The Spatial Structure of Commercial Rents by Retailing and the Changes of Land Value

Abstract

In the latter half of 1980's, major Japanese cities experienced tremendous rise in land values that caused serious social problems. But there have so far been few contributions by geographers which dealt with commercial land values, still less rent. This paper has three purposes. The first is to estimate rent on retail activities by means of available statistics. The second is to illustrate its spatial organization in an urban area within the framework of location theory (Fig. 2). The third is to explore the changes of commercial rent and land values focusing on their gaps.Figure 3 shows the Osaka Metropolitan Area for this case study which includes 34 cities lying within about 20km from the CBD. Although it is necessary to examine closely individual management in order to calculate the rent correctly, the author proposes here a simple method of estimating aggregated rent by business type and district on the basis of the statistical data. This approach makes it possible to carry out an investigation into the variation of rent level easily from the spatial point of view. Two business types are selected, i. e. retail trade-dry goods, apparel and accessories (abbreviated to retailing-apparel) as relatively high-order goods, and retail trade-food and beverages (retailing food) as low-order goods. The average rent is defined as excess profit calculated by deducting buying cost and wages for employees from annual sale per floor area as shown in Table 1.In the central district intensive land use results in higher rent than other areas, and retailing-apparel exceeds retailing-food in terms of rent. In the suburban area the rent level of retailing-food is often higher than that of apparel. The difference of rent level of retailing-apparel between those cities tends to grow large, while that of retailing-food becomes smaller as a whole (Fig. 4). The higher rent the business earns, the more its floor area increases geometrically. To a certain degree this reflects the mutual determinative relation between rent and location which forms the main thesis in location theory.Commercial rent is likely to depend not upon the distance to CBD as the so-called rent gradient supposes, but upon the size and nature of the market area in which a firm is located. Central place theory illustrates that its size is represented by a circle with a radius of range of goods. Assuming the size is constant, total demand within the market area is given by the product of demand per capita and consumer density. The range, however, varies by kind of goods, thus hierarchical central place systems are built up. With this theory in mind, three indices were chosen as the appropriate factors expected to influence commercial rent, i. e. population density, income level and centrality. Their spatial distributions present a concentric pattern (Figure 5), a sector pattern (Figure 6), and an hieralchical order pattern (Figure 6) respectively.The results of correlation analysis are shown in Table 2. Centrality has a significant positive coefficient for the rent of retailing-apparel, and income level for that of retailing-food, because customers usually purchase food at the nearest center, while in the case of apparel whose range is much longer they prefer centers in higher order. The indicator of population density has only limited significance for each rent against expectation. This is probably due to the fact that location changes of the retail activities and movements of population synchronize toward an equilibrium. Therefore it is possible to understand that the spatial organization of determinants of commercial rent in the metropolitan area is a compound of the three patterns (Table 3).According to the hypothesis of competitive bidding, the higher one of the rent by retailing-apparel and -food was chosen. And it was capitalized by the general long-term rate of government loans, using a formula of market fundamentals.