7 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 Original scientific article UDK 316.472.47(497 .4)"1987/2018" Hajdeja Iglič SLOVENIAN SOCIAL SUPPORT NETWORKS IN A LONG-TERM PERSPECTIVE, 1987–2018 ABSTRACT This article examines the effects of macro-level processes – economic recession, growth of the service economy, and deindustrialisation – on the social support networks of residents of Slovenia in the period between 1987 and 2018. Before the transition, Slovenia enjoyed highly functional social support networks. The eco- nomic downturns of the early 1990s and late 2000s increased the probability of insufficient social support. The growth of the service economy saw the importance grow of formal sources of support. Socio-demographic changes captured by the notion of the post-industrialisation of society brought various effects: the increased educational level and diminished size of households impacted the growing pres- ence of formal sources and mixed social ties in support networks, while the ageing population contributed to the increased presence of networks with insufficient social support. We conclude that Slovenian social support networks have experienced a distinct transformation over the last three decades. KEYWORDS: social support networks, Slovenia Slovenska oporna omrežja v dolgoročni perspektivi, 1987–2018 IZVLEČEK Članek obravnava vpliv makro družbenih procesov – ekonomske recesije, rasti storitvene ekonomije in deindustrializacije – na oporna omrežja prebivalcev Slovenije v obdobju od 1987 do 2018. Pred tranzicijo so imeli prebivalci Slovenije visokofunkcionalna oporna omrežja. Ekonomski recesiji v zgodnjih 1990-ih in poznih 2000-ih sta povečali tveganje nezadostne socialne opore. Rast storitvene ekonomije je pripeljala do povečanega pomena formalnih virov 8 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 Hajdeja Iglič opore. Družbene spremembe, ki spremljajo deindustrializacijo družbe, so imele različne učinke: povečevanje izobrazbene ravni prebivalstva in zmanjševanje velikosti gospodinjstev sta vplivala na povečano prisotnost formalnih virov in večjo raznolikost neformalnih virov v opornih omrežjih, medtem ko je staranje prebivalstva prispevalo k povečevanju deleža prebivalstva brez zadostne op- pore. Zaključujemo z ugotovitvijo, da so slovenska oporna omrežja v zadnjih treh desetletjih doživela večjo transformacijo. KLJUČNE BESEDE: oporna omrežja, Slovenija 1 Introduction The paper examines the effects of the macro-level processes on Slovenian social support networks. The study comprises the period from 1987 to 2018. This is the time marked in Slovenia by the political transition to democracy and economic transition to the market economy followed by the rapid development of the service sector and institutional changes in the provision of welfare towards the increased role of the market. In this same period the country went through two economic recessions, namely the economic shock that took place in the first half of the 1990s, immediately after the foundation of the new state, and the economic recession starting in the 2008 after the collapse of the Lehman Brothers bank in New York. The Slovenian society also experienced significant socio-demographic changes usually captured with the term post-industrial society that included ageing of the population, declining size of the households, and increasing level of education. The empirical analysis foregrounds the following research questions: What was the impact of the growth of the service economy, economic recessions, and deindustrialization of society on the availability of social support and restructuring of the sources of social support? The article thus aims to explore whether access to social support has decreased or increased due to the aforementioned social and economic processes, and whether they caused the re-orientation from one source of social support to another and to which. Socio-demographic change like the increasing level of education is expected to have transformed social support networks according to the middle-class so - ciability. The increased importance of the service economy presumably opened up access to market sources of support and in turn, diminished dependence on the informal sources of help offered by the networks of kin, friends, neighbours and other personal ties in which social support stem from the notion of solidar- ity, independent of the cash nexus or citizenship status. We believe these two changes combined have resulted in the increased use of formal sources and more 9 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 SLOVENIAN SOCIAL SUPPORT NETWORKS IN A LONG-TERM PERSPECTIVE, ... diversed support network, along with a reduced significance of kin ties. However, it is likely that this trend was slowed down by the instances of two economic downturns. In both cases, the aggrevation of macro economic situation (increased unemployment, lower incomes, increased poverty) is expected to have led to an increased share of those who had insufficient social support since economic crisis had double effect: it increased the need for more demanding forms of support and at the same time decreased the ability of usual support providers to offer such help. The insufficient social support has probably increased also as a result of the ageing of the population and declining size of the households. The long-term changes in the Slovenian social support networks are examined using data from the Stratification and Level of Living in Yugoslavia (Stratifikacija in kvaliteta živjenja) study in 1987 (Boh et al. 1987), and two waves of the Inter- national Social Survey Programme: Social Relations and Support Systems (ISSP 2001) and Social Networks and Social Resources (ISSP 2017). The surveys are representative of the Slovenian population. The social support network instrument used in all three surveys asked about to whom the respondents would turn in the case they needed to obtain various kinds of minor and major help. 2 Theoretical framework The theoretical chapter is divided into three sections. The first section briefly presents the concept of social support networks, emphasises their relative stabil - ity and highlights the most important differences between countries. The second section addresses the concept of middle-class sociability and its relations to so - cial support. The third section discusses how the deterioration of the individual’s material wellbeing due to the economic downturn affects the support networks. 