Effect of 2013 SNAP Benefit Cut on Food Security

On November 1, 2013, SNAP benefits were reduced for the first time in the program’s history, as the temporary boost to the benefits from 2009 ARRA expired. This cut caused a family of 4 to lose about $36 a month in benefits. Even though this does not seem like much, it is indeed a significant reduction for a food insecure household as it amounts to less than $1.40 per meal per person (Dean et al. 2013). The temporary boost to the SNAP benefits in 2009 has shown to have improved the food security of the SNAP participants and also have increased their food expenditures (Nord and Prell, 2011). As studies find a large increase in benefits induces households to increase their food expenditures (Nord and Prell, 2011; Beatty and Tuttle, 2014), we can speculate that a cut to benefits may, in turn, induce households to decrease household food expenditure, and therefore, food security.

We examine the effects of the SNAP cut on SNAP participation, food spending, and food security, using Current Population Survey Food Security Supplement (CPS-FSS) for 2012-2014. The December module of CPS-FSS contains information about the ability to fulfill basic food needs, especially for the participants in the Federal and community food assistance programs as well as food spending. The main analysis compares SNAP participation, food spending, and food security in December 2013 (about a month after the SNAP cut) or in December 2014 (about 13 month after the SNAP cut) with the corresponding statistics for December 2012 (about 11 month before the SNAP cut). With difference-in-differences strategy, we conduct analyses for all low-income households, for low-income households by SNAP participation status, and for households with incomes above the SNAP eligibility range but below the U.S. median income. For robustness check, we implement a placebo test, using different years other than 2012-2014 to rule out any significant changes in food security in the absence of change in SNAP benefits. However, direct comparison of SNAP recipients and non-recipients can be confounded by self-selection and under-reporting of SNAP receipts. Given that disadvantageous social economic status such as low education, single or unemployed head predicts SNAP participation, we use these characteristics to define treatment groups to evaluate the impact of the SNAP cut on a wide range of disadvantaged groups.

Previous studies have examined the effects of SNAP benefit increase, which occurred on April 1, 2009, on household’s economic behaviors. However, this analysis is likely polluted by the households who are induced to participate after the benefit boost. The benefit cut poses fewer issues with “selection” into the treatment group because fewer households would drop out due to a decrease than would join due to an increase. We hypothesis that reduced SNAP benefits would significantly reduce food expenditure and thus threaten food security for low-income households as they become further lack of access to sufficient, safe, and nutritious food. This paper is the first that documents the impact of the SNAP benefit cut on participants’ food security and food spending.