Association for Public Policy Analysis & Management

In this paper we leverage the random assignment of a need-based financial aid grant offer—the Fund for Wisconsin Scholars (FFWS) grant—and several sets of administrative records to provide among the first experimental evidence on the effect of need-based financial aid on students’ economic outcomes in their post-college years. Specifically, we estimate the effect of the need-based aid offer on students’ in-state employment, earnings, and participation in public assistance programs, namely Temporary Assistance for Needy Families (TANF), the Supplemental Nutrition Assistance Program (SNAP), and Unemployment Insurance (UI). We estimate these effects separately for students attending 4-year universities and two-year institutions, both colleges and technical schools. In addition to estimating the effect of the financial aid offer—the intention-to-treat (ITT) parameter—we also estimate the effect of receiving need-based aid, the treatment-on-the-treated (TOT) parameter. We estimate this parameter in an instrumental variables (IV) framework, using the randomized aid offer as an instrument for receipt of need-based aid.

Our preliminary results suggest that, for students in four-year universities, the FFWS grant offer reduces in-state employment both during students’ college years as well as during the time period where they would typically transition into the labor market. Moreover, we find evidence that the aid offer to reduce in-state earnings throughout the full eight-year period we study—the magnitude of these reductions are typically in the range of five percent of mean control group earnings, although they are slightly larger in some years. In further analysis, we show that the aid offer increases student grade point average (GPA), suggesting that the employment and earnings reductions during students’ in-college years are attributable to a reallocation of time and effort away from employment and toward coursework. For students’ post-college years, we provide suggestive evidence that the reductions are attributable to a combination of two mechanisms: 1) Offer-induced outstate migration, and 2) Offer-induced employment in lower paying jobs, which may be facilitated by our finding that the FFWS offer reduces students’ cumulative loan debt by $2,000-$3,000. We find little evidence that the FFWS grant offer affects the labor market outcomes of students in two-year institutions or the receipt of means-tested benefits for students in either two- or four-year institutions.