Behavioral Price Discrimination in the Presence of Switching Costs

We study the strategic impacts of behavioral price discrimination on manufacturers and retailers in a distribution channel when there are switching costs in consumer demand. Unlike previous empirical studies of behavioral price discrimination, which rely only on differences in price elasticity across customers, our pricing model allows the strategies of channel members to additionally account for differences in price elasticity across time (due to switching costs). In our dynamic structural model, both wholesale and retail prices respond to past demand, as well as influence future demand, of all brands in the category. We apply our pricing model to empirical data from the cola category, which is a duopoly involving brands Coke and Pepsi. From the retailer’s price discrimination standpoint, we find that exploiting a consumer’s most recent brand choice (which is easily observed in the customer database) for pricing purposes yields as much profit dividend as exploiting the consumer’s latent segment membership information (which requires predictive analytics). We also find that the retailer should outsource the data analytics and price customization to manufacturers, by selling its customer database to them, in order to further improve retail profit. Interestingly, our results show that, while both manufacturers’ profits improve when they price discriminate based on segment membership, the weaker brand’s profit decreases when price discrimination takes in to account switching costs, when compared to the case of no price discrimination. This happens because of the incentive for each brand to “poach” its rival’s customers using lower prices. We also find that both manufacturers end up worse off under price discrimination when the retailer can sell its customer database to manufacturers, illustrating that customer information is a potent source of channel power to the retailer in markets with switching costs.