Milk Advertising, Vending Machine Purchases, and Their Health Implications

The United States of America is facing a disease ? obesity. In order to combat or attempt to combat this problem, studying specific food groups and who consumes them and why is of upmost importance. This series of papers addresses the ?why? and ?who? of consumption for two industries: fluid milk and vending machine. Each industry will be analyzed using demand analysis methods to answer questions regarding what influences consumption of goods within these two industries.

The fluid milk industry has four fluid milk types, differentiated by milk fat percentage. Advertising strategies focus on generic fluid milk consumption rather than specific fluid milk types. Understanding how generic milk advertising affects specific milk type consumption is necessary to see if advertising improvements can be made. The fluid milk industry data are in time series format and complete (QUAIDS and Barten Synthetic) and incomplete demand systems are used to understand various relationships among prices, income, seasonality, and generic advertising. The incorporation of a polynomial distributed lag advertising variable in each demand model specification shows that generic milk advertising affects fluid milk type consumption differently. Compensated elasticities show that low-fat milk and skim milk and whole milk and skim milk are substitutes. Income elasticities show that each fluid milk type is a normal good. Catering advertising efforts towards specific milk type consumption may result in higher sales as long-term advertising affects milk type consumption differently. Further, Government programs separate milk types in regards to what qualifies for specific types of food assistance programs. If the fluid milk industry caters to such separation, fluid milk consumption, particularly whole milk, may increase.

The vending machine industry is an easy access provider of snacks and sodas. The vending machine industry is analyzed with cross-sectional data over a four year period from 2009 to 2012. With these data we analyze household characteristics that influence the decision to purchase from a vending machine through the use of a Tobit model and a probit model. We examine how socio-demographic characteristics and other purchasing habits affect vending machine purchases both through conditional and unconditional effects and likelihoods. Results indicate that socio-demographic characteristics significantly affect whether or not a purchase is made from a vending machine. Further, other purchasing habits, such as food away from home, chips and colas for at home consumption, and tobacco products positively and significantly affect a household?s vending machine purchases. Perhaps offering a larger variety of goods will attract a larger consumer base.

With the combination of the industries and methods, we are able to answer several questions and provide policy recommendations in regard to marketing strategies that target consumption habits. Further, we add to the current literature through both theoretical and applicable contributions.