ABSTRACT:   The Obama Administration has proposed the Healthy Food Financing Initiative to spur the development of supermarkets and other food retail outlets in neighborhoods that lack access to healthy foods—food deserts.   Several state and local governments along with private organizations have already implemented similar programs. These policy initiatives rely on measures of food access to describe how many people are affected, target policy interventions, and monitor whether food access is improving over time.  To support the Federal effort, the U.S. Department of Agriculture has developed a national level measure of food deserts that has been released through the Food Desert Locator, a web-based mapping tool.  Other researchers and organizations have developed alternative approaches to measure food deserts to help target policy interventions. 

            There is little consensus on what it means to lack access to healthy food—whether that means it is far away and hard to get to without a vehicle or whether supermarkets and sources of healthy food should be more prevalent to make healthy choices easier.  The focus on neighborhoods instead of individuals may result in overlooking the barriers some individuals face in accessing healthy food, while overestimating the problem in some neighborhoods, and, indirectly leading to a narrower set of policy options—ones that focus only on neighborhoods, not individuals. Further, there is a tension between the need for definitions and data that can be applied in a standard way across the nation and the need to detect variation in local conditions with local data.  Each of the national level measures has different purposes, some of which may be appropriate for policy, while others are not.  Each also faces important definitional, data, and methodological limitations to identifying food desert neighborhoods.

            This paper raises several concerns about national-level food desert measures, reviews and critiques currently available measures, and suggests improvements to these methods.  While data, methods and analytical tools to measure food access continue to improve, there are still many gaps in measurement that should be addressed to further inform policy decisions. Improving these measures is essential to more efficiently target scarce funds to the people and neighborhoods most critically in need. The paper urges that individual measures of food store access be estimated in addition to area-based measures, because individual measures provide information about need, changes in need over time and valuable information for monitoring and evaluating policies that can be masked or ignored when only area-based measures are used.  The paper also argues that because of the conceptual and methodological problems in measuring food access limitations, it is important to estimate a broad set of measures that examine multiple facets of food access.