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

85 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 2713_C006.fm Page 85 Monday, September 26, 2005 2:05 PM 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. 2713_C006.fm Page 86 Monday, September 26, 2005 2:05 PM 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 2713_C006.fm Page 87 Monday, September 26, 2005 2:05 PM 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 2713_C006.fm Page 88 Monday, September 26, 2005 2:05 PM 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 2713_C006.fm Page 89 Monday, September 26, 2005 2:05 PM Copyright © 2006 Taylor & Francis Group, LLC 90 GIS-based Studies in the Humanities and Social Sciences reflect the higher densities and more complicated commuting patterns in Japanese cities. In defining metropolitan areas for Japan, we must take the following four conditions as given. 1. The building blocks of metropolitan areas are municipalities (cities, towns, and villages), because most statistical data are only available up to the municipality level. 2. The prevalence of commuting by car in smaller metropolitan areas has increased the numbers commuting to cities from areas of low population density. The use of population-density standards in defining outlying areas is no longer practical. 3. In identifying densely inhabited urban areas, we use DID populations. 4. The inclusion of outlying suburban areas is determined by commut- ing patterns between municipalities. After examining many alternatives, Kanemoto and Tokuoka (2002) pro- posed the following definition of a UEA. First, local municipalities (cities, towns, and villages) are the building blocks of UEAs. The core of a UEA is a collection of densely settled municipalities (i.e., those with DID pop- ulations of at least 10,000) that do not constitute the “outlying municipal- ities” (suburbs) of any other core. The outlying municipalities of a UEA are defined mainly by the requirement that at least 10 percent of employed workers commute to the core. An MEA is a UEA whose core has a DID population of at least 50,000. An McEA is a UEA whose core has a DID population of at least 10,000 and less than 50,000. More specifically, they use the following standards. 6.5.1 Requirements for a Core 1. In the first round, potential cores are municipalities with DID pop- ulations of at least 10,000. 2. A municipality that is an outlying area of another central city is excluded from being a core. 3. For a pair of municipalities that each satisfy the requirement for being in a core and also satisfy the commuting-ratio requirement for being an outlying area of the other municipality, the one with the lowest commuting ratio is in the core, and the other is its outlying area. 4. An outlying municipality is included in the core if the following two requirements are satisfied. A “major” city (known as a Seirei Shitei Toshi ) for which data on its wards are available is included in the core if at least one of its wards satisfies the following requirements. 2713_C006.fm Page 90 Monday, September 26, 2005 2:05 PM Copyright © 2006 Taylor & Francis Group, LLC Urban Employment Areas: Defining Japanese Metropolitan Areas 91 a. The employees-to-residents ratio (i.e., the ratio of the number of employees to the number of residents) is at least unity. b. The DID population is at least 100,000 or one-third of the core. Condition 4 implies that a core may contain more than one municipality. The reason for adding the condition on the size of the DID in 4b is that the largest metropolitan areas, such as Tokyo and Osaka, have multiple central cities, and some very small municipalities satisfy the employees- to-residents ratio. It is inappropriate to include a municipality of 5,000 in the core alongside the central city of Tokyo, which has more than 7 million employees. 6.5.2 Requirements for an Outlying Area 1. A municipality is an outlying area of a core if at least 10 percent of its employed residents work in the core. 2. If a municipality satisfies condition 1 for more than one core, it is included in the outlying area of the core with which it has the strongest commuting ties. 3. A second-order outlying municipality that is an outlying area of another outlying municipality is included in a UEA. Higher-order outlying municipalities (i.e., third-order, fourth-order, etc.) are also included in UEAs. The criterion for a second-order municipality is that, of all the target municipalities, its commuting ratio to a first- order outlying municipality is the highest and satisfies the 10 per- cent criterion. Higher-order outlying municipalities are defined analogously. 4. If a municipality simultaneously satisfies the requirement for being an outlying area of a core and the requirement for being another outlying municipality, it is classified as an outlying area of the one with which it has the highest commuting ratio. That is, if 16 percent of the employed residents in city A work in core B and 17 percent of them work in city C, which is an outlying municipality of core B, then city A is an outlying area of city C. 6.5.3 The Iterative Procedure for Defining UEAs UEAs are determined by the following iterative procedure. 6.5.3.1 The First Iteration 1. Choose municipalities with DID populations of at least 10,000 as potential central cities. 