Association for Public Policy Analysis & Management

The Supplemental Nutrition Assistance Program (SNAP) is the largest nutrition assistance program in the United States. Once per month, each SNAP participant is issued a lump sum allocation of benefits that can be redeemed for food. Standard economic theory predicts that SNAP participants will ration their benefits throughout the month in order to smooth consumption and avoid having to skip a meal. However, if participants are present-biased they may consume too much at the beginning of the benefits cycle, leaving themselves unable to afford enough food at the month’s end.

Recently, policymakers have called for states to revamp their SNAP programs so that participants receive benefits on different days from one another. There are two common rationales for such reforms. First, staggering issuance could reduce hassles for consumers and retailers caused by too many customers shopping at the same time. Second, staggering issuance could shift the incidence of the benefit from retailers to consumers. That is, if SNAP participants receive their benefits on different days in the month, retailers will be unable to capture the program's surplus by raising prices during periods of predictably high food demand.

In this paper, we study whether the timing of SNAP issuance affects demand for food among SNAP recipients and, in turn, the pricing decisions of retailers by taking advantage of the substantial variation that exists in SNAP issuance policies by state and across time. For example, in Nevada all SNAP participants receive their benefits on the first of the month, whereas in Missouri the day on which SNAP participants receive their benefits varies by person – a Missourian’s issuance day may fall anywhere between the first and the 22nd of the month. We utilize a large panel of consumption and transaction-level retail data which contains daily expenditure data on food purchases for over 75,000 households as well as scanner data on weekly volume and price information from over 10,000 grocery stores to estimate the impact of SNAP benefit issuance staggering.

We find that food expenditure among SNAP-eligible households is tied closely to the state’s SNAP issuance policy: food expenditures among SNAP participants are cyclical in states that issue all SNAP benefits at the beginning of the month, but relatively constant in states that stagger issuance over the month. However, we find that the timing of SNAP issuance has no effect on the cyclicality of food prices. Our null effect is precisely estimated – a 95% confidence interval excludes price increases greater than 0.2 percent in the week that benefits are issued. Even in areas with high SNAP prevalence and low market concentration, we can reject that prices are at more than 0.5 percent higher in benefit delivery weeks than in other weeks.

 Taken together, our results suggest that while staggering SNAP issuance days over the course of the month is likely to reduce complications associated with surges in customer traffic, such as long lines or difficulty stocking shelves or staffing stores, the policy is unlikely to impact the incidence of the program.