust in different social groups, such as other nations, ethnic and religious groups, and people we know personally. The drop in generalised trust is thus expected to result in the contraction of social networks in the intermediary social sphere, namely networks that are formed in work and voluntary organisations, in local communities and in wider circles of friends and acquaintances.
Argument 1b. While the second argument also speaks about the influence of a crisis at the level of bridging social capital, it predicts the positive rather than the negative development of social capital. This scenario can be expected in the situation of an existing high level of social capital in society, and in the situation in which the crisis is being managed in a way which enables people to maintain their trust in institutions. Institutional trust spills over onto generalised trust that provides the basis for the building of wider social networks.
Argument 2a. This involves the assumption that a crisis has a positive impact on the power of bonding social capital through increased needs for larger and more demanding forms of social support, such as financial support during unemployment. Social support that requires larger investments in terms of time and financial resources is usually provided by strong social bonds (Wellman 1990). However, the intensification of social support within networks of strong bonds is not necessarily correlated with
broader social responsibility and, as we have already elaborated, in the conditions of insufficient generalised trust family networks might start closing off and complying with the syndrome of "amoral familialism" (Banfield 1955).
Argument 2b. Similar to the above argument, this argument also presumes support networks at the level of family and close friends become stronger while contending, however, that in the conditions of high level of generalised trust social responsibility and the building of social networks also occur on a wider scale, within civil society and local communities. This argument differs from Argument 2a by predicting that the complementarity of networks at different levels.
3. Results
The presentation of the results of the analysis is arranged so that we move from bridging to bonding social capital. After reviewing changes at different levels of social capital - from generalised trust to trust in different social groups to contacts in the private sphere - it will be established which of the above scenarios corresponds to the Slovenian situation.
3.1 Generalised trust and attitudes towards immigrants
Since 2002, the European Social Survey has offered a biennial measurement of generalised trust on a ten-level scale. Generalised trust is operationalised by the following question: "Generally speaking, would you say that most people can be trusted, or that you can't be too careful in dealing with people?". Moreover, the attitude towards other people is measured by two more questions: "Do you think that most people would try to take advantage of you if they got the chance, or would they try to be fair?" and "Most of the time people try to be helpful or that they are mostly looking out for themselves?" The generalised trust variable used in the analysis expresses the mean value of the answers to all three questions.4
According to the argument presented in the previous section, a drop in generalised trust is expected during an economic crisis due to the fall in GDP and increased inequality. This drop could be stopped by institutional trust provided the public would share its belief in the political elites responding adequately and justly to the crisis. The World Bank World Development Indicators data show that in Slovenian the GDP per capita dropped from USD 27.02 in 2008 to USD 24.05 in 2009 and USD 22.89 in 2010. At the same time, according to data from the Standardised World Income Inequality Database (Solt 2013) income inequality before taxes and transfers, measured as the Gini index, rose from 33.91 in 2008 to 34.60 in 2010, which is the highest Slovenian inequality rate after the year 2000.5 This rise in income inequality was not alleviated
4.	For the problems related to the measuring of generalised trust in cross-national survey research see Adam (2008) and Volf (2005).
5.	During the first half of the 2000s, income inequality after taxes and transfers was falling steadily, dropping from 23.42 in 2002 to 22.84 in 2007, when it started rising again. Its biggest growth occurred between the crisis years of 2009 and 2010 (Solt 2013).
by the taxation policy and social transfers. Trust in political institutions and actors6 also decreased in this period. After the year 2000 political trust was rising until it reached its peak value of 3.76 in 2008, after which it dropped to 2.49 in 2010.
Figure 2: The movement of generalised trust in Slovenia (2002-2010)
Figure 2 shows the changes in generalised trust in Slovenia since 2002. Generalised trust grew from 2002 to 2008, until its rate dropped in 2008-2010 below the level from the early 2000s. The 2008-2010 drop in trust was assessed by regression analysis7 which also included several variables at the individual level: education, generation and gender, political trust and income8. The impact of the time variable (or "crises" variable) is statistically significant in the bivariate model, but its effect decreases when we include the economic variable, and disappears when the political trust variable is
6.	Political trust is the mean trust expressed by the respondents towards the National Assembly, politicians and political parties. It is measured on a scale from 0 "do not trust at all" to 10 "trust completely". The data come from the European Social Survey.
