CIRJE-F-125. Abe, Makoto, "Investigating Behavioral Explanations for Asymmetric Price Competition", July 2001.

Asymmetric price competition in consumer packaged goods is a well-known phenomenon. Researchers have proposed three behavioral explanations: (1) heterogeneity in consumer preference, (2) the income effect, and (3) the reference and loss-aversion effects. These explanations have been offered independently by different researchers using different types of data with different methodology. Despite the interest in asymmetric price competition by the marketing community, no attempt has been made to compare across the three explanations and draw an inference on which one is most likely. The objectives of this paper are three-folds. Firstly, important factors when studying asymmetric price competition are discussed. These include, (1) confounding of the supply-side and demand-side factors when analyzing aggregate data, such as market share, and (2) an appropriate measure for asymmetric price competition as the change in market share of a brand for a unit price change of a competing brand. Second, based on these considerations, theoretical arguments against previous explanations are provided.

(1) The Heterogeneity Effect: Aggregate data is analyzed and cross-price elasticity is used. The explanation is inferred indirectly.

(2) Income effect: Infinite divisibility of goods is assumed.

(3) Loss-aversion effect: Household heterogeneity is not well accounted for.

These claims are then supported empirically using scanner-panel data from four categories. Third, an alternative explanation is proposed to show that a basic assumption of consumer utility in Microeconomic theory, the diminishing-marginal-return effect in price, results in asymmetric competition. It was shown that the concavity (i.e., the diminishing-marginal-return effect) in price was stronger than that of logarithm in all four product categories.