Impacts of School Choice Expansion on Public School Finance, Inputs, Voter Tax Preferences, and Achievement

• In Event: The Gap Between Promise and Practice: Three Logics of School Choice Under Scrutiny

Abstract

Over the last several decades, policymakers have worked to expand families’ access to K-12 educational options beyond their neighborhood public school. Much of the scholarship on the impacts of “school choice” has focused on student outcomes. A related, growing empirical literature extends this research on outcomes by also investigating impacts on finances and cost efficiency. But almost no rigorous evidence exists on changes to individuals’ attitudes towards taxation and government spending, even though increasingly “marketized” education systems may decrease enthusiasm for public goods (and the tax policies that fund them). 

In this study, we thus aim to address these gaps in the literature and ask: What is the impact of expanded school choice, specifically through private school vouchers and charter schools, on traditional school districts’ financial expenditures? Do shifts in key school operation inputs explain these expenditure impacts? What are the impacts on districts’ financial revenues? Do shifts in individuals’ preferences regarding government spending and taxation explain these revenue impacts? Finally, what are the impacts on the test achievement of districts’ students?  

To answer these research questions, we focus on Indiana, where 2011 reforms substantially altered the state’s education landscape. Specifically, during the legislative cycle state-level policymakers introduced both a private school voucher program and passed a bill that increased the number of charter school authorizers across Indiana. Consequently, this study uses 2011 as the key inflection point for a difference-in-differences research design, and we identify districts that experience a large policy-induced uptake in private school vouchers and charter school enrollment as “treatment” observations.

Key data come from NCES, SEDA, and the Private School Universe Survey and include school finance, operations (e.g., class size), test achievement, and sociodemographic data. Baseline variables are used in in an elastic net machine learning model to determine which districts in Indiana are highly exposed to school choice. Finally, publicly available resources identify whether and when a district held a referendum to raise property taxes and increase local revenue streams, as well as the outcome of the referendum—we considered these outcomes to capture local voters’ (revealed) preferences for taxation. 

We document four main findings. First, for treatment districts, per-pupil spending decreases across the board. These results appear to be explained in part by increases in class size. Second, per-pupil revenues go down significantly from all sources, mostly from local revenues. Third, local ballot measures that increase property taxes to fund traditional public K-12 schools occur and pass at higher rates in treatment contexts. Combined with the revenue findings, however, it appears that these “revealed” preferences for changes in government taxation do not sufficiently combat declining appropriations. Finally, we show that choice does not make districts more productive. During the post period, though (less funded) treatment contexts initially score higher on standardized tests, this trajectory reverses and becomes substantially negative and significant over time—for several student populations.

Author

• Mark Junxuan Chin, Vanderbilt University
  Presenting Author