Acessibilidade / Reportar erro

Sexual orientation in Brazil using unconditional quantile regression* * The authors are grateful for the CNPq support through the CNPq Universal Call/MCTI n. 01/2016 for Applied Humanities, Social and Social Sciences, Process n. 406564/2016-7.

Abstract

This paper seeks to complement the underexplored dimension of sexual orientation-based differences in Brazilian incomes. Studies addressing the theme in Brazil have not recognized hitherto the main determinants of wage inequality along the wage curve, given that only data from the 2010 Brazilian Census conducted by the Institute of Geography and Statistics (IBGE) has been used up until now. Thus, importantly, this is the first paper using pooled data from the Continuous National Household Sample Survey (Continuous PNAD) of 2012 to 2016 to evaluate income differentials based on sexual orientation, employing the general Oxaca-Blinder (1973) decomposition method, which in turn is based on Recentered Influence Function (RIF) regressions proposed by Firpo, Fortin and Lemieux (2007FIRPO, S.; FORTIN, N.; LEMIEUX, T. Decomposing wage distributions using recentered influence function regressions. Mimeo. University of British Columbia, 2007.). Findings: the composition effect presents favorable results for homosexuals as a group effect (both gays and lesbians), regarding demographic changes and human capital as well as activity and occupational sorting. On the other hand, there is no evidence of discriminatory effects on homosexuals in the context of the labor market. This is in sharp contrast with the effect of sexual orientation found in the existing literature.

Keywords:
Sexual orientation; Continuous PNAD; Recentered influence function

Instituto de Economia da Universidade Estadual de Campinas, Publicações Rua Pitágoras, 353 - CEP 13083-857, Tel.: +55 19 3521-5708 - Campinas - SP - Brazil
E-mail: publicie@unicamp.br