Where Do the Profits from College Football and Basketball Actually Go?

Garthwaite et al. (2025) examine who benefits from the economic rents generated by college sports under amateurism. They ask how revenue from football and men’s basketball is redistributed within athletic departments. They analyze panel data on revenues and expenditures from Power Five athletic programs (2006–2019), along with player-level demographic data. They find substantial rent-sharing: about $0.31 of each additional dollar is reinvested in revenue sports, while roughly $0.11 goes to non-revenue sports and up to $0.20 to facilities. They show these transfers disproportionately shift resources away from athletes who are more likely to be Black and from lower-income backgrounds.

Why This Article Was Selected for The Policy Scientist

This article addresses a central policy question: how institutions allocate economic rents when formal compensation is constrained. That issue extends well beyond college athletics to labor markets, regulation, and public finance, where rules often shape who captures value. It is especially timely given recent changes to athlete compensation rules and ongoing legal challenges to amateurism. Craig Garthwaite and his coauthors have an established record in health and labor economics, lending credibility to the analysis. The dataset is unusually rich, combining longitudinal financial records with demographic data, though it is limited to Power Five programs. The fixed-effects and IV strategies support causal interpretation.

Full Citation and Link to Article

Garthwaite, C., Holz, N., Keener, J., & Notowidigdo, M. J. (forthcoming). Who profits from amateurism? Rent-sharing in modern college sports. American Economic Journal: Applied Economics. https://www.aeaweb.org/articles?id=10.1257/app.20220595

Central Research QuestionThis article examines how the economic rents generated by college football and men’s basketball are distributed within athletic departments operating under an amateurism model. Specifically, it asks whether increases in revenue from these “revenue sports” are retained within those programs or shared with other components of the athletic department, including non-revenue sports, coaching staff, administrative personnel, and facilities. A related dimension of the research question concerns the distributional consequences of this rent-sharing: which groups of athletes benefit from these reallocations, and whether the structure of amateurism results in systematic transfers across athletes with different socioeconomic and racial backgrounds. The study situates this question within a broader labor economics framework, treating college athletics as a setting where compensation constraints generate rents that must be allocated through institutional mechanisms rather than market wages.

Previous LiteratureThe paper builds on a well-established literature examining the economics of sports and the concept of rent-sharing in labor markets. Earlier work, such as Gerald Scully (1974), established the concept of marginal revenue product (MRP) in professional sports, showing that athletes often generate value exceeding their compensation. Subsequent research extended this logic to college athletics, demonstrating that football and basketball players at elite programs produce substantial economic surplus relative to scholarship compensation. More recent labor economics research, including work by Patrick Kline and Arnaud Lamadon, has developed empirical strategies to estimate rent-sharing elasticities using firm-level data. The current study integrates these approaches by applying modern rent-sharing methods to intercollegiate athletics. It also relates to literature on Title IX and gender equity, as well as studies of racial and socioeconomic disparities in college sports participation. Relative to prior work, the article advances the literature by quantifying how rents are redistributed within athletic departments and by linking those redistributions to observable demographic patterns among athletes.

DataThe analysis relies on multiple high-quality datasets that collectively provide a detailed picture of athletic department finances and athlete characteristics. The primary financial data come from the Equity in Athletics Data Analysis (EADA) and the Knight Commission’s College Athletics Financial Information (CAFI) database, covering Power Five conference schools from 2006 to 2019. These data include sport-level revenues and expenditures, as well as detailed breakdowns of spending on coaches, administration, and facilities. A key strength of the dataset is its longitudinal structure, which allows for within-school comparisons over time.

To examine distributional effects, the authors construct a novel dataset by scraping athlete rosters and linking them to high school and neighborhood-level socioeconomic characteristics derived from U.S. Census data. This enables the identification of differences in background characteristics across athletes in revenue versus non-revenue sports. While the financial data are comprehensive for public institutions, coverage is somewhat more limited for private universities, and the analysis focuses primarily on Power Five programs, which may constrain generalizability to smaller or less commercially intensive athletic programs. Nonetheless, the data represent one of the most detailed and integrated sources available for studying the economics of college athletics.

MethodsThe empirical strategy is grounded in panel data analysis using fixed-effects regression models that exploit within-school variation over time. The primary specification estimates rent-sharing elasticities by regressing various categories of spending on revenue generated by football and men’s basketball, controlling for school and year fixed effects. This approach isolates how changes in revenue within a given institution translate into changes in spending across different components of the athletic department.

To strengthen causal interpretation, the authors implement several complementary strategies. First, they conduct event-study analyses within a difference-in-differences framework to assess whether changes in spending follow increases in revenue without evidence of pre-existing trends. Second, they employ an instrumental variables (IV) approach using conference-level revenue shocks—such as media rights payments and postseason distributions—as exogenous sources of variation in revenue. These instruments are plausibly unrelated to individual school decisions about spending, providing a stronger basis for causal inference.

While the study does not use randomized controlled trials, it incorporates multiple quasi-experimental techniques that are standard in applied microeconomics. The combination of fixed effects, event studies, and IV estimation represents a rigorous attempt to address endogeneity concerns, though the reliance on observational data means that causal claims remain contingent on identifying assumptions.

Findings/Size EffectsThe results provide clear evidence of substantial rent-sharing within college athletic departments. The authors estimate an own-sport spending elasticity of approximately 0.82, implying that roughly $0.31 of each additional dollar generated by football and men’s basketball is reinvested in those same sports. At the same time, significant portions of revenue are redistributed to other areas: approximately $0.11 per dollar is allocated to non-revenue sports, including both women’s sports and other men’s sports.

Additional rent-sharing occurs through increased spending on coaches’ salaries, administrative compensation, and athletic facilities. The estimated elasticities imply that about $0.03 per dollar goes to football coaches, $0.03 to other coaches, $0.09 to administrative personnel, and $0.20 to facilities. These findings indicate that a large share of incremental revenue is absorbed by the broader athletic department rather than accruing to athletes themselves.

Importantly, the study finds minimal effects on institutional support from the university, suggesting that increased athletic revenue does not substantially alter transfers between the athletic department and the broader institution. The distributional analysis reveals that these patterns have meaningful demographic implications: athletes in revenue sports, who are more likely to come from lower-income backgrounds and to be Black, effectively generate resources that are redistributed to athletes in non-revenue sports, who are more likely to be White and from higher-income backgrounds.

ConclusionThe article provides a comprehensive account of how amateurism shapes the allocation of economic rents in college athletics. By documenting the magnitude and direction of rent-sharing, it demonstrates that the financial surplus generated by revenue sports is systematically redistributed across the athletic department rather than being retained by the athletes who generate it. This contributes to a broader understanding of how institutional constraints on compensation influence resource allocation in labor markets.

Methodologically, the study represents a strong application of modern causal inference techniques within an observational setting, combining fixed effects, event-study analysis, and instrumental variables. The consistency of results across these approaches strengthens confidence in the findings, although the absence of experimental variation leaves open the possibility of residual confounding.

In terms of external validity, the focus on Power Five programs suggests that the results are most applicable to highly commercialized athletic environments, though the underlying mechanisms may extend to other settings where compensation constraints generate rents. Overall, the study advances the literature by integrating detailed financial data with distributional analysis, offering a rigorous and policy-relevant examination of the economic structure of college sports.a