Association for Public Policy Analysis & Management

Rapid technological advancements have renewed the emphasis on lifelong learning to ensure employability in an ever-shifting workplace (National Academies of Sciences & Medicine, 2017). Improving educational outcomes for individuals with weaker academic credentials is a key component of workforce development, as fewer than half of all adults have earned a postsecondary degree (Ryan & Bauman, 2016). Increasing human capital development among working adults could help to address many issues confronting the economy as a whole: increasing income inequality income between educated and non-educated workers; technological shifts that are pushing older, unprepared workers into low-skilled jobs; and rising levels of postsecondary debt, driven particularly by college dropouts with poor job prospects (Autor, Katz, & Kearney, 2008; Autor, Levy, & Murnane, 2003; Goldin & Katz, 2007; Looney & Yannelis, 2015).

This paper examines whether financial aid impacts educational investment or labor market outcomes for “non-traditional” students, a broad definition that typically includes those who may be older, independent, enrolled part-time, or working significant hours. I test whether tuition vouchers and cash payments alters non-traditional students’ outcomes using application data from the California Student Aid Commission’s Competitive Cal Grant program, a state-based grant program that explicitly targets older adults. There are two aspects of the program that make it an ideal site for producing evidence that is both causal and generalizable to other settings. First, I can credibly estimate causal effects using a regression discontinuity design that compares essentially identical students on the margins of award eligibility. Second, this paper is the largest and most comprehensive study of non-traditional students to date, using hundreds of thousands of applicants who are weighing enrollment decisions across all postsecondary sectors. This allows produce precise estimates of program impacts, and I can estimate heterogeneous impacts for various populations of interest.

I find that Competitive award eligibility increases bachelor’s degree completion by four percentage points among students interested in for-profit colleges. This impact constitutes a 17% increase over baseline, and instrumental variable estimates based on award utilization are roughly twice as large. Linking to California’s unemployment insurance (UI) wage data, I find quarterly earnings of for-profit students increase roughly $140 beginning roughly three years after application, the moment at which most of their increases in degree completion have finished. This constitutes a 2% increase in wages over baseline, which implies sizeable returns to degree completion. Impacts on degree attainment or labor market outcomes for students interested in community colleges or four-year colleges – who constitute the majority of applicants – are indistinguishable from zero, and I can generally eliminate any meaningful effects. This research highlights the challenges governments face in promoting human capital investment for older, less-educated adults, with a potential role for the for-profit sector in developing marketable skills for adult learners. Nonetheless, even when I restrict to for-profit students that exhibit the largest gains, the cost-benefit impacts on degree completion are significantly lower than observed in other studies of financial aid.