Effects of Cash Subsidies Embedded in Student Loans

Abstract

Every year, governments allocate considerable amounts of resources to support students’ access and performance in higher education. Grants and loans are directed to those students for whom college would seem unaffordable otherwise. These programs are justified on the grounds of how they help students access, persist and graduate from higher education. However, considering the time it takes to observe academic or labor market performance, the literature evaluating student loans and its additional embedded programs is scarce. This article wants to contribute filling this gap. Using a Regression Discontinuity Design (RDD), I evaluate a cash subsidy program embedded into the main governmental student loan program in Colombia. I exploit the allocation rule of these additional cash subsidies, to evaluate the effect of these transfers over dropout and graduation rates, and also over early default rates. My initial results indicate that the cash subsidies have substantial and significant effects on dropout rates at the cutoff point. Over graduation rates the effects are substantial, but barely significant. With respect to the default rates, cash subsidy beneficiaries tend to have lower rates at the cutoff, but this result is not statistically significant. However, the measure of the subsidy effect on the latter two outcomes is affected by the statistical power loss at the proximity of the cutoff scores, a limitation of the RDD.