Job search is generally described by an intensive effort margin such as the number of applications sent or of hours devoted. Using rich online job board data and a novel network method to determine relevant sets of ads for each applicant, we also investigate the job search selective margin, i.e. why workers apply to or
Using proprietary data from a Chilean online job board, we find strong, positive assortative matching at the worker-position level, both along observed dimensions and on unobserved characteristics (OLS Mincer residual wages). We also find that this positive assortative matching is robustly procyclical. Then, we use the generalized deferredacceptance algorithm to simulate ex post matches to