This paper uses firm-level panel data of on-the-job training to estimate its impact on productivity and wages. To this end, we apply and extend the control function approach for estimating production functions, which allows us to correct for the endogeneity of input factors and training. We find that the productivity premium of a trained worker is substantially higher compared to the wage premium. Our results are consistent with recent theories that explain work-related training by imperfect competition in the labor market. Url or DOI http://dx.doi.org/10.1162/REST_a_00460 Authors Jozef Konings Stijn Vanormelingen Country Belgium Publication Year 2015 Ranges Intervention Intervention Start Year 1997 Intervention End Year 2006 Evaluation Evaluation Start Year 1997 Evaluation End Year 2006 Policy field Training On-the-job training Target group Labour market status Employed Details Funding Source Other Outcome Variable Income/wages Data Source Administrative / survey Evaluation Method Regression
