Are close-knit networks good for employment?

How do short network cycles–the building component of close-knit network neighborhoods–affect diffusion? We study this issue in labor markets by explicitly modeling the flow of information about vacancies in social networks. We show that short network cycles induce stochastic affiliation in the transmission of job information, generating diffusion inefficiencies with important short- and long-run micro- and macroeconomic consequences. Short network cycles affect employment and inequality patterns within and across networked societies. In particular, they organize employment probabilities in the sense of the first-order stochastic dominance. People in close-knit neighborhoods and dense networks exhibit worse labor-market outcomes. Since dense, overlapping neighborhoods is one aspect of strong relationships, this uncovers an alternative mechanism behind the strength of weak ties (Granovetter, 1973). Moreover, short network cycles reinforce spatial and temporal correlations in employment status, shaping labor-market transition rates and aggregate employment fluctuations.