‘I am not alone’: Understanding public support for the welfare state

Abstract

Spanish

French

This article explores to what extent and how individuals’ welfare state attitudes relate to their subjective assessment of the available social support. Using various sociological and sociopsychological theories the authors first provide a theoretical analysis of the micro–macro links between perceived social support (micro), social trust in support availability (macro) and public attitudes towards welfare states (micro). An empirical test based on a large cross-country dataset of 31,122 respondents in 25 European countries shows that the more welfare is provided by the state, the less of it is desired in countries where individuals have the general belief that they can rely on each other for support. Importantly, only when considered jointly, do welfare state provision and social trust in support availability become essential in explaining welfare state attitudes.

Keywords

‘Deservingness’European comparison social support social trust welfare state attitudes

Introduction

The foundation of the modern welfare state, which has its origins in 19th-century Europe, lies in the responsibility taken by the state to provide economic equity, social protection and solidarity (Gangl, 2007). Welfare states offer services and benefits to meet people’s basic needs for education, health, income and housing. While formal institutions assembled by the state are a necessary condition for the proper functioning of a welfare state, the informal or public support of these institutions is also necessary to make the implementation of a welfare state sustainable. The literature shows that public attitudes towards welfare state policies are increasingly taken into account in the formal design of those policies (e.g. Crespi, 1997; Lewin-Epstein et al., 2003). Yet, much is still unknown about the determinants of this support.

In forming their opinion about the welfare state individuals are embedded in an institutional and social context. They are repeatedly involved in an exchange relationship with the welfare state and its institutions, but also with other individuals (e.g. friends, family, colleagues, neighbours) and the society as a whole. Social relations and social support are known to facilitate the provision of basic services and benefits (Cohen et al., 2000). Social relations leading to social capital through interactions between individuals (Lin, 2001) have been a frequent focus of studies looking at the relationship between social capital and the welfare state (e.g. Kääriäinen and Lehtonen, 2006; Rothstein, 2001; Scheepers et al., 2003; Van Oorschot et al., 2006). This relationship can be either positive or negative, known as crowding-in and crowding-out effects.

Individuals’ support for the welfare state may also depend on the extent to which they feel that they can rely on others in times of need. While there are studies analysing social capital and its relationship with welfare state provisions, the impact social capital has on public attitudes has not yet been researched. Moreover, though a number of these studies have taken ‘social support’ into account, they do so only indirectly, that is by considering it as one of the aspects of social capital (Kääriäinen and Lehtonen, 2006; Pichler and Wallace, 2007). The concept used refers to enacted social support, that is the extent to which people actually give support. This aspect of social support does not, however, capture the variety of ways in which available support is perceived in different societies. For any given level of objective support, some may perceive high level of support availability within their society, while others may not.

Individuals’ subjective assessment of the available social support has been overlooked in the literature on welfare state support. We aim at filling this gap by introducing elements of various theories to the analysis of public opinion towards the welfare state. Social support theory emphasizes the protective and buffering features of social support in the case of undesirable physical, mental and social situations. Within this theory, not only the support itself, but mainly the perception that it is available plays an essential role. At the individual level, such a perception increases individuals’ optimism and confidence in dealing successfully with various situations they may be confronted with (Barrera, 1986; Cohen et al., 2000). Sociopsychological theories argue that, in turn, this leads to social trust, societal integration and social cohesion at the societal level (Delhey and Newton, 2003; Lin and Dean, 1984; Uslaner, 2000). This societal cohesion is experienced either through a mutual sense of belonging and shared identity or through shared utility and obligations to cope with various situations; which will, according to sociological theories on social solidarity and deservingness criteria, lead to more solidarity and contribution to collective goods (see Van Oorschot, 2000b for an extensive discussion). In short, individuals living in societies characterized by a generalized belief of available support experience more social trust and social cohesion, and may therefore have different opinions about the welfare state than individuals living in societies where they do not expect social support from others when in need.

This article combines sociopsychological theories with the deservingness criteria perspective to specify how the macro level of perceived social support relates to the micro level of welfare state attitudes.¹ In doing so we explore an important missing link in the welfare state literature, which is the role in shaping public attitudes towards welfare states of social trust in the supportiveness of others if needed. Moreover, we investigate how the interaction between, on the one hand, this ‘social trust in support availability’ and, on the other, welfare state provision relates to individuals’ attitudes towards the welfare state. By looking at this important but under-researched aspect of social capital (i.e. social trust in support availability), this study complements the current research focusing on the causal relationship between the welfare state and social capital. In addition, it takes this research a step further by also investigating how this relationship correlates with public attitudes. An important question we answer is whether the extent of welfare state provisions strengthens or weakens the relationship between social trust in support availability and public opinions on welfare.

Public attitudes towards the welfare state: A brief overview

‘The welfare arrangements and institutions that serve the collective interest of modern societies have a stronger legitimacy to the degree that more people are motivated to contribute to the arrangements and people have more reasons to contribute’ (Van Oorschot, 2000a: 18). Various theoretical perspectives contribute to understanding welfare states attitudes. The first perspective is class related and explains public support by individuals’ personal interests (Galbraith, 1992; Halla et al., 2010; Heinemann, 2008; Wilensky, 1975). The second perspective is based on two important social ideologies: economic individualism and social equality (Blekesaune and Quadagno, 2003; Hasenfeld and Rafferty, 1989; Lewin-Epstein et al., 2003; Lipsmeyer and Nordstrom, 2003; Mau, 2004; Shapiro and Young, 1989). The third perspective emphasizes the role of cultural factors like morals and values (e.g. Kluegel and Smith, 1986; Lockhart, 2001; Van Oorschot, 2000a). The fourth perspective focuses on welfare systems as institutions that not only affect individuals’ personal interests, but also have a norm-shaping function (Andress and Heien, 2001; Arts and Gelissen, 2001; Bean and Papadakis, 1993; Blomberg and Kroll, 1999; Esping-Andersen, 1990; Jæger, 2006; Larsen, 2008; Rothstein, 1998; Svallfors, 1997).

