Job search and over-education in the Australian graduate labour market

Purpose: This paper examines the association between job search and over-education in the Australian graduate labour market. Other determinants of over-education are also investigated. Originality: While the existence of over-educated graduates in the labour market is well documented, there is only limited evidence on the determinants of over-education. One likely determinant that has received little attention is job search, which has been shown in other studies to influence employment outcomes. Design/methodology/approach: This paper is based on pooled cross sectional data from the Australian Graduate Survey (spanning 2003-11) and panel data from the 2011 Beyond Graduation Survey. Both data sets focus on recent graduates. Determinants are investigated using Heckman probit models (pooled data) and Mundlak-corrected random-effects models (panel data). Findings: A substantial proportion of new graduates are over-educated; however the rate of over-education fell in the three years after course completion. Obtaining a job via a university-based search method was associated with a lower over-education probability, even after addressing omitted variable bias. Other search methods had a small or insignificant effect. Research limitations/implications: Our data do not allow us to control for the concurrent use of multiple job search methods, nor other measures of search intensity. These results suggest that over-education may be reduced through wider use of university careers services.