The Impact of SNAP Food and Beverage Restriction Waivers

• In Event: SNAP Participation

Abstract

Changes to Supplemental Nutrition Assistance Program (SNAP) to restrict purchasing sugary beverages and sugary foods with SNAP funds were implemented for the first time on January 1, 2026. Despite proposals dating back to 1977 to restrict purchases using SNAP funds (USDA, 2007), a critical knowledge gap remains: there is no real world-evidence on the impact of these restriction waivers. This paper provides new evidence to fill this gap and inform policymakers of the overall and heterogeneous impact of restricting sugary beverage purchases with SNAP funds. 

Sugary beverage consumption remains widespread despite increased policy attention and public health campaigns with more than half of youth ages six and older consuming a sugary beverage within a 24-hour period (HER, 2025). Consumption patterns vary across demographic groups, with higher prevalence among non-Hispanic Black and Hispanic youth compared to non-Hispanic White youth (HER, 2025). These disparities raise important equity considerations in evaluating nutrition policy interventions. 

Restricting purchases of sugary beverages with SNAP benefits may reduce consumption by increasing the effective relative price of restricted goods, reinforcing health messaging, and mental accounting.  Alternatively, most SNAP households spend more on food than the benefit level (Hoynes et al., 2015), and these inframarginal households could use SNAP benefits to purchase eligible foods and pay for sugary beverages with other resources, offsetting the intended effect. Retailer compliance, consumer reactance, and substitution toward other high-calorie products may also attenuate impacts. Thus, the USDA has previously argued that restricting the purchase of sugary beverages would increase the costs of the program without changing the purchases or improving the health of SNAP recipients (USDA, 2007). 

To address these competing hypotheses, this paper uses household-level purchase data from the NielsenIQ Omnishopper panel from July 2025 through June 2026 to estimate the causal effect of SNAP food restriction waivers on beverage purchases.  Eleven states implemented restrictions through April 2026 and another 11 states have waivers approved by the USDA that will soon be implemented.  These waivers vary in scope, but all waivers restrict the purchase of soft drinks with SNAP benefits. The staggered implementation across states creates quasi-experimental variation that we leverage to identify causal policy effects. 

Preliminary evidence demonstrates that the average ounces of soft drinks purchased per month in states that implemented waivers between January and April 2026 trended similarly in 2025 to the average in states that will implement waivers later and in states without a waiver. As a result, we will implement a difference-in-differences research design by estimating a Poisson extended two-way fixed effects regression model with month and household fixed effects that accounts for staggered timing of the implementation of the waivers from January through April 2026 (Wooldridge, 2021, 2023; Deb et al., 2024).  We will examine overall effects on ounces of soft drinks purchased and heterogeneous impacts by race, ethnicity, rural/urban residence, and household composition. By providing the first real-world evidence on SNAP purchase restrictions, this project will inform debates on the design of food assistance programs, nutrition policy, public finance sustainability, and health equity.

Authors

• David Frisvold, University of Iowa
  Presenting Author

• FELIPE A. LOZANO-ROJAS, University of Georgia
  Non-Presenting Co-Author