Três ensaios sobre educação, desigualdades raciais e políticas de ações afirmativas

Abstract

This thesis is composed of three essays on education, racial inequalities and affirmative action policies. In the preparation of statistical tests and model estimates, the following microdata were used: National Household Sample Survey (PNAD) of the Brazilian Institute of Geography and Statistics (IBGE) for 2014, National Student Performance Examination (ENADE) of the National Institute of Educational Studies and Research Anísio Teixeira (INEP) for the years 2015 and 2017. The first essay sought to find evidence of the effect of the family background as a determining component for the success of individuals. The study investigated the effect of family origin composed of variables such as family monthly income and the history of parents' schooling on the performance of students in the National Student Performance Examination (ENADE). For this, the model of simultaneous equations was estimated, combining with the procedures adopted by Klein and Vella (2010). The results show that the family background exerts a great influence on student performance. In the second essay, we sought to investigate the different factors that explain the income gap between blacks and whites in the Brazilian labor market. Using the Oaxaca-Blinder decomposition (1973) applying an econometric model based on the income equation of Mincer (1974). The results showed evidence that wage inequality in the labor market explained by differences in productive attributes and by the weight of the component of racial discrimination. In the third essay, the objective was to analyze the effect of Affirmative Action Policies for blacks and poor people in public universities, having as impact variable the students' score in the National Student Performance Examination (ENADE) in the year 2015. Identify the impact of the policy means assessing the effects of the program on the target population and intends to establish a causal relationship between policy implementation and changes in social conditions. The quasi-experimental method was used as a strategy to construct the treatment group, with the aid of econometric techniques, taking into account some observable variables that enter into the model as control variables, such as sex, age, race, mother's education, schooling work, receive scholarship, study hours and people live. The results show that affirmative action policies have the effect of raising the performance of quota students, when compared to non-quota holders.