Spatial pricing in interdependent markets: testing assumptions and modeling price variation. A case study of gasoline retailing in St Cloud, Minnesota

In this paper we present an empirical evaluation of assumptions about consumer purchasing behavior for gasoline. Previous research has developed a theoretical model of spatial pricing in oligopolistically competitive markets in which it is hypothesized that retail prices vary because of both consumer price sensitivity and the choice sets available to consumers as well as awareness of prices at competing locations. With the use of household survey data collected from St Cloud, Minnesota we evaluate the plausibility of these assumptions, finding evidence to support the consumer purchasing behavior assumed in the theoretical model. By means of a spatial time series of gasoline price data for the same metropolitan area, we develop an empirical model of spatial price variation that incorporates some of the hypotheses of the original model. The results suggest support for the proposition that spatial price variations depend on the service characteristics of individual retailers and the accessibility or locational advantage of individual gasoline stations within the spatial configuration of the urban market. There also is empirical support for the conjecture that those sites which are more accessible, have larger choice sets, and charge lower prices tend to be those which attract the most sales from other retailers.