Bayesian Estimation of The Impacts of Food Safety Information on Household Demand for Meat and Poultry

Consumer reaction to changes in the amount of food safety information on beef, pork, and poultry available in the media is the focus of this study. Specifically, any differences in consumer reactions due to heterogeneous household characteristics are investigated. The data used in this study are monthly data from the Nielsen Homescan panel and cover the time period January 1998 to December 2005. These panel data contain information on household purchases of fresh meat and poultry as well as demographic characteristics of the participating households. The data used to describe food safety information were obtained from searches of newspapers using the Lexis-Nexis academic search engine. Consumer reactions are modeled in this study using a demand system that allows for both discrete and continuous choice situations. A seemingly unrelated regression (SUR) tobit model is estimated using a Gibbs sampler with data augmentation. A component error structure (random effects model) is incorporated into the SUR tobit model to account for unobserved heterogeneity of households making repeated purchases over time. Estimates of food safety elasticities calculated from the random effects SUR tobit model suggest that food safety information does not have a statistically or economically significant effect on household purchases of meat and poultry.food safety, panel data, Gibbs sampler, component error, Agricultural and Food Policy, Consumer/Household Economics, Food Consumption/Nutrition/Food Safety,