Turbulent Networks: Social Capital, Employer Hiring Preferences and Labor Market Outcomes.

Empirical evidence shows that many workers search for jobs through their networks, and that employers often prefer to hire through referrals. As a result, many have expected that who use their social networks will enjoy a labor market advantage over those who do not. However, as Mouw has documented, this pattern is not consistently observed empirically (Mouw 2003). In this paper we argue that networks might play a role in producing or buffering against various types of occupational instability (career turbulence) through the mechanisms of information and influence. Our results suggest that the perplexing lack of empirical support for the social network advantage is explained in part by differences in the local and global structure of networks and by variation in the strategies and preferences of employers. Using an agent-based simulation, we show that broader networks reduce the frequency of turbulent events in careers, and that dense networks can trap workers into unstable cycles of job holding. However, these results can be both amplified and attenuated by characteristics of global network structure of the network, and by employers' preferences for hiring through referral. We discuss the implications of these findings and make specific suggestions for important areas of future empirical research. [ABSTRACT FROM AUTHOR]