The Impacts of SNAP Participation on Mortality Rates

Prior research has examined the effects of the Supplemental Nutrition Assistance Program (SNAP, formerly the Food Stamp Program) on many different outcomes, but there are relatively few studies investigating how SNAP receipt impacts adult health outcomes. Given recent research finding a significantly higher risk of death for food stamp recipients compared to eligible non-recipients, these health effects may be significant. However, no prior study has satisfactorily examined the causal effects of food stamps on adult mortality rates.

This paper examines the effects of SNAP participation on adult health as measured by various mortality rates. Specifically, we use an instrumental variables approach to estimate the effects of county-level SNAP participation rates on various county-level mortality rates with a focus on adult rates. Using a nationally representative sample of households from the Survey of Income and Program Participation (SIPP), we construct a novel simulated SNAP eligibility variable making use of variation in many kinds of state SNAP eligibility policies. This variable is the proportion of the sample eligible for SNAP under the rules in each state and year and represents the overall “generosity” of each state’s SNAP policies. We use this simulated variable to instrument for SNAP participation. Information on SNAP participation rates and overall, subpopulation, and cause-specific mortality rates from 1996 to 2015 are gathered from the Census Small Area Income and Poverty Estimates (SAIPE) and Centers for Disease Control Wide-ranging Online Data for Epidemiologic Research (CDC WONDER) systems, respectively. We conduct analyses using the whole county sample and samples stratified by pre-treatment poverty rates in order to investigate the possible heterogeneity in effects of aggregate SNAP participation.

Our results indicate that higher SNAP participation reduces most mortality rates, with generally stronger reductions in the mortality rates of adults aged 20 to 64 and in lower-poverty subsamples. Overall, we find that a one percentage point increase in SNAP participation reduces the total mortality rate by 4.8 deaths per 100,000 population.