Mean food spending by food stamp households peaks sharply in the first three days after benefits are received. For those who conduct major grocery shopping trips only once per month (42% of all food stamp households), mean food energy intake drops significantly by the fourth week of the month. For the remaining households, intake remains steady over the course of the month. These patterns motivate an empirical model that simultaneously accounts for the shopping frequency and food intake decisions over time. Results have implications for policies that may affect the frequency of grocery shopping by food stamp households.

We derive a joint continuous/censored commodity demand system for panel data applications. Unobserved heterogeneity is controlled for using a correlated random effects specification and a generalized method of moments framework used to estimate the model. While relatively small differences in elasticity estimates are found between a flexible random effects specification and one that restricts the random effect coefficient to be time invariant, larger differences are observed when comparing the flexible model to a pooled cross-sectional estimator. The results suggest the limited ability of such estimators to control for preference heterogeneity and unit-value endogeneity leads to parameter bias. Copyright 2005, Oxford University Press.

This article employs new Food Guide Pyramid servings data to measure how food intake is affected by income and two food programs. The analysis uses a maximum likelihood estimator that combines the seemingly unrelated regression and one-way error component models, to account for two types of intrahousehold correlations. Participation in the Food Stamp Program is associated with higher intake of meats, added sugars, and total fats. Participation in the Special Supplemental Nutrition Program for Women, Infants, and Children is associated with lower intake of added sugars. Copyright 1999, Oxford University Press.

No abstract