This brief was developed using Microsoft Copilot and edited by Charlotte Sutcliffe, Duke undergraduate research assistant; for full text and references see
Gilraine, M., Petronijevic, U., & Singleton, J. D. (2023). School choice, competition, and Aggregate School Quality. Retrieved from https://www.nber.org/papers/w31328
Background:
This study investigates the cumulative effects of charter school expansion on student achievement, focusing on North Carolina’s 2011 removal of its statewide cap on charter schools. The authors analyze direct attendance effects and indirect competitive responses from traditional public schools. Using geocoded student-level data from North Carolina’s three largest commuting zones, the model incorporates heterogeneity in household preferences, school quality (measured via value-added to test scores), and school types (traditional vs. non-traditional curricula). The study estimates how public schools adjust quality in response to charter competition.
Findings:
The removal of the charter cap led to an increase in average public school quality (0.01 standard deviations in test scores), translating to a $2,000 increase in lifetime earnings per student. This gain was driven not by the quality of new charter schools - many of which were lower-performing - but by competitive pressure that incentivized public schools to improve. Notably, public schools only responded positively when nearby charter entrants offered traditional curricula. Non-traditional charters, which often drew students from private schools, did not elicit similar public school improvements.
The model reveals that students most sensitive to school quality were least likely to choose charter schools, especially non-traditional ones. Simulations show that the average charter school would lose fewer than 5% of its students if public schools improved quality by 0.05 standard deviations, with traditional charters being more vulnerable to such competition. Removing all charter schools would reduce average student test scores by 0.02 standard deviations - comparable to the effect of replacing the bottom 5% of teachers.
Further simulations estimate the expected return of approving a marginal charter school. While the direct effect on students attending the new charter may be negative, the competitive response from public schools yields a net positive effect. These returns are significantly higher when charters are located in low-income or underserved areas and when they follow traditional curricula.
Takeaways:
Charter school expansion can yield positive outcomes primarily through competitive effects on public schools, even when new charters are not higher quality. Policymakers should consider not only the quality of charter applicants but also their curriculum type and location. Strategic placement of traditional charters in disadvantaged areas can amplify benefits. Authorization frameworks that incorporate these factors can enhance the effectiveness of school choice policies.