Food Stamps: More than a Safety Net?
Social insurance and means-tested programs have a substantial impact on poverty, reducing poverty rates by approximately 14 percentage points. In the short run, these programs – such as Social Security, Unemployment Insurance, Medicaid, Earned Income Tax Credit (EITC), and the Supplemental Nutrition Assistance Program (SNAP) – are effective in alleviating poverty. A growing body of research examines how these US safety net programs impact lives in the long run. In “Long Run Impacts of Childhood Access to the Safety Net,” a National Bureau of Economic Research paper, authors Hilary Hoynes, Diane Whitmore Schanzenbach, and Douglas Almond examine the long-term effects of in utero and childhood exposure to the largest US hunger safety net program. They find positive effects on health and economic self-sufficiency, decades after initial exposure to the program.
Given the current controversy within Congress over the Farm Bill, these research findings are particularly timely. The Farm Bill serves as the US federal government’s primary vehicle for shaping food and agricultural policy. The predominant sparring point among policymakers is the level at which to fund SNAP. Given that upwards of 47.5 million people depend on nutrition assistance through SNAP, research on the long-run effects of the program can bring value to the policy analysis and decision-making process.
Hoynes et al. use the introduction of the Food Stamp program – currently referred to as SNAP – to examine the impact of a positive, policy-driven change in economic resources accessible in utero and during childhood. Rolled out across US counties between 1961-1975, the Food Stamp program expanded resources for low-income households by providing vouchers for the purchase of food. The researchers use the Panel Study of Income Dynamics data to connect family background and county of residence in early childhood to adult health and economic outcomes. They employ a difference-in-differences model using variation in exposure to the Food Stamp program across counties and over birth cohorts to estimate program impacts on health and economic indicators.
Results show that the health and economic impacts of expanding resources in utero and during childhood follow an individual into adulthood. Notably, Hoynes et al. find that childhood access to the program in a high-impact sample of males and females led to robust and statistically significant improvements in metabolic syndrome conditions (i.e., obesity, high blood pressure, heart disease, diabetes). Specific to women, researchers find that exposure to the Food Stamp program in early childhood is statistically associated with an increase in adult economic self-sufficiency (i.e., increase in educational attainment, income, earnings and decrease in receipt of public assistance).
Furthermore, alternative specifications are introduced to test the relationship between the age of exposure and the impact of the Food Stamp program. Findings demonstrate that the timing of the introduction of additional resources is relevant; gains in adult health outcomes are large and increase with exposure to age five. Initial exposure to the program after one is five-years-old leads to minimal improvement in adult health outcomes; however, access to the Food Stamp program in both early and later childhood leads to improved economic outcomes in adult life.
US safety net programs go beyond reducing poverty in the short run. Hoynes et al. show how exposure to the Food Stamp program in childhood leads to improved health and economic outcomes in the long run – findings that are particularly relevant, considering the current Farm Bill negotiations. Quantifying impacts of US safety net programs – both in the short and long run – can be informative to policymakers as they continue to debate how to stabilize long-term budget deficits and strike a balance between generating new revenue and implementing spending cuts.
Feature Photo: cc/(Heinrock)
One thought on “Food Stamps: More than a Safety Net?”
Sorry, comments are closed.