|
Article Excerpt
Conventional wisdom suggests that one of the goals of manufacturer advertising is to reduce the cross-price elasticity between products (make one's own and rivals' products appear to be less substitutable in the eyes of consumers). Conventional wisdom also suggests that, all else being equal, retailers will be able to obtain better terms of trade from manufacturers the more substitutable are the manufacturers' products. It follows that retailers should be opposed to advertising that has the effect of reducing cross-price elasticities and thus that manufacturer advertising can be a source of channel conflict. We show that these conventional wisdoms need not hold when only some consumers are exposed to the advertising messages. Using a Hotelling model of demand, we show that (1) manufacturers can be worse off from advertising that reduces the cross-price elasticities between their products, (2) channel conflict need not arise, even when the sole purpose of advertising is to affect cross-price elasticities, and (3) depending on its bargaining power, a retailer can be better off when the manufacturers' products are perceived to be less substitutable.
Key words: advertising; differentiation; channel conflict; bargaining; channel coordination; distribution channel; game theory
History: This paper was received November 16, 2002, and was with the authors 6 months for 1 revision; processed by Duncan Simester.
1. Introduction
A common theme in marketing is that manufacturers can earn higher profit by making their products less substitutable, thereby reducing cross-price elasticities. These higher profits allegedly come at the expense of consumers in the form of higher retail prices and at the expense of retailers in the form of better terms of trade. Not surprisingly, the instruments available to reduce cross-price elasticities are well known. Firms can design their products to appeal to some consumer groups but not others. They can offer products of different qualities. They can locate in different areas, and they can offer different service levels to consumers. In addition, marketing communications and advertising messages can be used to influence consumers' perceptions of product substitution. While all these strategies are at the disposal of manufacturers in the long run, in the short run things like product design, location, and service infrastructure are fixed. Much of the role in shaping consumers' perceptions of the substitutability among competing products in the short run thus falls to marketing communications and advertising.
It follows from conventional wisdom that manufacturers should advertise in a way that reduces cross-price elasticities; retailers should advertise with an eye towards increasing cross-price elasticities (e.g., with comparison charts emphasizing similarities); and manufacturers and retailers should be opposed to each other's advertising. Thus, it follows that advertising--whether conducted by the manufacturer or by the retailer--can be a source of channel conflict. In this paper, we show that conventional wisdom about persuasive advertising need not hold when only some consumers are exposed to the advertising messages. In particular, we show that (1) competing manufacturers can be worse off from advertising that reduces the cross-price elasticities between their products, (2) channel conflict need not arise even when the sole purpose of advertising is to affect cross-price elasticities, and (3) depending on its bargaining power, a retailer can be better off when the manufacturers' products are less substitutable.
The main theme of this paper is that the choice of advertising medium, and thus the extent of ad exposure (which consumers receive the ads and which do not), drives firms' optimal strategies. Intuitively, when a retailer sells the products of competing manufacturers, changes in a manufacturer's profit are driven by how the perceived substitution among products affects equilibrium prices (determined by the marginal consumers) relative to the hypothetical prices a retailer would set in the absence of the manufacturer's product. Conventional wisdom holds when the advertising messages reach all consumers, because then such ads always affect both the equilibrium prices and the hypothetical prices. However, when the ads reach only a subset of consumers, they may affect the equilibrium prices of the retailer but not the hypothetical prices a retailer would set if it did not sell the manufacturer's product. We show that this can change the distribution of profits in the channel in a way that reverses conventional wisdom.
The rest of the paper proceeds as follows. Section 2 discusses related literature. Section 3 introduces the notation and describes the Hotelling model of demand in the familiar case when firms sell directly to final consumers. It then extends the model to include a retail sector in which a common retailer sells the products of competing manufacturers. Section 4 analyzes how marketing communications and advertising messages that affect cross-price elasticities influence firms' profits when only a subset of consumers are exposed to the advertising messages. Section 5 discusses some extensions and alternative model formulations. Section 6 concludes.
2. Related Literature
This paper contributes to the advertising and channels literatures. One branch of the modeling-oriented literature on advertising examines the informative role of advertising, where consumers are alerted to a product's existence, price, or product quality (see, for example, Grossman and Shapiro 1984, Milgrom and Roberts 1986, Zhao 2000, Dukes and Gal-Or 2003). A second branch considers the persuasive role of advertising, where consumers' preferences and valuations are affected by the content of the advertising itself (see, for example, Dixit and Norman 1978, Lal and Narasimhan 1996, Agrawal 1996, Bloch and Manceau 1999). A third branch considers how advertising can be used as a signalling and coordination tool between channel members (see, for example, Desai 1997, 2000). In this paper we focus on the persuasive role of advertising. Thus, we will assume that consumers already know about the product's existence, price, and quality, and that only consumers' willingness to pay are affected by the advertising communications.
Although it is common in the literature on informative advertising to assume that not all consumers are exposed to a firm's advertising, none of the previous literature on persuasive advertising has considered this possibility. As a result, we are the first to challenge the conventional wisdom that one of the goals of manufacturer advertising should be to decrease consumers' cross-price elasticities between the firm's own and its rivals' products.
The papers closest to ours are Lal and Narasimhan (1996), Agrawal (1996), and Shaffer and Zettelmeyer (2002). Lal and Narasimhan (1996) are concerned with whether advertising increases or decreases the margins of manufacturers and retailers. Agrawal (1996) is concerned with optimal levels of advertising and trade promotions as a function of consumers' brand loyalty. We focus instead on whether firms are better or worse off with ads that reduce cross-price elasticities. We also differ in that we allow for the possibility of nonlinear contracts (wholesale prices and fixed fees) and consider manufacturer-retailer bargaining. This eliminates potential problems such as double marginalization and allows us to place the model's emphasis on understanding how each firm might try to manipulate its share of the total channel profits.
Shaffer and Zettelmeyer (2002) are concerned with the effects of information provision on the profits of channel members when the information is supplied by third parties. Although not about advertising per se, their model can be adapted in a straightforward way to analyze advertising issues. However, in their model, all consumers are assumed to receive the advertising messages, whereas in our model consumers can be selectively exposed. This difference is crucial for our results (compare Observation 2 and Proposition 1 below). We also differ in that our focus is on cross-price elasticities, whereas the focus in their model is on core and noncore demand shocks, and we differ in that we allow for the possibility of manufacturer-retailer bargaining.
There is also a large, related literature on firm conduct in distribution channels (see, for example, Jeuland and Shugan 1983, Mathewson and Winter 1984,...
|
