Impact of Access and Value on Fresh Food Consumption: Policy Implications
Posted: 7 Feb 2019
Date Written: January 14, 2019
A growing number of strategies are being proposed and piloted to increase consumption of healthy food in underserved neighborhoods. While there is agreement that food environments are linked to diet and health outcomes, there is surprisingly little agreement about the factors (e.g., access to grocery stores) that affect food choice, due in part to the observational and sparse nature of most available data, as well as a lack of understanding of causal mechanisms. This paper studies the impact of access to grocery stores and value of nutrition on consumer consumption, with a goal of improving operational aspects of consumer-level food policy for low-income households. Consumer-level food policy and food choice has received little attention from the OM community but presents a significant opportunity for impact particularly because of the complex interplay between supply chain, financial, and consumer behavior factors. We analyze the Food Acquisition and Purchase Survey dataset, employing the causal inference technique of matching to estimate the causal effect of a household's access to grocery stores and value of nutrition on fresh food spending among SNAP households, as well as two secondary outcomes. We find evidence that value and access have an effect on fruit and vegetable spending among certain populations, but that they impact household's decisions differently. Distance to the nearest grocery store affects a household's frequency of grocery store visits. Value can affect both frequency as well as food choices at the store. We also find that value and access interventions are substitutable, meaning that receiving both interventions has a similar impact to receiving either one. These effect estimates inform a proposal of the underlying causal mechanisms responsible for the observed results, based on a novel consumer choice model that explains food choice and shopping frequency. These results have several implications for policy-making. Contrary to recent literature, this paper concludes that access-related interventions (such as building new grocery stores) are worth pursuing, but relevant features of the neighborhood should be assessed prior to implementation. Additionally, there is evidence that the standard distance threshold used to define access and food deserts (1 mile) may be inappropriate and misrepresenting the true impact of access. Relatedly, because access and value-related interventions appear to be substitutable, policymakers should make informed, data-driven decisions about which type of intervention to pursue in a given neighborhood to avoid investing in redundant interventions.
Keywords: food access, food policy, matching, observational study
Suggested Citation: Suggested Citation