Purpose - The aim of this paper is to analyse the effects of overeducation on wages in the Italian graduate labour market. Design/methodology/approach - The relationship between overeducation and wages is analysed using a double selection approach. Two basic individuals' decisions are explicitly considered in the analysis: the decision to work and the choice of occupation. The first one could create a problem of sample selection bias, whereas the second one could generate an endogeneity bias. Findings - Although in line with previous studies it was found that overeducated workers earn less than their appropriately educated peers, the results suggest also that this pay gap is significantly higher when a double selectivity approach is used than when an OLS approach is employed. The main reason why OLS techniques significantly underestimate the wage penalty associated with overeducation appears to be the bias introduced by the endogeneity of overeducation. Originality/value - Recent studies have examined the impact of overeducation on earnings using a single selection framework. This paper extends previous work by analysing the relationship between overeducation and wages using a double selection approach. Additionally, little research has focused on overeducation in Italy. |