2004 Poverty Research Small Grants
Funded grants
The Effect of Food Stamps on Food Security: Public Policy and Household Well-Being in the Economic Downturn of 2001-2002
Parke Wilde, Assistant Professor, Friedman School of Nutrition Science and Policy, Tufts University
Description
The purpose of this research is to measure the effect of Food Stamp Program (FSP) participation on food security during the recent economic downturn of200l -2002. The FSP provides a targeted food benefit for Americans with incomes below or near the poverty line. It is the Federal Government's largest intervention against hunger and food insecurity. By controlling for both observed and unobserved confounding variables, this project promises to offer improved estimates of the Food Stamp Program's true impact on its most directly relevant outcome.
The research will for the first time exploit the longitudinal structure of the Federal Government's flagship food security survey module to control for household-level "fixed effects" -- constant unobserved factors that may influence both program participation and food security. The survey module is fielded as a supplement to the Current Population Survey (CPS). Four of the 8 rotation groups in the CPS were interviewed in both December 2001 and December 2002, and thus (after subtraction for attrition and migration) we have longitudinal data for approximately 20,000 households.
This research also controls for observable time-varying household conditions, such as employment status and a rough measure of income. Our methodology includes a simple fixed-effects logit model for the probability of being food secure, and also a novel variant of the Rasch model (another statistical model in the logit family of models, which has guided the Federal Government's food security research in the past).
