Whether charter school expansion has historically improved overall school quality remains a subject of debate. More critically, much of the existing literature does not provide ex ante policy analysis capable of informing charter authorization decisions, such as whether to approve new charter applications or impose caps on charter growth.
Michael Gilraine, Uroš Petronijević, and John D. Singleton address this gap by developing and estimating an equilibrium model that links charter school presence and characteristics to both the level and distribution of student achievement. The model incorporates household heterogeneity and variation in school quality that shape enrollment choices, while also accounting for competitive responses by traditional public schools.
Using this framework, the authors conduct counterfactual simulations to evaluate the aggregate academic effects of charter expansion in North Carolina. They also examine how policy can influence these effects by shaping the type, quality, and geographic placement of charter schools. Their estimates indicate that charter expansion increased the average public school’s value-added by roughly 0.01 standard deviations in test scores.
Demand estimates suggest that the average student is willing to travel approximately 0.1 miles for a 10-percentile increase in school quality. The analysis further shows that if public school quality rose by 0.05 standard deviations, the average charter school would lose fewer than 5 percent of its students, with traditional charters experiencing larger enrollment losses than non-traditional, specialized charters.
Finally, the authors evaluate whether authorizers should approve marginal charter applications. They find consistently positive academic returns to approval, particularly when charters are steered toward disadvantaged neighborhoods, where expected gains are roughly twice as large as in higher-income areas.
- Working Paper: https://www.nber.org/papers/w3...