The purpose of this paper is to make an overview of the development of the Japanese urban system after the Second World War by means of comparing urban dimensions in 1950, 1960, and 1970.
The principal component analysis, which reduces a wide array of descriptive measures of objects to a series of fundamental factors, is used to extract urban dimensions. The cities for the 1950 analysis are 153 cities with the population of over 50, 000 in their administrative areas. The cities for the 1960 and 1970 analysis are all cities with the population of over 30, 000 in their DID's (Densely Inhabited District), 185 cities in 1960 and 242 cities in 1970 respectively. The input variables for each of the three time periods comprise five types of data: 1) populaton size and change, 2) population structure, and 3) social, 4) economic, and 5) locational characteristics. The numbers of the input variables for the three time periods are 34 variables in 1950, 50 variables in 1960, and 54 variables in 1970. And then, the tabulation of data for cities in the three time periods are based on the administrative areas of cities in 1970, in order to exclude the influence of the formal expansions of the administrative areas of cities.
As a result of the principal component analysis of the input data matrices for the three time periods, the writer extracted seven components in 1950 and eleven components in 1960 and in 1970, all of which had the eigenvalue of over unity. Major components (urban dimensions) in 1950, 1960, and 1970 are as follows:
Major components in 1950……the social status (eigenvalue; 11.1, percentage of the total variance explained; 32.5%), the rurality (5.8, 17.0%), the manufacturing (3.1, 8.8%), the urban size (1.0, 5.6%), the sex ratio (1.6, 4.6%), the transportation and communication (1.4, 4.1 %), the unnamed component (1.1, 3.4%) (Table. 1).
Urban components in 1960……the social status (11.5, 23.0%), the manufacturing (7.6, 15.2%), the centrality (5.9, 11.7%), the rurality (4.5, 9.1 %), the transporation and communication (2.5, 5.0%), the services (2.0, 3, 9%), the urban growth (1.8, 3.6%), the government (1.6, 3.1%), the urban size (1.3, 2.7%), the population density (1.1, 2.3%), the sex ratio (1.0, 2.0%) (Table. 2).
Major dimensions in 1970……the social status (14.0, 26.0%), the manufacturing (9.6, 17.8%), the transportation and communication (4.9, 9.0%), the urban size (3.9, 7.2%), the mining (2.4, 4.5%), the age structure (2.2, 4.0%), the goverment (2.0, 3.6%), the services (1.5, 2.8%), the sex ratio (1.5, 2.7%), the unnamed component (1.2, 2.2%), the unnamed component (1.1, 2.0%) (Table. 3).
From the comparison of the above components, the following points are recognized as the changes of the Japanese urban system after the Second world war.
1) The social status was always extracted as the first component in the three time periods, but the spatial distribution of the component scores has showed a remarkable contrast between cities located in the two largest Metropolitan Areas and cities outside them (Fig. 1). (The two largest Metropolitan Areas are the Keihin Metropolitan Area whose central cities are Tokyo, Yokohama, and Kawasaki, and the Keihanshin Metropolitan Area whose central cities are Osaka, Kyoto, and Kobe.) This result is interpreted as follows: the white-collar workers have been concentrated in the two largest Metropolitan Areas since 1950, especially after 1960. This interpretation is supported by the fact that the percentage of the white-collar workers in these areas to the nation has consistently risen since 1950 (Table. 4). As the concentration of the white-collar workers is nothing else but that of the management function, this result means that the management function has been concentrated in the two largest Metropolitan Areas after 1950, especially after 1960.
View full abstract