Source: Josefa Ramoni-Perazzi and Don Bellante, Journal of Labor Studies, Vol. 28 no. 1, Winter 2007
Using data from the Health and Retirement Study, we apply propensity score matching methods to examine evidence on the rent paid to public sector workers in the United States. Traditionally, wage differentials are computed assuming that workers from both public and private sectors are comparable, without actually controlling for the comparability of the units. Using this method, we are able to control for selection bias and, at the same time, select a subsample of comparable workers in terms of their conditional probability of choosing to work in the public sector on which to estimate separate wage equations.