Do Performance-Based Funding Policies Affect Underrepresented Student Enrollment?

ABSTRACT

More states are using performance-based funding (PBF) systems in an effort to incentivize public colleges to operate more effectively. Responding to concerns about equity, states are also adopting provisions that encourage colleges to serve more students who at risk of not completing college. In this paper, I examine whether PBF policies in general—and more specifically policies that have provisions designed to incentivize colleges to enroll underrepresented students—are associated with the number of lower-income, underrepresented minority, and adult students enrolled at four-year public colleges. Using a generalized difference-in-difference framework, I find little evidence that PBF systems as a whole meaningfully affect underrepresented student enrollment. However, the presence of bonuses for serving at-risk students appears to help mitigate any efforts to enroll a more advantaged student body that may be present in other PBF systems.

KEYWORDS:

• Accountability
• enrollment
• equity
• finance
• performance funding

Acknowledgments

I would like to thank my research assistant Olga Komissarova for her assistance developing a dataset of colleges subject to various performance funding systems, Nick Hillman and David Tandberg for helpful comments on an earlier version of this paper, and Dennis Kramer II for helpful advice regarding methodology. All errors remain my own.

Notes

^1. Because Colorado (which had PBF for part of the panel) does not grant traditional appropriations to its public colleges, I ran models including and excluding the state. Results excluding Colorado were substantively similar and are available upon request from the author.

^2. The number of states participating in the State Higher Education Executive Officers (SHEEO) surveys varied from 35 in 2012–2013 to 50 in 2005–2006.

^3. For state appropriations, I added 1 to the dollar values before doing the logarithmic transformation, so the smallest logged value was 0 in the case of colleges with no appropriations in a given year.

^4. I used the 2010 Carnegie Classifications to classify all associate and baccalaureate/associate colleges as 2-year colleges, although IPEDS classified them as 4-year colleges. Therefore, community colleges offering a small number of bachelor’s degrees were excluded.

^5. The number of Pell Grant completions was added to IPEDS in October 2017, but there was only 1 year of data available as of late 2017.

^6. The College Scorecard has the percentage of Free Application for Federal Student Aid (FAFSA) filers who are first-generation students, but it was not possible to get the number of first-generation students from that information. Additionally, states do not appear to use consistent definitions of “first-generation,” making analyses difficult.