Do wages and job satisfaction really depend on educational mismatch? Evidence from an international sample of Master graduates
The purpose of this paper is to find econometric evidence of a negative influence of educational mismatch on either wage or job satisfaction, once potential sources of bias are adequately considered. The analysis attempts to answer the question: do wage or job satisfaction really depends on educational mismatch? The paper uses a panel data of 1690 early career Master graduates from Università della Svizzera italiana (USI), Switzerland. First, a wage equation with dummies representing educational mismatch and other control variables is estimated. On the other hand, a regression in which the dependent variable is the degree of self-assessed job satisfaction is run in order to identify the effect of mismatch on job satisfaction. The analysis finds no robust econometric evidence of a negative influence of educational mismatch on either wage or job satisfaction, once potential sources of bias are adequately considered. The estimates have been conducted on a specific sub-population, i.e. a limited sample of Master graduates from a single Swiss university in the years 2006-2016; it is then not straightforward that results can be generalised to the whole population. The influence of educational mismatch on job satisfaction has been extensively studied in the previous literature; however, most of the existing studies are likely to report biased results due to unobserved heterogeneity and measurement error. We address these two serious econometric issues by proposing a new instrumental variable for a self-assessed mismatch, i.e. time spent in job search after graduation.
Education + Training
Start page number
End page number
educational mismatch, instrumental variable, job satisfaction, ordered model, measurement error