2.1 Social support networks The term social support network denotes a web of personal relationships that are expected to or actually provide social support, with which they influence the individuals’ health and wellbeing, as well as their chances of finding a job and being ‘successful in life’ (among others: Litwak and Szelenyi 1969; Berkman and Syme 1979; House, Umberson and Landis 1988; van Vaux 1988; Welman and Wortley 1990; Thoits 1992; van der Poel 1993; Kogovšek et al. 2003; Alonso 2012). When offering social support, people put their resources (such as time, money, skills, information) at the disposal of others within a framework of ongo - ing social exchange. In this economy of personal gifts and favors, it is essential that people be connected to others who possess the required resources and are willing to share them. 10 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 Hajdeja Iglič While numerous studies have examined the adaptability of support networks in providing help during individuals’ personal crises and changed life situations, only rare studies have analyzed the change in support networks that emerges over a longer period and as the result of macro-level social processes. Thus, a significant part of the literature that we draw upon includes comparative studies conducted on the representative samples of countries and regions in which a small number of the selected forms of social support is observed. One of them is the 1987 ISSP survey carried out in seven countries, among which are also Slovenia’s neighbouring countries (the USA, GB, Denmark, Australia, Austria, Italy, and Hungary). The results of the analysis of this data set have been re - ported by Hollinger and Haller (1990). The authors find a high rate of similarity in support networks between developed countries despite the fact that they have different historical trajectories and political and socio-cultural traditions. The most important source of help for adults is their partner, followed by friends, mother, and adult children. The latter are typically mainly a source of instrumental rather than emotional assistance. Instrumental support is in the function of geographical distance; the larger the distance, the more the kin ties are replaced by non-kin ties and formal sources of support. Emotional support is more difficult to find than instrumental support. However, the lack of mention of the father as well as the role of partner as the source of social support reflect the degree of patriarchalism within the family relations and represent one of the dimensions of variability between the countries. Although in the Anglo-Saxon and Western European countries geographical distance to the relatives is larger than in Southern and Eastern European ones, this does not reduce the share of their emotional support, which even increases, mainly in relationships between the partners and between the father and his children. Countries also show differences in the presence of friendship and other non-kin ties. In some countries friends are substitute for kin, in others they supple - ment kin ties, while in still others they are almost non-existent, even when people lack access to social support within kinship network. Although the authors do not discuss this issue further, the inclusion of friends in support networks does seem to be related to the distinction between the instrumental and emotional support. Friends tend to be perceived as potential source of instrumental support in so - cieties where formal sources are absent, in this case they complement already strong and supportive kin network, while in societies where kin networks have weakened and formal sources of support are abundant they to a certain extent substitute kin ties, especially with respect to emotional support. The second comparative analysis, which uses data similar to our study, focuses on two Asian cities: Beijing and Hong Kong (Lee et al. 2005). The study offers 11 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 SLOVENIAN SOCIAL SUPPORT NETWORKS IN A LONG-TERM PERSPECTIVE, ... the opportunity to observe the differences in support networks within the same cultural tradition, but in contexts that differ in terms of the economic development and strength of the service economy. Here again, the authors find a high similarity of support networks, with differences only occurring in the presence of formal sources of help, especially with respect to financial support. Formal sources are used to a higher degree in Hong Kong than in Beijing. The temporal analysis (comparing Beijing in two time points) shows that with the decline of the role of the work place as provider of social support in the everyday life of Chinese, their social support networks acquired characteristics that made them similar to support networks in countries included in the abovementioned ISSP surveys. 2.2 Middle-class sociability In the second half of the 20 th century, social networks experienced significant changes. Sociological studies of personal networks conducted during the 1970s and early 1980s in the developed countries have shown that middle-class per- sonal communities are not ‘lost’ as would follow from the classical sociological works and analyses of early industrialisation, but are being transformed towards the “liberated” community. The expression comes from Wellman’s (1979; see also Drouhot 2016) naming of one of the three hypothetical forms of personal networks, which would typically have a sparsely knit, spatially dispersed and socially diverse network structure. In the same period, Fisher (1982) who ana - lysed the differences between personal networks in rural and urban communities concluded that urban personal networks allowed for the individual’s choice of their significant partners, based on the quality of the relationship, while the influ - ence of social roles in the structuring of personal networks has been decreasing. Comparing friendship networks in different social classes, Allan (1989) showed that middle-class friendship relations stand out in terms of decontextualisation and the inclusion of selected friends in the private sphere, where relationships are less constrained with normative expectations than in the public sphere. Finally, Paine (1969) argued that the main feature of middle-class sociability was that the boundaries of the relationships were defined contextually and discursively. There are several interpretations as to which characteristics of the middle-class social position have brought about the development of the new form of sociabil - ity that evolved around the notions of autonomy and choice. Some principally link the turn to the social ties outside the family and relatives with the crumbling away of these ties due to increased spatial mobility (Pahl and Prevalin 2005). This is a structural approach that connects new forms of sociability with the new opportunities for establishing and maintaining social contacts. The decreased opportunities for maintaining the familiarity of kin ties that comes from everyday 12 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 Hajdeja Iglič closeness increases the opportunity to achieve familiarity in non-kin, mainly friendship ties. A different interpretation claims that the change of sociability style occurs in the conditions in which interpersonal relationships are losing their productive function in the provision of economic well-being. Namely, strong normative element in interpersonal relationships which is traditionally found in kinship net - works helps maintain loyalty, predictability and stability (Silver 1990). This is a functionalist argument according to which the opening up of corporative family units to non-kin ties should be attributed to their decreased significance for the individual’s material well-being in the conditions of heightened social mobility and generous systems of social security. Regardless of the differences in the above arguments, they both claim that the transition from the ‘networks of loyalty’ to ‘networks of choice’ is linked to macro social changes taking place in the developed societies in the period of economic expansion after the Second World War. 2.3 Networks in the conditions of economic scarcity and insecurity During the past three decades, however, the discourse on personal networks began to be dominated by more pessimistic tones. The studies of the patterns of sociability and participation within the research programme on social capital, which were encouraged primarily by Putnam’s work (2000), call attention to the decreased informal sociability and civic participation and increased loneliness. Also, a renowned article (McPherson, Smith-Lovin and Brashears 2006) that compares the personal networks in the US between 1987 and 2004 finds that these have become smaller and more kin-oriented and that the number of people without a significant discussant partner has increased. Despite some alarming conclusions, empirical studies showed that these trends differ from country to country and that the change is less profound than first assumed (Clark 2015). According to earlier interpretations, the reasons for changes in networks were assumed to lie in new technologies, changes in the working participation and the replacement of civic-minded generations born between 1910 and 1940 with less civic-minded and more individualised generations. However, recent studies point to strong compositional effects (socio-demographic changes such as a change in age, education and household structures of the population) and period effects (declining real incomes and growing economic inequality) (Costa and Kahn 2003; Uslaner 2002; Clark 2015). The worsening of individu - als’ economic situation that falls under the period effect, entails their decreased resources for socialising, participation and networking (Iglic, Rozer and Volker 13 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 SLOVENIAN SOCIAL SUPPORT NETWORKS IN A LONG-TERM PERSPECTIVE, ... 2019). In the conditions of economic insecurity, individuals also have to redirect themselves from social to productive activities, which bring them extra income. And even if individuals keep maintaining extensive networks, these networks do not necessarily provide support, since they have been affected by the economic downturn as well (Letki and Mierina 2015). 3 Hypotheses We continue by presenting hypotheses about the effects of macro-social processes in Slovenia during the past three decades on the change of support networks. The hypotheses are formulated separately for three sets of factors: the development of the service economy, economic downturn, and socio-demo - graphic changes. 3.1 The development of the service economy in Slovenia Support networks perform the function of the provision of goods and services, which the individual either cannot or does not wish to obtain in other ways. Therefore, with the growth and improvement of the quality of the service sector, the role of the so-called formal sources in Slovenian support networks is expected to have increased. The service sector has experienced intensive growth since the late 1980s. In 1971, the share of those employed in the tertiary sector totalled 31 %, which increased to 51 % in 1991 and reached 65 % in 2005 (Drozg 2007). By the early 2000s, the share of services in the GDP exceeded 50 %. Particularly important for the support networks is the development of financial services. The liberalisation of the financial market, the development of new financial products, and easier access to them coupled with increased needs for financial as- sistance resulted in the higher indebtedness of the Slovenian population in the first decade following the transition. The burden of the population as the ratio between the size of bank loans and the income of the population increased between 1992 and 2000 from 0,39 to 3,50 (Ferk 2007). Besides the financial sector, growth also occurred in other services, such as in household help in the case of illness, help in the garden, help in the case of emotional problems and in the need for counselling. Most of these services can be bought on the market, while some are also accessible in the public sector or are at least partly co-funded from public resources. The growth of the service economy throughout the post-transition period sug - gests the following hypothesis: the period after 1987 saw the growing significance of formal sources of help, in particular with regard to the need for financial resources. Accordingly, the significance of informal sources in the provision of welfare was declining. (H1) 14 