Supply-Side Responses to SNAP Benefit Changes and the Effect on the Local Food Environment

The Supplemental Nutrition Assistance Program (SNAP) provides near cash-benefits to over 40 million American. Program benefits are redeemable for nearly any unprepared food or beverage at retailers who meet the United States Department of Agriculture’s (USDA) stocking requirements.

Few policy changes have impacted the local food environment for SNAP participants. In 2009, however, the American Recovery and Reinvestment Act (ARRA) funded a temporary 19 percent average increase in SNAP benefits. The change in benefit size has been established to have an impact on the number of SNAP-authorized store locations as well as the types of food items purchased by SNAP participants.

In this paper, we explore in detail the supply side response to changes in the SNAP benefit level. This is the first evaluation to examine the impact of SNAP-store authorization dynamics on food access and the food environment. To do so, we examine whether changes in the size of SNAP benefits impacts the number and the type of SNAP-authorized retailers within a county. We hypothesize that the supply-side response to SNAP policy changes is heterogeneous in both the type of retailer and across areas with low-food access (e.g., “food deserts”) and high food access areas. Specifically, we focus on changes in the number of healthful food retailers, defined here as supermarkets, full-service grocers, and fruit and vegetable retailers. We test our hypothesis using longitudinal administrative records of all SNAP authorized store locations in the United States from 2006-2017, linked to historical business data for all food retailers and Census tract level indicators of food deserts.

We find that the number of convenience stores and small grocery stores authorized to accept SNAP benefits rose dramatically in the years after the benefit increase due to ARRA, and fell beginning in 2013 when SNAP benefits reverted to their pre-ARRA levels. The vast majority of growth in SNAP-authorized stores was concentrated in non-food desert census tracts. However, our estimation strategy indicates a proportionally larger growth in the number of SNAP-authorized store locations in census tracts that are defined as a food desert. This research has important implications given that there are new, enhanced SNAP retailer requirements that increase the level of difficulty to become SNAP-authorized in areas where there is not significantly limited food access.