by Christian Belzil and Jorgen Hansen.
In this paper, we examine the determinants of training opportunities and the correlation between the incidence of various training activities (on-the-job and off-the-job) and accumulated schooling. We use two different data sources for the empirical analysis: the NLSY and the WES. Using the longitudinal information in NLSY, we assess the causality between training and education within a life cycle context. Using the rich information on both workers and workplaces in WES, we investigate the determinants training opportunities in Canada. The results indicate that there is a weak positive causal effect of schooling on training (on-the-job) after conditioning on unobserved abilities and tastes, and a weak negative correlation between unobserved abilities/tastes explaining schooling and on-the-job training. These results are consistent with the facts that more educated workers have lower costs of learning new skills and/or that schooling enhances the return to training. Our results from the WES data indicate that the incidence of training is stronger in skill intensive industries (or occupations) and positively correlated with workplace performance and technology use. Further, we found a positive correlation between training and education in the WES data as well.