Essays on Educational Inequality and Policy

Abstract

This dissertation is composed of three independent but related research articles, each one studying a specific Brazilian policy/program targeted to enhance access of vulnerable groups to higher education. While the articles investigate considerably different policies, they all essentially address the same issue, which is how the public sector can efficiently use its tools to reduce the barriers to higher education, and if the evaluated – or similar – policies may or not serve as a guide to policy makers worldwide who wish to increase the efficiency and equality of opportunity of their educational systems. In the first article, I analyze the effects of a law that implemented racial quotas in Brazilian federal universities on the pre-college academic performance of non-white individuals. The objective of this article is to provide a better understanding of the incentives provided by affirmative action in education – a matter that is ambiguous from a theoretical perspective and lacks empirical evidence. This research provides evidence that the Law of Quotas fostered incentives to pre-college human capital accumulation as it induced non-white students to attain higher scores on Brazil’s high school exit exam. Therefore, the implementation of racial quotas in higher education not only promotes equity, but also brings about efficiency gains, as it encourages non-white students to close the performance gap with white students by the end of secondary education. In the second article, I investigate the effects of a Brazilian federal program (the Prouni) that grants full and partial college scholarships to students from low-income families on access to HE. The objective of this article is to examine the effectiveness of financial aid to college students on promoting human capital accumulation. To this end, I exploit the introduction of the Prouni in 2005 by employing a difference-in-differences methodology, such as to evaluate the effects of the program on the higher education enrolment rates of low-income students. This research provides evidence that, by the third year after its implementation, the Prouni had increased the odds of attending higher education by 37% for individuals entitled to the full scholarship and by 20% for those entitled to the partial scholarship. Moreover, it is estimated that every US$ 1,000 spent per student generated an approximate 0.8 percentage points increase in the HE enrolment rate of academic age individuals - similar to those found in developed economies. The findings from this research also suggest that the impacts of the grants on access to higher education were greater for women and for non-white persons. In the third article, I study the welfare and labor supply effects of different student loan schemes in higher education, by developing a partial equilibrium microsimulation model in which graduates maximize their expected utilities under wage uncertainty, risk aversion and elastic labor supply. The model predicts that shifting from a mortgage loan to an income-contingent loan (ICL) scheme shall (i) decrease labor supply; (ii) increase graduates’ expected welfare; (iii) reduce repayment burdens, and (iv) increase the number of years until the debt is fully repaid. The model is then calibrated with real Brazilian data and the results confirm the predictions when changing the Brazilian government-backed student loan program to an ICL. I find that shifting to an ICL is especially welfare-enhancing for women and non-white people, two population groups who have lower initial earnings, flatter income growth curves throughout their working lifetimes and who also face greater unemployment risks.