Simultaneous Search and Recruitment∗

I present an equilibrium search model of the labor market in which workers send multiple applications and firms can interview multiple applicants. The potential number of interviews is determined by an ex ante investment in a recruitment technology. A better technology is more costly, but serves two purposes: it increases the probability that the firm will identify an applicant that is qualified for the job, and it allows the firm to make new job offers, in case earlier offers got rejected by the workers. I characterize the equilibrium and argue that many existing models are incorporated as special cases, including the standard directed search model with one application and one job offer and the stable matching outcome. Therefore, structural estimation of the the model can indicate how large the inefficiencies in the decentralized labor market are and what the potential gains from labor market policy are. Moreover, the estimation can shed light on the question whether ex ante frictions (i.e. on the worker side) or ex post frictions (i.e. on the firm side) contribute more to the output loss compared to the frictionless Walrasian world.