An Analysis of Public- and Private-Sector Wages Allowing for Endogenous Choices of Both Government and Union Status

Abstract

Studies of public/private sector wage differentials typically assume that the govenment and union status of a worker are exogenous variables. Recently, some studies have relaxed this assumption slightly by allowing the union status to be endogenous. In this paper, we consider a more general selection model in which a worker selects among four labor markets: private/nonunion, private/union, public/nonunion and public/union. A multinomial logit model is estimated to capture this selection decision. Consistent wage equation estimates are then derived using a generalization of the now familiar two-step estimation procedure. Some evidence is found for selection bias in the private/nonunion and the public/union sectors.The pattern of these selection effects produces larger union wage premiumsin the public as compared to the private sector. While this is in contrast to the standard findings, the standard errors on the public sector union wage differentials are quite high. In addition, the data indicates that the public/private sector wage differential is largest for federal workers despite the comparability process determining their wages.