2713_C006.fm Page 91 Monday, September 26, 2005 2:05 PM Copyright © 2006 Taylor & Francis Group, LLC 92 GIS-based Studies in the Humanities and Social Sciences 2. Exclude as potential central cities defined by 1, those municipalities that are outlying areas of other potential central cities. 3. Determine the outlying municipalities for the potential central cities by using the following procedure. a. Select municipalities for which the percentage of employed res- idents who work in a central municipality is at least 10 percent; and for each of them, determine the central municipality that has the highest commuting ratio. This identifies potential first-order outlying municipalities. b. Determine potential second-order outlying municipalities by choosing municipalities that satisfy the commuting-ratio criteri- on. c. Determine potential second-order outlying municipalities. d. Determine potential third-order outlying municipalities. e. Check for fourth-order outlying municipalities. (Currently, there are none.) f. If a municipality is simultaneously a first-order, second-order, or third-order outlying area, identify the target municipality with the highest commuting ratio. g. List the outlying municipalities for each central city. 6.5.3.2 The Second Iteration 1. Of the potential outlying municipalities identified in the first itera- tion, those that satisfy the following two requirements are included in the cores of the UEAs to which they belong. If a candidate is a “major” city, it is included in the core if at least one of its wards satisfies the following requirements. a. The employees-to-residents ratio is at least unity. b. The DID population is at least 100,000 or at least one-third of that of the central municipality. 2. Potential first-order outlying municipalities are those in which at least 10 percent of employed residents work in the central munici- pality. For each of these, choose the central municipality that has the highest commuting ratio. The procedures applied in the first itera- tion are applied to determine higher-order outlying municipalities. 6.5.3.3 Other Iterations The procedures used in the second iteration are used for subsequent itera- tions. Step 1b in the second iteration for adding an outlying municipality to the core remains the same. In particular, the central municipality is the one 2713_C006.fm Page 92 Monday, September 26, 2005 2:05 PM Copyright © 2006 Taylor & Francis Group, LLC Urban Employment Areas: Defining Japanese Metropolitan Areas 93 identified in the first iteration and does not include those added in the second iteration. 6.6 Urban-Employment Areas for the 1995 Population For the 1995 population census, three iterations were needed to define the UEAs. The total number of UEAs is 278, of which 118 are MEAs and 160 are McEAs. Figure 6.1 maps Japan’s MEAs. Table 6.1 presents a rough outline of the requirements of UEAs. Figure 6.2 shows MEAs, McEAs, DIDs, and municipality boundaries in Ibaraki Prefecture. The darkest gray areas within MEAs and McEAs are DIDs. The second-darkest areas are the cores of MEAs and McEAs, and light-gray areas are their outlying areas. Light-gray curves show boundaries of municipalities (cities, towns, and villages). FIGURE 6.1 The Metropolitan Employment Areas for the 1995 population census. Sapporo Sendai Tokyo Nagoya Osaka Hiroshima Fukuoka 2713_C006.fm Page 93 Monday, September 26, 2005 2:05 PM Copyright © 2006 Taylor & Francis Group, LLC 94 GIS-based Studies in the Humanities and Social Sciences 6.7 The Construction of the MEA Economic Database Since the building blocks of the UEAs are local municipalities, the UEA data can be obtained by summing the municipality-level data. However, data on many important economic indicators are not available at the municipality level. For example, most production data are available only at the prefectural level. Kurima and Ohkawara (2001) constructed MEA data for total produc- tion (value added), private capital, and social-overhead capital. TABLE 6.1 Requirements of Urban Employment Areas Requirement Urban Employment Area Categories Metropolitan Employment Area: The DID population of the core is at least 50,000. Micropolitan Employment Area: The DID population of the core is at least 10,000 and less than 50,000. Qualification of Areas City of at least 10,000 DID people. Qualification of Central Municipalities (Cores) Municipalities that satisfy either of the following two requirements are included in the core. (The core may include more than one municipality.) (a) The DID population is at least 10,000, and the municipality is not an outlying area of another core. (b) The requirements for an outlying area are satisfied, and the following two requirements are also satisfied. (i) The employees-to-residents ratio is at least unity. (ii) The DID population is at least 100,000 or one-third of the core. For a pair of municipalities, each of which satisfies the requirement for being in a core and also satisfies the commuting-ratio requirement for being an outlying area of the other municipality, the one with the lowest commuting ratio is in the core, and the other is its outlying area. Qualification of Outlying Municipalities (a) A municipality is a first-order outlying area of a core if at least 10 percent of its employed residents work in the core. (b) A municipality is a second-order outlying area of a core if at least 10 percent of its employed residents work in an outlying municipality and the commuting ratio to the municipality is the highest among all target municipalities. Higher-order (i.e., third-order, fourth-order, etc.) outlying municipalities are defined analogously. (c) If a municipality satisfies requirement (a) for more than one core, it is included in the outlying area of the core with which it has the strongest commuting ties. (d) If a municipality satisfies the requirement for an outlying area of a core, as well as for another outlying municipality, it is an outlying area of the one with which it has the highest commuting ratio. 