Can Local Scholarship Programs Improve College Finances and Academic Success for Students?

Bueno, Mawi, Page, and Smith (2026) examine how place-based scholarships affect student borrowing and early academic outcomes. They ask whether receiving the Achieve Atlanta scholarship changes loan use and first-semester performance. Using linked administrative data from Atlanta Public Schools, Achieve Atlanta, and Georgia’s public college systems, they compare recipients to similar non-recipients. They find recipients are 7 percentage points less likely to borrow (18% reduction) and borrow about $608 less (43% lower). Academically, they earn GPAs about 0.11 points higher (5%) and complete roughly one additional credit hour, indicating modest but meaningful improvements in early college success.

Why This Article Was Selected for The Policy Scientist

This article addresses a central policy question: whether financial aid can reduce debt burdens while improving academic progress, a concern that has become more salient as tuition growth and student borrowing continue to shape access to higher education. Understanding these dual effects is critical for evaluating whether place-based interventions can alter both financial and human capital trajectories. The authors have contributed extensively to this literature, and this study extends prior work on persistence by examining mechanisms. The linked administrative dataset is unusually strong, enabling detailed financial and academic measurement. However, reliance on OLS limits causal inference; future work using quasi-experimental or experimental designs would strengthen confidence. Generalizability is plausible but context-dependent.

Full Citation and Link to Article

Bueno, C., Mawi, Z. E., Page, L., & Smith, J. (2026). How do place-based scholarships affect student borrowing and academic outcomes? Lessons from Atlanta. Education Finance and Policy. https://doi.org/10.1162/EDFP.a.444

Central Research QuestionThis article examines whether and how receipt of a place-based scholarship—specifically the Achieve Atlanta program—affects students’ financial behavior and academic performance during college. The core inquiry extends beyond well-established effects on access and persistence to focus on mechanisms: do scholarship recipients borrow less, rely on different types of financial aid, and perform better academically in their initial college experience? The authors center their analysis on first-semester outcomes, motivated by prior evidence that scholarship recipients exhibit higher persistence and completion rates. By isolating early financial and academic indicators, the study seeks to clarify whether reductions in borrowing and improvements in academic performance plausibly contribute to longer-term attainment effects. The research question is therefore both descriptive and explanatory, aiming to illuminate how financial aid structures influence student decision-making and early academic trajectories.

Previous LiteratureThe study situates itself within a substantial body of research on financial aid and postsecondary outcomes. Prior work has consistently found that grant aid, including place-based “Promise” scholarships, increases college enrollment, persistence, and degree completion. Foundational studies on programs such as the Kalamazoo Promise and Tennessee Promise demonstrate positive effects on access and attainment, while meta-analyses indicate modest but consistent gains across a range of grant aid interventions. However, the literature has largely emphasized enrollment and completion, leaving less clarity regarding the mechanisms that produce these outcomes. In particular, there is limited evidence on how scholarships affect borrowing behavior and academic performance during college. Existing studies on borrowing suggest that some aid programs reduce student debt, but findings are sparse and context-specific. Similarly, research on academic outcomes shows mixed or limited effects on GPA and credit accumulation. This article contributes by integrating these dimensions—financial and academic—within a single empirical framework, thereby extending the literature from outcomes to processes.

DataThe analysis draws on a comprehensive, linked administrative dataset combining records from Atlanta Public Schools (APS), the Achieve Atlanta program (AATL), the University System of Georgia (USG), and the Technical College System of Georgia (TCSG). This integrated dataset allows the authors to observe student demographics, high school performance, scholarship receipt, detailed financial aid packages, loan borrowing behavior, and college academic outcomes. The sample includes 5,273 students graduating from APS between 2016 and 2021 who enrolled in a public postsecondary institution in Georgia, capturing approximately 71 percent of scholarship recipients in the relevant cohorts. The dataset is notable for its granularity, particularly in linking financial aid components—such as subsidized, unsubsidized, and PLUS loans—with academic outcomes like GPA and credit accumulation. Such comprehensive administrative linkage is relatively rare in higher education research and addresses a common limitation in prior studies that rely on fragmented or single-institution data sources. The richness of the dataset enhances the precision of measurement and enables a more detailed examination of student financial and academic behavior.

MethodsThe empirical strategy relies primarily on ordinary least squares (OLS) regression and linear probability models to estimate the relationship between scholarship receipt and outcomes of interest. The authors include an extensive set of controls, including high school GPA (modeled flexibly), demographic characteristics, financial need indicators (such as Pell Grant receipt and Expected Family Contribution), and fixed effects for high school, cohort, and college attended. This specification is designed to compare scholarship recipients to observationally similar non-recipients within the same institutional and cohort contexts. To assess robustness, the authors conduct several supplementary analyses, including restricting samples based on full-time enrollment and financial need thresholds, as well as implementing propensity score matching. They also employ coefficient stability tests to evaluate the potential influence of unobserved confounders. Despite these efforts, the authors explicitly characterize their findings as descriptive rather than causal, acknowledging the absence of a credible identification strategy such as a regression discontinuity or randomized design. This limitation reflects constraints in the available data but is transparently addressed.

Findings/Size EffectsThe results indicate that scholarship receipt is associated with meaningful differences in both financial and academic outcomes during the first semester of college. Financially, recipients are approximately 7 percentage points less likely to take out any student loans, representing an 18 percent reduction relative to non-recipients. Among those who borrow, recipients take on about $608 less debt on average, a 43 percent decrease. These reductions are driven primarily by declines in unsubsidized and PLUS loans, which carry less favorable repayment terms. The findings also suggest that recipients are offered fewer loans and accept a smaller share of the loans available to them, indicating both supply-side and demand-side mechanisms.

Academically, scholarship recipients outperform comparable peers across several metrics. They earn GPAs that are approximately 0.11 points higher, equivalent to a 5 percent increase relative to the control mean. In addition, they attempt and complete more credit hours, with recipients earning roughly one additional credit hour in their first semester. These effects, while modest in absolute terms, are substantively meaningful given their concentration in the initial college period. The combination of reduced borrowing and improved academic performance provides a plausible explanation for previously observed gains in persistence and completion. The results are consistent across multiple model specifications, reinforcing their robustness within the constraints of the research design.

ConclusionThis study provides a detailed examination of the financial and academic mechanisms through which place-based scholarships may influence college success. By leveraging a uniquely comprehensive administrative dataset, the authors extend the literature beyond enrollment and completion outcomes to capture early behavioral responses to financial aid. The findings suggest that scholarship support reduces reliance on student loans—particularly higher-cost borrowing—and is associated with modest improvements in academic performance. These patterns offer insight into how financial aid may alter both the economic and academic dimensions of the college experience. However, the reliance on multivariate regression limits the ability to draw causal conclusions. While the authors employ extensive controls and robustness checks, future research would benefit from designs that provide stronger causal identification, such as randomized controlled trials or quasi-experimental approaches. Nonetheless, the study represents a substantive contribution by clarifying the pathways through which financial aid policies may operate, particularly in urban, place-based contexts.