In this article, we focus on public attitudes towards a welfare state by considering public views on whether the responsibility of providing for everyone should lay on the individuals themselves or on the government. In doing so we incorporate Rothstein’s (1998) view that debates on welfare policy should be about the level of general fairness regarding the relation between citizens and the state. In other words, do we want to solve our common issues individually as citizens or do we want the state to take over and provide for them? A more detailed discussion is provided later.

Sociopsychological theories on social support, social trust and deservingness

Social support theory distinguishes between three broad categories of related concepts: social embeddedness, enacted support and perceived social support (e.g. Barrera, 1986; Hagan et al., 1996). While social embeddedness refers to the actual connections between individuals and enacted support to actual support provided, perceived social support refers to generalized beliefs about the supportiveness of others (Cohen et al., 2000). Empirically it has been shown that the perception of available support has a ‘much stronger influence on outcomes than the actual receipt of social support’ (Lincoln, 2007: 2). The rationale for this finding is that the beneficial outcomes on success and well-being are primarily driven by a general positivity within social relations rather than by specific aspects of relationships (Stansfeld, 2002).

Perceived social support is derived from subjective information (based on interactions) that leads an individual to believe (1) that she/he is cared for and loved (emotional support); (2) that she/he is esteemed and valued (esteem support); and (3) that she/he belongs to a network of mutual obligations (network support) (e.g. Lin and Dean, 1984). As indicated in link [1] Figure 1, there are two ways through which this information may be beneficial. One that is more individual-related predicts that one’s beliefs in the availability of support affect appraisals of stressful and difficult situations which in turn buffer the effect of such situations on one’s well-being and success. The other mechanism is through self-esteem and social cognition. The more one believes that one can rely on the supportiveness of others, the more one will observe supportive behaviour and be supportive oneself, in turn reinforcing others’ perceptions of social support. The more these individuals believe that others are to be trusted in making their support available if needed, the more they will be motivated to be supportive themselves, thus reinforcing the virtual circle of trust. This process creates a ‘generalized’ belief about others’ support (e.g. Cohen et al., 2000; Pierce et al., 1997), which will further increase self-esteem and optimism, and thereby will positively affect an individual’s well-being and success.

Figure 1.

A theoretical conceptualization of social trust in support availability, welfare state provision and public attitudes towards the welfare state.

Personality traits like optimism and self-confidence have been argued to lead to higher levels of social trust (Allport, 1961; Uslaner, 2000). Social trust (also known as generalized trust) is a belief in the trustworthiness of strangers in one’s society. Sociopsychologists view social trust as a product of individual characteristics developed through one’s life experiences. The societal approach to trust, on the other hand, emphasizes the role of societies and their formal institutions on the formation of generalized trust (Berggren and Jordahl, 2006; Newton, 2001; Putnam, 2000). Empirical studies seem to support the sociopsychological view, however, by showing that optimism leads to more success and well-being, and that the latter are positively associated with social trust (e.g. Delhey and Newton, 2003).²

Above, we derived from social support theory that success and well-being are also positively affected by a self-enforcing generalized belief that support will be available if needed (link [1] in Figure 1). This means that there is a link from individual trust in others’ supportiveness leading to a generalized belief in such support to increasing success and well-being, and finally yielding social trust (link [2] in Figure 1). Thus, individuals’ perceived social support becomes a social property in the form of social trust as long as their confidence in the supportiveness of others leads to a general optimism and associated satisfaction and well-being (links [1] and [2]).³ In this sense, in addition to personal traits and societal institutions, social trust is also the product of confidence and optimism about living in a ‘helpful and cohesive environment’ (Barrera, 1986). Trusting behaviour becomes thus a norm at the societal level (Letki, 2006). We conclude that a measure of generalized belief in the availability of support (i.e. a measure at the macro level) is needed for our analysis.

Returning to the issue of the welfare state and its public support, this discussion leads to the following questions: Does the generalized belief in a society that others are to be trusted in making their social support available if needed affect individuals’ attitudes towards the welfare state?; and if so, how?

Sociological theories on social solidarity, with their origin in the work of Durkheim and Weber, argue that when individuals experience a common fate they will be more willing to contribute to collective goods. Individuals’ feelings and perceptions seem to be an important factor underlying this shared identity and utility (Cook, 1979). ‘The degree to which people feel attracted to one another and are loyal at the micro level, and the degree to which they perceive a collective identity and we-feeling at the meso level are decisive for their willingness to contribute to the common good’ (Van Oorschot, 2000a: 17). We know from sociopsychological theories that the stronger the perception of the supportiveness of others (micro level), the higher is the level of social trust, social cohesion and a collective identity of being supportive (meso/macro level) (links [1] and [2] in Figure 1). We can thus theoretically predict that a higher perception of social support at the macro level leads to more solidarity and higher support for welfare states (Koster, 2007). This link can be best explained by engaging the literature on deservingness criteria (De Swaan, 1988; Van Oorschot, 2000a, 2000b). The main contribution of this literature is that in deciding whether to support welfare provisions⁴ individuals consider various criteria, among which identity and reciprocity are the mechanisms driving the link from the social trust in a helpful and cohesive society (denoted here as ‘social trust in support availability’) to more individual support for welfare (link [3] in Figure 1). A helpful and cohesive society contributes to a we-feeling or a collective identity, which in turn makes it justifiable to contribute to the welfare state. In addition, living in a society characterized by social trust in support availability will lead to a general confidence of people being reciprocal in times of need. The society will then be characterized by ‘generalized reciprocity’ (Szreter, 2002), which will make it easier for individuals to consider more government responsibility for the needy as deserving.

The following macro–micro link (link [3] in Figure 1) can be hypothesized:

Hypothesis 1: The level of social trust in support availability is positively related to public support for more state responsibility.

