Chi guadagna di più dalla formazione professionale in azienda? Impatto su produttività e salari, dalla teoria alla stima empirica

This paper analyses the impact of training on productivity and wages, in order to verify who gains more from training, the firm or the employees, and to assert then who has the greater incentives to invest on it. We start analysing the main theoretical models for training, with particular attention to Becker (1962) and Acemoglu and Pischke (1998). These two models entail quite opposite implications, so we want to verify using real data which one better approximates reality. Our analysis will consist on estimating the impact of training on productivity and wage using a panel of Italian firms from the RIL-AIDA datebase, for the years 2005-2007-2009. The econometric analysis follows the literature in assuming that technology at firm-level can be characterized by a Cobb-Douglas production function. We then estimate the production and wage functions using different techniques, starting with the simple OLS method and then moving to panel regressions with Fixed-Effect or First-Difference estimation, in order to exploit the longitudinal nature of the dataset and avoid possible omitted variable or endogeneity bias. We found a positive impact of training intensity on both labour productivity and wages, with a percentage effect of 0.12 on productivity and 0.05 on wages with our best OLS estimation and an effect of 0.04 on productivity and 0.03 on wages with FE estimation. Unobserved firm heterogeneity therefore partially explains the higher estimates of the OLS regressions, but FE estimations are likely affected by attenuation bias, so the real effect will probably lie in between. In general, the effect on productivity is greater than the effect on wages in all the specifications, irrespectively of the estimation technique. Therefore we can conclude that firms and employees share the returns from human capital and hence suppose that there are labour market imperfection and firms may have sufficient incentives to invest in training, even without external financing.