Abstract

In this paper we are testing the hypothesis that job seekers with networks that are strong in social capital will obtain higher wages in subsequent jobs. In contrast to the largest part of the literature, we do not focus on the dyad of jobseeker and successful contact, i.e. the contact associated with the job that was taken up (Granovetter 1974). Instead, we focuses on characteristics of the whole network. To avoid bias due to homophily in network formation, we employ Mouw`s (2003) two-step indirect test of causality, that was designed to distinguish between genuine and spurious causal effects of social capital on wages. In a first step, we analyze the effect of a specific network characteristic on wages. In a second step we analyze the effect of that characteristic on the probability that a job seeker finds a job via a personal contact. Employing the two-step test to a broad variety of different network characteristics, we find social capital in the form of weak ties and influential contacts to pass the test and thus be candidates for having a causal influence on wages, where strong ties, employed contacts and job search support are likely to be correlated with wages only due to homophilous networks.