Welfare state, social capital and perceived social support

Various studies have analysed the relationship between the welfare state and social capital (e.g. Kääriäinen and Lehtonen, 2006; Scheepers et al., 2003). The main discussion is whether the formal institutions of a welfare state stimulate or erode a more informal way of meeting needs. One line of research predicts an erosive effect, also known as the crowding-out hypothesis (e.g. Arts et al., 2003; Etzioni, 1995; Fukuyama, 2001), while another, known as the crowding-in hypothesis, argues that social capital can develop and prosper in a well-institutionalized welfare state (Kuhnle and Alestalo, 2000; Rothstein, 2001; Szreter, 2002; Taylor-Gooby, 1985). Empirical tests find more support for the latter, especially at the individual level (Van Oorschot and Arts, 2005).

Because of its multidimensional character, social capital is usually captured either by membership in associations, generalized trust, social norms, social networks, or a combination of these aspects. Though positive, the correlations between these aspects are generally quite low (e.g. Rothstein, 2001; Van Oorschot et al., 2006), leading to the conclusion that one must carefully distinguish between the various aspects of social capital in order to better understand its relationship with the welfare state (Van Oorschot and Finsveen, 2009).

In this article, we introduce and explore another important aspect of social capital –social trust in support availability.⁵ Importantly, this differs from generalized trust in social institutions and in other people (e.g. Narayan and Cassidy, 2001; Putnam, 2000; Rothstein, 2001). While the standard measure of generalized trust refers to the extent to which there is a belief that people can generally be trusted, the measure used in this article refers to social trust related to a specific aspect, namely social support being available when needed. We believe that due to this specific focus, our measure of social trust deals with an often-heard criticism of the standard measure, which is its vague and imprecise information on the radius of trust within a society (Fukuyama, 2001).

We also take the analysis a step further by exploring the combined effect on public attitudes of welfare state provision and social trust in support availability. There are two aspects to be distinguished here. First, the way the welfare state and social trust in support availability are related to each other (link [4] in Figure 1); and second, the way the relationship between the two affects public attitudes towards welfare states (link [5] in Figure 1). As mentioned earlier, the first aspect (link [4]) has to some extent been previously addressed in the literature (i.e. indirectly through measures of trust in institutions and other people). The main conclusion is that a strong welfare state will increase levels of trust and cooperation by creating a societal environment of national solidarity, protection and empowerment (Rothstein, 1998; Uslaner, 2000; Van Oorschot and Finsveen, 2009). We argued earlier that social trust in support availability also contributes to a helpful and cohesive society through mechanisms such as optimism, well-being and belonging. Therefore, in creating and maintaining a helpful and safe environment, the welfare state and social trust in support availability will support each other:

Hypothesis 2: The extent of welfare provision is positively related to the level of social trust in support availability.

Regarding the second aspect (link [5] in Figure 1), to the best of our knowledge there is no research considering the way in which an interaction between welfare state and social trust in support availability relates to public attitudes towards welfare. In this article, we present such an analysis. Being in a society that is perceived as cohesive, supportive and reciprocal based on the availability of social support will increase individuals’ sense of fairness to share the responsibility of supporting the needy. The sharing of this responsibility can, however, take two forms. On the one hand, the higher the level of social trust in support availability, the more the individuals living in such a society will be confident of self- organization or -governance (Ostrom, 1990). In this case, individuals may simply believe that they have no need for a welfare state because living in a society perceived as cohesive and helpful will give them the confidence that they can provide for the needy themselves. This weakens the positive relationship between trust in support availability and support for the welfare state (Hypothesis 1) and leads to the following hypothesis:

Hypothesis 3a: The extent of welfare state provision weakens the relationship between the social trust in support availability and public support for more state responsibility.

On the other hand, though living in a cohesive and helpful society, individuals may prefer to share the responsibility of supporting the needy through the state (i.e. third-party governance) rather than counting on their self-governance skills. In this case, they may prefer more state responsibility to support the needy. Hence the alternative hypothesis:

Hypothesis 3b: The extent of welfare state provision strengthens the relationship between the social trust in support availability and public support for more state responsibility.

Data, operationalization and method

Data

Three datasets are used to test the hypotheses. The individual-level data are obtained from the European Values Study (EVS) (Halman, 2001). More specifically, we make use of the information gathered in 1999–2000 from the third wave of this ongoing project, conducted among 39,797 respondents in 32 European countries. These data are combined with welfare state data at the national level, which are available through Eurostat. Finally, data about levels of social trust in support availability within a country are provided by Fidrmuc and Gërxhani (2008), based on the Eurobarometer Survey 1998, 1999 and 2001 and the Candidate Countries Eurobarometer Survey 2002 (see also Koster, 2007). These three different data sources are merged into one dataset. National-level data from seven of the countries surveyed in the EVS are missing and are therefore excluded from the analyses. The final dataset includes 31,122 respondents in 25 countries.

Operationalization

Dependent variable

The dependent variable public attitudes towards the welfare state is measured as follows. Respondents are asked to rate on a scale from 1 to 10 whether they think individuals should take more responsibility for providing for themselves (1) or that the state should take more responsibility to ensure that everyone is provided for (10). The main reason for using this measure is the fact that it best captures individuals’ views on a ‘fair’ balance with respect to the relation between citizens and the state in solving common problems (Rothstein, 1998). For example, in solving problems like healthcare, education, pensions and so forth, do we want the state or the individuals to take the responsibility for it?

Independent variables

The variable social trust in support availability includes two measures based on the following survey question: ‘If you had any of the following problems (you were feeling depressed; you needed help finding a job for yourself or a member of your family) is there anyone you could rely on to help you, from outside your own household?’ One measure captures the fraction of people that perceive support from others (outside the own household) if they needed a job and the other measures perceived support when depressed. These variables are coded 0 for people who believe they do not have someone outside own household to rely on and 1 for people who believe they do. As explained in the previous sections, our interest lies in a macro variable measuring a generalized belief in the availability of social support. This is obtained by aggregating the individual-level data to the country-level.⁶ The information used is based on Fidrmuc and Gërxhani (2008), who use the Eurobarometer survey to calculate the fraction of people with perceived access to different social resources within a country.

