The effect of educational mismatch on wages for 25 countries

By making use of the Duncan&Hoffman model, the paper estimates returns to educational mismatch using comparable microdata for 25 European countries. Our aim is to investigate the extent to which the main empirical regularities produced by other papers on the subject are confirmed by our data base. On the basis of tests proposed by Hartog&Oosterbeek, we also consider whether the observed empirical patterns are in line with the Mincerian basic human capital model and Thurow's job competition model. Using Heckman's sample-selection estimator, we find that results are rather consistent with those found in the literature, and that the job-competition model could be accepted, whereas the Mincerian human capital model could be rejected for most of the countries.