Retail Concentration, Food Deserts, and Food Disadvantaged Communities

Troy C. Blanchard, Mississippi State University
Thomas A. Lyson, Cornell University

In this report we apply the concept of food deserts to U.S. non-metropolitan areas, in an effort to understand inequalities in food access for non-metropolitan residents. A food desert is a place where residents have limited access to supermarkets and supercenter stores. The objective of our study is to:

• Develop a measure of food deserts for U.S. non-metropolitan counties
• Describe the characteristics of food desert populations
• Assess the impact of food deserts on the consumption of fruits and vegetables.

These research objectives are a central issue to both researchers and policy makers alike because food retailing has changed dramatically over the past twenty years. As large food retailers have entered small town markets, many smaller grocers in nearby towns have gone out of business leaving residents with few local food options that offer nutritious and inexpensive groceries. This process presents an important health policy issue because of the potential dietary implications of food deserts on childhood and adult overweight and obesity in the U.S.

Current Research on U.S. Food Deserts

To date few studies have examined U.S. food deserts and little is known about the nutrition related consequences of the availability of low cost quality foods. Although the term "food deserts" has been used by researchers to describe places in Europe with few food retailers, U.S. researchers have only recently begun to apply this concept to U.S. rural areas. Our work contributes to this growing body of literature by developing a measure of U.S. food deserts with publicly available data. We also take this line of study one step further by examining the dietary consequences of living in a food desert.

Research Methods

To complete our research objectives, we employ data on both food retailers and persons from the U.S. Bureau of the Census and the Center for Disease Control Behavioral Risk Factor Surveillance System surveys. Using Geographical Information System (GIS) technology, we identified the proportion of the population for each county that resided in a food desert. We define food deserts as places where persons travel at least 10 miles to access a supermarket or supercenter food retailer. We linked our food desert measure to data from the 2000 Census of Population and Housing to characterize the population of food desert counties. We completed our objectives by geographically linking information on food retailer access to survey respondents from the Behavioral Risk Factor Surveillance System (BRFSS) survey. Because BRFSS data contain information fruit and vegetable intake, we examined the degree to which food desert residents consumed fewer servings of fruits and vegetables.

Findings

The findings from our study suggest a number of interesting trends for the implications of food deserts in the U.S. We detail these below:

Food Desert Choices
A central finding for our study was that food deserts contain a higher number of small grocery and convenience stores. Because these stores often sell lower quality groceries at higher process, food desert residents must travel long distances to access the quality, low priced groceries at a supermarket or supercenter. Additionally, food deserts are less likely to have fruit and vegetable markets (farmer's markets).

Food Desert Populations
A second key finding from our study is that food desert counties contain a higher percentage of poor persons, lower median family incomes, a less educated population, and higher rates of unemployment.

Dietary Intake
Our findings on dietary intake suggest that food deserts are associated with poorer dietary intake. Residents of food deserts are 23.4% less likely to consume five or more servings of fruits and vegetable after adjusting our estimates by age, race, sex, and education. An additional finding of our study is that food desert residents with a post high school education are less likely to consume five or more servings of fruits and vegetables per day than post high school educated persons in non-food desert areas.

Discussion

Our study points to a central finding regarding the prevalence and severity of food deserts in U.S. non-metropolitan areas. Primarily, individuals living in food desert areas will pay higher prices for groceries or incur a greater travel cost to access the large food retailer that may offset the savings available at these stores. More importantly, healthy alternatives, such as fruit and vegetable markets, are less prevalent in food deserts. This is especially troubling for vulnerable segments of the population such as poor, low-income individuals and the disabled, who comprise a greater share of the population in food deserts. For these persons it may not be feasible to shop at a large food retailer because of travel cost and time considerations.