Universal Vouchers, Private-School Pricing, and School-Finance Challenges

Abstract

This thesis examines whether universal school vouchers incentivize private schools to raise their tuitions, testing the “Bennett Hypothesis” in the K-12 context. We conduct an event study of Florida’s Family Empowerment Scholarship, the nation’s largest voucher program. Because no centralized database for tuition exists, we construct an original dataset by manually scraping current and archived school websites for 350 schools across Florida and Texas over an eight-year period. Supplementing these data with enrollment and demographic data from the Private School Survey and American Community Survey, we use a difference-in-difference framework to estimate the voucher’s effect on tuition, using Texas as an untreated control. We estimate this effect across a range of specifications, including weighted and fixed-effect regressions and regressions stratified by religious affiliation, neighborhood income, and school capacity.

In the aggregate, we find no statistically significant effect of voucher introduction on private-school tuition. This null result masks heterogeneity: schools in low-income neighborhoods raise tuitions at greater rates than those in wealthier areas, and voucher onset is correlated with a highly significant rate of application-fee introduction. Together, these findings suggest that voucher funding passes through to families rather than being captured by schools. Rather than raising sticker prices, private schools appear to be managing voucher-induced demand at least in part through pre- matriculation application fees. These findings on voucher incidence have meaningful implications in policy debates amid widespread, partisan voucher introduction across the United States.