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 Hajdeja Iglič 3.2 Economic downturns The first survey examining support networks was carried out in 1987 when Slovenia began a six-year period of negative economic growth. Throughout the 1980s, real wages declined compared to the decade before (Kordež 2018), and inflation reached 181,1 % (SURS). By 1989, it had increased to 2.762,7 % (SURS). Slovenia’s independence and the interruption of the economic flows with the rest of the former Yugoslav republics resulted in unemployment that until then had been at an insignificant level. Between 1987 and 1993, the number of unemployed increased from 17 ,826 to 137 ,142 (ZRSZ). In 1993, for the first time since the transition had begun, Slovenia again recorded positive economic growth, and the number of the unemployed started to decline. However, it re - mained above 100,000 until 2001 (ZRSZ). Although Slovenia was among the transition countries that in the early 1990s were among the first to have overcome the transformation depression (Lorenčič 2012), it was only later, in the early 2000s, that the real income reached the levels from the mid-1980s and inflation dropped below 5 %. The great global crisis of the late 2000s had similar dimensions. The number of unemployed rose until 2013, amounting to 124.015. It is expected to fall to the pre-crisis level only by 2021 (ZRSZ). The wage growth that slowed down during the crisis completely stopped in 2012 and 2013. With the exception of the minimum wage, which increased by 20 % in 2010, wages started to grow again only in 2017 (SURS). Both cases of the economic downturn that marked Slovenia in the first three decades after the transition saw a large increase in the risk of unemployment and the decline of household income. They are expected to be reflected in support networks in terms of the increased share of individuals who have no one to turn to for help. (H2) This hypothesis builds on the idea that economic crisis leads to the contraction of networks since it reduces the resources for socialization and participation. Also, in economic crisis the need for social support increases but since it affects all partners in the network the overall capacity of network to of - fer support decreases. We expect that in the economic downturn the informal support networks contract with respect to both kin and non-kin ties. (H3) 3.3. Socio-demographic changes Support networks vary with the size of household, education, age and gen - der. In the observed period from 1987 to 2018, the size of the household in Slovenia was diminishing, its average size being 3,06 in 1991 and 2,48 in 201 1 (SURS). The main increase was observed in the number of one- and two-member 15 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 SLOVENIAN SOCIAL SUPPORT NETWORKS IN A LONG-TERM PERSPECTIVE, ... households in urban settlements. From the aspect of support networks, the pres- ence of informal support is expected to decrease and the presence of formal help increase with the diminishing household size. (H4) In addition, decreasing size of the households is conceived as a structural cause for an increase in insuf- ficient social support (H5) as well as the mobilization of non-kin social ties and consequently more diverse networks (H6). The Slovenian population is ageing. In 1991, the average age was 35,9 years, and persons older than 65 years represented 1 1 % of the entire population (SURS). By 201 1, the average age was 43,2 years, and the share of those older than 65 amounted to 19 % of the entire population (SURS). With the elderly’s support networks in general offering less support ( Hlebec 2003; Kavčič 201 1), the share of those who have no one to turn to for help is expected to rise. (H7) During the past three decades, the education of the population has changed significantly. In 1991, 47 ,2 % of the population had completed elementary school or less with this share dropping to 30 % by 201 1 (SURS). In the period from 1991 to 2001, the share of the population that had completed high school increased from 43,1 % to 54 % (SURS). Throughout this period, the share of the population with tertiary education was also growing, increasing from 8,9 % in 1991 to 17 ,5 % in 201 1 (SURS). Education is the most frequently used indicator of the changes in the style of sociability. With the increasing educational level, the share of ‘liberated’ personal communities is expected to increase, which means that the networks had become more diverse (H8). 4 Data The analysis is based on three surveys that use the same approach to the study of personal support networks. Namely, the respondents were asked whom they would turn to for help in different situations of need or emergency. For each type of need or emergency, they list one or two sources of social support that they identify in terms of role relationship (for example, husband/wife/partner, mother, father, etc.). Although they use the same general approach, the surveys differ with respect to a number of different situations for which they ask about the potential source of social support. In the three surveys from the mid 1980s (in the USA in the GSS 1985, in Slovenia in the YULOL 1987 , and in Australia, the UK, Germany, Austria, Italy, Hungary in the ISSP 1986) the respondents were asked about the sources of social support in six different situations: (1) help in the household and garden; (2) help with shopping and around the home when having flu and hav - ing to stay in bed; (3) help when needing to borrow a large sum of money; (4) 16 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 Hajdeja Iglič advice when feeling upset and having a problem with one’s husband, wife or partner; (5) advice when feeling a bit down or depressed; (6) advice regarding important changes in life, for example about the job or moving to another city. For each situation, they were asked to elicit two sources of support (so-called 1 st and 2 nd choice). The list of role relationships was long and very specific. It included: “Husband/wife/partner”, “Mother”, “Father”, “Daughter”, “Son”, “Sister”, “Brother”, “Other relative including in-laws”, “Closest friend”, “Other friend”, “Neighbour”, “Co-worker”, different formal sources of support, “Other”, and “No one”. The second wave of the ISSS module on social support networks, titled Social Relations and support systems, was conducted in 2001. The