Location Theory

J. Gabszewicz

• Publisher: Taylor & Francis
• Year: 2013
• ISBN:
  9781136472824
• Edition: 1
• Topic: Betriebswirtschaft
• Subtopic: Business allgemein

Analyses the economic theory of urban land use in both its positive and normative aspects.

Spatial Competition and the Location of Firms

JEAN JASKOLD GABSZEWICZ and JACQUES-FRANÇOIS THISSE†

Université Catholique de Louvain, Louvain la Neuve, Belgium

On ne fait point de l’Industrie entre del et terre; il faut se poser quelque part sur le sol.

L. Walras, Eléments d’économie politique pure.

1. INTRODUCTION

A competitive market is often viewed as a central exchange place where a large number of buyers and sellers are in free intercourse with one another. Arbitrage among sellers and buyers then leads to the competitive price. Examples are the Stock Exchange, the Chicago Wheat Market, and the periodic markets where traders meet and transfer commodities on the spot. Let us call such a market a “concentrated” market.

In the real world, however, many market activities are performed at dispersed points in space, even though these activities concern the same product. The market now operates over a complex network of scattered buyers and sellers, due to the topography of space and the location decisions of firms and households. Examples of such “dispersed” markets abound in real life: think of department stores located in a town, gasoline stations located along a road, cement factories located in a country, etc.

Once the dispersed nature of many markets is recognized, then the natural companion assumption of a concentrated market, namely that it embodies numerous sellers and buyers, also needs revision. For a dispersed market it implies that each firm finds only a few rivals in its immediate neighborhood; further away there are more competitors but their influence is lessened by transportation costs. Similarly, not all consumers are alike to the firm; those who are far away will not buy from this firm because they have to pay too high a transportation cost. Thus “the market is commonly subdivided into regions within each of which one seller is in a quasi-monopolistic position” (Hotelling [50], p. 41). Under these circumstances, the possibilities of arbitrage are substantially reduced and firms enjoy some monopoly power. Consequently, in a dispersed market, the competitive assumption becomes hardly tenable and, as repeatedly advocated from Sraffa [91] to Greenhut [40], alternative approaches must be explored to describe the conscious interactions amongst few separated sellers and many scattered buyers.

The difference between the concepts of concentrated and dispersed markets has an analog, in industrial economics, in the difference between a market for a homogeneous product and an industry with differentiated products. In a market with a homogeneous product, substitutes are bunched into a single point of the space of characteristics, and sellers of this product may be numerous. In an industry with differentiated products, substitutes are dispersed in the space of characteristics, and the seller of a particular variety enjoys a quasimonopolistic position relative to the buyers who most preferred it. Thus the interest in modelling spatial competition which arises from dispersed markets immediately extends to the process of competition amongst differentiated commodities.

The standard models of microeconomics do not seem well-suited to analyze dispersed markets (or industry with differentiated products). The partial equilibrium model is designed to characterize market solutions when the product is homogeneous; the general equilibrium model is designed to explain price determination, taking into account the interdependencies among all commodities. Between these two extremes lies an intermediate degree of complexity, represented by the model of spatial competition. Unlike partial equilibrium analysis, this model allows for substitution; unlike general equilibrium analysis, it deals only with the range of products which are taken as belonging to the same industry.

In a pathbreaking paper, Hotelling [50] provided the framework for the basic model of spatial competition. The present article rests upon this work, and aims to provide an integrated overview of the contributions rooted in the same tradition. It is organized as follows. After reviewing the main results obtained in the theory of spatial monopoly (Section 2), we propose a unified framework describing the basic ingredients of the process of competition in a dispersed market (Section 3). Then we investigate the market solution for a given number of firms (Section 4) and in the case of entry of new firms (Section 5). Finally, we briefly discuss some possible reformulations of the basic model and present some concludi...