Public-Private Wage Differential and Economic Cycle in Brazil

This paper analyzes the wage premium paid to Brazilian state government employees compared to private sector ones. This differential is estimated with Mincerian equations using 1995 to 2015 annual data from the National Household Sample Survey (PNAD/IBGE) and corrected for selection bias through a bivariate probit model for the probabilities of being employed and being a public employee. Once the public-private wage differentials were obtained for each year and state government, they were adopted asthe dependent variable in a panel data approach to test the hypothesis of a countercyclical behavior. The covariates used are the unemployment rate, public finance indicators, the percentage of public employees in total employment, industry share in GDP, and variables that capture electoral cycles. Results show that the wage differential has a countercyclical behavior associated with rigid hiring and dismissal rules in the public sector. However, the relationship between the wage differential and the unemployment rate was found to be nonlinear, and for high levels of unemployment, the earnings of public employees are also affected.