number of situations in which the respondents turn to types of people for help was reduced to three: (1) help in case of illness; (2) advice when depressed or a bit down; and (3) borrowing a large sum of money. Respondents were again asked to whom they would turn first and second for help, and the list of available role relationships was longer than in the mid-1980s. The third wave of ISSP on social support networks, Social network and social resources, was conducted recently. In Slovenia, the data were collected in 2018. The questions about potential sources of help refer to a long list of situations. The original questionnaire from 1986–1987 was complemented with additional questions asking about persons with whom people associate in their free time and those who offer them help with administrative problems or official paperwork, or look after them if they were seriously ill. The response categories were changed significantly again. The described changes in the measurement instrument present a serious chal - lenge for the analysis, which aims to focus on the changes over time. In order to cope with this problem, we compare all three time points simultaneously by observing the distribution of the types of social support networks. The network types are defined on the basis of the small number of response categories and for the small number of helping situations. 5 Analysis of trends 5.1 Sources of social support in 1987 Before we turn to the analysis of trends we offer a short presentation of the Slovenian social support networks before the transition, in 1987 . Table 1 shows the share of respondents who would turn to different social ties if they needed a certain kind of help. The last column shows the average share of respondents turning to different social ties for six kinds of help together. Different social ties 17 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 SLOVENIAN SOCIAL SUPPORT NETWORKS IN A LONG-TERM PERSPECTIVE, ... were divided into three categories: “partner”, “kin”, and “non-kin”, with the ad - dition of “formal” sources of help as well as the possibility to answer ‘no one’. Only the first choice was taken into account. Table 1: Sources of social support, 1987: by different kinds of social support and overall (percentages). House- work Illness Family pro- blems Depres- sion Borrow money Advice Overall Partner 50,1 47,2 14,1 45,8 16,1 54,3 40,1 Kin 39,0 44,8 47,3 22,1 60,1 30,9 39,9 Non-kin 9,9 5,6 23,1 29,3 12,5 0,1 14,1 Formal support 0,5 1,1 3,4 0,9 8,1 1,5 2,2 No one 0,7 1,2 12,1 1,9 3,4 4,6 3,7 N 285 284 281 282 281 283 281 Slovenian support networks in the period just before the transition show a significant role of partner and kin ties, which presented the source of support in 80 % of the situations. They are followed by non-kin ties, which are important especially with respect to help in case of emotional problems (family problems and depression) although they also offer financial support. Relying on sources is extremely low and the share of those who would not be able to turn to any source for one or another kind of difficulty is also observed to be low. The family and kinship networks with some help of friendship networks seemed to successfully perform the function of social support. 5.2 Changes in the 1987–2001 period Table 2 shows changes in Slovenian social support networks in the first ob - served period, 1987–2001. As opposed to the previous table, only three kinds of help are considered here – help in case of illness, depression, and financial problems which is the reason why figures for overall support in 1987 in Table 2 differ from Table 1. The results are presented for the 1 st choice only, and for the 1 st and 2 nd choice together. Individual cells show percentages that again tell how often a certain source of help was chosen among all the choices. Descriptive analysis shows an increased percentage of partnership ties and formal sources in 2001 compared to 1987 . Other informal ties, both kin and non-kin, have lost their significance. The drop is especially strong when considering the role of kin ties among the 1 st choice and non-kin ties among the 1 st and 2 nd choice to- gether. The share of those who have nobody to turn to for help was significantly increased. 18 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 Hajdeja Iglič Table 2: Sources of social support, Slovenia 1987–2001: overall (column percentages). 1. choice 1. & 2. choice 1987 2001 1987 2001 Partner 35,9 41,2 ** 20,0 23,6 *** Kin 43,7 34,7 *** 49,9 46,8 ** Non-kin 15,4 12,3 ** 21,6 14,5 *** Formal 3,1 7,1 *** 3,6 7,5 *** No one 1,9 4,5 *** 5,0 7,6 *** N 281 1077 281 1077 * p < 0,05 ** p < 0,01 *** p <0,001 The results suggest that the first period saw the contraction of informal support networks to close family members, whose help was complemented with newly emerging formal sources. These changes could be interpreted in terms of the rise of the service economy and they confirm Hypothesis 1. The increased share of the population without enough support also support Hypothesis 2 about the impact of the economic downturn. The support networks also exhibit contraction with respect to kin and non-kin ties, supporting Hypothesis 3. Obviously, the formal sources failed to replace the reduced role of informal ties which means that there were also other reasons for the contraction of the informal support networks than just their replacement with formal sources. We claim these reasons were related to the deteriorating economic situation of the population. 5.3 Changes in the 2001–2018 period This section presents the analysis of support networks in the period 2001– 2018. Due to the changes in the measuring instrument, adaptation in the analysis is needed again. This time the results are reported together for help in the cases of illness and depression, and separately for help in the case of financial short - age. The lack of information is marked as “n.a.” (information not available) in Table 3. 