Abstract
Little is known about the effects of need-based financial aid disbursed late into college and how students respond when they approach lifetime limits for receiving aid. I exploit changes to federal Pell Grant eligibility rules that reduced the lifetime availability for grant aid from 9 to 6 full-time-equivalent years to examine these questions. Using data from the University System of Georgia and a matched difference-in-differences research design, I compare student outcomes before versus after the rule change for Pell recipients affected and unaffected by the new policy. Risk of aid exhaustion due to the policy change led students to increase their academic effort, as measured by term-over-term re-enrollment and term credits attempted and earned. I find weak evidence that the policy change accelerated time to completion and no evidence that it increased or decreased degree attainment overall. These findings indicate that aid disbursement policies and lifetime aid limits can impact the cost-effectiveness of aid expenditures and the efficiency of college degree production.
This is a preview of subscription content, log in via an institution to check access.
Source: University System of Georgia administrative records
Source: University System of Georgia administrative records
Source: University System of Georgia administrative records
Similar content being viewed by others
Notes
Author’s calculations using the U.S. Department of Education’s National Center for Education Statistics 2004/2009 Beginning Postsecondary Students Survey.
In addition to lifetime eligibility limits on federal Pell Grants, states including California, Florida, and New York cap the duration of state need-based aid that students are eligible to receive. For example, in California and New York, students are eligible to receive the Cal Grant and the New York Tuition Assistance Program, respectively, for up to four full-time-equivalent years.
I define High- and Low-Pell students in Sect. 3.
More than one-quarter of undergraduates also take out subsidized and unsubsidized federal Stafford loans to pay for college; in 2011–2012, students borrowed approximately $6300 in federal student loans (Baum et al. 2017).
As I describe in more detail in Sect. 3.1.1, the FTE provision means that lifetime Pell use is determined by both the number of years of aid received and by a student’s enrollment intensity (i.e., full-time, part-time, etc.) in aid-receiving years.
For example, the California State University System predicts that 4% of its total undergraduate population lost eligibility as a result of the new lifetime limit (Nelson 2012). If this percentage is nationally representative, then 374,000 of the 9.35 million students enrolled in 4-year degree programs are predicted to have been affected. I find that 4% of students in the USG dataset were also potentially affected by the new lifetime limit. However, the USG dataset excludes students who entered as transfer students, and as a result, the 400,000 estimate may be conservative.
Additional evidence that the rule disproportionately affected students attending 4-year institutions is evident from the patterns of persistence across college sectors. While nearly 40% of Pell Grant recipients take more than 6 years to earn a bachelor’s degree (Wei and Horn 2009), only 10% of Pell recipients who began at a community college remain enrolled in the 2-year sector after 5 years (Cho et al. 2013). Most Pell recipients still enrolled in college after 5 years are therefore working towards bachelor’s degrees at 4-year institutions.
Since 2013, the Board of Regents of the University System of Georgia has consolidated eighteen institutions into nine for cost-saving purposes. However, because the GDW data include institutional identifiers prior to consolidation, the dataset in practice covers student enrollments across thirty-seven unique campuses.
The USG dataset also does not include the cost of attendance charged to each student. However, this does not preclude using the disbursement schedules to identify eligible award amounts because the cost to attend USG institutions is sufficiently high that eligible award amounts in practice are based solely on EFC for the vast majority of students in the system. For example, in 2006–2007, eligible award amounts were based solely on EFC once cost of attendance exceeded $4049. In that year, the lowest cost of attendance across all USG institutions for in-state students living off-campus was $9799 and the average cost of attendance exceeded $14,000. This pattern holds, with very few exceptions, for students across all school years, institutions, and living arrangements in this study.
Less-than-half-time enrollment is defined as attempting fewer than 6 credits per term. Half-time enrollment is defined as attempting at least 6 but less than 9 credits per term. Three-quarters-time enrollment is defined as attempted at least 9 but less than 12 credits per term.
For example, a student with an EFC of $1000 who received exactly 5 FTE years of Pell through 2011–2012 would have been eligible for a Pell Grant of $4600 in 2012–2013, whereas a student with the same financial need who accumulated 5.5 FTE years of Pell would have been eligible for $2300.
More specifically, because ED sent communication to all students with Pell receipt in excess of 4.5 FTE years in April 2012, restricting the treated group to students with Pell receipt in excess of 4.5 FTE years would include some students who were not contacted. For example, a student who received 4.25 FTE years of Pell through fall 2011 and enrolled full-time in spring 2012 would have received 4.75 FTE years of Pell at the end of the year but would not have crossed the 4.5 FTE threshold until after ED notified students. The 5 Pell FTE cutoff avoids this potential source of identification error. Furthermore, if some students who received fewer than 5 FTE years of Pell were actually influenced by the new lifetime limit, results using the 5 Pell FTE cutoff will capture conservative effect estimates because some students in the Low-Pell group would be assigned incorrectly.