Comparative welfare state research has generated a number of indicators for welfare state provisions. The most common measure is the size of the welfare state in terms of social spending relative to GDP. Social spending includes among others unemployment benefits, disability, sickness and disease benefits (see e.g. Van Oorschot et al., 2006). Other approaches focus on identifying welfare state regimes (Esping-Andersen, 1990; Larsen, 2008) and the associated level of decommodification, for example the extent to which citizens depend on the market for their individual welfare (Fraile and Ferrer, 2005). This is usually measured with the so-called ‘generosity data’ (see e.g. Scruggs, 2004). All these measures relate to a certain extent to the fairness issue captured by our variable of interest public attitudes towards the welfare state. However, given that focusing on regimes or welfare state generosity severely decreases the number of countries that can be included in the analyses, we decided to use the indicator of social spending relative to GDP, which is available for most European countries.

Finally, in order to capture the interaction between welfare state provision and social trust in support availability, we combine these two variables by calculating the product of the standardized scores on welfare state provision and the score on social trust in support availability (see Gërxhani, 2004 for more details).

Control variables

We control for several variables found to be related to welfare state support in earlier studies (see section on Public attitudes towards the welfare state). They include sociodemographic characteristics like the respondents’ education, age, gender, being in a stable relationship with a partner and employment, but also proxies for one’s social ideology like a variable measuring a person’s left–right self-placement (ranging from 1 = left to 10 = right). We also include control variables that are often used in the literature as direct or indirect measures of individuals’ access to social resources through potential contacts with others.

The indirect variables include one’s religious denomination (a dummy variable) and town size. These variables may affect individual access to support and therefore may influence people’s attitudes towards the welfare state because, for example, individuals belonging to a religious denomination may be more in favour of individual responsibility as long as they can receive social support from contacts established through their religious denomination. Individuals living in large towns may be more in favour of state responsibility because they lack the social resources more available in smaller villages. The direct control variables include: one’s membership of a voluntary organization (measured as a dummy variable), the level of generalized trust (indicating whether the respondents think that most people can be trusted or that you cannot be too careful) and whether one spends time with friends (ranging from 1 = not at all to 4 = every week). The first two are expected to contribute to a feeling of shared collectivity and fate, and thus will lead to more support for state responsibility to ensure that everyone is provided for (Van Oorschot et al., 2006).⁷ The third, spending time with friends, captures more directly the resources individuals will be able to access to provide for themselves if needed. Hence, the more time individuals spend with friends, the less they will be in favour of state responsibility.

Finally, to assess the impact of country-level economic environment, we also control for the GDP per capita. The main argument for including such a variable is that since GDP development has a direct effect on a country’s social spending (Van Oorschot and Arts, 2005), we expect public attitudes towards welfare states to vary according to the economic development of a country.⁸

Method

The dependent variable is measured at the individual level, whereas the independent variables are measured at both individual and country level. Multilevel regression analysis (or hierarchical linear modelling) allows us to investigate effects at different levels of analysis and at the same time (Snijders and Bosker, 1999). It does so by showing that the parameters at the micro level are a function of the macro level and that this relationship can be expressed in terms of the macro-level variables (DiPrete and Forristal, 1994). In its general form, the multilevel model has a fixed part (the linear function of the independent variables) and a random part (in this particular case the unexplained variation at the individual level and the unexplained variation between the countries).

All variables, except the dummy variables, are grand mean centred.⁹

The multilevel analysis is performed in the following steps. First an empty model is computed (Model 0). The empty model is an unconditional model without independent variables and serves as a baseline by which the other models are evaluated. The control variables are added in Model I. Given our theoretical argument that welfare state provision and social trust in support availability are considered jointly when formulating an opinion on a welfare state, we specify several models accounting for their separate and combined relationship with welfare state attitudes and compare them to Model I. Model IIa includes the variable welfare state provision without the indicator of social trust in support availability, whereas Model IIb includes the measure for social trust in support availability but not the variable welfare state provision. Model III includes both welfare state provision and social trust in support availability. Finally, Model IV examines how the interaction between welfare state provision and social trust in support availability is related to welfare state attitudes. The parameters in these models are estimated by the maximum likelihood method and the regression coefficients are tested by Wald tests (Snijders and Bosker, 1999). The deviance between the models is used to evaluate the fit of the different models.

Results

Table 1 gives an overview of the country-level means of the dependent variable public attitudes towards the welfare state (individual vs state responsibility) and the independent variables welfare state provision, social trust in support availability and GDP per capita.

Table 1.

Descriptive statistics per country.

	Number of respondents	Individual vs state responsibility^a	Welfare state provision^b	Social trust in support availability^c	GDP per capita
Austria	1522	4.00 (2.57)	55.66	0.79	24,195
Belgium	1912	4.99 (2.73)	47.06	0.73	22,623
Bulgaria	1000	5.11 (2.89)	44.98	0.54	1563
Czech Republic	1908	4.86 (2.57)	44.75	0.80	5521
Denmark	1032	4.39 (2.15)	52.10	0.82	29,993
Estonia	1005	6.05 (2.42)	40.04	0.63	4106
Finland	1038	4.57 (2.46)	53.87	0.73	23,292
France	1615	3.99 (2.51)	53.77	0.78	22,548
Germany	2036	4.73 (2.70)	56.94	0.66	23,114
Great Britain	1000	4.45 (2.39)	47.40	0.80	24,151
Greece	1142	5.63 (2.58)	44.90	0.63	10,497
Hungary	1000	6.09 (2.84)	33.30	0.72	4697
Ireland	1012	4.54 (2.54)	42.76	0.90	25,271
Italy	2000	5.63 (2.68)	50.90	0.78	19,269
Latvia	1013	6.68 (2.67)	43.18	0.56	3302
Lithuania	1018	5.42 (2.83)	45.00	0.64	3263
Luxembourg	1211	4.47 (2.47)	53.50	0.80	46,278
Malta	1002	5.12 (2.76)	20.60	0.67	9932
Netherlands	1003	4.68 (2.11)	46.06	0.86	24,270
Poland	1095	5.73 (2.62)	51.56	0.68	4455
Portugal	1000	4.83 (2.73)	46.86	0.79	14,422
Romania	1146	4.82 (3.23)	42.19	0.59	1651
Spain	1200	5.65 (2.40)	46.86	0.86	14,422
Sweden	1015	4.22 (2.22)	52.87	0.79	27,287
Turkey	1206	5.26 (3.42)	12.46	0.60	2956
Mean	1245	5.01 (2.72)	46.20	0.73	16,270

Note: A total of 31,122 respondents in 25 countries.

