Research

I document the extent to which students have biased beliefs about college loans, finding that the vast majority of students overestimate their price, and that the information gap is wider for poorer students. An information treatment in the form of a chatbot that talks to high-school seniors in Colombia around the time of college application has a significant effect in correcting biased beliefs about college loans. The information treatment is found to significantly increase the probability of taking a loan, which is entirely driven by a higher probability of application.

How do universities’ differ in their preferences over the composition, diversity and academic ability of their student body? A large literature in education economics has remained mostly silent about this question due to difficulties in measurement and the opacity of university admission processes in countries such as the U.S. We explore this question in the Chilean higher education setting, one of many countries that use centralized assignments for university admissions. We study the effect of a reform that allowed universities to select students based on their ranking relative to their high school peers, in addition to standardized tests and GPA. Using an array of empirical designs and machine learning prediction tools, we find significant heterogeneity in universities’ willingness to admit students based on their predicted academic performance and demographic characteristics. Our results shed light on the importance of accounting for institutions’ responses to policies aimed at increasing equity in access to higher education. We also emphasize and quantify trade-offs between guaranteeing diversity and inclusion in admissions—hence addressing structural inequality and fostering upwards mobility (Chetty et al., 2020)—and leveraging institution-specific knowledge on predictors of academic success.

In this article, I discuss the misfits problem, a practical issue that arises in random treatment assignment whenever observations cannot be neatly distributed among treatments. I also introduce the randtreat command, which performs random assignment of unequal treatment fractions and provides several methods to deal with misfits.

Teacher allocation—both within and across schools—is a major source of inefficiency in school districts. The school teachers market is particular because teachers cannot freely supply any number of labor hours, because they also have to fit each school’s schedule. These schedules are determined via a costly process which is inflexible and produces sub-optimal results. In this paper I investigate how inflexible scheduling of labor hours is an important source of frictions in the school market, which leads to inefficient assignment of teachers across school districts.