School Choice Policies and Market Competition: Implications for the Public Education Sector

This dissertation studies the effect of school choice policies with a focus on charter schools and school vouchers. Exploiting the time-varying charter competition driven by the Michigan 2011 cap-lifting policy, the first chapter examines the effect of charter competition on traditional public school districts’ cost efficiency in two ways. The first part estimates the effect of the cap-lifting policy on districts’ resource allocation. The second part extends the policy analysis to disentangle the separate effects of potential competition and actual competition so that we can understand the dynamic effects of the policy. Results show that the cap-lifting policy on average pressures school districts to reduce non-instructional expenditure. The dynamic analysis indicates that such a change in resource allocation is the districts' attempt to deter future competition and extend sustainability. Further analysis also shows that actual charter competition changes districts’ student composition where the percentage of students in the English Language Learner program increases 2.5 percentage points. Little evidence shows that such student sorting results in a significant change in overall academic performance in the short term. The findings suggest that districts attempt to respond to the policy-driven charter competition by prioritizing instruction and reducing cost inefficiency. The second chapter investigates the effect of a Chilean targeted school voucher program. In 2008, Chile implemented a targeted voucher program that increased funding for disadvantaged students in public and participating private schools by approximately 50%. Evidence shows that disadvantaged students did make substantial fourth-grade test score gains exceeding 0.2 standard deviations. However, focused on contemporaneous changes in family background, as well as limited school input, market competition, and school switching responses, the analysis raises doubts that a program-induced improvement in school quality accounts for this convergence.