Much of the research on food deserts has focused on the relationship between the food retail environment and nutrition and health outcomes. Intermediary differences in food shopping patterns are often implicitly assumed to drive these relationships (environment hypothetically affects shopping, which hypothetically affects consumption). Research is limited, though, on whether these food shopping discrepancies exist.

This article investigates whether a number of food shopping outcomes and the food retail environment are in fact associated and in which kinds of neighborhoods, using the Electronic Benefit Transfer (EBT) records of over 40 000 households receiving Supplemental Nutrition Assistance Program (SNAP) benefits in western Massachusetts.

In some, though not all, food retail environments, we find a small but statistically significant negative association between continuous distance and both the percentage of SNAP redemptions spent at supermarkets and the number of benefit-spending trips taken to supermarkets. Nonetheless, SNAP households located in neighborhoods with what would be considered poor access to supermarkets still spent, on average, more than 75% of their redemptions at these retailers, only 5 percentage points lower than households located one block from a supermarket.

These results suggest that SNAP participants’ inability to reach healthy food retailers is at most a minor driver of geographic disparities in nutrition and health outcomes.

Learn more about our work in food assistance and nutrition policy.