FARC-EP (1994-2006): de su Plataforma y la Agenda Común por el Cambio para una Nueva

Ideological differences are recognized as influencing trust levels. Inglehart (1999) and La Porta et al. (1997) point to the significance of religious traditions in their analysis of the

WVS data. La Porta et al. (1997) suggest that hierarchical religions such as Catholicism and Islam are associated with low levels of trust, while Inglehart (1999) found that Protestant and Confucian societies are generally more trusting. Branas-Garza et al. (2006) however found that Catholicism reinforces both horizontal and vertical trust more than other religions. Moreover, Catholic affiliation reinforces the link between religiosity and social capital. Uslaner (2000) demonstrates that protestant societies are more trusting due to their individualistic nature while Muslims tend to trust less because of their more collectivistic culture. Bjornskov (2005), like Uslaner, argues that Protestants are more trusting than other religions. He distinguishes between two possible explanations. One stresses that Protestantism is not a hierarchical religion. The second emphasizes that in Protestantism, the responsibility of ones actions is individualized so that actions that are considered morally wrong will somehow be penalized in the afterlife.5 _{On the other hand, Catholics believe it to be possible}

to be absolved of one’s sins by the church. The practice of absolution thus releases the subjects of the Pope of individual responsibility for their worst deeds, which could lead people to be more wary of trusting their fellow citizens. Another joint problem for hierarchical religions may be the potential tendency for individuals to place part of the responsibility for their actions on a supreme power, leaving this God-given uncertainty to naturally lead to lower degrees of trust in fellow citizens. Another possible effect of hierarchies is that people come to live according to strict rules. They may therefore fail to develop trust because following rules does not induce any social learning about what people would do in the absence of any enforced formal rules.

The WVS data provides evidence that generally supports the above given argumentation. Protestants are found to be most trusting with an average interpersonal trust score of 0.50. Catholics appeared to be least trusting since their trust index is only 0.33, considerably smaller than that of Protestants. Other religions and atheists seem to differ little in their trust levels. Both groups had interpersonal trust score estimated at 0.41. It should also be noted that religion must conduct a certain influence on the relationship between social trust and social policy since inclusion of religion in the model considerably changes the sign on social spending. It is however difficult to explain the direct or indirect effects of religion on the relationship between social trust and social spending. But one thing is sure here: religion should be included in the model.

Apart from one’s specific religion, social trust is influenced by the religiousness of an

individual (Branas-Garza et al, 2006; Van Oorschot and Arts, 2005). More religious people

5 _{However, in his further analysis, he comes to the conclusion that the positive effects of Protestantism are most}

likely due to positive effects specific to the Nordic countries that potentially might be traced back to particular Viking norms.

are found to be associated with higher levels of social capital of all forms. On the one hand, faith in people and faith in a Supreme Being both promote civic engagement. People with more faith participate more often in civic affairs – especially in more demanding activities such as volunteering. On the other hand, faith leads people to put less emphasis on materialistic values and more on helping others. Whiteley (1999) also suggests that religious beliefs should generally create an ethos which is trusting, altruistic, and favourably inclined towards cooperation with other people.

The empirical data give little support to this argument since we obtained results based on the WVS which advocate for the absence of any difference in social trust between more and less religious people. Both groups have interpersonal trust scores that were estimated at the level of 0.41 while institutional trust was found to equal 14. Although they seem to differ little in trust levels, we find strong spurious effects of religiosity on the relationship between social spending and social trust. Inclusion of the religiosity variable in the equation drastically changes the coefficient on social spending. The latter can be perceived as the necessity to control for religiosity if one wants to obtain the direct effect of social spending on social trust. Moreover, religiosity was found to explain slightly more than one percent of total variation in social trust, which is an additional argument for including religiosity in the model.

The control for ideology reflects the fact that basic ideological beliefs may well influence an individual’s willingness to trust other people. Whiteley (1999) asserts that ideologies from the left, which emphasize cooperation, solidarity, and fraternity, are more likely to result in an ethos of trust in other people than ideologies of the right, which stress individualism, competition, and a social Darwinism struggle for survival. Another explanation that supports this view is provided by Triandis (1995 in Simpson, 2006), who suggests that, given a tension between individual and collective interests, actors in collectivist societies tend to give greater priority to group goals leading to higher levels of trust among individuals. Van Oorschot and Arts (2005) also insist on the significance of political effects on the different elements of social capital, in particular on interpersonal trust. Their empirical research shows as well that left-leaning people tend to have more trust in other people than right-leaning people.

However, when looking at the aggregated level of collectivist traditions, the research finds that collectivist societies are usually less trusting than individualist societies (Simpson, 2006; Yamagishi et al., 1998). In spite of wide argumentation about the effects of political stance on social trust, we see little variation in trust levels across left-leaning and right- leaning groups. Both possess interpersonal trust just that slightly exceeds 0.4 and institutional trust of about 14 unites. Moreover, the explanatory power of political stance is too small (about 0.2 percent) to be included in the final equation. Besides that, political stance causes no

spurious effect on the relationship between social spending and social trust, which again is a good reason to omit this variable from the analysis.