The Impact of the Medicaid “Free Care” Reversal on School District Finances and Staffing

• In Event: Investigation of Policies for Increasing Children’s Receipt of Medicaid-funded School-based Health Services

Abstract

The federal Centers for Medicare & Medicaid Services opened the door to expanded school health funding and services in 2014 when they reversed their “free care” policy. The policy effectively barred schools from seeking Medicaid reimbursement for services outside of special education plans. The 2014 free care policy reversal (henceforth FCR) enabled schools to bill for services provided to any Medicaid-enrolled student, paving the way for potential student health service expansions.

Before schools can take advantage of FCR, states must update their Medicaid rules. States began these updates in 2015, and 26 states have now implemented FCR.

Despite FCR’s growing adoption, little research has examined its effects. The current study provides the first comprehensive analysis of the impact of FCR on school finances and staffing. Specifically, we investigate the impact of FCR on the following four outcomes: school district revenue, student support expenditures, student-to-support staff ratios and student-to-counselor ratios.

We use administrative panel data on the universe of U.S. public K-12 school districts from 2010-11 to 2022-23. We use a group-time average treatment effect (ATTgt) difference-in-differences (DD) modeling framework to estimate FCR effects. The ATTgt approach estimates FCR effects separately for each group of states that adopted the policy in the same year, allowing us to examine heterogeneity by adoption year and aggregate effects by both adoption-relative time and calendar time. The approach also avoids bias driven by staggered treatment adoption. Additionally, we estimate triple-difference (DDD) models in this framework, comparing outcome differences between districts in the top and bottom quintiles of Medicaid enrollment (50% and 20% respectively) within each state.

DDD models provide evidence of increased student support expenditures in high-Medicaid districts among states that implemented FCR pre-pandemic. The overall ATT shows an increase of $22 per student in support expenditures among states that adopted FCR from 2015 to 2019, p < .10. Event study estimates with compositionally balanced pre- and post-adoption periods find increases of $13 to $22 per student across the first three years of policy adoption, with the latter two years statistically significant at the p < .05 level. No pre-trends are statistically significant. Calendar time aggregation reveals that effect estimates for these pre-pandemic adopters remained fairly stable through the pandemic, ranging from $24 to $30 per student from 2017-18 to 2021-22, aside from a slight dip to $16 during the 2019-20 school year.

Estimates from other models were small and noisy. Coefficients for staffing outcomes were in the expected direction, i.e. reductions in the student-staff ratios, but were not statistically significant. School districts commonly partner with community health providers for student health services which may explain why school staffing models were noisier than expenditure models.

Additionally, substantial heterogeneity exists among FCR-adopter states. We are currently exploring this heterogeneity and working to obtain more detailed state staffing and Medicaid revenue data to gain a clearer understanding of state-specific FCR responses.

Authors

• David Loeb, University of Pennsylvania
  Presenting Author

• Katie Pullom, University of Pennsylvania
  Non-Presenting Co-Author