Optimal timing of college subsidies: Enrollment, graduation, and the skill premium

The large increase in the skill premium has led many policy makers to propose increasing college subsidies as a strategy to increase college enrollment and reduce inequality. Existing analyses typically assume that there is no distinction between enrollment and graduation, but in fact more than 50% of college enrollees do not graduate. This highlights the need to explicitly model the decision to drop out in assessing the effect of subsidies. I build a general equilibrium model of education choice that features an explicit dropout decision, calibrate it to match key features of the US economy, and use it to evaluate the impact of college subsidies on the supply of college-educated labor. I show that moving to a system in which subsidies are back-loaded—i.e., higher for juniors and seniors than for freshmen and sophomores—will increase the supply of college-educated labor, even holding total subsides constant. In fact, this effect is larger than what would be accomplished by a uniform 50% increase in subsidies.