This paper examines the influence of educational mismatch on wages according to workers' region of birth, taking advantage of our access to rich matched employer-employee data for the Belgian private sector for the period 1999-2010. Using a fine-grained approach to measuring educational mismatch...

A large body of literature shows that first-generation immigrants born in developing countries experience a higher likelihood of being overeducated than natives (i.e. immigrant overeducation). However, evidence is remarkably scarce when it comes to the overeducation of second-generation...

This paper is among the firsts to investigate the impact of overeducation and overskilling on workers' wages using a unique pan-European database covering twenty-eight countries for the year 2014, namely the CEDEFOP's European Skills and Jobs (ESJ) survey. Overall, the results suggest a wage...

While the literature on the incidence and wage effects of over-education is substantial, specific results for doctoral graduates are surprisingly scarce. This article aims to fill this gap, not only by measuring the prevalence of over-educated PhD holders in Europe (i.e. in EU Member States and...

The authors use matched employer-employee panel data on Belgian private-sector firms to estimate the relationship between wage/productivity differentials and the firm's labor composition in terms of part-time and sex. Findings suggest that the groups of women and part-timers generate employer...

This paper is one of the first to use employer-employee data on wages and labor productivity to measure discrimination against immigrants. We build on an identification strategy proposed by Bartolucci (2014) and address firm fixed effects and endogeneity issues through a diff GMM-IV estimator....

Different empirical studies suggest that the structure of employment in the U.S. and Great Britain tends to polarise into "good" and "bad" jobs. We provide updated evidence that polarisation also occurred in Germany since the mid-1980s until 2008. Using representative panel data, we show that...