The first chapter studies the effects of a teacher performance bonus program implemented in Brazil in 2008. The program covered all schools directly managed by the State of Sao Paulo government, and was based on a standardized test run by the state education authority. I use high-school exit exams organized by the federal government (ENEM) to examine the impact of the bonus system on students' achievement using different comparison groups. In all cases I find statistically significant effects between 5% and 10% of a standard deviation. Moreover, effects are larger in the objective section of the test as compared to writing, a subjects not appraised by the merit pay scheme. The second chapter investigates if there are systematic subject-specific gender differences between scores in proficiency exams, which are graded electronically, and marks assigned by teachers, taking into account the influence of a wide array of potentially confounding factors. The results show that teachers consistently reward girls more generously than boys. Although the inclusion of controls for confounding factors (especially proximate measures of behavior) reduces the magnitude of the estimated coefficient, results are robust and significant. I also find that female teachers inflate grades of female students more than male teachers, and that girls are more over-assessed in language than in math as compared to boys. This finding points to relative grade distortion in addition to the absolute disparity often found in the literature. Finally, the third measures the degree of segmentation in the Brazilian labor market--the wage differential between formal and informal workers--with data from the Monthly Employment Survey (PME), a rotating panel of households in six metropolitan regions. Controlling for observable and unobservable characteristics, I find that workers earn more in the formal sector, which supports the segmentation hypothesis. Contrary to other studies that rely on two-step procedures to correct for self-selection bias or repeated cross-sections to account for individual fixed effects, my true panel allows me to solve both problems simultaneously and obtain a better estimate of the wage differential. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]