Using five waves of the European Labour Force Survey for the period 2011-2016 we analyse the differential incidence of overeducation between natives and migrants in twenty-four EU Member States. We look separately at secondary and tertiary educated individuals and for the latter group we apply two separate methods to measure overeducation: the Eurostat method and the realized matches method. We also look at how the likelihood of being over/undereducated is influenced by the length of stay in the host country.

In the first part of our analysis we present simple descriptive statistics, and document that, on average, non-EU born (NEB) and European migrants are less well matched than natives with comparable (i.e. secondary and tertiary) education.  However, these basic descriptive statistics are likely to be affected by how individual characteristics (potentially affecting overeducation) are distributed among native and migrant workers in the sample used in the analysis. Hence, in the second part, we exploit the rich set of information provided by the LFS and probe our data further. Applying standard econometric techniques, which allow us to control for observable characteristics (year, country and industry fixed effects, age and its squared term, the degree of urbanization in the area of workers’ residence and gender), we test whether being a migrant, per se, can significantly affect the likelihood of being over/undereducated. Our results confirm that EU migrants and NEBs (ceteris paribus) are more likely to be overeducated (and less likely to be undereducated) compared to natives with the same educational level.

Our data also allows us to check whether the quality of the match improves or degrades as time of residence in the host country increases. We find that the negative gap for both secondary and tertiary educated NEBs and EU migrants increases with the length of stay in the host country with the exception of EU migrants with short tertiary attainment, for which the opposite holds.

Several not mutually exclusive interpretations can be advanced for our results, for example: even if equipped with the same amount of formal education, NEB migrants might lack in other dimensions of their human capital; their social network might be less extensive than those of locals negatively affecting their chance of finding a good match on the labour market; some discrimination on the part of local employers might be at play. Each possible explanation calls for a different mix of public policies, but our analysis does not allow for a casual interpretation of the findings which goes beyond the scope of this report.       

Nonetheless, the results presented here are important for at least two reasons: first, they document how the measurement of overeducation is to some extent affected by the methodology adopted; second, they suggest that overeducation among migrants, especially the non-European ones, is persistent and does not seem to disappear with their permanence in the host country.