The role of individual social capital in occupational choice and earnings

In recent years, the concept of social capital has been used by social scientists as a panacea for all kinds of problems facing individuals or societies. In the process, it is riddled with a weak formalization, a great deal of definitional ambiguity, poor measurement and casual empiricism, to say the least. Within the economics discipline, some efforts are underway to formalize the concept and make it gain wide acceptance in similarity to the development of human capital. Glaeser and al. (2002) is the first attempt to an exhaustive microeconomic theory of social capital investment. ^ This thesis investigates in a rigorous manner some of the implications of the theory above. Given Glaeser and al. (2002) definition of individual social capital which includes intrinsic social abilities, I find that social capital explains the sorting of individuals into social occupations. I also find that ceteris paribus on average an additional hour per week spent on social activities in childhood translates to a 0.37% increase in earnings in early adulthood, i.e. a tenth of the return to an additional year of schooling. However, these results do not hold when intrinsic social abilities are excluded. Indeed, I propose an alternative definition of individual social capital which centers on the investments made by decision makers. My definition is driven by theory and linked to clear and empirically sound proxies. By using an instrumental variables approach, I find no causal effects of social capital investment on either earnings or the selection of sociable occupations. Based on Lee\u27s (1983) two-step method to correct for potential self-selectivity bias, I confirm that while expected earnings affect occupational choice, it does not explain the sorting of individuals into sociable occupations. Instead, non pecuniary factors like the relationship to a father who works in a sociable occupation are at play.