I present evidence that the parallel trends assumption is violated in the full sample in “Appendix 1”.
Before restricting the data to 5 + FTE students, I also conditioned the sample on students who: (1) graduated from high school, (2) qualified for Pell awards above the minimum amounts, and (3) never left and later transferred back into a USG institution. The first two restrictions eliminate students potentially affected by other changes to Pell Grant eligibility introduced at the same time as the new lifetime limit. I impose the third restriction because it is possible that transfer students received Pell awards during enrollment spells I do not observe in the data. Eighteen percent of first-time, degree-seeking students left the USG system and subsequently returned as a transfer student. By excluding those students, I observe Pell receipt over 8 years as near-completely as possible in the study sample.
I used a modified version of Eq. (1) to calculate FTE status for all students, where I ignored students’ EFC-eligibility status and determined FTE status solely by enrollment intensity. For example, a student who attempted 12 credits during fall 2010 and 6 credits during spring 2011 would be assigned an FTE of 0.75 for that year (i.e., 0.5 for full-time enrollment in the fall and 0.25 for part-time enrollment in the spring). An attractive feature of this approach is that Pell FTE years and FTE years are derived from a consistent set of rules.
I did not coarsen the entry cohort or years to attain 5 FTE status variables to ensure that matches were only made among students who: (1) entered college at the same time, and (2) progressed through college at a similar cadence over several years. This ensures that High- and Low-Pell students experienced the same policy environment at initial enrollment and that the parallel trends assumption is plausible in the matched sample. Except for the indicator variables, which cannot be coarsened further, I coarsened all other variables in the matching procedure to ensure a high match rate among High-Pell students.
The upper bounds of the bottom three attempted credit quartiles are: 126, 136, and 146 credits.
The numbers of High- and Low-Pell students in the matched sample are not identical because, as mentioned above, CEM constructs groups of matched students rather than matched pairs. The procedure identified 1468 unique strata containing at least one High- and one Low-Pell student, but on average the strata contain slightly more High-Pell than Low-Pell students (5.9 versus 5.4 students, respectively).
This mean difference in SAT achievement in the unmatched sample is 95 points. The percent reduction in bias is therefore equal to \(1-\left(\frac{22}{95}\right)=0.77.\)
4.5% of Low-Pell students in the matched sample also never received Pell aid. A potential concern is that students who never received Pell aid or received aid intermittently may be fundamentally different from students who receive Pell aid routinely, and thus serve as a poor comparison group. To examine this, I also restricted to students who received three or more years of Pell before matching High-to Low-Pell students. As shown in Appendix Table 13, I find no evidence that conditioning the sample on more consistent Pell recipients yields substantively different results from the main matched sample, although the estimates on degree completion are slightly attenuated and less precise in the more restricted sample.
To be clear, the High-Pell group in this analysis is still restricted to students who received 5 or more FTE years of Pell, but I allow for enrollment impacts to materialize after treated students received 4.5 FTE years of Pell to match the timing of the Department of Education’s direct-to-student communication. See footnote 12 for discussion of the choice to define treated students using a 5 Pell FTE cutoff instead of a 4.5 Pell FTE cutoff.
Missing SAT scores and EFC at entry are predicted using the full set of non-missing baseline characteristics. In all results, I present estimates from multiple imputation regressions that account for uncertainty in the imputed values for students with missing data.
I cluster standard errors by the 37 unique campuses prior to USG consolidation activities.
In 2010–2011, approximately one-third of USG students received HOPE Scholarships. On average, award amounts the following year declined by $300 per semester for students who no longer qualified for full HOPE scholarships (Suggs 2016).
Forty-eight% of treated students in the matched sample entered a new aid award year having received exactly 5 FTE years of Pell and were therefore eligible for full Pell awards that year.
This in part reflects the fact that Georgia does not offer state need-based aid to students.
I report on degree outcomes within 8 years given that this is the longest I observe students across all entry cohorts. However, I also report on BA attainment ever, which captures degree completion through spring 2016. This outcome therefore captures degree attainment over a longer timeframe, but the time horizon varies across cohorts. For the earliest entrance cohort (fall 2002), the BA ever outcome captures degree attainment through 14 years following entry. For the last cohort (fall 2008), this outcome captures degree attainment through 8 years.
