QUANTIFYING THE EFFECTS OF JOB MATCHING THROUGH SOCIAL NETWORKS

The recent literature explains the theoretical implications of the matching of workers to jobs through social networks. These insights are obtained for extremely simplified economies or rely on unrealistically simple social networks. Therefore, it is difficult to obtain a sense for the quantitative importance of effects generated by real life social networks. In this paper, I augment a labor market matching model to allow for information transmission through social networks. I illustrate the effects of social networks and I use simulations to quantify the predictions of the model for complex and realistic social networks. Information transmission through social contacts reduces the steady state unemployment rate from a hypothetical 6.5% to 5%. Social referrals can explain 1/5th of the observed duration dependence of unemployment. They cannot explain much of the variation in wages of otherwise homogeneous workers and do not substantially influence aggregate outcomes over the business cycle.