Breaking Boundaries: Competitive Effects of Inter-District School Choice

• In Event: School Choice, Student Mobility, and the Geography of Opportunity

Abstract

Inter-district school choice (IDSC), which allows students to attend public schools outside their residential districts, is supported by policymakers who believe it expands access to high-quality education through two channels: allowing participants to attend higher-quality schools and creating competitive pressure for school districts to improve. While a large literature estimates the direct effects of IDSC participation (Bifulco et al., 2009; Cowen & Creed, 2017; Deming et al., 2014), less is known about how school districts respond to IDSC enrollment competition. Because most students remain enrolled in their residential districts, these competitive effects are central to understanding broader policy consequences. This paper examines how reducing household barriers to IDSC participation in Indiana affects school districts' enrollment and academic quality.

House Enrolled Act 1001 shifted responsibility for school operating funds from local property taxes to the state and, as an unintended consequence, eliminated transfer tuition for nonresident students beginning in the 2010 school year, thereby removing a major household barrier to IDSC participation. Because districts can both lose resident students and attract incoming transfers, removing transfer tuition may increase, decrease, or leave a district’s net enrollment demand unchanged. To measure each school district’s IDSC enrollment-threat exposure, I compute the enrollment-weighted average difference between neighboring districts’ pre-policy achievement and the district’s own achievement. Higher exposure indicates that a district is surrounded by higher-achieving competitors and faces stronger predicted enrollment-loss pressure.

I estimate continuous-treatment difference-in-differences models that compare post-policy changes in enrollment and value added across districts with different levels of IDSC enrollment-threat exposure. The primary specification includes district and year fixed effects, while accounting for cross-market spillovers by controlling for the average exposure faced by neighboring districts. Because year fixed effects absorb any response common to all districts, the estimates identify how the policy’s effects vary across the exposure distribution rather than its aggregate effect on Indiana school districts. I also estimate event-study specifications of my primary model, which test for differential pre-trends and trace post-policy dynamics.

A one-standard-deviation increase in exposure leads to a 2.05-percentage-point decline in enrollment due to shifts in net transfer students, while value added increases by 0.0077 test-score standard deviations. The value-added effect is concentrated in elementary grades, where a one-standard-deviation increase in exposure shifts value added upward by 0.0154 standard deviations. Moving from the 25th to the 75th percentile of the exposure distribution implies an elementary value-added effect equal to 8 percent of the baseline achievement gap between districts at those percentiles. Event studies show no differential pre-policy trends; after transfer tuition is removed, higher exposure predicts increasingly negative enrollment changes over time, while the elementary value-added effect emerges immediately and persists throughout the post-policy period.

Consistent with an enrollment-threat mechanism, elementary value-added effects are largest among districts with below-median pre-policy enrollment, where enrollment losses are more consequential relative to district size, and below-median acreage, where shorter travel distances reduce households’ remaining transfer costs. Together, these findings show that competitive responses to IDSC depend on the extent expanded access creates credible enrollment threats.

Author

• Pietro B Pellerito, University of Notre Dame
  Presenting Author