An Empirical Analysis of Cigarette Addiction

We use a framework suggested by a model of rational addiction to analyze empirically the demand for cigarettes. The data consist of per capita cigarettes sales (in packs) annually by state for the period 1955 through 1985. The empirical results provide support for the implications of a rational addiction model that cross price effects are negative (consumption in different periods are complements), that long-run price responses exceed short-run responses, and that permanent price effects exceed temporary price effects. A 10 percent permanent increase in the price of cigarettes reduces current consumption by 4 percent in the short run and by 7.5 percent in the long run. In contrast, a 10 percent increase in the price for only one period decreases consumption by only 3 percent.