Social network externalities and price dispersion in online markets
Since 1998 several empirical studies have addressed the large price dispersion in online markets, large as it is in conventional offline retailing, it is persistent and it is not fully explained by observed e-retailers attributes. Moreover, some consumer online markets presents high degrees of sales concentration, making that the largest e-retailers have large market shares, as it occurs in the online bookstore industry where only two e/retailers Amazon and Barnes and Noble, dominate almost 85% of the market. The main contribution of this research is to propose another driver of price dispersion in online environments: the social network externalities. Internet is not a spot market. Buying in internet market can be risky in both space and time: (a) involves a delayed time between the time consumer pays and the time he or she receives the order, and (b) the product and the consumers are spatially separated. Like any other innovation diffusion, online market expansion involves a consumer adoption process that is fed by the interaction of consumers' experiences in the form of references, word of mouth, digital word of mouth or just by imitating behavior. The interaction of consumer choices introduces network externalities in the demand for e-commerce products. Two theoretical models of social network externalities and the supply and demand of e-services are developed in this research. The second model addresses two issues: how social network externalities operate into the consumers demand function and what prediction can be derived from the demand model. The model takes into account some stylized facts of the purchasing experience of buying in internet. For example, that buying in internet involves some risk, that consumers are risk averse, and that they try to assess the probability of a satisfactory delivery from the e-retailers sampling opinions and purchase experiences from others consumers. The main hypothesis of this research is that high network size sellers in e-markets are more attractive to consumers because they can assess the seller's probability of a satisfactory delivery more precisely compared with sellers with low activity in the market. For this reason consumers are willing to pay a price premium to large network e-retailers. The rest of the derived hypotheses in this research are that this gain in price premiums is a diminishing function of the network size, and that the network size effect is more important for high-value products than for low-value ones. Finally, consumers are also willing to pay a premium to e-retailers with a higher probability of a satisfactory delivery, and that this delivery effect is, again, more important for high-value product. These five hypotheses derived from this model are tested using auctions of a low value product -DVD box set of a TV show- and a high-value product -a middle price digital camera-, in the electronic auction market e-Bay. In all econometric estimations, the sample evidence is consistent with the hypotheses of this research. The second model examines the implications of social network externalities in the framework of a duopoly market structure game of an incumbent and a follower firm game. The model shows that when social network externalities are present in the demand equation, the firm that enters the market in the first period will profit the network advantage building a large customer installed base producing an equilibrium, which is characterized by price dispersion.