Files
Abstract
This dissertation aims to estimate a comprehensive food demand system and apply the analysis to food assistance programs. The primary objectives of this study involve estimating different food demand systems and examining the price and income elasticities of households enrolled in the Supplemental Nutrition Assistance Program (SNAP) and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC). Moreover, this study develops a novel approach to efficiently estimate a large non-linear demand system accounting for price endogeneity.In Chapter one, we examine whether there was a structural change in WIC households’ demand when the revision was implemented. We find that demand for skim/low-fat milk and whole grain became less price elastic post revision. The model also detects a spillover effect to demand for carbonated beverages that are not included in WIC food packages. These results are consistent with the hypothesis that the 2009 revision increased WIC households’ preferences for healthier foods.
In Chapter two, we investigate the change in food demand of SNAP households during the benefit month and how food demand responds to changes in food prices. By using two-way Exact Affine Stone Index demand, we compare the difference in food demand sensitivity between households in the first two weeks of the SNAP benefit month and the last two weeks of the benefit month.
In Chapter three, we derive the asymptotic properties for a tractable iterated three stage least squares (I3SLS) estimator and apply it to estimate a large quadratic almost ideal demand (QUAID) system with endogenous prices. Then we calculate an exact price index from the estimates and compare it with several commonly used price indexes.