The Complicated Relationship Between Generalized Trust and Democracy

R. Putnam (2001) insists that social capital is made up out of generalized trust, reciprocal norms, and social activities. Social capital, he points out, has a strong and positive influence on democracy. An analysis of the 2005 World Values Survey Data, however, does not show us such a relationship between generalized trust and democracy. This fact seems to imply that Putnam’s thesis can be applied to only advanced democracies. In this presentation, I will discuss the reason such a phenomenon emerges against the prediction of Putnam’s social capital theory, and show that the social capital theory does not need to be rejected, but needs to be complemented by new theoretical assumptions.
In order to explain the weak correlation between generalized trust and democracy, I distinguish between trust based on democratic values and trust based on authoritarian values. I then demonstrate a simple mathematical model assuming individual’s rational choice and different network effects of the two types of generalized trust. This model enables us to explain a non-linear relationship between generalized trust and democracy, as it allows us to see that generalized trust has both positive and negative influences on democratic values because it intermediates with educational achievement. Thus, we can say that Putnam focused on only the positive influence of generalized trust on democracy in advanced democratic societies, and ignored the negative influence of generalized trust on developing democracies.
In order to test this theoretical explanation of the relationship between generalized trust and democracy, I analyze the 2005 SSM (Social Stratification and Social Mobility Survey in Japan) Data and 2005 World Values Survey Data. An analysis using multilevel SEM shows that predictions of non-linear correlation between generalized trust and democracy at both individual and nation levels are accurate.