State and Racialized Heterogeneity in Medicaid Burdens and Economic Security around Childbirth

• In Event: Beyond Heterogeneity: Methodological Challenges and Opportunities Researching Policy at the Nexus of Race and Place

Abstract

In 2022, 41% of births in the U.S. were covered by Medicaid: a federal-state partnership program that provides health insurance to low-income Americans. While federal statute sets parameters around administration, states are given autonomy to decide which populations and services to cover, how to deliver care, how much to reimburse providers, and more (Rudowitz et al., 2024). State decision-making around program eligibility and processes is consequential for low-income pregnant women and children; for example, those living in states with fewer administrative burdens—like shorter waiting times, waived face-to-face interview requirements, and longer recertification periods—are more likely to participate in Medicaid (Fox et al., 2023). These rules are both distributional and consequential (Herd & Moynihan, 2013). They are consequential because they may disrupt access to an entitlement program that reduces economic hardship around childbirth (Allen et al., 2024). They are also distributive in that they disproportionately burden racialized groups and those with fewer social and economic resources (Christensen et al., 2020; Ray et al., 2019).  

This study asks two questions: what are the effects of state Medicaid program generosity on women’s economic security around childbirth? How do these effects differ by mothers’ race and ethnicity? I will measure state Medicaid burdens using an index capturing state-year variation in Medicaid eligibility, administrative burdens, and benefit levels (Fox et al., 2020). I use data from the 2008-2023 panels of the U.S. Survey of Income and Program Participation, a nationally-representative and longitudinal survey, to identify women who gave birth while participating in the survey and met their state’s income eligibility requirements for Medicaid prior to pregnancy. I create multiple indicators of economic security including the Economic Security Index, indicators of food insecurity and fiscal hardship, and indicators of income poverty using the official and supplemental poverty measures. 

I use different quasi-experimental methods to estimate the effects of Medicaid generosity on women’s economic security by leveraging variation within states and over time in Medicaid program rules. One uses a differences-in-differences approach with a continuous treatment (Callaway & Sant’Anna, 2024) to examine whether low-income women giving birth in states with more generous administrative rules following the Affordable Care Act (“higher-dose” states”) have different outcomes than women living in states with more restrictive rules (“lower-dose” states). The first set of models predicts Medicaid enrollment as a first stage. The second set of models predicts economic security around childbirth. I include state and time fixed effects and control for a host of time-varying person- and state-level covariates. To answer my second research question, I estimate a series of models interacting the treatment indicator with an indicator of identifying as non-Hispanic White, non-Hispanic-Black, Hispanic of any race, or non-Hispanic Asian.  

Altogether, this study engages with larger questions around how researchers might leverage survey data and measures of state policy variation to understand the distributive and consequential outcomes of state policy administration. It also makes a substantive contribution to research on Medicaid and the safety net for low-income women during a critical time period: the transition to parenthood.

Author

• Anna Walther, University of Wisconsin-Madison
  Presenting Author