67884 East Asia and Pacific Economic Management and Poverty Reduction an eye on east asia and pacific 6 China Urbanization Review: Balancing Urban Transformation and Spatial Inclusion by Somik Lall and Hyoung Gun Wang Roughly one out of every 10 people in the world today is a resident of Box 1: The World Bank’s Urbanization a Chinese city. The United Nations projects that the number of Chinese Review Flagship Study urban dwellers will increase from over 622 million today to over 1 billion The study is part of a series of in circa 2030. How China manages its urban transformation will have country pilots under a global product considerable bearing not only for the Chinese economy and society, (Urbanization Review) being developed by the Finance, Economics and Urban but for the world at large. Has China’s urbanization amplified the pace Development Department (FEU) of the of economic prosperity? Has urbanization connected poor people World Bank. The objective of this product with prosperity and enhanced living conditions across the country’s is to provide diagnostic tools to inform vast landscape? What are priorities for public policy and investment policy dialogue and investment priorities to generate win wins for urban efficiency and social inclusion? This on urbanization by operationalizing the “Urbanization Review� helps answer these questions by diagnosing the framework for urbanization policies developed by the World Development health of China’s urban economy and urbanization process, highlighting Report (2009) – Reshaping Economic how policy challenges vary across the urban portfolio. Geography, as well as the Bank’s new Urban and Local Government China has embraced the urbanization process that has accompanied Strategy “Systems of Cities: Harnessing its dramatic economic transformation. By investing in the institutional Urbanization for Growth and Poverty Reduction�. To test the relevance and foundations, Chinese policymakers have sown the seeds to amplify flexibility of the core diagnostic tools the gains from urbanization. An urban land market has been created, across countries at incipient, intermediate and regulations standardizing the assignment of land use rights have and advanced states of urbanization, been established. And while the urban planning law in the 1980s country pilots are being initiated in China, aimed at controlling the size of large cities, the 10th Five-Year Plan Colombia, India, Indonesia, and Vietnam, (2001–05) and onwards instead chose to emphasize the synergistic as well as a case study of South Korea. Ideally, a sequence of policies that start development of China’s large, medium, and small cities.1 In places by developing the ‘soft structures’ or such as Changsha, Zhuzhou, and Xiangtan, which have a combined common institutions followed by the population of 13 million, planning guidelines have enforced land use ‘hard structures’ focusing on connective rights and promoted intensification in central areas, while expressways infrastructures can generate win-wins for and railways have improved connectivity among these cities and with efficient and inclusive development. their rural hinterlands. 1. Rapid urbanization accompanied by spatial efficiency and economic prosperity Spatial Transformation: Coastal areas are dominated by large cities, the hinterland by small and medium sized cities. Urbanization has been rapid along the coastline and the Yellow River (Figure 1). Table 1 shows the “system� of Chinese cities in this Urbanization Review. On average, the 12 cities in the largest group are over 10 times as large as the 133 cities in the 1 World Bank (2009). World Development Report – Reshaping Economic Geography, World Bank. 1 smallest group. And out of the 38 cities with more than 2.5 million residents, 28 are in Eastern part of the country. In contrast, the landscape of the central region is dominated by medium sized cities between 1 and 2.5 million people and the western region is dominated by smaller cities of less than 1 million. Figure 1. China’s urban footprint and urban system Urban footprint Urban system Source: Reproduced using GIS data by Schneider et al. (2009). Source: Authors’ own definition based on the 2000 population census. Table 1. China’s system of cities City group All cities 1 (>4M) 2 (2.5-4M) 3 (1.5-2.5M) 4 (1-1.5M) 5 (<1M) Number % Number % Number % Number % Number % Number % Eastern 8 66.7 20 76.9 20 44.4 29 40.8 24 18.0 101 35.2 Central 2 16.7 4 15.4 16 35.6 30 42.3 49 36.8 101 35.2 Western 2 16.7 2 7.7 9 20.0 12 16.9 60 45.1 85 29.6 Total 12 100.0 26 100.0 45 100.0 71 100.0 133 100.0 287 100.0 Source: the 2000 Population Census. Note: The city classification is based on urban population of permanent residents in circa 2000. The study focuses on 287 prefecture-level cities. Economic Transformation: Economic prosperity is concentrated in large cities with good access to international markets. Urbanization—particularly urban concentration in the largest cities has amplified economic progress. 