2713_C006.fm Page 94 Monday, September 26, 2005 2:05 PM Copyright © 2006 Taylor & Francis Group, LLC [...]... then Y ( A) = N ∑ ∑ Y( I , i) N((Ii,,jj)) j i Second, the private capital in each MEA is obtained by using proportional allotment on the basis of the production shares in manufacturing and nonmanufacturing industries (rather than employment shares) Copyright © 20 06 Taylor & Francis Group, LLC 2713_C0 06. fm Page 96 Monday, September 26, 2005 2:05 PM 96 GIS- based Studies in the Humanities and Social Sciences. .. used for each of the four types of social- overhead capital In agriculture, forestry, and fishing, social- overhead capital is allocated on the basis of production shares in the agricultural sector In industrial infrastructure, allocation is based on the production shares in the manufacturing industry In telecommunications and railways, allocation is based on total production Allocation of infrastructure... Kurima and Ohkawara (2001) allocate the prefecture-level data to municipalities by using proportional allotment and then aggregate them to obtain MEA data In prefectures that include the major cities, they construct the data for the remaining areas and allocate them to municipalities outside those cities The production data are available for 14 industrial categories (12 industries, the public sector, and. .. others) The output of each industry is allocated to municipalities on the basis of the industry employment shares For example, consider an MEA denoted by A Denote the number of workers in industry i in municipality j by N ( i , j) , and denote the total production and employment of the prefecture (denoted by I) that contains the MEA by Y ( I , i) and N ( I , i) , respectively Total production in the. .. 2002 (in Japanese) Kurima, R and T Ohkawara, Construction of MEA -based Economic Data, mimeo, 2001 (in Japanese) Office of Management and Budget, Recommendations From the Metropolitan Area Standards Review Committee to the Office of Management and Budget Concerning Changes to the Standards for Defining Metropolitan Areas, Federal Register, 64 (202), October 20, 1999 Office of Management and Budget, Standards... researchers and government agencies have already used the UEA Chapter 8 of this book contains an example of empirical studies that use the MEA economic database Acknowledgment This research was supported by Grant -in- Aid for Scientific Research No 10202202 and no. 166 1002 from the Ministry of Education, Culture, Sports, Science, and Technology References Kanemoto, Y and K Tokuoka, Proposal for the Standards... urban-area definition called the UEA for Japanese cities The UEAs are divided between MEAs (large UEAs) and McEAs (small UEAs) according to their sizes We have also constructed an economic database for the MEAs The UEAs for the population censuses of 1980, 1990, 1995, and 2000 and the MEA economic database for the 1995 definition can be found on the UEA Web site (www.urban.e.u-tokyo.ac.jp/UEA/ index_e.htm)... Report on the Prefecture Accounts Private Capital Stock and Social Overhead Capital: Estimates by the Central Research Institute of the Electric Power Industry (CRIEPI) (The method of estimation is explained in Ohkawara et al., 1985.) Copyright © 20 06 Taylor & Francis Group, LLC 2713_C0 06. fm Page 97 Monday, September 26, 2005 2:05 PM Urban Employment Areas: Defining Japanese Metropolitan Areas 6. 8 97 Conclusion... data for three industry categories — namely, primary, secondary, and tertiary — but these are only available for prefectures and 12 large cities We use these data for Q in the procedure described above The original data sources for the MEA economic database are as follows Number of employees and population: Population Censuses of 1980, 1985, 1990, and 1995, and Basic Resident Registers (Jumin Kihon Daicho)... Registers data to modify the simple linear interpolations of the census data Let P denote the census population data and let Q denote the Resident Registers data The annual ˆ data that we use, Pn+ i , are then i Q ˆ Pn+ i = Pn + ( Pn+ 5 − Pn ) n+ i ˆ 5 Q n+ i where n is the census year, i = 1, 2 , 3 , 4 , and i ˆ Q n+ i = Q n + (Q n+ 5 − Q n ) 5 The Annual Report on Prefecture Accounts contains annual employment . for being an outlying area of the other municipality, the one with the lowest commuting ratio is in the core, and the other is its outlying area. 4. An outlying municipality is included in the. 20 06 Taylor & Francis Group, LLC 90 GIS- based Studies in the Humanities and Social Sciences reflect the higher densities and more complicated commuting patterns in Japanese cities. In. 2713_C0 06. fm Page 93 Monday, September 26, 2005 2:05 PM Copyright © 20 06 Taylor & Francis Group, LLC 94 GIS- based Studies in the Humanities and Social Sciences 6. 7 The Construction of the