ACA Expansion and the Joint Processing of Medicaid and SNAP: Spillover Effects on Food Stamps Participation

A primary aim of the ACA was to reduce uninsurance rates. Expanding Medicaid to low-income adults was envisioned as an important tool to achieve this, and the deadline for states to implement expansion was 2014. A Supreme Court decision rendered the policy an option, however. Not all states participated in the expansion, and there was uneven timing among states that did (Kaiser Family Foundation 2017). Compared to non-expanding states, expansion states saw greater reductions in uninsurance, including among hard-to-reach subpopulations (Kaester et al. 2017) (McMorrow et al. 2015). Beyond insurance coverage, however, Medicaid expansion could produce effects in other programs that target low-income populations.

The Supplemental Nutrition Assistance Program (SNAP), known as food stamps, is a federal entitlement that provides monthly food purchasing assistance to families with income under 130 percent of federal poverty. Since welfare reform in the late 1990s, SNAP has emerged as a key counter-cyclical program (Bitler and Hoynes 2016). As with other public programs, however, takeup is never complete among the eligible (Blank and Ruggles 1996) (Leftin and Wolkwitz 2009). Eligibility requirements for SNAP are determined at the federal level, but states vary in the details of how they administer the program. In particular, some states jointly process SNAP and Medicaid applications, ensuring that individuals applying for one of the programs are considered for the other. We hypothesize that the Medicaid expansion under the ACA is a potential channel of access to food stamps for low-income adults in states with joint processing. The hypothesized effect would operate through reduced transaction costs and improved information for applicants.

We use data from the Current Population Survey December Basic Survey and the Food Security Supplement from 2010 to 2016 (IPUMS-CPS). In the period covered by this study, 32 states including DC expanded Medicaid. Our population includes approximately 129,000 individuals across all states with incomes under 138 percent of the federal poverty line, the threshold for adult Medicaid under the ACA. We use a triple-differences strategy to identify the combined effect of state Medicaid expansion and joint processing of Medicaid and SNAP on food stamps participation above and beyond the Medicaid expansion’s effect for states that do not process jointly. Our preferred model controls for individual characteristics, state-by-year unemployment, and includes full sets of state and year fixed effects. Our estimate finds a 4 percentage point increase in food stamps participation attributable to joint processing in Medicaid expansion states, or an increase of about 8 percent above the 46 mean percent participation rate in our study population. This estimate is significant at the 95 percent level of confidence using standard errors robust to heteroskedasticity and clustered by state.

Our results suggest that the Medicaid expansion—in addition to increasing health insurance coverage—connects low-income adults and their families to food stamps. There are likely to be resulting positive consequences for poverty alleviation and nutrition. Next steps include an analysis of how these effects differed for families with children versus those without, as well as stratified estimates for different low-income groups.