The incidence and wage effects of overeducation using the vertical and horizontal mismatch in skills : Evidence from Switzerland

Purpose – The purpose of this paper is to propose an improved concept of educational mismatch that combines a statistical measure of over- and undereducation with the worker’s self-assessment of skill utilization. The novelty of this measurement approach consists in identifying the vertical and horizontal nature of skills mismatch, that is, a mismatch in which skills are either over/underutilized or not utilized. Design/methodology/approach – Cross-sectional data from the Swiss Household Panel survey for the years 1999 and 2004 are used to determine the true extent of educational mismatch among workers. Moroever, different versions of the Duncan and Hoffman wage equation are estimated depending on whether basic or alternative measures of educational mismatch are included. Findings – The empirical analyses provide the following results: first, at least two-third of the statistically defined overeducated workers perceive their skills as adequate for the job they hold and are then apparently overeducated; second, overeducated workers whose skills are not related to the job do not receive any payoff to years of surplus education; and third, apparently overeducated workers have similar wage returns compared to others with the same schooling level but who are statistically matched. All in all, these findings confirm that most of those overeducated according to the statistical measure have unobserved skills that allow them to work in a job for which they are well-matched. Originality/value – The paper indicates the need to consider both vertical and horizontal skill mismatches when measuring educational mismatch in the labour market. In that way, it is possible to account for worker heterogeneity in skills whose omission may generate biased estimates of the incidence and wage effects of over- and undereducation.