| dc.description.abstract | Persistent food insecurity and hunger increase the risk of illness, psychological dysfunction and lower educational achievement. Even though these burdens affect society at large, they are most acutely felt by the individuals and households living in poverty. To address the costs of hunger and food insecurity, policies have been designed, many by urban planners, to increase access to healthy food. Because low-income populations are assumed to shop at the nearest store that sells food, most policies have focused on opening new supermarkets in “food deserts.” However, there is little evidence that such assumptions are true or that nearby supermarkets make a difference.
This dissertation presents a 7-year panel model for 207 counties in Texas as a tool to test the consequences of commuting patterns and the structure of the retail grocery market on food dollars spent by people living in poverty. The model uses longitudinal data from the Supplemental Nutrition Assistance Program (SNAP). The model uses publicly available geocoded data on SNAP benefits and redemptions, retail locations, and commuting patterns. The model explicitly examines the consequences of commuting patterns and retail markets, both local and in surrounding counties, for SNAP redemption.
Results show that commuting patterns and the grocery retail market are important factors for predicting SNAP redemptions. Specifically, workers that commute out of a county have a negative effect on the amount of SNAP dollars redeemed in a county, and workers that commute into a county have the opposite effect. Large SNAP retailers, such as super stores or chain stores, have the largest positive effect. The number of supermarkets in a neighboring county does not affect the net SNAP dollars redeemed within a county, but the number of neighboring super stores or chain stores does. SNAP redemptions decrease significantly when counties do not have large retailers and when counties have more outbound workers than inbound workers. The factors identified in this research that influence redemption patterns may have implications for policies that attempt to enhance SNAP redemptions. In the broader picture, such policies may have a significant impact on food access for people living in poverty. | en |