19 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 SLOVENIAN SOCIAL SUPPORT NETWORKS IN A LONG-TERM PERSPECTIVE, ... Table 3: Sources of social support, Slovenia 2001–2018: illness & depression, and borrow money (column percentages). Illness Depression Borrow money 2001 2018 2001 2018 Kin 83,2 79,5 ** n.a. n.a. Non-kin 13,7 18,6 *** n.a. n.a. No one 3,1 1,9 * 8,5 7,6 Formal n.a. n.a. 19,2 28,2 *** Personal n.a. n.a. 72,2 63,9 *** N 1077 1047 1077 1047 * p < 0,05 ** p < 0,01 *** p <0,001 Situations of illness and depression show the decreasing presence of kin ties and the increased presence of non-kin ties. The share of situations in which individuals have nobody to turn to for help was slightly diminished and is on the border of statistical significance. For these two situations of help, we do not have any information regarding formal sources. However, there was a high increase in the percentage of financial help that respondents would ask for from formal sources. As many as 10 % more respondents than in the previous period would turn to formal sources, such as banks and savings institutions if they needed fi - nancial help. Accordingly, this would lower the percentage of those who would use informal sources. The percentage of those who, in this case, would not find help remains high and unchanged. The 2001–2018 period can be seen as the continuation of the process of post-industrialisation of the economy, mainly with regard to financial needs, sup - porting Hypothesis 1. At the same time, the share of the situations of insufficient or absent support is not diminishing, which probably reflects the economic crisis of the 2008 as predicted by Hypothesis 2. However, the share of non-kin ties is on the rise in the case of small instrumental (illness) and emotional (depression) support, while kin ties decline, indicating the transition to the pattern of middle- class sociability. Of course, despite a decline, kinship ties still remain prevalent. But due to the nature of the 2018 measurement instrument, kin ties cannot be separated from partnership ties. So we do not know if the trend of the growing importance of partnership ties increased also into this period, and how strong was really a decline in the kin ties different from the partnership ties. The Hypothesis 3 thus does not hold for the period from 2001 on. 20 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 Hajdeja Iglič 5.4 Trends in social support networks and socio-demographic variables This section presents the analysis of both periods simultaneously, adding a larger number of respondents’ socio-demographic characteristics. To this purpose, we again changed the analytical approach and first delineated different types of social support networks on the basis of very broadly defined categories of the sources of social support. The following types of support networks were formed with the help of the classification analysis (we used the two-step cluster procedure in the SPPS): “kin- centred”, “mixed”, “formal” and “insufficient” (Table 4). As its name suggests, a “kin- centred” network mainly includes kinship ties (66,5 %) as a support source in case of illness and depression, financial support is expected from personal ties and social support is available in all cases. A “mixed” type network comprises an approximately equal share of different non-kin and kin ties (29,7 % and 36,1 %, respectively), financial support is again expected from informal sources and social support is in general widely available. A “formal” network type indicates a high presence of formal sources in case of financial needs, otherwise it is kin- centered. Finally, the “insufficient” type reveals a high absence of both formal and informal sources of help. In all types of support networks kinship ties are predominant source of support. Table 4: Prevalence of network types by wave (row percentages) and their characteristics (column percentages). Type 1 Kin-centered support Type 2 Formal support Type 3 Insufficient support Type 4 Mixed support N Wave Wave 1987 60,4 7,4 5,5 26,7 281 Wave 2001 51,9 17,5 11, 3 19,3 1061 Wave 2018 43,2 27,0 7,8 22,0 1034 Network characteristics Kin 66,5 53,9 40,7 29,7 Non-kin 0,0 12,2 8,9 36,1 Personal 33,5 0,0 7,1 33,4 Formal 0,0 33,8 3,3 0,0 No one 0,0 0,0 38,7 0,0 N 117 9 489 243 465 % 51,7 17,5 8,4 22,5 21 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 SLOVENIAN SOCIAL SUPPORT NETWORKS IN A LONG-TERM PERSPECTIVE, ... Regardless of the year of the survey, the most common type is a “kin-centred” support network, with, however, its presence tending to diminish. Between 1987 and 2018, the presence of this type of network fell from 60,4 % to 51,9 % and further to 43,2 %. At the same time, the presence of a “formal” type of support network dramatically increased. Its presence first grew from 7 ,4 % to 17 ,5 %, and further on to 27 % in the following period. In the first transition period, the presence of a “mixed” network type fell from 26,7 % to 19,3 %, and was slightly increased in the second period, but did not reach the 1987 level. The presence of an “insufficient” network type mainly grew in the first period: from 5,5 % to 1 1,3 %; while declining in the next period, it still remained higher than in 1987 . Table 5 shows the results of multinomial logistic regression of four network types on socio-demographic variables and different time points (or waves of study). This analytical approach was used because the answers are mutually co-dependent (the respondents could only select one social role for each type of help; therefore, they only had one choice). The results are expressed in Exp (B), which means that the value lower than 1 indicates lower, and the value higher than 1 higher probability of the respondent having a certain type of support network compared to a “kin-centred” network, which was defined as the base category. 