Association for Public Policy Analysis & Management

Summary of Theory and Potential Contributions to the Literature: Food deserts are often defined as deprived areas with poor access to retail food outlets, but scrutiny of the typical definitions reveal that there is much ambiguity on the phenomenon (Rose et al., 2009), specifically in terms of access to food retailers. While food deserts conceptually put an emphasis on proximity to grocery stores, geographical access varies, depending on distance and aggregation methods (Hewko, Smoyer-Tomic & Hodgwon, 2002). Many previous studies assumed that grocery shopping is done within a walkable distance of one’s home, with the boundary for relevant neighborhoods at one-half to one mile from a centroid (Allard, 2013). This study, however, includes more distance parameters to accommodate automobile trips, which is typical behavior for grocery shopping in the U.S. context, and the study develops a discrepancy index that indicates the discrepancy between expected needs and access to grocery stores (Clarke, Eyre & Guy, 2002; Lee & Lim, 2009; Apparicio et al., 2008).

Many studies in the literature have examined the relationship between socio-demographic characteristics and food deserts, finding that people who live in poor communities have limited access to healthy foods, with black and Hispanic households more likely to be food-insecure than white households (Scott, 2013; Coleman-Jensen, McFall, & Nord 2013). There is also growing literature spotlighting modes of available transportation, including vehicle ownership or access to mass transit, as factors affecting a household’s food-store access (Allard, 2013). Thus, the primary socio-economic factors of this study are income, race/ethnicity, and vehicle ownership. Ultimately, we seek to better understand possible differences in healthy-food accessibility across socioeconomic groups.

Research Questions: The Atlanta Regional Commission reported that 500,000 residents in the Atlanta area are living without access to grocery stores as of 2016. Our preliminary analysis shows that neighborhoods without any grocery stores have increased over the last decade. Thus, this paper raises the following research questions to understand the changes in the community’s food-access environment by exploring the dimensions of accessibility and socio-economic characteristics: 1) How has access to healthy food changed over the past decade, and what are the new neighborhoods with limited access to healthy foods? 2) How do socio-economic factors explain the different food environments in various communities, and have the changes in accessibility differed in majority minority, majority white, extreme-poverty, and high-poverty neighborhoods?

Data and Methods: The scope of this analysis is limited to the metro Atlanta area. This study makes use of the ReferenceUSA database to get food retailer information, including detailed data on the types of businesses, their locations, number of employees, and sales revenue. Also, the primary data source for demographic, socioeconomic, and commuting variables for households and individuals is the U.S. Census’ Public Use Micro Sample (PUMS). This study will analyze geographical accessibility to grocery stores based on the dimensions of distance and discrepancy indicators, and will further examine associations among minority population composition, deprivation, and access to healthy foods.