The Effects of Participation In the Supplemental Nutrition Assistance Program On the Material Hardship of Low-Income Families with Children

BACKGROUND AND PURPOSE: This study examines the effects of participation in the Supplemental Nutrition Assistance Program (SNAP) on measures of the non-food material hardship of low-income families with children. SNAP may reduce non-food material hardship by allowing recipients to reallocate resources originally directed toward food to other expenses. Wu and Eamon (2010) find that public benefits receipt, including SNAP, is associated with increased material hardship among low-income households with children, such as inadequate housing and difficulty meeting essential expenses. Their analysis does not, however, address selection issues related to program participation, the likely source of these positive correlations. Selection issues are well-documented in studies on SNAP and food insecurity, a key USDA outcome. Only recently have studies using an instrumental variables approach or other sophisticated techniques found a negative relationship between SNAP and food insecurity (Ratcliffe & McKernan, 2010). We believe ours is the first study to use an instrumental variables approach to examine the effects of SNAP on non-food material hardship. We analyze the ability of low-income families with children to pay essential expenses including housing and utility costs. We hypothesize that SNAP should reduce non-food material hardship. METHODS: Data are drawn from the Survey of Income and Program Participation, a nationally representative survey collected by the U.S. Census Bureau. SIPP gathers data on demographics, income, program participation, family structure, and employment. Our study period is 1998-2005. Our sample is families with children at or below 150% of poverty. We use an instrumental variables method, which attempts to correct for unobservable differences between participants and non-participants. Because our key variables are endogenous dummies, we utilize a bivariate probit model. As instruments, we use changes in state SNAP policies including recertification period lengths and biometric requirements. RESULTS: We find that SNAP reduces the risk that families will experience difficulty meeting essential household expenses by 59%. In particular, SNAP reduces the risk that families will fall behind in their rent/mortgage by 39%. We find some evidence that SNAP decreases the probability that families will fall behind on their utility bills. Our results are robust to numerous sensitivity tests regarding selection of sample and instruments and definitions of SNAP participation. CONCLUSIONS: SNAP participation allows families to reallocate resources otherwise directed toward food to other essential expenses, broadly affecting family well-being. SNAP's prominence among means-tested programs suggests it should be evaluated using a broader set of material hardship outcomes than previously done. Future research might examine whether families change the allocation of SNAP resources based on external circumstances: do they allocate more toward utilities during winter? Future research could also examine whether SNAP buffers families from material hardship when they experience job loss. Overall, we hope this study begins a broader examination of the effects of SNAP on non-food material hardship outcomes. Ratcliffe & McKernan. (2010). How much does SNAP reduce food insecurity? ERS-USDA, Contractor and Cooperator Report No. 60. Wu & Eamon. (2010). Does receipt of public benefits reduce material hardship in low-income families with children? Children and youth services review, 32(10).