22 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 Hajdeja Iglič Table 5: Multinomial regression of network types on socio-demographic characteristics and wave. Model 1 Wave Model 2 Sex Model 3 House- hold Model 4 Educa- tion Model 5 Age Model 6 Full model Exp (B) Exp (B) Exp (B) Exp (B) Exp (B) Exp (B) Type 2 (Formal Wave 1 (1987) Wave 2 (2001) base 2,751*** base 3,1 14*** support) Wave 3 (2018) 5,087*** 5,604*** Female base base Male 1,321** 1,203 + Household size 0,775** 0,790** Edu low/middle 0,749* 1,706 Edu higher base base Edu interactions: low/middle * wave 2 / 0,665 higher * wave 2 base low/middle * wave 3 0,685 higher * wave 3 base Age 15-24 0,366*** 0,395*** Age 24-64 base base Age over 65 0,745 0,528*** Type 3 Wave 1 base base (Insufficient Wave 2 2,371*** 2,684*** support) Wave 3 1,960** 1,698** Female base base Male 1,073 1,237 Household size 0,469*** 0,555*** Edu low/middle 1,347 0,585 Edu higher base base Edu interactions: / low/middle * wave 2 3,801** higher * wave 2 base low/middle * wave 3 2,149* higher * wave 3 base Age 15-24 0,457** 0,530* Age 24-64 base base Age over 65 3,010*** 1,920** 23 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 SLOVENIAN SOCIAL SUPPORT NETWORKS IN A LONG-TERM PERSPECTIVE, ... Type 4 Wave 1 base base (Mixed Wave 2 0,838 1,253 support) Wave 3 1,150 1,013 Female base base Male 0,785** , 0,809* Household size 0,719** 0,583*** Edu low/middle 0,973 1,290 Edu higher base base Edu interactions: / low/middle * wave 2 0,901 higher * wave 2 base low/middle * wave 3 0,526* higher * wave3 base Age 15-24 2,232*** 2,607*** Age 24-64 base base Age over 65 0,770 0,514*** R 2 (Nagel- kerke) 0,061 0,007 0,025 0,003 0,087 0,171 χ 2 164,525 19,137 64,562 8,888 236,473 472,932 d.f. 6 3 3 3 6 27 N 2376 2369 2361 2352 2349 2315 Note: Reference category is network Type 1 (“kin-centered” support network) * p < 0,05 ** p < 0,01 *** p <0,001 + borderline significance The first column (Model 1) in the Table 5 shows the effects of three time points. The next four columns (Model 2 to Model 5) show bivariate effects of different socio-demographic variables (sex, household size, education and age), and the last column (Model 6) presents the multivariate effects of all variables simultane - ously. Also, interaction effects between education and wave are included. Multivariate analysis corroborates that the probability of respondents having a ‘formal’ type of support network is growing from one period to the other and that this trend is relatively independent of the demographic changes, supporting again the Hypothesis 1. With respect to socio-demographics, the probability of the use of “formal” support network compared to “kin-centred” network is indicated to be higher among men, respondents with higher education, those living in smaller households and middle-aged people. We thus find support for Hypothesis 4 that smaller households tend to rely more on formal sources of sup - port than bigger households. The increase in the presence of smaller households in the Slovenian society should be regarded as one of the reasons for larger 24 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 Hajdeja Iglič reliance on formal sources. Education turn out insignificant in the multivariate analysis, and the ageing of the population works against the transition towards more “formal” support networks. As was established in previous analyses, the probability for “insufficient” social support increased after 1987 . It is slightly lower in the 2001–2018 period, but still statistically significantly higher than in 1987 . These results are explained in terms of economic downturns that characterised both periods supporting Hypothesis 2. Socio-demographic changes also had a significant impact on the rise of “insufficient” type of networks. Again, size of household is important as smaller households are more likely than bigger ones to have “insufficient” compared to “kin-centered” support networks confirming Hypothesis 5. Ageing of the population brings about more of ‘insufficient’ type of support networks as proposed by Hypothesis 7 . Education affects the probability of the respondent having “insufficient” support network only in the interaction with the wave: in the time of economic downturns lower and middle educational categories are significantly more likely to have “insufficient” support than higher educational category confirming that the increase in insufficient network type is related to the stress posed by the economic recession on networks of people with less economic resources. The results for the impact of time on “mixed” type of support networks are statistically insignificant. The economic processes and the development of the service economy did not affect the structure of support networks in terms of kin and non-kin ties as much as they did affect the emergence of formal sources of support and the increase in insufficient support. The dynamics of the changes in the likelihood of ‘mixed’ network type depends solely on socio-demographic factors. They are more probable among men than women and among those liv - ing in smaller households confirming Hypothesis 6 that the increasing distance between kin ties presents the opportunity for non-kin ties to take over the function of providing social support. The increasing education, especially in the second post-transition period (after 2001) which saw an increase of graduates at the tertiary level working in the service sector, brought about more mixed “network type”, which can be explained in terms of middle-class sociability and confirms Hypothesis 8. But as ageing of the population contributes negatively to the likelihood of “mixed” type of support networks, the overall picture of Slovenian networks in terms of the kin/non-kin ties structure is not significantly different from 1987. 25 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 SLOVENIAN SOCIAL SUPPORT NETWORKS IN A LONG-TERM PERSPECTIVE, ... 6 Conclusion This article examines social support networks of the residents of Slovenia in the period between 1987 and 2018. Although network microstructures usually exhibit high levels of stability, they nevertheless change in response to the large- scale processes. In the article we focused on the impact of processes such as the growth of the service economy, economic downturn and socio-demographic changes. The analysis started with the analysis of the Slovenian support networks in 1987 . The results show that the former were relatively diverse and functional despite the insufficiently developed formal sources of support. The percentage of respondents with insufficient support was small. In the two post-transition periods, 1987–2001 and 2001–2018, the kin- centered character of support networks diminished and networks became more diverse in terms of sources of social support. The role of formal sources increased, but due to the economic downturns the needs for social support increased more than the access to formal sources, resulting in the increased proportion of the population with “insufficient” type of networks. The economic recession in the first post-transition period affected in particular those with lower and middle levels of education who experienced a contraction in support networks with respect to both kin and non-kin