Retail Pricing of Substitutable Products under Logit Demand.

Abstract

This dissertation discusses substitution effects on retail pricing problems. In the presence of substitutable products, the inventory levels and prices of different products interact in complicated ways and, therefore, insights developed in the context of single-product pricing problems do not necessarily carry over. Unfortunately, pricing problems that involve multiple products and multiple sellers tend to be analytically intractable, which has so far limited the extent of analytical results for such problems. Motivated by this gap in the literature, this dissertation studies two related problems. First, a single retailer which sells substitutable products and, second, competition among multiple sellers each offering a single product. The first problem is explored in two separate chapters, which analyze dynamic pricing in the presence of limited inventories and an extension to personalized pricing. In the analysis of the problem with a single retailer selling limited inventories of substitutable products, we consider both price-based substitution and stock-out based substitution. When both types of substitution are allowed, the existing literature offers little in the way of analytical results. We add to the literature by developing a series of results about the marginal values of products, which then leads to results about the structure of the optimal pricing policy. We extend our analysis to personalized pricing where we show the analytical properties of the revenue function. In addition, we conduct numerical analysis to study the interactions between dynamic pricing and personalized pricing. For the problem with multiple sellers, we assume a major retailer allows smaller merchants to sell through it. In this setting, the smaller merchants gain access to the store traffic of the major retailer, in return for commission fees. We study the characteristics of the equilibrium and explore the conditions under which this business model is viable.