1 = individuals should take more responsibility for providing for themselves; 10 = the state should take more responsibility to ensure that everyone is provided for. Mean score, standard deviation between brackets.

As percentage of GDP.

Mean shares of people who believe that other citizens will be supportive in times of need.

Sources: EVS, Eurostat and Eurobarometer.

Next, we ran a correlation analysis, indicating that welfare state provision and social trust in support availability are positively correlated (r = .39; p = .06, N = 25). As shown in Figure 2, countries with high levels of welfare state provision tend to have high levels of social trust in support availability. Though only weakly significant, this result provides direct support for Hypothesis 2. A strong welfare state and a helpful and cohesive society with high levels of social trust in support availability support each other.

Figure 2.

The relationship between welfare state provision and social trust in support availability.

Finally, we test Hypotheses 1 and 3. The former predicts a direct relationship between social trust in support availability and welfare state attitudes, while the latter predicts an indirect relationship through an interaction between welfare state provision and social trust in support availability. The results of these empirical tests are reported in Table 2. Model I investigates the relationship between the control variables and public opinion towards the welfare state. All results, except for ‘belonging to a voluntary organization’ and having a ‘stable relationship’, are statistically significant and supportive of previous findings in the literature. This model significantly improves compared to the empty model (deviance = 616.65, p < .01). The results show that older people and women are in favour of more state responsibility to ensure that everyone is provided for. On the other hand, higher educated people, the employed and those who declare they are politically right-wing are more supportive of the idea of individual responsibility regarding one’s own welfare. These results confirm that the preferences people have towards a welfare state are strongly related to a combination of self-interest and ideology. Furthermore, the results presented in Model I show that individuals belonging to a religious denomination, individuals spending more time with friends and those who are more trusting towards strangers are more in favour of individual responsibility, whereas individuals living in large towns express a higher preference for state responsibility. In general terms, this confirms that individual access to social support (i.e. measured at the micro level) does indeed negatively affect welfare state support.¹⁰ GDP per capita seems to be negatively related to attitudes towards the welfare state, such that the higher the GDP, the less people support the welfare state. This result may be related to the thermostat function of public opinion meaning that institutions can lose public support as they pass a certain threshold (Wlezien, 1995). As a result, the public may prefer a decrease in the level of social spending in wealthier countries, which also have more welfare state arrangements.

Table 2.

Individual and aggregate determinants of public attitudes towards the welfare state.

	(I)	(IIa)	(IIb)	(III)	(IV)
National level (Level 2)
Welfare state provision		−0.01(0.01)		−0.00(0.01)	−0.03(0.02)
Social trust in support availability			−0.27(0.73)	−0.27(0.73)	−0.71(0.73)
Welfare state provision × Social trust in support availability					−0.16*(0.07)
Control variables
National level (Level 2)
GDP per capita	−0.03**(0.01)	−0.03**(0.01)	−0.03**(0.01)	−0.03*(0.01)	−0.02(0.01)
Individual level (Level 1)
Educational level	−0.08**(0.01)	−0.08**(0.01)	−0.08**(0.01)	−0.08**(0.01)	−0.08**(0.01)
Age	0.01**(0.00)	0.01**(0.00)	0.01**(0.0)	0.01**(0.01)	0.01**(0.01)
Gender (1 = female)	0.15**(0.03)	0.15**(0.03)	0.15**(0.03)	0.15**(0.03)	0.15**(0.03)
Stable relationship (1 = yes)	−0.06(0.04)	−0.06(0.04)	−0.06(0.04)	−0.06(0.04)	−0.06(0.04)
Employed (1 = yes)	−0.15**(0.01)	−0.15**(0.01)	−0.15**(0.01)	−0.15**(0.01)	−0.15**(0.01)
Left–right self-placement (1 = left to 10 = right)	−0.31**(0.04)	−0.31**(0.04)	−0.31**(0.04)	−0.31**(0.04)	−0.31**(0.04)
Religious denomination (1 = yes)	−0.11*(0.04)	−0.11*(0.04)	−0.11*(0.04)	−0.11*(0.04)	−0.11*(0.04)
Town size	0.02**(0.01)	0.02**(0.01)	0.02**(0.01)	0.02**(0.01)	0.02**(0.01)
Belong to voluntary organization (1 = yes)	−0.02(0.04)	−0.02(0.04)	−0.02(0.04)	−0.02(0.04)	−0.02(0.04)
Spend time with friends (1 = not at all to 4 = every week)	−0.10**(0.02)	−0.10**(0.02)	−0.10**(0.02)	−0.10**(0.02)	−0.10**(0.02)
Generalized trust (1 = most people can be trusted)	−0.14**(0.04)	−0.14**(0.04)	−0.14**(0.04)	−0.14**(0.04)	−0.14**(0.04)
Intercept	3.91**(0.15)	3.91**(0.15)	3.91**(0.15)	3.91**(0.15)	4.01**(0.16)
Deviance	616.65**	0.08	0.13	0.22	3.42*
ICC	0.03	0.03	0.03	0.03	0.03

Notes: A total of 31,122 respondents in 25 countries.

Standardized regression coefficients are reported, standard errors in parentheses.

Empty model: –2 log likelihood = 109,255.80; Intraclass correlation = 0.06.

p < .05; **p < .01.

Sources: EVS, Eurostat and Eurobarometer.

