Food deserts drive diet-related illness in low-income communities

The claim

The food desert thesis holds that the geography of grocery retail in the United States is not neutral — that supermarkets have systematically retreated from low-income and minority neighborhoods over decades of disinvestment, leaving residents without realistic access to fresh produce, lean proteins, and whole grains. In this framing, the higher rates of obesity, type 2 diabetes, and cardiovascular disease concentrated in these communities are not the product of individual dietary preference but of constrained choice architecture: when the only food retail within walking distance or a short transit ride is a dollar store, a fast food outlet, or a corner store stocked primarily with ultra-processed products, the resulting diet is structurally determined, not freely chosen.

The claim is supported with important qualifications. The spatial maldistribution of food retail is real and well-documented. Its connection to diet quality and chronic disease is supported by the evidence. However, the research record consistently shows that proximity alone is insufficient — affordability and SNAP purchasing power operate as co-equal structural constraints, and improving physical access without addressing price fails to produce dietary improvement. The food desert frame is correct as a structural diagnosis but incomplete as a policy prescription.

The mechanism

Supermarket disinvestment tracks redlining. Decades of racially discriminatory mortgage and lending policy — formally documented in HOLC redlining maps from the 1930s — concentrated poverty in specific urban neighborhoods. As white residents and capital migrated to suburbs following World War II, supermarket chains followed purchasing power outward. The resulting geography is not random: USDA Food Access Research Atlas data show that census tracts with high poverty rates and high minority population share have systematically lower supermarket density even after controlling for population size. Larson et al.’s 2009 systematic review of 55 studies found that low-income neighborhoods had 25% fewer supermarkets than high-income neighborhoods, and predominantly Black neighborhoods had half the number of supermarkets as predominantly white neighborhoods.

The price mechanism operates independently of proximity. Even when low-income residents live near a supermarket, healthy food is more expensive in absolute terms. Rao et al.’s 2013 meta-analysis of 27 studies across 10 countries found that eating a healthy diet cost approximately $1.50 more per day than an unhealthy diet — a difference of roughly $550 per year per adult that is economically prohibitive for households near the poverty line. This price differential means that proximity to a full-service grocery store does not, by itself, resolve the constraint. The binding limitation for many food-insecure households is affordability, not mere geographic access.

SNAP purchasing power has eroded relative to healthy food costs. The Supplemental Nutrition Assistance Program is the primary federal mechanism for food purchasing power in low-income households, but maximum benefit levels have historically been calibrated to the USDA Thrifty Food Plan — a dietary standard that food policy researchers have documented as unrealistically austere. Even after the 2021 Thrifty Food Plan update, which raised benefits by approximately 21%, independent analyses by the Center on Budget and Policy Priorities found that SNAP maximum benefits cover roughly 68% of actual food costs in high-cost urban areas. When households exhaust SNAP benefits before month end — a phenomenon documented in USDA panel data — they shift toward cheaper, calorically dense, nutritionally poor foods not because of preference but because of arithmetic.

Corner stores and dollar stores fill the vacuum but do not substitute. The retreat of supermarkets has been partially filled by dollar stores, convenience stores, and small corner stores whose inventory economics preclude stocking fresh produce at competitive prices. A 2019 analysis by Hunt et al. found that while food desert census tracts had seen modest supermarket entry over the prior decade, the dominant entrant category was dollar stores, which carry primarily shelf-stable ultra-processed products. The result is a food environment that has improved in terms of raw store count while remaining poor in terms of nutritional availability.

The WIC natural experiment. The Women, Infants, and Children program provides specific food packages to low-income pregnant women, postpartum women, infants, and young children. A 2009 revision to WIC food packages — which added fruits, vegetables, and whole grains while reducing juice and whole milk — provides quasi-experimental evidence of a structural food access intervention. Studies by Whaley et al. and subsequent analyses documented measurable improvements in fruit and vegetable consumption among WIC-participating women following the package revision. This is a structural food access change — not a behavioral intervention, not a nutrition education campaign — that produced dietary improvement.

Who benefits

Convenience store chains and dollar store corporations — Dollar General, Dollar Tree, and Family Dollar — have aggressively expanded into low-income neighborhoods where they face limited competition from full-service grocery chains. Their business models depend on high-margin shelf-stable ultra-processed inventory and do not support perishable fresh produce logistics. These chains have actively lobbied against SNAP eligibility restrictions that would limit which retailers can accept benefits.

Ultra-processed food manufacturers — including Kraft Heinz, Conagra, and PepsiCo — benefit from the food desert structure because their products dominate the retail channels that fill food access gaps. Their products require no cold chain, have long shelf lives, and carry high margins relative to fresh produce.

Agricultural commodity lobbies, particularly corn and soybean processors, benefit from federal farm subsidy structures that make commodity-derived ingredients (high-fructose corn syrup, refined soy oils) systematically cheaper than fresh produce, structurally pricing healthy food out of reach for the lowest-income households.

