Consumer Search and Alternative Market Equilibria

Abstract

The connection between imperfect information and imperfect competition has received much attention in recent literature. A variety of equilibrium outcomes have been obtained, for example perfect competition in Fisher (1972), pure monopoly in Diamond (1971), and price dispersion in Salop and Stiglitz (1977). This reflects corresponding differences in assumptions regarding consumers' information costs and demand functions, and firms' production costs and oligopolistic interactions. The purpose of this paper is to build a model sufficiently general to encompass these earlier results as special cases, and so bring out their mutual relationships. In doing so we compare the methodology involved in generating monopolistic competition due to consumers' imperfect information, with the methodology involved in generating monopolistic competition due to product differentiation originated by Chamberlin (1948). The present model considers a particular problem of limited price information concerning a homogeneous product. It is supposed that identical consumers know the distribution of prices charged in the market, but do not know which store charges which price. This information may be obtained at a cost which differs among consumers. The probability distribution of information costs over consumers is known to the stores. Given the stores' price distribution, each consumer decides whether to become informed. He enters the market only once. Informed consumers go to the lowest-price store, and uninformed ones choose a store at random. A consumer becomes informed if the utility to be had from paying the information cost and buying at the lowest price is higher than the expected utility from remaining uninformed and purchasing randomly. If purchasing information generates the same utility as random selection, the consumer chooses the latter. (Diamond's (1971) equilibrium can occur as a special case of this information structure.) Each store sets its price to maximize its profit and assumes in the BertrandNash manner that other stores will not change their prices in response. However, it calculates the effect of its actions on the consumers' information-gathering and, hence, on its sales (i.e. the equilibrium is a Stackelberg equilibrium between producers and consumers). All stores have identical U-shaped cost curves (i.e. increasing marginal cost is assumed, but the implications of assuming constant marginal cost are discussed as well), and there is free entry. [ABSTRACT FROM AUTHOR]