The Internet has led to a large number of third-party sources that offer high-quality information about firms's products at little or no cost to consumers. As a result, many of these sources have grown in popularity, extending well beyond the usual reach of traditional third parties such as Consumer Reports and Kelly's Blue Book. For example, the online version of Edmunds offers, at no cost to consumers, information about new products, existing products, long-term tests, and buyes' guides, all relating to the automotive industry. AvWeb.com delivers weekly aviation news and new product reviews to its readers, and a large number of websites follow developments on computer platforms such as the Apple Macintosh. In this paper we analyze how the provision of third-party information affects the division of profits in a multiproduct distribution channel. To illustrate, consider the competition between Microsoft and Apple in the operating systems (OS) market and their channel relationship to CompUSA, a retailer that sells both Macs and Windows-based PCs. Consider two pieces of third-party information. First, suppose that CNET, an Internet technology site, reviews the newest upgrade of the MacOS and writes that the new user interface is even easier to use than previously. Second, suppose that an article in the technology section of the Wall Street Journal notes that changes in Apple's networking support now enable Macs to be better integrated into PC networks. These two pieces of information are similar in the sense that they both express good news about the MacOS and thus they both can be expected to benefit Apple by increasing consumer demand for Macs. One might also expect that in both cases CompUSA will capture some of the gains that come from the increased demand for Macs and that Microsoft will lose because the good news about the MacOS will induce some consumers to choose Macs over Windows-based PCs. However, we will show that this intuition is incorrect. The two reviews can have surprisingly different implications for the profits of Microsoft and CompUSA. The reason is that the two reviews differ on one crucial dimension: the group of customers for whom they are primarily relevant. The CNET review talks about improvements in the customer interface – precisely what Apple's core consumers care about. The Wall Street Journal review talks about compatibility with prevailing PC standards – important to consumers who care relatively more about compatibility and who are thus more likely to prefer Windows (Apple's noncore consumers). We show that good news about the MacOS that is more relevant to Apple's core consumers (the CNET review) benefits Microsoft but harms CompUSA, while good news about the MacOS that is more relevant to Apple's noncore consumers (the Wall Street Journal review) has the opposite effect. It harms Microsoft but benefits CompUSA. Stated more generally, our main result is that when third-party information affects consumers' product valuations, the type of information that induces the change is critical to understanding which firms gain and which firms lose. In particular, depending on the type of third-party information, we find that (1) a retailer can be harmed by good news about a product that it carries; (2) a manufacturer can gain from good news about a rival's product; and (3) good news about a product category need not benefit all the manufacturers in that category. There are three novel features of the analysis. First, we derive the equilibrium division of profit among firms when a retailer sells the products of competing manufacturers, and we have done so while placing few restrictions on the feasible set of contracts. Second, we show how this equilibrium division of profit lends itself to a simple graphical interpretation that depicts which firms gain and which firms lose from third-party information. Third, we provide a taxonomy of information types and identify the key features of each type that cause profit incentives to vary. In particular, we conceptualize information as having three components, namely (1) the products to which the information pertains, (2) whether the information is positive or negative, and (3) the consumers to whom the information is relevant. We show that all three information components play a role in determining the change in each firm's profit. Our framework can also be used to analyze a variety of other settings of interest; for example, it can be used to analyze profit incentives when the retailer has bargaining power, when there is downstream competition, and when there are non-information-based changes in consumers' valuations. In addition, our framework may be used both to analyze the effects on profits of persuasive advertising and to predict advertising content.
