Panel Paper: Understanding State Differences in Food Security: The Role of the State Food Security Infrastructure

Thursday, November 6, 2014 : 8:30 AM
Acoma (Convention Center)

*Names in bold indicate Presenter

Judi Bartfeld, University of Wisconsin Madison and Fei Men, University of Wisconsin - Madison
Food insecurity increased sharply during the Great Recession, and has remained persistently elevated since then.  Nonetheless, there is considerable variation in food insecurity among states—from a low of 8.7% in North Dakota to a high of 20.9% in Mississippi, based on the most recently available data.  Whereas considerable research has focused on household predictors of food insecurity, as well as on the potentially ameliorative role of food assistance programs, we lack a comprehensive understanding of whether and to what extent demographic and contextual factors explain cross-state variation in food security.  The existing research in this area largely precedes the Great Recession and the concomitant spike in both food insecurity and food assistance program enrollments. This paper addresses this gap by examining the role of a wide range of state characteristics—broadly conceptualized as the food security infrastructure—in contributing to food security outcomes from 2001-2012. 

                We link data from 12 years of household data from the Current Population Survey’s Food Security Supplement (CPS-FSS) to a host of variables describing the state policy, economic, social, and political environments.    Policy variables describe time and state-varying attributes of food assistance programs, but also include broader policies impacting economic wellbeing, including minimum wage policies and the Earned Income tax Credit; economic variables capture factors such as unemployment, prevailing wages, and housing costs; the social and political environment is reflected in variables such as residential stability, urbanicity, and dominant political party.  Collectively, these domains comprise a set of attributes that may directly and indirectly impact households’ ability to consistently meet food needs.  Food insecurity is modeled at various severity thresholds, and multilevel modeling strategies are used to reflect the nested structure of households situated within particular state-year contexts.  Preliminary analyses suggest impacts of all domains – policy, economic, and social/political – on the risk of household food insecurity.  Results have important implications for policymakers in that they highlight multiple potential points for intervention, both within and outside of the food assistance policy arena.  Furthermore, while these analyses focus on the state-level environment, the domains considered are in some cases also subject to local variation and influence, and should thus be of interest to communities seeking to address food security concerns.