The relationship between the individual-level control variables and welfare attitudes remains the same after including the variables at the national level, implying that this relationship is not mediated by social trust in support availability and welfare state provision. In particular, it is noteworthy that the correlation between the micro-level individual access to support and welfare attitudes does not diminish if one takes into account the country’s macro-level social trust in support availability. It is the latter (and especially in interaction with the welfare state) that we are interested in, however.

Hypothesis 1 predicts a positive relationship between social trust in support availability and preference for government responsibility. As previously shown, there is a positive relationship between social trust in support availability and welfare state provision, hence these variables are first entered individually in Models IIa and IIb and then together in Model III to account for multicollinearity. The results, presented in Table 2, show that both welfare state provision and social trust in support availability are negatively related to preference for government responsibility, but that these effects are not statistically significant. Hypothesis 1 is therefore not supported by our data.

Hypotheses 3a and 3b predict either a positive or a negative interaction effect between social trust in support availability and welfare state provision on public support for more government responsibility to provide for everyone. These hypotheses are tested in Model IV. Adding the interaction term increases the fit of the model significantly (deviance = 3.42, p < .05). As shown in Table 2 the interaction term is negatively related to support for more government responsibility (b = −0.16, p < .05). To help interpret this interaction term, Figure 3 displays the estimated marginal effect of social trust in support availability on public attitudes, dependent on the extent of welfare state provision.

Figure 3.

The interaction effect of social trust in support availability and extent of welfare provision on attitudes to the welfare state.

Investigated separately, neither welfare state provision nor social trust in support availability are significantly correlated with public attitudes towards the welfare state. By taking a closer look at this relationship, we conclude that social trust in support availability and welfare state provision are in fact interdependently related to public attitudes towards the welfare state.

As shown in Figure 3, the relationship between social trust in support availability and attitudes towards the welfare state depends on the extent of welfare provision. In countries with relatively low levels of welfare state provision, public preference for government responsibility for welfare decreases somewhat with social trust in support availability. In countries with relatively high levels of welfare state provision, government responsibility is favoured much less, as the social trust in support availability increases. Hence, Hypothesis 3a, predicting that the extent of the welfare state provision weakens the relationship between social trust in support availability and public support for more state responsibility, is supported by the data, while Hypothesis 3b (predicting the reverse) is not supported.¹¹

Our results show that, when considered separately, social trust in support availability and welfare state provision are not significantly related to public attitudes towards the welfare state. When they are jointly taken into account they matter more for welfare state attitudes.

Discussion and conclusion

Numerous studies have made useful contributions to our understanding of welfare states, their public support and the role of social capital. This article tries to bridge these three aspects. It does so by focusing on one under-researched aspect of social capital, the social trust in support availability and its combined effect with welfare state provision on individual attitudes towards the welfare state.

The starting point of this research was that in forming and expressing their preference for smaller or larger welfare states, individuals not only take their personal interests and ideology into account but also the institutional and social context they are embedded in. They do so because they are repeatedly involved in an exchange relationship with a welfare state and its institutions (i.e. either through contributing or receiving provisions), but also with other individuals, known or unknown to them. In this article we have first explored the latter aspect, which is the perception that one has about relying on unknown others in times of distress and need. According to social support theory, knowing that one can rely on others leads to individuals’ optimism and confidence, which in turn increases their success and well-being, and is thus perceived as social support. Sociopsychological theories argue further that the more optimistic, happy and successful individuals are as a result of a high level of perceived social support, the higher their level of social trust will be in the supportiveness of unknown others in the society they live in. This in turn will contribute to a feeling of belonging and collective identity within the society, which according to sociological theories on social solidarity and deservingness criteria will encourage individuals to support more government responsibility rather than individual responsibility to provide for everyone.

We have considered all these theories and provided a detailed theoretical analysis of the mechanisms underlying the relationship, on the one hand, between perceived social support at the micro level and social trust in support availability at the macro level, and, on the other hand, the relationship between the latter and welfare states attitudes. Moreover, we have taken the analysis one step further by arguing that the relationship between social trust in the supportiveness of others and public attitudes towards welfare states will be dependent on the institutional setup with respect to the extent of welfare state provision.

Our empirical results show that social trust in support availability and welfare state provision are indeed interdependently related to public attitudes towards the welfare state. The interaction between the two shows that social trust in support availability matters for welfare state attitudes when the extent of welfare provision is high.

The mirror image of this conclusion is that in countries characterized by relatively high levels of social trust in support availability, public support will be lower the larger the welfare state becomes. This finding implies that being in a society that is perceived as cohesive, supportive and reciprocal based on the availability of social support encourages individuals to share the responsibility of supporting the needy themselves (i.e. self-governance) rather than calling upon the state’s responsibility, the larger the welfare state becomes. Hence, the more welfare is provided by the state, the less of it is desired in societies where individuals believe they can rely on each other for support. Though the data show that welfare state provision and social trust in support availability strengthen each other, the more there is of both simultaneously the more individuals will find it fair and justifiable to call upon their own responsibility to provide for themselves. This is primarily driven by their belief that other individuals in the society they live in will be helpful in times of need.

Further contributions of this study are: first, it complements existing research on welfare state attitudes by exploring a macro-level aspect of social capital, which is social trust in support availability by others if needed. This is an important aspect because it captures the variety in which available support is perceived in different societies. As our empirical analysis shows, such a variety has important consequences at the micro level. Even though researchers have repeatedly stressed the importance of moving beyond individual-level explanations, relatively few studies have done so. This may, to a large extent, be related to lack of international comparative data and therefore it is likely that more studies will be carried out as such data become more available. As this field of enquiry develops, it will become clearer how institutional variables and various aspects of social capital shape public support for the welfare state. The second contribution of this study concerns the finding that institutional (i.e. welfare provision) and social forces (i.e. social trust in support availability) do not function independently of each other. In their relationship with welfare state attitudes, they are in fact mutually dependent. Though the relationship between welfare provision and other aspects of social capital has been discussed before, especially by focusing on the crowding-in and/or crowding-out hypotheses, the way such relationship correlates with public support for the welfare state has remained unknown, until now.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

1.