social ties. In the second post-transition period we see a sign of the spread of the mid - dle-class sociability as a result of the larger number of graduates with tertiary education whose jobs are more dependant on the service economy than in the previous periods, and the changing size of the households which are becoming smaller. Both trends contributed to the “mixed” type of support networks which are more diverse, with kin ties co-existing with non-kin ties. However, they have been countered with the ageing of the population. The picture of the Slovenian support networks that derives from the analysis is that after the transition sources of support have become more diverse. In par- ticular, people rely more on formal sources. But support is at the same also more insufficient than before. This is partly due to the economic distress experienced by the families and households during the two recessions, and partly due to socio-demographic changes, especially ageing of the population and smaller size of the households. Although we can expect that the impact of both – smaller household size and older population – will be in the future partly ameliorated by the increasing education, it nevertheless represents a challenge for the social policy. 26 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 Hajdeja Iglič References Allan, Graham (1989): Friendship: Developing a sociological perspective. London: Harvester/Wheatsheaf. Alonso, Mercedes F. (2012): Social support networks in Spain: The factors that determine models of choice. International Sociology, 27 (3): 384–402. Berkman, Lisa F., and Syme, Leonard S. (1979): Social networks, host resistance, and mortality: a nine-year follow-up study of Alameda County residents. American Journal of Epidemiology, 109 (2): 1 186–2014. Clark, April K. (2015): Rethinking the decline in social capital. American Politics Rese - arch, 43 (4): 569-601. Costa, Dora L. and Kahn, Matthew E. (2003): Undestanding the American decline in social capital, 1952-1998. Kyklos, 56 (1): 17–46. Drouhot, Lucas G. (2006): Reconsidering ‘community liberated’: How class and the natio - nal context shape personal support networks. Social Networks, 48 (January): 57–77. Drozg, Vladimir (2007): Prispevek h gospodarski geografiji Slovenije. Revija za geo - grafijo, 2 (1): 67–88. Ferk, Barbara (2007): Zadolževanje in prezadolženost prebivalstva. UMAR: Delovni zvezki, 16. Kraj: Založba. Fisher, Claue S. (1982): To dwell among friends. Chicago: University of Chicago Press. Hlebec, Valentina (2003): Omrežja starostnikov v Sloveniji. Družboslovne razprave, 43 (19): 171–182. Hollinger, Franz, and Haller, Max (1990): Kinship and social networks in modern so - cieties: a cross-cultural comparison among seven nations. European Sociological Review, 6 (2): 103–123. House, James S., Umberson, Debra, and Landis, Karl R. (1988): Structures and processes of social support. Annual Review of Sociology, 14: 293–318. Iglič, Hajdeja, Rözer, Jesper, and Volker, Beate (2019): Economic crisis and social capital: The role of politics in understanding short-term changes in social capital. Unpublished manuscript. Kavčič, Matic (201 1): Družbena in individualna tveganja starejših - vloga omrežij socialne opore. Ljubljana: Fakulteta za družbene vede, UL: Doctoral dissertation. Kogovšek, Tina, Hlebec, Valentina, Dremelj, Polona, and Ferligoj, Anuška (2003): Omrežja socialne opore Ljubljančanov. Družboslovne razprave, XIX (43): 183–204. Kordež, Bine (2018): Od česa je odvisna višina naše pokojnine. Available from: https:// damijan.org/2018/09/22/od-cesa-je-odvisna-visina-nase-pokojnine/ . (Accessed 30.09.2019). Lee, Rance P.L., Ruan, Danching, and Lai, Gina (2005): Social structure and support networks in Beijing and Hong Kong. Social Networks, 27 (3): 249–274. Letki, Natalia and Mierina, Inta (2015): Getting support in polarized societies: Income, social networks, and socio-economic context. Social Science Research, 49 (Janu - ary): 217–233. 27 DRUŽBOSLOVNE RAZPRAVE, XXXV (2019), 91: 7 - 27 SLOVENIAN SOCIAL SUPPORT NETWORKS IN A LONG-TERM PERSPECTIVE, ... Litwak, Eugene, and Szelenyi, Ivan (1969): Primary Group Structures and Their Functi- ons: Kin, Neighbors, and Friends. American Sociological Review, 34 (4): 465–481. Lorenčič, Aleksander (2012): Prelom s starim in začetek novega. Tranzicija slovenskega gospodarstva iz socializma v kapitalizem (1990-2004). Ljubljana: Inštitut za novejšo zgodovino. McPherson, Miller, Smith-Lovin, Lynn, and Brashears, Matthew E. (2006): Social isola- tion in America: Changes in core discussion networks over two decades. American Sociological Review, 71 (3): 353–375. Pahl, Ray, and Prevalin, David J. (2005): Between family and friends: a longitundinal study of friendship choices. The British Journal of Sociology, 56 (1): 433–450. Paine, Robert (1969). In search of friendship: An exploratory analysis in ‘middle-class’ culture. Man, 4 (4): 505–524. Putnam, Robert (2000): Bowling alone: The collapse and revival of American community. New York: Simon & Schuster. Silver, Allan (1990): Friendship in Commercial Society: Eighteenth-Century Social The- ory and Modern Sociology. American Journal of Sociology, 95 (6): 1474–1504. SURS. Statistični urad Republike Slovenije (Statistical Office of the Republic of Slovenia). Available from: https://www.stat.si/StatWeb/en (Accessed 30.09.2019). Thoits, Peggy A. (1992): Stress, Coping, and Social Support Processes: Where Are We? What Next? Journal of Health and Social Behavior, 35 (Extra issue): 53–79. Uslaner, Eric M. (2002): The moral foundations of trust. Cambridge: Cambridge Uni - versity Press. van der Poel, Matt G.M. (1993): Delineating personal support networks. Social Net - works, 15 (1): 49–70. van Vaux, Alan (1988): Social support: theory, research, and intervention. New York: Praeger Publishers. Wellman, Barry (1979): The community question. American Journal of Sociology, 84 (5): 1201–1231. Welman, Barry, and Wortley, Scott (1990): Different strokes from different folks: Com - munity ties and social support. American Journal of Sociology, 96 (3): 6558–6588. ZRSZ. Zavod Republike Slovenije za zaposlovanje (Employment Office of the Republic of Slovenia). Available from: http://english.ess.gov.si (Accessed 30.09.2019). Authors’ data: Assoc. Prof. Dr. Hajdeja Iglič Faculty of Social Sciences, UL e-mail: hajdeja.iglic@fdv.uni-lj.si tel.: 051 3161 1 1