GIS Based Studies in the Humanities and Social Sciences - Chpater 6

6 Urban Employment Areas: Defining Japanese Metropolitan Areas and Constructing the Statistical Database for Them Yoshitsugu Kanemoto and Reiji Kurima CONTENTS 6.1 Introduction..................................................................................................85 6.2 The Need for a New Metropolitan-Area Definition..............................86 6.3 Metropolitan-Area Definitions in the U.S. ..............................................87 6.4 The Structure of Japanese Metropolitan Areas.......................................89 6.5 Defining Urban-Employment Areas.........................................................89 6.5.1 Requirements for a Core................................................................90 6.5.2 Requirements for an Outlying Area.............................................91 6.5.3 The Iterative Procedure for Defining UEAs...............................91 6.5.3.1 The First Iteration.............................................................91 6.5.3.2 The Second Iteration ........................................................92 6.5.3.3 Other Iterations.................................................................92 6.6 Urban-Employment Areas for the 1995 Population..............................93 6.7 The Construction of the MEA Economic Database ...............................94 6.8 Conclusion ....................................................................................................97 Acknowledgment..................................................................................................97 References...............................................................................................................97 6.1 Introduction For those interested in analyzing urban activity, the first task should be to define urban areas. The legal definition of a city is a natural starting point, but many urban activities extend beyond jurisdictional boundaries. For example, many workers in large metropolitan areas commute from suburban 85 Copyright © 2006 Taylor & Francis Group, LLC 86 GIS-based Studies in the Humanities and Social Sciences jurisdictions to central cities. We therefore need a definition of an urban area within which most everyday activities are undertaken. An urban area typi-cally comprises a core area that has significant concentrations of employ-ment, which is surrounded by densely settled areas that have close commuting ties to the core. In the United States, the federal government has defined metropolitan areas since 1947 and provides a variety of statistical data relating to them. There is no counterpart in Japan, and the only definitions of metropolitan areas available are those proposed by a small number of researchers. Most of these adopt standards similar to the Standard Metropolitan Statistical Area (SMSA), which was in use in the 1960s and 1970s. However, in the U.S., two major changes have occurred since then, which reflect changes in the pop-ulation distribution and activity patterns. First, in the 1980s, the Consolidated Metropolitan Statistical Area (CMSA) was introduced, which connects met-ropolitan areas that have significant interactions. Second, a new definition known as the Core-Based Statistical Area (CBSA) was introduced for the 2000 population census. In Japan, changes in metropolitan areas motivated a revision of the first generation of metropolitan-area definitions. Kanemoto and Tokuoka (2002) proposed a new metropolitan-area definition to deal with complicated inter-action patterns in Japanese metropolitan areas. The newly defined metro-politan areas are known as Urban-Employment Areas (UEAs), because they are based on employment patterns. The UEAs are divided between Metro-politan-Employment Areas (MEAs) and Micropolitan-Employment Areas (McEAs) according to their sizes. Researchers affiliated with the Center for Spatial Information Science at the University of Tokyo have been construct-ing a database for UEAs. In this chapter, we explain the definition of the UEA and a method of constructing an economic database for them. 6.2 The Need for a New Metropolitan-Area Definition As noted above, a number of researchers have developed their own defini-tions of metropolitan areas. Examples are the Regional Economic Cluster (REC) of Glickman, the Functional Urban Core (FUC) of Kawashima, and the Standard Metropolitan Employment Area (SMEA) of Yamada and Tokuoka. These SMSA-type definitions apply different standards to central cities and suburban areas. According to the SMEA, a central city requires a population of at least 50,000, a percentage of nonagricultural workers of at least 75 percent, at least as many daytime occupants as nighttime ones, no more than 30 percent of the population commuting out, and no more than 15 percent commuting to another central city. A suburban municipality requires a percentage of nonagricultural workers of at least 75 percent and at least 10 percent of the population commuting to the central city. Copyright © 2006 Taylor & Francis Group, LLC Urban Employment Areas: Defining Japanese Metropolitan Areas 87 The idea of defining central cities and suburban areas separately is attrac-tive because of its simplicity. It first defines central cities, and then finds suburban areas for each of them, and the process does not involve iteration. However, it has shortcomings. For example, a city with a high population density may not be included in a metropolitan area. For example, Yamaguchi city, which is the capital of the Yamaguchi prefecture, did not belong to an SMEA until 1985. It was not classed as a central city because it had fewer daytime occupants than nighttime occupants and at the same time did not satisfy the conditions for being a suburb of another city. Recently, this problem has become increasingly serious because of the emergence of a large number of subcenters and because of increasingly complex commuting patterns. If we use commuting ties to define a suburban area in relation to a particular central city, a city that is close to more than one central city may not belong to a metropolitan area. In the 1995 population census, there were 441 cities with populations of at least 50,000, of which 60 could not be classed as either central cities or suburbs of an SMEA. Of these 60 cities, 16 have populations of at least 100,000. Given that a single city with a population of 100,000 can itself be classed as an SMEA, excluding these cities from metropolitan areas is not consistent. Many cities that do not belong to an SMEA are suburban areas from which at least 30 percent of the population commutes out. Typically, commuters have more than one city to commute to. Almost 50 percent of these cities are located on the periphery of the Tokyo SMEA. To deal with these problems, we can relax either the requirements for central cities or those for suburban areas. For example, the core of a metro-politan area may include subcenters with sufficiently large concentrations of employment even if they satisfy the requirements for classification as suburban areas of a central city. In the Tokyo metropolitan area, Yokohama, in which employment was about 1.4 million in 1995, could be included in the core area. Another possibility is to modify the requirements for suburban areas so that they take account of commuting to other suburban cities. SMEAs have three types of requirement for central cities: namely, popu-lation size, urban characteristics, and the employment core. Of these ele-ments, the employment-core requirements should be reexamined first so that areas with significant population concentrations are not excluded from met-ropolitan areas. Another problem with SMEA relates to the requirements for urban characteristics. The percentage of nonagricultural workers represents this element, but it is no longer an effective index of urbanization. 