Estimating SNAP purchasing power and its effect on participation

The Supplemental Nutrition Assistance Program (SNAP) aims to improve food security among low-income households and is one of the largest safety net programs in the United States. SNAP benefits are legislated at the federal level and fixed across the contiguous United States. Hence, due to regional food price differences, the real value of SNAP benefits is unequal and could have consequential impacts on program participation. This paper estimates the extent to which SNAP purchasing power differs across states by constructing the first spatially and temporally consistent Thrifty Food Plan (TFP) price index using retail scanner data for the period 2006–2016. We find that the difference in the real value of SNAP benefits between the highest and the lowest cost states ranges from 5% to 9% for a household of four during the sample period. Subsequently, we estimate the effect of real SNAP benefits on program participation. Our results show that a 10% increase in SNAP purchasing power leads to a 0.9 percentage point increase in the SNAP caseload per capita and an 8.1 percentage point increase in the SNAP caseload per eligible individual. We show that these effects would be overlooked if the TFP price index is not corrected for expenditure and outlet biases.