7.	The "crisis" variable is defined as a dichotomous variable with the value 0 for "the year 2008" and 1 for "the year 2010."
8.	The level of education achieved is measured on a four-level scale and, for the purpose of this analysis, the following four dichotomous variables were formed from the original variable: primary education or lower, lower secondary education, higher secondary education and tertiary education. Generations are defined according to the year of birth in the following way: born before 1946, born between 1946 and 1960, born between 1961 and 1975, born between 1976 and 1980, and born after 1980. Gender is a dichotomy in which the value 0 is attributed to the female gender. The income was expressed as the monthly available household income defined by the upper value of the decile to which the respondent or his/ her household belong. The data on the upper limit of the deciles were collected by the SILC research and refer to the year prior to that in which the European Social Study was carried out. Political trust is operationalised as trust in political institutions and actors (national assembly, political parties and policies) and trust in state systems (police, judiciary).
added to the model. In Slovenia, generalised trust has mainly decreased due to the low political trust accompanying the economic crisis, and only to a smaller extent due to the decrease in income and household living standards. Rather than by changes in material wellbeing as such, social capital has more significantly been influenced by the way in which the economic crisis is managed.
Generational differences are noteworthy in this regard: it is the youngest generation that has experienced the highest drop in generalised trust. Young people are also those whose generalised trust has been taken away due to their distrust in politics as opposed to older generations whose generalised trust has diminished due to the worsening of the material conditions of their lives.
Table 1: Multivariate regression analysis of generalised trust and attitudes towards immigration (standardised coefficients)
	Generalised trust	Immigration
Wave (1=2010)	.016	.030
Gender (1=male)	.069**	.006
Education, primary	-.020	-.004
Education, lower secondary	(base)	(base)
Education, higher secondary	.036	.037
Education, tertiary	.151***	.164***
Cohort born before 1946	.058*	.013
Cohort born 1946-1960	-.022	.012
Cohort born 1961-1975	(base)	(base)
Cohort born 1976-1990	.002	.025
Cohort born after 1990	.035	.042
Household income	.157***	.071**
Trust - political institutions	.215***	.110***
Trust - state institutions	.180***	.085**
Active in vol. organisations	-.016	
Generalised trust		.167***
R2 adj.	.211	.137
NOTE: *sig. <0.05 **sig. < 0.01 ***sig. < 0.001
The next step in our analysis involves observation of the attitudes towards social groups which are different from the respondent's. In the ESS the question regarding the attitude to immigrants which we used in our study reads: "Is your country made a worse or a better place to live by people coming to live here from other countries?"9 A positive attitude to immigrants was growing up until 2006, when it started to fall. The analysis shows that the change in attitude to immigrants between the two time points is again statistically significant, and that this change can be explained by both decline in the political trust and income although the effect of political variables is
9. Answers to this question were measured on a 10-level scale in which the lower values express a critical attitude to immigrants.
again stronger than the effect of economic variable.10 The young and highly educated have a more positive attitude to immigrants than other demographic categories and the changes in the attitude of young people are mainly a result of their increasing political distrust. The attitude to immigrants is also strongly influenced by generalised trust spilling over onto specific objects of trust, in our case the group of immigrants.11
3.2 Everyday contacts
Everyday contacts include social ties with people individuals meet in formal and informal social contexts, from work and voluntary organisations to local community and circles of friends and acquaintances. In the WVS, the respondents were asked about their membership and active participation in ten types of voluntary organisations: church and religious organisations; sports and recreational organisations; artistic, educational and musical organisations; trade union organisations; political parties; environmental organisations; professional organisations; humanitarian and charity organisations; consumers' organisations and others. The data show that membership in organisations dropped from 1.48 to 1.39, and active membership from 0.72 to 0.67.12 Membership fall principally occurred in traditional voluntary organisations, i.e. church and religious organisations and trade unions, which have older membership.