Although the coefficient on overall degree attainment for students who attained 5 FTE status after 2012–2013 is non-trivial (0.028), this appears to be overestimated since most students in this group are members of the 2008 entry cohort and I only observe degree completion through 8 years for this group. The point estimates on overall attainment are closer to zero for students I observe over a longer time horizon. It is therefore likely that the estimate for the post 2012–2013 group would attenuate if I were able to track the 2008 entry cohort over a longer period.
In Appendix Table 12, I report analogous estimates by EFC status after dropping students whose EFC status may have been affected by the change to the family income restriction. The results are substantively identical to those in Table 7, which reinforces that the family income restriction is not driving the main results, nor the variation in effects by zero-EFC status.
For this procedure, I used the same covariates as in the main CEM matching solution to estimate the probability of being a High-Pell student separately for those who were exposed and not exposed to the policy change. I then matched High-Pell to Low-Pell students with the same exposure status and similar predicted probability of High-Pell status (i.e., within ± 0.05% points).
Fifty-one% of 5 Pell FTE students not enrolled post-2011 graduated in years 6–8 versus 54% of treated students enrolled post-2011. As a result, 12,000 students [(400,000*0.54) − (400,000*.51)] are estimated to have graduated more quickly. See footnote 7 for how the estimated number of affected students is derived.
References
Anderson, C. B., & Steele, P. E. (2016). Foiling the drop-out trap: Completion grant practices for retaining and graduating students. Washington, DC: Coalition of Urban Serving Universities and the Association of Public & Land-Grant Universities.
Angrist, J., Autor, D., Hudson, S., & Pallais, A. (2016). Evaluating post-secondary aid: Enrollment, persistence, and projected completion effects (NBER working paper no. 23015). Cambridge: National Bureau of Economic Research.
Barr, A. (2019). Fighting for education: Veterans and financial aid. Journal of Labor Economics, 37(2), 509–544.
Baum, S., Ma, J., Pender, M., & Welch, M. (2017). Trends in student aid 2017. Washington, DC: College Board.
Carruthers, C. K., & Özek, U. (2016). Losing HOPE: Financial aid and the line between college and work. Economics of Education Review, 53, 1–15.
Castleman, B. L., & Long, B. T. (2016). Looking beyond enrollment: The causal effect of need-based grants on college access, persistence, and graduation. Journal of Labor Economics, 34(4), 1023–1073.
Cho, S., Jacobs, J., & Zhang, C. (2013). Demographic and academic characteristics of pell grant recipients at community colleges (CCRC working paper series no. 65). New York: Columbia University, Teacher’s College, Community College Research Center.
Cornwell, C. M., Lee, K. H., & Mustard, D. B. (2005). Student responses to merit scholarship retention rules. The Journal of Human Resources, 40(4), 895–917.
Delisle, J., & Miller, B. (2015). Myths & misunderstandings: The undeserved legacy of year-round Pell Grants. Washington, DC: New America.
Denning, J. T. (2019). Born under a lucky star: Financial aid, college completion, labor supply, and credit constraints. Journal of Human Resources, 54(3), 760–784.
Denning, J. T., Marx, B. M., & Turner, L. J. (2017). Propelled: The effects of grants on graduation, earnings, and welfare (NBER working paper no. 23860). Cambridge: National Bureau of Economic Research.
Dynarski, S. M. (2003). Does aid matter? Measuring the effect of student aid on college attendance and completion. American Economic Review, 93(1), 279–288.
Dynarski, S., & Scott-Clayton, J. (2013). Financial aid policy: Lessons from research. Future of Children, 23(1), 67–91.
Garibaldi, P., Giavazzi, F., Ichino, A., & Rettore, E. (2012). College cost and time to complete a degree: Evidence from tuition discontinuities. Review of Economics and Statistics, 94(3), 699–711.
Gichevu, D., Ionescu, F., & Simpson, N. (2012). The effects of credit status on college attainment and college completion (No. IZA discussion paper no. 6719). Bonn: Institute of Labor Economics.
Goldrick-Rab, S., Kelchen, R., Harris, D. N., & Benson, J. (2016). Reducing income inequality in educational attainment: Experimental evidence on the impact of financial aid on college completion. American Journal of Sociology, 121(6), 1762–1817.
Iacus, S. M., King, G., & Porro, G. (2012). Causal inference without balance checking: Coarsened exact matching. Political Analysis, 20(1), 1–24.
King, G., Nielsen, R., Coberley, C., & Pope, J. E. (2011). Comparative effectiveness of matching methods for causal inference. Cambridge: Harvard University.
Marx, B. M., & Turner, L. J. (2018). Borrowing trouble? Human capital investment with opt-in costs and implications for the effectiveness of grant aid. American Economic Journal: Applied Economics, 10(2), 163–201.
