Ownership Concentration and Market Power in Urban Land Markets

In many North American cities the recent period of inflation has been accompanied by a very rapid increase in land prices, resulting in a significant increase in the relative price of housing.' One of the popular targets for " blame " for this phenomenon is large developers and landowners who have been accused of restricting development and raising prices for their own gain. However, an adequate analysis of the determinants and consequences of the existence of market power in urban land markets cannot be found in the urban literature, where competitive conditions are assumed to prevail. In this paper we develop a theoretical analysis of the effect of concentration of ownership on the allocation of resources in urban land markets, focusing on the relationship between market structure and the spatial distribution of occupancy and land prices in an urban area. We will show that because of the spatial heterogeneity of land, a natural entry barrier exists which implies that concentrated ownership always confers potential market power on large landowners, and that exercised market power necessarily distorts the spatial distribution of occupancy and land prices. In particular, we show that leapfrog development is one possible result of exercised market power, providing a possible explanation for this common urban phenomenon. 1. THE MODEL We will use a model of a circular city which is adapted from Solow's version (1973) of the standard urban location model. Since we wish to examine the relationship between concentration, per se, and market power, we will assume that there is an arbitrarily large amount of homogeneous land so that there are no " entry " barriers resulting from limited supplies of specialized land. For simplicity, we will assume the Central Business District (CBD) is a circular region of fixed size with radius R. We will ignore the production aspect of the CBD other than to assume a composite consumption good and composite housing good are produced. The consumption good is produced by a perfectly competitive industry at constant costs. The composite housing good, also produced by a perfectly competitive industry, is a flow of housing serviceg consisting of housing and land in fixed proportions, so that the housing composite is measured by lot size. The supply price of a unit of housing services at distance x from the CBD, excluding the cost of land, is C(x), where C'(x) > 0. Thus C(x) includes the cost of providing and maintaining serviced land and housing at distance x from the CBD, but not the (rental) cost of the raw land. With no loss in generality, we assume that land outside the CBD has no economic use other than for housing. Households thus consume a single composite consumption good and a composite