How do food retail choices vary within and between food retail environments?

There is mixed evidence on whether the local food environment explains differences in food choices. It is valuable in such research to use strong research designs, taking account of the endogeneity of household residential decisions. This article offers a literature review, emphasizing two promising approaches in the most recent research: (1) longitudinal studies of supermarket or retailer entry and exit within particular locations and (2) studies of variation in food choices across multiple households within particular locations. We review how these approaches address the endogeneity problem. Because this literature review indicates the importance of understanding variation across households within locations, our empirical analysis uses nationally representative data from USDA's National Household Food Acquisition and Purchase Survey (FoodAPS) to measure the expected values, variances, and intracluster correlations for selected outcome variables and explanatory variables typical of those used in studies of the food retail environment. We find that geographic variation in supermarket proximity does not account for most of the variation of a household's choice of a primary food retailer and does not account for much of the high-level of variation across households in food security and dietary quality. These results and our review of the literature suggest that food store access may modestly influence food choices, but that other household-level factors may matter more.