10.	Reduced material wellbeing causes the feeling of economic deprivation and cultural threat due to immigration. This is testified to by answers to the following questions: "Is immigration bad or good for our country's economy?" and "Do immigrants pose a threat to our culture?" which highly correlate with the basic question of whether immigration is good or bad for a society.
11.	As opposed to the previous model, this time the equation also includes the variable of generalised trust, in accordance with the thesis that the impact of generalised trust trickles down: those who trust people in general also trust specific social groups.
12.	The numbers express the average number of memberships and active memberships per capita in ten groups of organisations.
Table 2: Multivariate analysis of participation in voluntary organisations, neighbourhood trust and importance of friends (standardised regression coefficients)
	All membership	Active membership	Neighbourhood trust	Importance friends
Wave (1=2010)	-.029	.028	-.036	-.083***
Gender (1=male)	.049*	.083***	.058**	.005
Education, primary	-.045	-.049*	.020	.023
Education, lower secondary	(base)	(base)	(base)	(base)
Education, higher secondary	-.042	.045*	-.010	.016
Education, tertiary	.076**	.070**	-.043	-.012
Cohort born before 1946	-.131**	-.068**	.146***	.048
Cohort born 1946-1960	-.111**	-.037	.036	.031
Cohort born 1961-1975	(base)	(base)	(base)	(base)
Cohort born 1976-1990	-.035	.007	-.038	.135***
Cohort born after 1990	.019	.056**	.020	.082***
Household income	.044	.048*	.005	.049*
Generalised trust	.039	.071*	.084**	.060**
Trust in the family	.021	.005	.217***	.041
R2 adj.	.042	.050	.102	.045
NOTE: *sig. <0.05 **sig. < 0.01 ***sig. < 0.001
The difference between the years before and after the beginning of economic crises is statistically significant only when we examine the decrease in active participation in organisations. The inclusion of different groups of variables in the model shows that the drop is influenced mainly by changes in generalised trust and income: the effect of the time is completely eliminated when all variables are included in the model (Table 2). These results tell us that parallel to the reduction of the available income and generalised trust occurs the reduction of people's active participation in voluntary organisations, and the re-directing of their resources and time to some other activities and social networks.
In the next step, our focus moved from formal organisational contexts towards the local community. The question we used asks about trust in the neighbourhood and can be found in WVS questionnaire.13 The average level of neighbourhood trust was slightly lowered during the economic crises due to the reduction of the share of those stating that they »completely trust« their neighbourhood from 17% to 14%, while there was a corresponding rise in the share of those who only »trust partly«.
Lowering of neighbourhood trust is directly related to decrease in generalised trust - those who trust people in general also trust individual categories of others including
13. The question "How much do you trust people in your local community?" is measured on a scale from 1 "trust completely" to 4 "do not trust at all". For the needs of this analysis, the variable was dichotomised into 1 "trust completely" and 0 "trust less or not at all".
neighbours and acquaintances from the local community. But trust in neighbourhood also highly correlates with trust in the family which means that the economic crises has double effect on neighbourhood trust. While the decreasing generalised trust supresses neighbourhood trust, increases in the bonding social capital positively effect trust in the communal networks. This positive spill-over effect of bonding social capital was not discerned in the case of participation in associational networks.
The next question refers to informal networks of friends. We measure changes in friendship networks with the question asking about the importance of friends in one's life. This question is available in the WVS where it is contained in a set of questions including the family, free time, politics, work and religion.14 Table 5 shows that the valuation of friends in terms of their importance in the individual's life was reduced with the economic crises. Not only friends but most other activities included in the survey's question (work, religion, free time) are less important after the crises than before the crises. The only exception is family which remained equally important as before the crisis; we address this phenomenon in the next section. Although younger generations consider friends more important than the older generations, this generational change is not strong enough to reverse the trend of falling importance of friends. Decrease in generalised trust and changes in income have direct impact on the weakening of the social bond with friends.
The analyses presented so far have shown that during the crisis the levels of social capital diminished both at the level of generalised trust, trust in specific social groups as well as in every day social networks. While the changes are small, they are telling enough to allow us to conclude about the economic crisis' negative impact on bridging social capital and weak social ties. Thus, Argument 1a is confirmed and Argument 1b is refuted. In the next section, Arguments 2a and 2b will be addressed with the aim of identifying the impact of the crisis on the networks of strong ties.
