ProPelled: The Effects of Grants on Graduation, Earnings, and Welfare

We estimate the effect of grant aid on poor college students’ attainment and earnings using student-level administrative data from four-year public colleges in Texas. To identify these effects, we exploit a discontinuity in grant generosity as a function of family income. Eligibility for the maximum Pell Grant significantly increases degree receipt and earnings beginning four years after entry. Within 10 years, imputed taxes on eligible students’ earnings gains fully recoup total government expenditures generated by initial eligibility. To clarify how these estimates relate to social welfare, we develop a general theoretical model and derive sufficient statistics for the welfare implications of changes in the price of college. Whether additional grant aid increases welfare depends on (1) net externalities from recipients’ behavioral responses and (2) a direct effect of mitigating credit constraints or other frictions that inflate students’ in-school marginal utility. Calibrating our model using nationally representative consumption data suggests that increasing grant aid for the average college student by $1 could generate negative externalities as high as $0.50 and still improve welfare. Applying our welfare formula and estimated direct effects to our setting and others suggests considerable welfare gains from grants that target low-income students.