“Crowding in” or “crowding out”? An examination of the impact of the welfare state on generalized social trust

Abstract

Generalized social trust is correlated with increased levels of civic engagement, lower crime rates, and greater economic growth. Many scholars believe that equality provides the conditions in which social trust can flourish. Thus, welfare programs might be one way to generate social trust. However, the relationship between social spending and trust is contested: Some argue it is negative, while others argue it is positive. This study examined the effects of total social welfare expenditures on social trust in 18 OECD countries, holding constant individual characteristics, country characteristics, and country and year effects. Fixed effects analyses indicate that every additional percent of gross domestic product spent on social expenditures 5 years prior is associated with a 4.7 percent increased likelihood that respondents of that country will endorse trusting other people. Further testing for reverse causality found no significant association between trust and later social expenditures, supporting the claim that expenditures drive trust instead of the reverse.

Education and Life Satisfaction in Relation to the Probability of Social Trust: a Conceptual Framework and Empirical Analysis

“…5). The significant and positive relationship between education and social trust is also confirmed in other recent studies (see for example, Huang et al 2009Huang et al , 2011Zanin et al 2013;Brewer et al 2014). • Life satisfaction and social trust Life satisfaction is one of the most common qualiquantitative indicators of subjective wellbeing that is employed in studies of quality of life.…”

Dualized trust: risk, social trust and the welfare state

“…Gelissen et al, 2012), although the direction of the causal arrow remains under debate (c.f. Bergh and Bjørnskov, 2011;Brewer et al, 2014). There is a consensus, however, that welfare states are not all created equal: social policy programmes may either produce or destroy social capital, depending on their design.…”

“…The choice of controls is based upon the standards of pre-existing work examining generalised trust (e.g. Brewer et al, 2014). At the individual level, these include: education level (using the five-category harmonised ISCED-97 scheme); household income decile; household size, which may have an indirect effect (by changing the meaning of household income) or a direct one (insofar as individuals living alone may have lower trust); gender (with males coded as 1); age and its square, to incorporate a potentially curvilinear effect; trade-union membership; marital status; (self-described) status as a minority (minorities coded as 1); religion, included primarily to capture the Catholic/Protestant divide (with Catholics generally less trusting than Protestants); labour market status controls (unemployment, part-time employment, self-employment; nonemployment (i.e.…”

Does trust increase willingness to pay higher taxes to help the needy?

“…These models are designed to estimate the effects of trust vis‐à‐vis country‐level explanations while controlling for the same set of individual‐level explanations as in the second model. All country‐level variables are lagged in time (Brewer, Oh and Sharma, ).…”

“…Similarly, the results of the structural equations model developed by Bergh and Bjørnskov () do not suggest any significant effect with respect to transfers and subsidies as a percentage of GDP on increases in trust. Finally, Brewer, Oh and Sharma () and van Oorschot and Arts () warn the effect may run in the opposite direction, namely, from support for redistribution to trust.…”