Real estate developers and municipal authorities in historically disinvested areas benefit when food access failures are framed as individual choice problems, because the structural alternative — requiring food retail as a condition of development subsidies, or zoning for food retail in residential areas — imposes costs on development interests.

The counter

The most important corrective to the naive food desert thesis comes from the Pittsburgh RAND study. Dubowitz et al. (2015) tracked a low-income urban cohort in Pittsburgh over time after a new supermarket opened nearby. If physical proximity were the binding constraint, diet quality should have improved among residents who gained supermarket access. It did not, in any statistically significant way. The finding was not that food environment is irrelevant — it is that proximity alone, without addressing affordability, is not sufficient. This result has been replicated in other settings and is now the research consensus: food access interventions that address location but not price produce limited dietary improvement.

There is also an important distinction between food deserts and food insecurity. Food deserts are a geographic construct. Food insecurity — the lived experience of not having reliable access to sufficient food — is a distinct but related phenomenon. Some food-insecure households live outside food deserts; some residents of food deserts are not food insecure. Policy that addresses food desert geography without addressing food insecurity (i.e., SNAP purchasing power) will miss the households with the highest disease burden.

The food sovereignty critique, developed in part through work by the Alana Institute and Indigenous food systems scholars, argues that the food desert frame itself is too narrowly focused on supermarket proximity and Western dietary standards. Communities with traditional food cultures — Indigenous communities, recent immigrant communities — may have alternative food access structures (community gardens, ethnic markets, home food production, social food sharing) that supermarket-centric mapping fails to capture. Urban agriculture initiatives in cities including Detroit, Cleveland, and Philadelphia have demonstrated that food production capacity can be built within disinvested areas, partially substituting for retail access while rebuilding food sovereignty.

The strongest version of the counter is not that food environments are irrelevant but that the intervention implication of the food desert frame — build more supermarkets — is too narrow. The evidence supports a multicomponent model: supermarket access, SNAP purchasing power, WIC food package standards, corner store conversion programs, and urban agriculture, operating simultaneously.

References

Dubowitz, T., Ghosh-Dastidar, M., Cohen, D. A., Beckman, R., Steiner, E. D., Hunter, G. P., Florez, K. R., Huang, C., Vaughan, C. A., Sloan, J. C., Zenk, S. N., Cummins, S., & Collins, R. L. (2015). Diet and perceptions change with supermarket introduction in a food desert, but not because of the supermarket. American Journal of Public Health, 105(S3), e119–e125. https://doi.org/10.2105/AJPH.2014.302392

Hunt, A. R., Browne, D. C., Bhattacharjee, S., & Dykstra, E. (2019). Dollar stores are not improving food access in US food deserts. Applied Geography, 110, 102036. https://doi.org/10.1016/j.apgeog.2019.102036

Larson, N. I., Story, M. T., & Nelson, M. C. (2009). Neighborhood environments: Disparities in access to healthy foods in the U.S. American Journal of Preventive Medicine, 36(1), 74–81. https://doi.org/10.1016/j.amepre.2008.09.025

Leung, C. W., Ding, E. L., Catalano, P. J., Villamor, E., Rimm, E. B., & Willett, W. C. (2012). Dietary intake and dietary quality of low-income adults in the Supplemental Nutrition Assistance Program. American Journal of Clinical Nutrition, 96(5), 977–988. https://doi.org/10.3945/ajcn.112.040014

Rao, M., Afshin, A., Singh, G., & Mozaffarian, D. (2013). Do healthier foods and diet patterns cost more than less healthy options? A systematic review and meta-analysis. BMJ Open, 3(12), e004277. https://doi.org/10.1136/bmjopen-2013-004277

United States Department of Agriculture, Economic Research Service. (2019). Food access research atlas. USDA ERS. https://www.ers.usda.gov/data-products/food-access-research-atlas/

Ver Ploeg, M., Mancino, L., Todd, J. E., Clay, D. M., & Scharadin, B. (2015). Where do Americans usually shop for food and how do they travel to get there? Initial findings from the National Household Food Acquisition and Purchase Survey (Economic Information Bulletin No. 138). USDA Economic Research Service.

Whaley, S. E., Ritchie, L. D., Spector, P., & Gomez, J. (2012). Revised WIC food package improves diets of WIC families. Journal of Nutrition Education and Behavior, 44(3), 204–209. https://doi.org/10.1016/j.jneb.2011.09.011

Zenk, S. N., Schulz, A. J., Israel, B. A., James, S. A., Bao, S., & Wilson, M. L. (2005). Neighborhood racial composition, neighborhood poverty, and the spatial accessibility of supermarkets in metropolitan Detroit. American Journal of Public Health, 95(4), 660–667. https://doi.org/10.2105/AJPH.2004.042150