3.3 Close social contacts
Strong ties networks will be studied through the questions about the trust in the family and access to persons with whom one can talk about intimate and personal matters.15 In Slovenia, 86% of respondents express a high level of trust in the family in 2010 and about 89% of respondents see the family as very important, both testifying to the great centrality of the family for the Slovenians compared to Western and Northern European countries.
14.	The importance of friends is measured on a scale from 1 "very important" to 4 "not at all important", which was in our analysis dichotomised into 1 "very important" and 0 "all other".
15.	In the WVS, the survey participants answered the question about the importance of the family in their lives on a scale of 1 "very important" to 4 "not at all important", which was transformed into the dichotomous variable: 1 "very important" and 0 "all others". Also the question about trust in the family which was measured on a scale from 1 "trust completely" to 4 "do not trust at all" is taken from the WVS. The respondents answered with 0 "no" and 1 "yes" to the question from the ESS asking if they had anybody with whom they can talk about intimate or personal matters.
Table 3: Multivariate analysis of the family relations and the availability of intimate bonds (standardised regression coefficients)
	Trust family	Intimate support
Wave (1=2010)	.055**	.031
Gender (1=male)	.005	-.013
Education, primary	-.010	.002
Education, lower secondary	(base)	(base)
Education, higher secondary	-.001	.091**
Education, tertiary	.018	.116***
Cohort, born before 1946	.037	-.135***
Cohort, born 1946-1960	.024	-.026
Cohort, born 1961-1975	(base)	(base)
Cohort, born 1976-1990	.012	.033
Cohort, born after 1990	.031	.062**
Household income	.080**	.058*
Generalized trust	-.046*	.076**
R2 adj.	.010	.053
NOTE: *sig. <0.05 **sig. < 0.01 ***sig. < 0.001
The crisis has not brought any dramatic changes in the individual's relations with his or her family, although there are some indications that these bonds have even been reinforced as trust in the family has grown (Table 3). At the same time, the results of the analysis show a generally negative relationship between trust in the family and trust in other people. This tells us that trust in the family does not coincide with generalised trust and that the two represent opposing patterns of social integration where strong integration into the family networks is accompanied by weak integration into wider networks and generalised trust. Thus the Slovenian case resembles most closely the scenario 2a where bonding social capital increases at the same time as bridging social capital decreases. In this process reinforcement of family ties helps strengthen the dividing line with weak ties outside the family circle, in particular with ties established in the contexts of voluntary associations and networks of friends and acquaintances. Also lowering of the income during the crises did not bring about less trust in the family despite the generally positive relationship between the family trust and income. This means that although low income, controlled for age and education, usually implies weaker family integration, lowering of the income during the period of economic crises did not undermine family trust.
As suggested in the literature, the intimate support is positively related with bridging social capital. Intimate support is to a larger extent available to those with higher level of generalised trust, and to younger and more educated people. The availability of intimate support has not changed during the economic crises, largely because the affect of crises which runs through the variable of generalised trust has been offset
by the generational changes. Without the generational change which is reflected in different patterns of socialising between the younger and older generation, Slovenian respondents would have less intimate contacts as a result of the economic crises. The analysis of the close social ties thus confirms the 2a scenario.
4. Discussion
The economic crisis that started in 2008 has brought about reduced access to material resources of households and increased social inequality in Slovenia. It has also lowered political trust since people were not satisfied with the ways the political elites managed the consequences of the crises. Both decrease in the well-being and in the political trust have had important consequences for social capital, understood in terms of social networks and interpersonal trust.
Previous research has drawn attention to the impact of the economic crisis at the level of individuals including stress, mental disease, and exposure to risky ways of living. But economic crisis also produces cracks in the social fabric since it, along with worries related to questions of everyday survival, decreases trust in others and discourages people from participating in voluntary association and building social ties in various formal and informal networks.
