Testing means-tested aid

We estimate the effect of an unexpected institutional financial aid (IFA) award on student outcomes using administrative data collected from nine universities, exploiting variation in IFA schedules within and across university-entry cohorts. Each £1,000 of IFA during the first year of college increases the chances of completing that year by 1.4 percentage points, improves test scores by 0.059 standard de- viations, and increases the chances of graduating with a good degree by 3.4 percentage points. We find that high-ability and low-income students benefit the most and calculate outcome-maximizing and cost-minimizing IFA schedules for each university.