Empirical Essays on Political Economy and Inequality

Abstract

The rst chapter uses a unique dataset on Brazilian party members and variation from mayoral elections to examine the determinants of party membership in Brazil. It starts by examining the effect of winning office on the membership of political parties at the local level. The effect of interest is identi ed using a differences-indifferences approach that compares changes in membership of parties that assume office with changes in membership of all other political parties registered in a municipality. The results indicate that winning office increases the membership of the party of the mayor by 0:5%. In addition, political alignment with higher levels of government has a signifi cant effect on the membership of the mayoral candidate party. Finally, the paper documents that party switching is one of the drivers of the estimated increase in membership. The paper offers evidence in favor of the hypothesis that party membership is driven by opportunistic motives in addition to ideology. The second chapter combines data on the universe of recipients of the Bolsa Famlia program from 2005 to 2015 with data on party membership to investigate the returns to political loyalty. Speci cally, it uses variation from mayoral elections to investigate whether members of political parties that assume office at the local level are more likely to receive social transfers. Regression results from an IV estimation show that indeed members of the party that gained access to municipal government are signifi cantly more likely to receive the benefi t. Additionally, it finds no evidence that members of parties that did not win office are more likely to lose the benefi t as a result of the electoral defeat. This chapter offers direct evidence of material rewards to party membership. The last chapter focuses on the impact of pay transparency on earnings inequality in the Brazilian public sector. Differences-in-differences estimates show that the disclosure of wages reduced the 90/50 decile wage gap across municipalities located in states that adopted wage transparency in comparison to those located in states that did not adopt the policy. There is also no evidence that earnings decile gaps below the median were affected by the salary transparency policy, which indicates that the effect of disclosure in the public sector was mainly concentrated at the upper tail of the log earnings distribution. Finally, evidence presented suggests that the effect on inequality compression is the result of lower returns to top paid occupations rather than changes in employment. The paper suggests that at the margin, top paid public sector employees are insensitive to changes in their earnings, indicating that there are rents that accrue to holding these positions.