In the Slovenian context, at the same time as bridging social capital is eroding, family networks, which are traditionally very strong, are becoming even stronger. The erosion of generalised trust coincides with the increase of trust in strong family bonds which offer material social support in the times of uncertainty and distress. In contrast to communitarian argument, the escalation of interpersonal dependence does not lead to a strengthening of wider social solidarity, but to the formation of islands of strong loyalty and trust. One of the results of strengthening of bonding social capital is also lower availability of intimate social support which is usually provided by the bridging social capital and social ties which emerge as a result of generalised trust. The Slovenian case in many ways testifies to the liberal argument about the importance of material security for the development of bridging social capital.
Changes perceived by surveys refer to a short period: in the case of the European Social Survey data to a period of two years, and in the case of the World Valus Survey to six years. Possibly, our research has identified the initial effects of the crisis that are a result of the huge drop in GDP in a short time period, and which will over time gradually turn in a positive direction. Yet, the results of our analysis clearly imply that the path to the positive spiral of social capital in times of a crisis is paved in political trust, including trust in political actors and institutions. Therefore, it could be concluded that the management of the crisis has just as big, if not a bigger, impact on social capital than the crisis itself.
Bibliography
Adam, Frane (2008): Mapping Social Capital Across Europe: Findings, trends and methodological Shortcoming of Cross-national Surveys. Social Science Information, 47: 159-186.
Algan, Yann, and Cahuc, Pierre (2010): Inherited Trust and Growth. American Economic Review, 100:2060-2092.
Banfield, Edward C. (1958): The Moral Basis of a Backward Society. Free Press, New York.
Berggren, Niclas, and Jordahl, Henrik (2006): Free to Trust? Economic Freedom and Social Capital. Kyklos, 59:141-169.
Besser, Terry R., Recker, Nicholas, and Agnitsch, Kerry (2008): The Impact of Economic Shocks on Quality of Life and Social Capital in Small Towns. Rural Sociology, 73:580-604.
Beugelsdijk, Sjoerd, de Groot, Henri, and van Schaik, Anton B.T.M. (2004): Trust and Economic Growth: A Robustness Analysis. Oxford Economic Papers - New Series, 56:118-134.
Bjornskov, Christian (2007): Determinants of Generalized Trust: A Cross-country Comparison. Public Choice, 130:1-21.
Bourdieu, Pierre (1986): The Forms of Capital. In J. Richardson (ed.): Handbook of Theory and Research for Sociology: 241-258. Greenwood, New York.
Costa, Dora, and Kahn, Matthew (2003): Understanding the American Decline in Social Capital, 1952-1998. Kyklos, 56:17-46.
Couch, Stephen R., and Kroll-Smith, Steve (1994): Environmental Controversies, Interfactial Resources, and Rural Communities: Siting Versus Exposure Disputes. Rural Sociology, 59:25-44.
Delhey, John, and Newton, Kenneth (2005): Predicting Cross-National Levels of Social Trust: Global Pattern or Nordic Exceptionalism? European Sociological Review, 21:311-327.
De Swann, Abram (1988): In Care of the State: State Formation and Collectivization of Healthcare Education and Welfare in Europe and Africa in the Modern Era. Polity Press, Oxford.
Etzioni, Amitai (1995): The Spirit of Community: Rights, Responsibilities and the Communitarian Agenda. Fontana Press, London.
EU (2007): Social Cohesion, Trust and Participation: Social Capital, Social Policy and Social Cohesion in the EU Countries and Candidate Countries. European Commission, Brussels.
Freudenberg, William, and Jones, Timothy R. (1991): Attitudes and Stress in the Presence of a Technological Risks: A Test ofthe Supreme Court Hypothesis. Social Forces, 69:1143-11 68.
Fukuyama, Francis (2000): Social Capital and Civil Society. IMF Working paper 00/47, Washington.
Gaskin, Katharine, and Smith, Justin D. (1995): A New Civic Europe? A Study of the Extent and Role of Volunteering. The Volunteer Center: London.
Gelissen, John P.T.M., van Oorschot, Wim J.H., and Finsween, Ellen (2012): How Does the Welfare State Influence Individual Social Capital? European Societies, 14:416-440.
Giddens, Anthony (1998): The Third Way: The Renewal of Social Democracy. The Polity Press, Cambridge.
