Is the over-education wage penalty permanent?

Abstract

Much has been written about the impact of over-education on wages using cross-sectional data, although there have been few studies that analyse the returns to over-education in a dynamic setting. This paper adds to the existing literature by using panel data to investigate the impact and permanence of over-education wage penalties, whilst controlling for unobserved individual heterogeneity. Our fixed effects estimates suggest that the over-education wage penalty cannot solely be explained by unobserved heterogeneity. Over-education is permanent for many workers since around 50 percent of workers over-educated in 1991 are still over-educated in 2005. However, we also show that these workers are of lower quality compared to around 25 percent who find a match within five years of being over-educated. Finally, there is a significant scarring effect for workers over-educated in 1991 since they never fully reach parity compared to those who were matched in 1991, although this is not the case for graduates who manage to find a match within 5 years.