Research

Using data on German firms between 2009 and 2019, I study whether catching-up to the productivity frontier changes firm's demand for apprenticeship training. I find that productivity convergence induces firms to train more apprentices with no effect on training of incumbent workers. These effects are robust to dynamic models, simultaneity, unobserved heterogeneity and product, labor, and apprentice market competition. Policies that stimulate productivity should account for the way heterogeneous firms participate in the apprenticeship market in Germany.   

When firms replace obsolete technology with newer models, how do they recruit skilled workers? Do they hire externally, train workers or train apprentices? We find that introduction of new technologies lead to an upscaling rather than an upskilling. More technologically intensive investments increase the role of young workers in a firm. Retraining opportunities for existing workers increase when firms update their older technologies. 

Does training at a high productive frontier firm matter for long run outcomes? If so, for whom? In this paper, we show that frontier firms provide a substantial earnings advantage to apprentices even after controlling for observed and unobserved selection. However, the earnings premium is limited to high latent ability individuals. We find heterogeneity by firm's training motive and degree of selectivity in their apprentice recruits. We argue that the firm where an apprentice trains has a substantial effect on careers even within narrowly defined apprentice groups.