UC Berkeley

Retailers are introducing new store brands at a rapid pace, and annual sales of store brands in the U.S. now exceeds $108 billion. In the literature on store brand decisions, it is commonly assumed that (1) the retailer is a downstream monopolist; (2) either the store brand quality level is fixed, or, the marginal cost of production is constant and independent of the quality level of the store brand; and (3) the retailer either produces the store brand in-house, or sources it from a non-strategic manufacturer. Although these assumptions significantly simplify the analysis, they do not capture what is commonly seen in practice. As a consequence, the insights from these studies may not apply more broadly. Assumption (1) needs to be relaxed in order to study retailers' product assortment decisions (in terms of store and national brands) and related pricing decisions at two competing retailers, along with pricing decisions of a leading national brand manufacturer. Assumption (1) and (2) need to be simultaneously relaxed in order to investigate a retailer's store brand quality-positioning decision when facing competition from another retailer that already carries a store brand. Assumptions (2) and (3) need to be relaxed simultaneously in order to study how a retailer's optimal quality-positioning strategy changes across various sourcing arrangements and various pricing power relationships among retailers and manufacturers. This dissertation contributes to the store brand literature by analyzing models based on more realistic assumptions than those in the literature.

This dissertation consists of three stand-alone papers. The first paper (in Chapter 2) investigates a retailer's product assortment and pricing problem when she has the option to carry a store brand, a national brand, or both. I compare her decision when she is a downstream monopolist and when she faces competition from another retailer who may also oer the same national brand and a competing store brand. Specifically, I assume the quality levels of the products are exogenous and analyze a manufacturer-Stackelberg game involving a national brand manufacturer and two competing retailers. The national brand manufacturer sets a wholesale price for the national brand product (the same for both retailers; I assume they are similar in size and can therefore secure the same wholesale price). Then, observing the wholesale price, the retailers engage in a Nash game in which they set the retail prices for the product(s) they choose to carry. Finally, customers decide whether and what to purchase. Customers are heterogeneous in two dimensions: location, which can be interpreted as the degree of loyalty to one retailer or the other, and willingness to pay per unit of quality. Each customer visits the retailer where he can obtain the maximum surplus (willingness to pay less purchasing and transportation costs) among the offered products. After the customer arrives at the selected retailer, the transportation cost is now sunk, so he buys the offered product with the larger difference between his willingness to pay for the product and its price, if it is non-negative.

The second paper (in Chapter 3) addresses store brand quality positioning decisions for retailers facing retail competition. Specifically, I assume one of the two retailers (Retailer 2) already carries a store brand product whose quality level is fixed, and both retailers may offer the national brand product with a fixed quality level. The representation of customer preferences and the resultant demands are the same as in the first paper. I model the dynamics via a two-stage game. In the first stage, Retailer 1 decides whether to introduce a store brand product, and if so, its quality level. Then the three parties engage in a manufacturer-Stackelberg pricing game. Finally, customers decide whether and what to purchase. In the first stage, Retailer 1 anticipates the outcome of the second-stage game. I analyzed the second stage game in the first paper; it is a subproblem in the second paper. I also analyze a setting in which both retailers may choose the quality levels of their store brand products simultaneously.

The third paper (in Chapter 4) studies a retailer's equilibrium quality-positioning strategy under three sourcing structures, and for each sourcing structure, I consider three types of channel price leadership. Specifically, I study games between (among) a retailer, a national brand manufacturer and a strategic third-party manufacturer, where applicable. The retailer carries a product (with a fixed quality) offered by the national brand manufacturer, and is considering introducing a store brand whose quality can be decided. Customers are heterogeneous in their willingness-to-pay (WTP) per unit of quality. The utility a customer derives from either product equals her WTP per unit of quality times the product quality. Each customer chooses the product that gives her the greatest surplus (utility less price), provided that it is non-negative. The unit production cost of both products is strictly convex and increasing in the quality level of the product. I derive the retailer's equilibrium store-brand quality decision under three sourcing arrangements and three pricing power scenarios. The three sourcing arrangements are in-house (IH), a leading national brand manufacturer (NM) (whose product the retailer also carries), and a strategic third-party manufacturer (SM). The three power scenarios are the ones most commonly seen in the literature: Manufacturer-Stackelberg (MS), Retailer-Stackelberg (RS), and Vertical Nash (VN). In sum, I examine nine (i.e., three times three) combinations of sourcing and pricing power (or game) scenarios, and compare the retailer's optimal quality positioning decision and other equilibrium results (including prices) across the nine scenarios. In all nine combinations of sourcing and pricing power scenarios, the retailer moves rst in setting the quality of her store-brand (during the product development phase) before any pricing decisions are made. I derive subgame perfect equilibria for all scenarios. To the best of my knowledge, I am the first to present a comparison of equilibria for these nine realistic combinations of sourcing and pricing power in this context.

This dissertation makes several contributions to the literature on store brand strategies. First, the majority of papers on store brand strategies consider a monopolist retailer. The few papers that consider retail competition are based on restrictive assumptions concerning factors such as product quality (e.g., assuming store brand products have equal quality levels) or product offering (e.g., both retailers must oer the national brand product). My work in papers 1 and 2 takes a first step in presenting a model that is general enough to allow me to study retailers' strategies in a context with store and national brands, and with retail competition. Second, prior research utilizes demand models that are limited in their ability to capture customers' joint selection of a retailer and a product. My work in papers 1 and 2 is based on a model of customer preferences that allows me to incorporate both quality differentiation among the products and the degree of customer loyalty to retailers, both of which are important in my problem context. This model is flexible enough to support a fairly rich representation of demands. Third, in paper 3, I take a first step in studying the interaction between store-brand sourcing and positioning decisions, and the interplay of these decisions with the retailer's pricing power. From a comparison of the retailer's equilibrium store brand quality levels for the nine combinations of sourcing and game structure, I obtain a full characterization of the ordering of store-brand quality, retailer's prot, retail prices and consumer welfare across the nine combinations. To the best of my knowledge, I am the first to present a comparison of equilibria for these nine realistic combinations of sourcing and pricing power in the store brand context. I also show that sourcing of store brands plays a key role in the competitive interaction between a retailer and a national brand manufacturer. Whereas the marketing and economics literatures have emphasized the role of store brands in helping retailers elicit price concessions from national brand manufacturers, I find that having a preferable sourcing arrangement for a store brand product is more valuable than having pricing power.