Note that our implementation of perceived social support is indeed at the macro level. Though it builds on micro-level variables (how individuals perceive support), the measure we are interested in is the ‘general’ perception of support at the level of society. The motivation for this perspective is extensively explained in the following text.

2.

also find that subjective measures of success and well-being (e.g. life satisfaction) are more closely associated with social trust than objective ones related to economic circumstances (e.g. occupation and income).

3.

In their sociopsychological study on social support, suggest that though measured at the individual level, social support can also be conceptualized in a larger context, for an entire community or society.

4.

emphasizes that welfare support should not only be related to the degree to which individuals are willing to pay for it but also to their preferences for the rationing of welfare.

5.

Due to its positive association with more expansive trust that encompasses people beyond one’s household, this aspect relates more to bridging- than to bonding-social capital (see Van Oorschot et al., 2006 for a detailed discussion).

6.

Note that while individual perception of available support is measured as a 0–1 variable, the aggregate confidence in supportiveness of others corresponds to the percentage of people who believe that other citizens will be supportive in times of need.

7.

Though membership in voluntary organizations is often used as a proxy for individual civic engagement and generalized trust (Putnam, 2000), one can also argue that the potential contacts established in these organizations may provide resources that one can utilize in times of need. In this case, being a member of a voluntary organization may provide the social support needed, rendering state responsibility unnecessary and thus less preferred.

8.

Note that this can be particularly relevant since our sample includes eight former Eastern European countries that may differ considerably from other countries in the sample with respect to wealth and welfare state spending. To check whether the sample setup affects our results, we constructed models with and without a dummy variable comparing the Eastern European countries with the rest of the sample. Since these models do not show an effect of the dummy variable, we decided to maintain the GDP level as the only control variable. The results can be provided upon request.

9.

For an overview of centring decisions in multilevel analysis, see .

10.

Note that though negative, generalized trust in strangers is an exception since it says something more about how individuals estimate the trustworthiness of the society around them, rather than about their access to actual support.

11.

As suggested by an anonymous referee, we tested the robustness of our results to exclusion of Turkey (which has an extremely low level of welfare state provision). Excluding Turkey does not seriously affect any of our conclusions.

Biographical notes

Klarita Gërxhani is an associate professor of economic sociology at the University of Amsterdam. She is the author of various articles published in the journals European Economic Review, Public Choice, Social Indicators Research, Journal of Economic Psychology, among others.

Ferry Koster is an associate professor of labour, organization and management at Erasmus University Rotterdam, and a researcher at AIAS/University of Amsterdam. His work has appeared in, among others, the European Sociological Review, European Union Politics and Journal of European Public Policy.

References

Allport

(1961) Pattern and Growth in Personality. New York: Holt, Rinehart and Winston.

Andress

Heien

(2001) Four worlds of welfare state attitudes? A comparison of Germany, Norway, and the US. European Sociological Review 17(4): 337–356.

Arts

Gelissen

(2001) Welfare states, solidarity and justice principles: Does the type really matter? Acta Sociologica 44(4): 283–300.

Arts

Halman

Van Oorschot

(2003) The welfare state: Villain or hero of the piece? In: Arts

Hagenaars

Halman

(eds) The Cultural Diversity of European Unity. Leiden: Brill.

Barrera

Jr (1986) Distinctions between social support concepts, measures, and models. American Journal of Community Psychology 14(4): 413–445.

Bean

Papadakis

(1993) Popular support for the welfare state: A comparison between institutional regimes. Journal of Public Policy 13(3): 227–254.

Berggren

Jordahl

(2006) Free to trust: Economic freedom and social capital. Kyklos 59(2): 141–169.

Blekesaune

Quadagno

(2003) Public attitudes toward welfare state politics: A comparative analysis of 24 nations. European Sociological Review 19(5): 415–427.

Blomberg

Kroll

(1999) Do structural contexts matter? Macro-sociological factors and popular attitudes towards public welfare services. Acta Sociologica 42(4): 300–318.

10.

Cohen

Underwood

Gottlieb

(2000) Social Support Measurement and Intervention. New York: Oxford University Press.

11.

Cook

(1979) Who Should be Helped: Public Support for Social Services. Beverly Hills, CA: Sage.

12.

Crespi

(1997) The Public Opinion Process: How the People Speak. Hillsdale, NJ: Lawrence Erlbaum.

13.

Delhey

Newton

(2003) Who trusts? The origins of social trust in seven societies. European Societies 5(2): 93–137.

14.

De Swaan

(1988) In Care of the State. Amsterdam: Bakker.

15.

DiPrete

Forristal

(1994) Multilevel models: Methods and substance. Annual Review of Sociology 20: 331–357.

16.

Esping-Andersen

(1990) Three Worlds of Welfare Capitalism. Cambridge: Polity Press.

17.

Etzioni

(1995) The Spirit of Community. London: Fontana Books.

18.

Fidrmuc

Gërxhani

(2008) Mind the gap: Social capital, East and West! Journal of Comparative Economics 36(2): 264–286.

19.

Fraile

Ferrer

(2005) Explaining the determinants of public support for cuts in unemployment spending across OECD countries. International Sociology 20(4): 459–481.

20.

Fukuyama

(2001) Social capital, civil society and development. Third World Quarterly 22(1): 7–20.

21.

Galbraith

(1992) The Culture of Contentment. Harmondsworth: Penguin.

22.

Gangl

(2007) Welfare state. In: Ritzer

(ed.) Blackwell Encyclopedia of Sociology Online. Oxford: Blackwell.

23.

Gërxhani

(2004) Tax evasion in transition: Outcome of an institutional clash? Testing Feige’s conjecture in Albania. European Economic Review 48(4): 729–745.

24.

Hagan

MacMillan

Wheaton

(1996) New kid in town: Social capital and the life course effects of family migration on children. American Sociological Review 61: 368–85.

