The Effect of SNAP on the Composition of Purchased Foods: Evidence and Implications

Overview

Justine Hastings, Ryan Kessler, and Jesse M. Shapiro of Brown University released a working paper analyzing the effects of the Supplemental Nutrition Assistance Program (SNAP) on nutritional quality of purchased foods. SNAP is the largest federal nutrition assistance program, giving approximately 40 million Americans benefits that can be used to purchase food for at-home consumption. SNAP benefits are calculated to cover approximately 70 percent of a family’s food budget and are provided monthly through an Electronic Benefit Transfer (EBT) card.

Findings

The research team examined detailed transaction records for nearly half a million customers at a large U.S. grocery chain. Chronicling purchases over nearly seven years, these records reflect thousands of transitions on and off SNAP, as most people who receive SNAP benefits only do so for the short term. The team then utilized specific product identifiers and various measures of food type and nutrient content to track the healthfulness of households’ grocery purchases over time, both when on and off SNAP.

They found that SNAP participation had only a small effect on the nutritional quality of purchased foods. Indeed, the program’s effect was small compared to the variation in nutritional quality across households.

Implications for Policy and Practice

SNAP is one of the federal government’s largest “safety net” programs and is meant to help low-income families increase their food expenditures and put more healthy foods on the table. The existing literature on household food budgets suggest that adults with higher socioeconomic status (SES) spend more on food for at-home consumption and eat diets richer in fruits and vegetables and with lower fat and sugar. However, even though SNAP appears to increase food spending, participation in the program may have only a small effect on the nutritional quality of purchased grocery foods. Therefore, closing the gap in food-at-home spending between households of different SES would not close the corresponding gap in the nutritional quality of purchased foods.