A more nuanced approach is needed to address the issue of access to food
For areas where residents have limited or no access to places to purchase fresh, affordable food, opening new grocery stores seems like an obvious solution. However, the effectiveness of this intervention strategy is surprisingly unclear.
In a study published in Applied Geography, an interdisciplinary team of researchers including UConn Geography, Sustainability, Community and Urban Studies associate professor Peter Chen, Department of Allied Health Sciences associate professor Ran Xu, and assistant professor in the Department of Geography at Binghamton University Congcong Miao Ph.D '26, evaluated the policy effect using nationwide human mobility data. They found that opening new grocery stores in USDA-defined low-income, low-access (LILA) areas, commonly known as "food deserts," increased store visits, but the effect was temporary, lasting only a month. This is the first study to investigate the relationship at the national scale, and it confirms that addressing the food desert issue is not as simple as adding new food retail locations.
Addressing food deserts became a focus of policy intervention and research following the 2014 Farm Bill, Chen explains. These policy interventions, such as the Healthy Food Financing Initiative, sought to invest in LILA areas to encourage healthy food purchases. However, previous studies analyzing such interventions at local scales found no significant influence on shopping habits or diet quality when new stores were opened.
"Our study is not the first to investigate this topic because there are, for example, experiments looking at opening grocery stores and studying how that changes local residents' diet quality, but those studies are pretty scattered and most only focus on one area or one single store," says Xu.
Miao says this study goes further by using national-scale human mobility data and focuses on whether opening new grocery stores influences the residents' shopping behavior, and if so, how long that effect lasted.
To answer these questions, the researchers analyzed mobility data representing 267 million grocery store visits in 2018 and 2019 coupled with data on grocery store openings across the United States. The data came from anonymized mobile device location records.
"From the data, we know which census tract the visitors came from, and which grocery store they visited, and the census tract is also the level where USDA identifies low-income, low-access communities," says Miao. "Then we compared how the residents' overall grocery shopping activity changed before and after the new store opening relative to the changes in census tracts with similar socioeconomic statuses. This allows us to estimate the effect."
The researchers saw an initial positive effect that lasted for the first month. Then, after the first month, the effect is not significant, and it gradually drops into no effect.
"Since we did observe a short-term increase in grocery shopping activity, one possible explanation could be that those new open stores attracted customers through grand opening promotions or other special discounts or activities," says Miao.
"Just adding a grocery store is not enough to solve the food desert issue," says Xu. "That's the counterintuitive part. It's much more nuanced than that."
These nuances will hopefully be the focus of further research. For example, financial access is likely a big part of the picture, says Chen. There are also cultural preferences or habits that contribute to personal shopping choices.
"Improving food access requires more than simply increasing the number of grocery stores, and policymakers also need to understand the residents' needs, preferences, and everyday shopping behaviors," says Miao. "For example, helping people better understand what a healthy diet looks like and making healthy food more affordable could be important complementary strategies in addressing food access issues."
These key takeaways can help inform policymaking decisions such as ensuring that funding is not being directed toward initiatives that do not have the intended impacts. The findings also show the need for further food access research.
"For example, in the 2014 Farm Bill, there was a strong incentive to invest in food desert neighborhoods, but if you look at recent initiatives like one called Healthy People 2030, which is a national directive by the U.S. Department of Health and Human Services, they didn't include food access as one of their measurable outcomes," says Chen. "There has been a national de-emphasis on physical food access, but our research provides strong and corroborating evidence to support these national initiatives."
The research was supported by a seed grant from UConn InCHIP and a generous contribution from Alan Bennett '69 (CLAS), for public health policy research at UConn.