The Journal of Industrial Economics
In this paper the revealed preference approach is used to construct a nonparametric test of the monopoly model and some simple generalizations of it, and the test is applied to data for the cigarette industry. The test exploits the maintained hypothesis that variations in the excise tax charged on a package of cigarettes allow us to assess seller reactions to common exogenous variations in product marginal cost. These easily implemented nonparametric tests indicate that the monopoly hypothesis and other simple models that do not embody at least a moderate amount of competition serve as poor predictors of the effects of excise tax changes on cigarette prices, sales, and revenues. We also find some evidence that excise tax increases do not consistently act to increase prices and decrease sales, which calls into question many other hypotheses about firm behavior in the industry.