The effects of accountability on higher education

This paper analyzes the effect of an accountability system in the Brazilian college market. For each discipline, colleges were assigned a grade that depended on the scores of their students on the ENC, an annual mandatory exam. Those grades were then disclosed to the public, giving applicants information about college quality. The system also established rewards and penalties based on the colleges’ grades. I find that the ENC had a substantial effect on different measures of college quality, such as faculty education and the proportion of full-time faculty. The detailed information from this unique dataset and the fact that the ENC started being required for different disciplines in different years allow me to control for time-specific effects, thus minimizing the bias caused by policy endogeneity. Indeed, I find strong evidence on the importance of controlling for time-specific effects: estimates of the impact of the ENC on college quality more than double when I do not take those effects into account. The ENC also affects positively the ratio between applicants and vacancies, and it decreases the faculty and the entering class sizes. The results suggest that its introduction fostered competition and favored colleges entering the market.

Barriers to entry and development

This paper analyzes the effect of an accountability system in the Brazilian college market. For each discipline, colleges were assigned a grade that depended on the scores of their students on the ENC, an annual mandatory exam. Those grades were then disclosed to the public, giving applicants information about college quality. The system also established rewards and penalties based on the colleges’ grades. I find that the ENC had a substantial effect on different measures of college quality, such as faculty education and the proportion of full-time faculty. The detailed information from this unique dataset and the fact that the ENC started being required for different disciplines in different years allow me to control for time-specific effects, thus minimizing the bias caused by policy endogeneity. Indeed, I find strong evidence on the importance of controlling for time-specific effects: estimates of the impact of the ENC on college quality more than double when I do not take those effects into account. The ENC also affects positively the ratio between applicants and vacancies, and it decreases the faculty and the entering class sizes. The results suggest that its introduction fostered competition and favored colleges entering the market.