Nelson, L. (2012). Some college, no degree, no Pell. Washington, DC: Inside Higher Ed.
Richburg-Hayes, L., Brock, T., LeBlanc, A., Paxson, C., Rouse, C. E., & Barrow, L. (2009). Rewarding persistence: Effects of a performance-based scholarship program for low-income parents. New York: MDRC.
Schneider, M., & Clark, K. (2018). Completion reforms that work: How leading colleges are improving the attainment of high-value degrees. Washington, DC: American Enterprise Institute and Third Way.
Schudde, L., & Scott-Clayton, J. (2016). Pell Grants as performance-based scholarships? An examination of Satisfactory Academic Progress requirements in the nation’s largest need-based aid program. Research in Higher Education, 57(8), 943–967.
Scott-Clayton, J. (2011). On money and motivation: A quasi-experimental analysis of financial incentives for college achievement. Journal of Human Resources, 46(3), 614–646.
Suggs, C. (2016). Troubling gaps in HOPE point to need-based aid solutions. Atlanta: Georgia Budget and Policy Institute.
U.S. Department of Education (2012). Implementation of the 12 Semester Lifetime Limit for Federal Pell Grants. Retrieved August 9, 2017, from https://ifap.ed.gov/eannouncements/021712PreliminaryInfoImpl12SemesterFedPellGrant.html.
Wei, C. C., & Horn, L. (2009). A profile of successful Pell grant recipients. Washington, DC: U.S. Department of Education, Institute for Education Sciences, National Center for Education Statistics.
Acknowledgements
I am grateful to the Office of Research and Policy Analysis at the Board of Regents of the University System of Georgia, and in particular to Angela Bell, Rachana Bhatt, and Ke Du for making this study possible by sharing their data. Thank you also to Christopher Avery, Celeste Carruthers, David Deming, John Hansen, Whitney Kozakowski, Anne Lamb, Bridget Terry Long, Adela Soliz, Martin West, participants at the EPPE Colloquium at the Harvard Graduate School of Education, APPAM Fall 2016 conference, SREE Spring 2016 conference, and anonymous reviewers for helpful comments and suggestions on earlier versions of this paper. The views in this paper do not reflect those of the College Board. All errors, omissions, and conclusions are mine.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix 1
In Fig. 4, I present evidence that the parallel trends assumption is violated in the full sample. The figure plots enrollment rates in the first 4 years of college for students who entered college several years before the Pell lifetime limit took effect and who enrolled for at least 5 FTE years in college. In panel A, the solid lines show re-enrollment rates for students who were exposed to the new lifetime limit and the dashed lines show the same for students who were not exposed. Students affected and unaffected by the new lifetime limit (based on their lifetime receipt of Pell aid) are denoted by white and black circles, respectively.
Panel A shows evidence of differential changes in early enrollment behavior by Pell receipt status and exposure to the new lifetime Pell limit status. During their first 4 years of college, students who would have been affected by the new lifetime limit but were not exposed re-enrolled at or below the rates of students who were neither affected, nor exposed. By comparison, beginning in term four, affected and exposed students consistently re-enrolled at higher rates than their unaffected but exposed peers. In panel B, I show that these differences are statistically significant.
Because all students in the analysis entered college four or more years before Congress enacted the new lifetime Pell limit, it is impossible that this policy change caused the observed enrollment differences in Fig. 4. Furthermore, in Table 9, I report estimates of compositional differences between High- and Low-Pell students in the full sample. Consistent with the enrollment trends, the results reveal different compositional changes in the two groups over time, which provides further evidence that the parallel trends assumption is rejected in the full sample.
Source: University System of Georgia administrative records
Re-enrollment rates in years 1–4 of college among all students enrolled for 5 or more full-time-equivalent years, by Pell receipt status and exposure to the new lifetime Pell limit status. a Enrollment by term. b Difference-in-differences estimate. The sample is comprised of all students who attended a USG institution for five or more FTE years within 8 years of entry. Within this group, High-Pell students are defined as those who received 5 or more FTE years of Pell and Low-Pell students are defined as those who received less than 5 FTE years of Pell. Summer terms are excluded. In panel B, each point denotes the average enrollment rate difference of High-Pell versus Low-Pell students in the post-policy versus pre-policy period. The dotted lines represent 95% confidence intervals around each estimate.
Appendix 2
Rights and permissions
About this article
Cite this article
Mabel, Z. Aiding or Dissuading? The Effects of Reducing Lifetime Eligibility Limits for Need-Based Aid on Bachelor’s Degree Attainment and Time to Completion. Res High Educ 61, 966–1001 (2020). https://doi.org/10.1007/s11162-020-09600-0
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11162-020-09600-0