Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Previous empirical work on job search has proposed different approaches to estimating mobility rates assumed in models of search. However, these methods either only work for specific models of wage determination, or else require that we know the initial distribution of productivity for workers in our sample. In this paper we show it is possible to estimate mobility rates without having to restrict attention to models in which wages are constant over the course of a job or to assume that the initial distribution of productivity is known. More generally, the approach we propose allows us to freely estimate one degree of unobserved heterogeneity, be it in initial conditions or mobility rates. Applying our results to data from the NLSY suggests that that the theoretical restrictions on the initial distribution of productivity implied by the standard model can overstate the extent of frictions to upward mobility. (Copyright: Elsevier)