Conventional analysis predicts that workers pay part of their on-the-job training costs by accepting a lower starting wage and subsequently realize a return to this investment in the form of greater wage growth. Missing from the conventional treatment of on-the-job training is a discussion of the process by which heterogeneous workers are matched to jobs requiring varying amounts of training. This matching process constitutes a key feature of the on-the-job training model presented in this article and tested with a unique data set containing extensive information concerning on-the-job training, employer search, wages, and wage and productivity growth.

This article specifies and estimates an empirical model of male labor supply based on an implicit market model of wage-hours determination. We discuss how moving from a standard labor supply model to an implicit market model affects model specification and choice of estimation technique. We find that average hourly earnings are not independent of hours worked and that OLS estimates of the wage-hours relationship are biased. We also show that a labor supply model that assumes wages to be independent of hours worked produces a positively biased estimate of the effect of the wage on labor supply.

Data on Naval Reserve recruiters are used to estimate the effects of on-the-job learning, experience, and individual characteristics on job performance. Generalizations of the Poisson distribution form the basis for estimating the effects of explanatory variables and control for individual heterogeneity and overdispersion. The findings show strong learning effects during the first 2 years on the job. Furthermore, lower pay-grade individuals have steeper learning curves than individuals in higher pay grades. Estimates of individual differences in productivity show a large variance in unobserved ability.