Gustavsson, Magnus, and Jordahl, Henrik (2008): Inequality and Trust in Sweden: Some Inequalities are More Harmful than Others. Journal of Public Economics, 92:348-365.
Halman, Loek (2003): Volunteering, Democracy, and Democratic Attitudes. In Paul Dekker and Loek Halman (eds): The Values of Volunteering: 179-198. Kluwer Publishers, New York.
Helliwell, John F., Huang, Haifang, and Shun Wang (2014): Social Capital and Well-Being in Times of Crises. Journal of Happiness Studies, 15:145-162.
Henriksen, Lars, and Bundesen, Peter (2004): The Moving Frontier in Denmark: Voluntary-State Relationships since 1850. Journal of Social Policy 33:605-625.
Hlebec, Valentina, Filipovič Hrast, Maša, and Kogovšek, Tina (2010): Social Networks in Slovenia: Changes During the Transition Period. European Societies, 12:697-717.
Hooghe, Marc, and Stolle, Dietlind (2003): Introduction: Generating Social Capital. In M. Hooghe and D. Stolle (eds.): Generating Social Capital. Civil Society and Institutions in Comparative Perspective. Palgrave Macmillan: New York.
Iglič, Hajdeja (2004): Dejavniki nizke stopnje zaupanja v Sloveniji. Družboslovne razprave, 20:149-175.
Iglič, Hajdeja (2004): Tri ravni socialnega kapitala v Sloveniji. In B. Malnar, I. Bernik (eds.): S Slovenkami in Slovenci na štiri oči. Ob 70-letnici sociologa Nika Toša. Dokumenti SJM, FDV, Ljubljana.
Inglehart, Ronald (2003): Modernization and Volunteering. In P. Dekker and L. Halman (eds.): The Values of Volunteering: 55-70. Kluwer Publishers, New York.
LeGrand, Julian (1997): Knights, Knaves or Pawns? Human Behavior and Social Policy. Journal of Social policy, 26:149-169.
Leigh, Andrew (2006): Trust, Inequality, and Ethnic Heterogeneity. Economic Record, 82:268280.
Kaasa, Anneli, and Parts, Eve (2008): Individual-level Determinants of Social Capital in Europe: Differences Between Country Groups. Acta Sociologica 51:145-168.
Kääriäinen, Juha, and Lehtonene, Heikki (2006): The Variety of Social Capital in Welfare State Regimes - a Comparative Study of 21 Countries. European Societies, 8:27-57.
Knack, Stephen, and Keefer, Philip (1997): Does Social Capital Have an Economic Pay-off? A Cross-country Investigation. Quarterly Journal of Economics, 112:1251-1288.
Knack, Stephen, and Zak, Paul J. (2002): Building Trust: Public Policy, Interpersonal Trust, and Economic Development. Supreme Court Economic Review, 10:91-107.
Kuhnle, Stein, and Alestalo, Matti (2000): Introduction: Growth, Adjustments and Survival of European Welfare States. In S. Kuhnle (ed.): Survival of the European Welfare State: 3-18. Routledge, London.
Larsen, Christian A. (2007): How Welfare Regimes Generate and Erode Social Capital: The Impact of Underclass Phenomena. Comparative Politics, 40:83-101.
Lin, Nan (2000): Inequality in Social Capital. Contemporary Sociology, 29:785-795.
Mandič, Srna, and Hlebec, Valentina (2005): Socialno omrežje kot okvir upravljanja s kakovostjo življenja in spremembe v Sloveniji med letoma 1987 in 2002. Družboslovne razprave, 21:263-285.
Patulny, Roger (2004): Social Capital and Welfare: Dependency or Division? Examining Bridging Trends by Welfare Regime, 1981-2000. Social Policy Research Centre, Sydney.
Putnam, Robert D. (2000): Bowling Alone: The Collapse and Revival of American Community. Simon & Schuster, New York.
Rothstein, Bob (2001): Social Capital in the Social Democratic Welfare State. Politics & Society, 29:207-241.
Rothstein, Bob, and Uslaner, Eric M. (2005): All for All: Equality, Corruption, and Social Trust. World Politics, 58:41-72.