25.

Halla

Lackner

Schneider

(2010) An empirical analysis of the dynamics of the welfare state: The case of benefit morale. Kyklos 63(1): 55–74.

26.

Halman

(2001) The European Values Study: 3rd Wave. Tilburg: EVS/WORC.

27.

Hasenfeld

Rafferty

(1989) The determinants of public attitudes toward the welfare state. Social Forces 64(4): 1027–1048.

28.

Heinemann

(2008) Is the welfare state self-destructive? A study of government benefit morale. Kyklos 61(2): 237–257.

29.

Jæger

(2006) Welfare regimes and attitudes towards redistribution: The regime hypothesis revisited. European Sociological Review 22(2): 157–170.

30.

Kääriäinen

Lehtonen

(2006) The variety of social capital in welfare state regimes: A comparative study of 21 countries. European Societies 8(1): 27–57.

31.

Kluegel

Smith

(1986) Beliefs about Inequality: American’s Views of What is and What Ought to be. New York: Aldine de Gruyter.

32.

Koster

(2007) Globalization, social structure and the willingness to help others: A multilevel analysis across 26 countries. European Sociological Review 23(4): 537–551.

33.

Kreft

De Leeuw

Aiken

(1995) The effect of different forms of centering in hierarchical linear models. Multivariate Behavioral Research 30(1): 1–22.

34.

Kuhnle

Alestalo

(2000) Introduction: Growth, adjustments and survival of European welfare states. In: Kuhnle

(ed.) Survival of the European Welfare State. London: Routledge, pp. 3–18.

35.

Larsen

(2008) The institutional logic of welfare attitudes: How welfare regimes influence public support. Comparative Political Studies 41(2): 145–168.

36.

Letki

(2006) Investigating the roots of civic morality: Trust, social capital, and institutional performance. Political Behavior 28(3): 305–325.

37.

Lewin-Epstein

Kaplan

Levanon

(2003) Distributive justice and attitudes toward the welfare state. Social Justice Research 16(1): 1–27.

38.

Lin

(2001) Social Capital: A Theory of Social Structure and Action. Cambridge: Cambridge University Press.

39.

Lin

Dean

(1984). Social support and depression. A panel study. Social Psychiatry 19: 83–91.

40.

Lincoln

(2007) Social support. In: Ritzer

(ed.) Blackwell Encyclopedia of Sociology Online. Oxford: Blackwell.

41.

Lipsmeyer

Nordstrom

(2003) East versus West: Comparing political attitudes and welfare preferences across European societies. Journal of European Public Policy 10(3): 339–364.

42.

Lockhart

(2001) Protecting the Elderly: How Culture Shapes Social Policy. University Park: Pennsylvania State University Press.

43.

Mau

(2004) Welfare regimes and the norms of social exchange. Current Sociology 52(1): 53–74.

44.

Narayan

Cassidy

(2001) A dimensional approach to measuring social capital: Development and validation of a social capital inventory. Current Sociology 49(2): 59–102.

45.

Newton

(2001) Trust, social capital, civil society, and democracy. International Political Science Review 22(2): 201–214.

46.

Ostrom

(1990) Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press.

47.

Pichler

Wallace

(2007) Patterns of formal and informal social capital in Europe. European Sociological Review 23(3): 423–435.

48.

Pierce

Sarason

(1997) Handbook of Social Support and the Family. New York: Plenum Press.

49.

Putnam

(2000) Bowling Alone. New York: Simon and Schuster.

50.

Rothstein

(1998) Just Institutions Matter: The Moral and Political Logic of the Universal Welfare State. Cambridge: Cambridge University Press.

51.

Rothstein

(2001) Social capital in the social democratic welfare state. Politics and Society 29(2): 207–241.

52.

Scheepers

Ten Grotenhuis

Gelissen

(2003) Welfare states and dimensions of social capital: Cross-national comparisons of social contacts in European countries. European Societies 4(2): 185–207.

53.

Scruggs

(2004) Welfare state entitlements data set: a comparative institutional analysis of eighteen welfare states. Version 11.

54.

Shapiro

Young

(1989) Public opinion and the welfare state: The US in comparative perspective. Political Science Quarterly 104(1): 59–89.

55.

Snijders

Bosker

(1999) Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling. London: Sage.

56.

Stansfeld

(2002) Review of Cohen, Underwood, Gottlieb (2000). Social Support Measurement and Intervention. New York: Oxford University Press.

57.

Svallfors

(1997) Worlds of welfare and attitudes to redistribution: a comparison of eight western nations. European Sociological Review 13(3): 283–304.

58.

Szreter

(2002) The state of social capital: Bringing back in power, politics, and history. Theory and Society 31(5): 573–621.

59.

Taylor-Gooby

(1985) Public Opinion, Ideology and State Welfare. London: Routledge.

60.

Uslaner

(2000) Producing and consuming trust. Political Science Quarterly 115(4): 569–590.

61.

Van Oorschot

(2000a) Why pay for welfare? A sociological analysis of reasons for welfare solidarity. The Netherlands Journal of Social Sciences 36(1): 15–36.

62.

Van Oorschot

(2000b) Who should get what, and why? On deservingness criteria and the conditionality of solidarity among the public. Policy and Politics 28(1): 33–49.

63.

Van Oorschot

Arts

(2005) The social capital of European welfare states: The crowding out hypothesis revisited. Journal of European Social Policy 15(1): 5–26.

64.

Van Oorschot

Finsveen

(2009) The welfare state and social capital inequality. European Societies 11(2): 189–210.

65.

Van Oorschot

Arts

Gelissen

(2006) Social capital in Europe: Measurement and social and regional distribution of a multifaceted phenomenon. Acta Sociologica 49(2): 149–167

66.

Wilensky

(1975) The Welfare State and Equality: Structural and Ideological Roots of Public Expenditure. Berkeley: University of California Press.

67.

Wlezien

(1995) The public as thermostat: Dynamics of preferences for spending. American Journal of Political Science 39(4): 981–1000.