Intended and unintended consequences of a proposed national tax on sugar-sweetened beverages to combat the U.S. obesity problem.

Monthly data derived from the Nielsen Homescan Panel for calendar years 1998 through 2003 are used to estimate the effects of a proposed tax on sugar-sweetened beverages (SSBs). Most arguments in describing the ramifications of a tax fail to consider demand interrelationships among various beverages. To circumvent this shortcoming we employ a variation of Quadratic Almost Ideal Demand System (QUAIDS) model. The consumption of isotonics, regular soft drinks and fruit drinks, the set of SSBs, is negatively impacted by the proposed tax, while the consumption of fruit juices, low-fat milk, coffee, and tea is positively affected. Diversion ratios are provided identifying where the volumes of the SSBs are directed as a result of the tax policy. The reduction in the body weight as a result of a 20% tax on SSBs is estimated to be between 1.54 and 2.55 lb per year. However, not considering demand interrelationships would result in higher weight loss. Unequivocally, it is necessary to consider interrelationships among non-alcoholic beverages in assessing the effect of the tax.