Salamon, Lester, and Sokolowski, Wojciech (2003): Institutional Roots of Volunteering. In P. Dekker and L. Halman (ed.): The Values of Volunteering, 71-90. Kluwer Publications, New York.
Searing, Elizabeth A.M. (2013): Love thy Neighbor? Recessions and Interpersonal Trust in Latin American. Journal of Economic Behavior & Organization, 94:68-79.
Skocpol, Theda (1996): Unraveling from Above. American Prospect, March-April: 20-25.
Steijn, Sander, and Lancee, Bram (2011): Does Income Inequality Negatively Affect General Trust? Examining Three Potential Problems with the Inequality-Trust Hypothesis. GINI Discussion Paper No. 20.
Taylor-Gooby, Peter (1998): Markets and Motives: Implications for Welfare. CCWS Working Paper, University of Kent.
Uslaner, Eric M. (2002): The Moral Foundations of Trust. Cambridge University Press, Cambridge.
Uslaner, Eric M. and M. Brown (2005): Inequality, Trust and Civic Engagement. American Politics Research, 33:868-894.
Pichler, Florian, and Wallace, Claire (2007): Patterns of Formal and Informal Social Capital in Europe. European Sociological Review, 23:423-435.
Van Oorschot, Wim, and Arts, Wil (2005): The Social Capital Of European Welfare States: the Crowding Out Hypothesis Revisited. Journal of European Social Policy, 15:5-26.
Van Oorschot, Wim, Arts, Wil, and Loek Halman (2005): Welfare State Effects on Social Capital and Informal Solidarity in the EU: Evidence form the 1999/2000 European Values Survey. Policy & Politics, 33:33-54.
Volf, Tomaž (2005): Problem zaupanja v sodobnih družbah (doktorska disertacija). Univerza v Ljubljani, Fakulteta za družbene vede.
Wall, Karin, Aboim, Sofia, Cunha, Vanessa, and Vasconcelos, Pablo (2001): Families and Informal Support Networks in Portugal: The Reproduction of Inequality. Journal of European Social Policy, 11:213-233.
Wellman, Barry, and Wortley, Scot (1990): Different Strokes from Different Folks: Community Ties and Social Support. American Journal of Sociology, 96:558-588.
Worms, Jean-Pierre (2002): Old and New Civic and Social Ties in France. In Putnam R.D. (ed.): Democracies in Flux: the Evolution of Social Capital in Contemporary Society. Oxford University Press, Oxford.
Wolfe, Alan (1989) Whose Keeper? Social Science and Moral Obligation. University of California Press, San Francisco.
Zak, Paul J., and Knack, Stephen (2001): Trust and Growth. Economic Journal, 111:291-321.
Data sources
ESS Round 5: European Social Survey Round 5 Data (2010). Data file edition 3.1. Norwegian Social Science Data Services, Norway - Data Archive and distributor of ESS data. http:// www.europeansocialsurvey.org/download.html?file=ESS5e03_1&y=2010 ESS Round 4: European Social Survey Round 4 Data (2008). Data file edition 4.2. Norwegian Social Science Data Services, Norway - Data Archive and distributor of ESS data. http:// www.europeansocialsurvey.org/download.html?file=ESS4e04_2&y=2008 WORLD VALUES SURVEY Wave 6 2010-2014 OFFICIAL AGGREGATE v.20140429. World Values Survey Association (www.worldvaluessurvey.org). Aggregate File Producer: Asep/ JDS, Madrid SPAIN. http://www.worldvaluessurvey.org/WVSDocumentationWV6.jsp WORLD VALUES SURVEY Wave 5 2005-2008 OFFICIAL AGGREGATE v.20140429. World Values Survey Association (www.worldvaluessurvey.org). Aggregate File Producer: Asep/ JDS, Madrid SPAIN. http://www.worldvaluessurvey.org/WVSDocumentationWV5.jsp
Author's data:
Associate Professor Hajdeja Iglič, PhD
Department of Sociology, Faculty of Social Sciences, University of Ljubljana
Kardeljeva ploščad 5, 1000 Ljubljana, Slovenia
Telephone: +386 1 5805 216
E-mail address: hajdeja.iglic@fdv.uni-lj.si