Demand for Food-Away-From-Home: A Multiple Discrete/Continuous Extreme Value Model

Obesity is a complex problem with many causes, from genetic and behavioral disorders to environmental factors, including access to calorie-dense fast food meals. Economists and epidemiologists disagree on the importance of access to fast food as a causal factor for obesity, but agree that any policy regulating access to fast food will likely use the price system, through taxes or other means to raise the relative cost of buying fast food. Yet, little is known of the structure of demand for food-away-from-home (FAFH). This study provides estimates of the price-elasticity of demand for four di¤erent types of FAFH using a novel new dataset from NPD, Inc. By including physiological measures of obesity, physical activity and health status as additional regressors in an instrumental variables framework, we control for important sources of observed heterogeneity. We nd that all types of FAFH are price elastic in demand, but ne dining is highly elastic while fast food is nearly unit elastic. Food-at-home (FAH), on the other hand, is relatively elastic. Critically, cross-price elasticities of demand show little willingness to substitute between FAH and any type of FAFH. When prices are rising, consumers prefer to change the type of restaurant they visit, rather than forego the experience entirely. As shown elsewhere in the literature, therefore, taxing fast food is likely to be counterproductive