Association for Public Policy Analysis & Management

Many federal and state programs that provide grants for postsecondary education tie grant generosity to a student’s family income or have strict income thresholds for eligibility. To target students who benefit the most from additional grant aid, it is important to understand the effects of an additional dollar of grant aid on the overall composition of a student’s financial aid awards and academic performance at different levels of family income. I investigate responses to grant aid using student-level data from the Maryland Longitudinal Data System, Maryland’s central repository for education and workforce data. Using the variation in the eligibility criteria of the Maryland Educational Assistance (EA) Grant during the 2008-2015 period, this paper uses a regression discontinuity framework to estimate the heterogeneous effects of an additional dollar of grant aid on a student’s persistence from the first to second year, as well as the composition of a student’s financial aid package during the first year. Using data on earnings, I also investigate whether receiving additional grant aid results in students earning less from employment during the academic year. I find that, on average, $1,000 in EA grant aid results in a 1.3 percentage point effect on persistence to the second year, and that these effects are strongest at lower incomes (with close to a 3 percentage point effect for students with family incomes around $45,000 -$52,000) and generally are small or close to zero for students at higher levels of income. These persistence effects are larger for these students despite lower institutional grant aid and student loan receipt in response to the EA grant. In general, I find no effect on earnings on average, though find suggestive (though imprecisely estimated) evidence that students at lower incomes may earn less during the academic year in response to the grant.