6.3 Metropolitan-Area Definitions in the U.S. In revising the Japanese metropolitan-area definition, it is useful to study other countries that have experienced a similar trend of increasingly complex Copyright © 2006 Taylor & Francis Group, LLC 88 GIS-based Studies in the Humanities and Social Sciences metropolitan areas. In the U.S., there was a major revision in 2000 with the introduction of a new metropolitan-area definition known the Core-Based Statistical Area (CBSA). According to the Office of Management and Budget (2000), a CBSA is a geographic entity associated with at least one core of 10,000 or more population, plus adjacent territory that has a high degree of social and economic integration with the core as measured by commuting ties. The standards designate and define two categories of CBSA: Metropol-itan Statistical Areas and Micropolitan Statistical Areas (Office of Manage-ment and Budget, 2000, p. 82, 236). A Metropolitan Statistical Area is associated with at least one urbanized area that has a population of at least 50,000, and a Micropolitan Statistical Area is associated with at least one urban cluster that has a population of at least 10,000 but less than 50,000. A CBSA is identified in four steps. First, a CBSA must contain sufficiently large urban (densely settled) areas. Specifically, it must have an urbanized area, as defined by the Census Bureau, of at least 50,000 people, or an urban cluster, as defined by the Census Bureau, of at least 10,000 people. Second, the core of a CBSA comprises a central county or counties asso-ciated with the urban areas. Specifically, a central county or counties must: (a) have at least 50 percent of its population in urban areas of at least 10,000 people; or (b) have within its boundaries a population of at least 5,000 located in a single urban area of at least 10,000 people. Third, outlying counties of a CBSA must satisfy the commuting require-ment that: (a) at least 25 percent of the employed residents of the county work in the central county or counties of the CBSA; or (b) at least 25 percent of the employment in the county is accounted for by workers who reside in the central county or counties of the CBSA. Fourth, closely connected CBSAs are merged into one CBSA. In particular, two adjacent CBSAs merge into one CBSA if the central county or counties (as a group) of one CBSA qualify as outlying counties to the central county or counties (as a group) of the other CBSA using the measures and thresholds stated in (a) and (b) above. Because of institutional differences, we cannot apply the U.S. definitions to Japanese cities. The most important difference is that the Japanese gov-ernment does not define urban areas that extend beyond jurisdictional boundaries. The nearest equivalent in Japan is a Densely Inhabited District (DID) defined within a local municipality. The DID is defined by the Statistics Bureau as an area that is a group of contiguous Basic Unit Blocks, each of which has a population density of 4,000 inhabitants or more per square kilometer, or which has public, industrial, educational, and recreational facil-ities, and whose total population is 5,000 or more within a local municipality. U.S. requirements for outlying areas have changed considerably. First, the measures of settlement structure, such as population density, that had been used to define outlying counties are no longer used; currently, commuting data are used. The reason for this change is that “as changes in settlement and commuting patterns as well as changes in communications technologies Copyright © 2006 Taylor & Francis Group, LLC Urban Employment Areas: Defining Japanese Metropolitan Areas 89 have occurred, settlement structure is no longer as reliable an indicator of metropolitan character as was previously the case” (Office of Management and Budget, 1999). In Japan, metropolitan areas have expanded into rural areas, and the use of settlement structure may no longer be relevant. Second, the percentage commuting out was raised from 15 percent to 25 percent, because the percentage of workers in the U.S. who commute to work outside their counties of residence increased from approximately 15 percent in 1960 to almost 25 percent in 1990. In Japan, we should also reconsider the com-muting ratio. However, it is not clear whether the ratio should be raised, because, as we explain later, commuting patterns in Japan are much more complicated than in the U.S. 6.4 The Structure of Japanese Metropolitan Areas According to the 1995 population census, 724 cities and towns have DID populations of at least 10,000, of which 440 have total populations of at least 50,000. The number of cities with DID populations of at least 50,000 is 297. Many urban areas are dormitory towns, and relatively few of these are employment centers. Of the 724 (297) cities and towns with DID populations of at least 10,000 (50,000), only 281 (120) have larger commuter inflows than outflows. Some large cities, such as Yokohama, Chiba, and Kawasaki, have larger outflows than inflows. However, the central wards of these cities are employment centers that have larger commuter inflows than outflows and DID populations of at least 50,000. Consider commuting patterns. The average commuter-outflow proportion is 32 percent, but employed residents commute to a wide variety of urban areas. Some cities have highly concentrated commuting patterns: More than 50 percent of the employed residents of 16 cities and towns (Tomiya, Wako, Urayasu, Fuchu, Komae, Hoya, Nagayo, Kokufu, Uchinada, Ichikawa, Sanwa, Kouyagi, Ishikari, Musashino, Hashikami, and Yakumo) commute to one city. Another extreme is Zama, where more than 5 percent of the employed residents commute to one of six other municipalities. 6.5 Defining Urban-Employment Areas Kanemoto and Tokuoka (2002) proposed a new metropolitan-area definition known as the Urban Employment Area (UEA). UEAs are divided between Metropolitan Employment Areas (MEAs) and Micropolitan Employment Areas (McEAs) according to their sizes. These are similar to the CBSA for U.S. cities, but there are substantial differences in specific requirements to Copyright © 2006 Taylor & Francis Group, LLC ... - tailieumienphi.vn