Central government versus private sector wages and cognitive skills: evidence using endogenous switching regression

Abstract

The importance of estimation techniques that allow for nonrandom selection of workers into the public and private sectors has been established in the theoretical and empirical literature. A separate body of work has explored the contribution of cognitive and other basic skills to earnings. This article brings together these two strands of empirical literature using Adult Literacy and Lifeskills (ALL) survey data for Norway and Bermuda. In the case of Norway, results from both Ordinary Least Squares (OLS) and a switching regression model agree that cognitive skills are rewarded more in the public sector and that, in both sectors, the main effect is the direct effect of skills on earnings. In the case of Bermuda, however, switching regression estimates are substantially different with respect to the how skills affect earnings; furthermore, controlling for cognitive skills changes the nature of selection and, hence, the estimates of sector wage differentials.