76 percent of urban exports are generated in cities with 2.5 million or more people. And 62 percent of overall foreign capital is utilized in these cities. In particular, as illustrated in Figure 2, proximity to international markets has been the main driver of urbanization and urban economic success. Figure 2. Proximity to international markets drives city growth Economic concentration, GDP, 2007 Long distance migrants from other provinces, 2000 The urban economic structure is more complex when looked at closely. First, big cities tend to generate higher incomes, but there are significant variations within each city size group, likely to be contributed by economic geography conditions and other China urbanization review: balancing urban transformation and spatial inclusion 2 Box 2: Viewing China’s urbanization through a “System of Cities� This study employs a “systems of cities� perspective to evaluate China’s urbanization process and its urban economy. Chinese cities are classified into five groups by urban population size. In contrast with previous studies, the city groups are based on size of the permanent urban population rather than registered population. This is very important as unregistered migrant workers account for a significant share of urban population in China. The main sources of Chinese socioeconomic data at the sub-national level are the Statistical Yearbooks compiled by the National Bureau of Statistics of China (NBS). Provincial and City Yearbooks are compiled and published annually. However, as raw data are from individual provincial and local governments without strong expertise in statistical data collection, the quality of data are always questionable. Even for a simple urban population count, each city applies different criteria: some cities report registered (hukou) population, and others permanent people with actual residence of more than 6 months or de facto population. The majority of scholars concur that China’s 2000 population census is a relatively reliable source of data for analytic work. In addition, the 2000 population census was the first census to include permanent residents living in urban areas for more than six months rather than relying on the registered population in hukou system. This study uses the 2000 population census as the main information source and complements data from Statistical Yearbooks. For example, local GDP, local fiscal revenues and expenditures, and detailed public service delivery data (rather than access to services) are only available in Statistical Yearbooks. Statistical analysis of the total urban population residing in urban districts, county-level cities and counties of each prefecture city is obtained from the 2000 Population Census. The total urban population residing in urban districts is computed for county-level cities and counties of each prefecture city. 287 prefecture-level cities are grouped by the size of their urban population, and assigned into five groups of equal population mass (approximately 20 percent of the total urban population). Group 1 contains the largest cities and Group 5 the smallest ones. On average, cities in the largest group are 10 times bigger than those in the smallest group. The year 2000 definition of 287 prefecture level cities is back-casted and fore-casted to different years to construct time-consistent panel data. The main benefit of this city grouping of equal urban mass is it helps identify specific drivers or bottlenecks of Chinese city performance. Other things being equal, each group of cities with the same urban mass should perform similarly. Any differences are driven by other city-specific factors commonly shared in the same group. Figure 3. Strong correlates of urban productivity (city GDP per capita), 2007 Average income by city size (yuan, in thousands), 2007 Average income (2007) by city manufacturing share (2000), in thousands 100 80 80 60 60 40 40 20 20 0 0 <800K 800K–1.3M 1.3M–2.2M 2.2M–8M >8M 1: low manufacturing 2 3 4 5: high manufacturing excludes outside values excludes outside values Average income (2007) by urban clustering within 50 km (2000), in thousands Average income (2007) by proximity to BSH, in thousands 100 100 80 80 60 60 40 40 20 20 0 0 1: low urban cluster 2 3 4 5: high urban cluster 1: close to BSH 2 3 4 5: remote from BSH excludes outside values excludes outside values Note: Average city income is measured by GDP per capita. China urbanization review: balancing urban transformation and spatial inclusion 3 local factors (Figure 3). A city’s manufacturing employment share, urban clustering 2 , and access to international market are highly correlated with urban productivity (city GDP per capita). Table 2. City size by itself does not induce city growth Second, econometric analysis in OLS (1) (2) (3) (4) Table 2 shows proximity to Beijing, dependent variable ln(gdp per capita, 2007) annual city pop growth, 2007–2000 Shanghai or Hong Kong, share of ln(city pop, 2000) 0.207*** -0.035 -0.000 -0.000 workforce in manufacturing, and (0.047) (0.052) (0.001) (0.001) ln(distance to Beijing, -0.077** -0.004*** clustering of urban population in close Shanghai, Hong Kong, km) (0.035) (0.001) proximity as important contributors Share of manufacturing 2.944*** 0.008** to a city’s success—both in terms of employment, 2000 (0.360) (0.004) urban productivity (GDP per capita) ln(urban clusters in 50km, 0.095* -0.003*** 2000) (0.055) (0.001) and population growth. A city’s size Constant yes yes yes yes by itself does not explain success— Observations 287 287 287 287 its linkage with international and local R-squared 0.057 0.486 0.003 0.321 markets does.3 Note: 1. Robust standard errors in parentheses; * significant at 10%; ** significant at 5%; *** significant at 1% Third, not surprisingly, large inland cities are lagging behind though possessing high-quality human capital and relatively good consumer amenities (Figure 4). Figure 4. Inland large cities are lagging behind City population growth, 2007–2000 City GDP per capita and its growth, in thousands 3 70 12% 2 60 11% 1 50 10% 0 40 9% -1 30 8% -2 20 -3 10 7% -4 0 6% Coastal Large Cities Coastal Small Cities Inland Large Cities Inland Small Cities Coastal Large Cities Coastal Small Cities Inland Large Cities Inland Small Cities City population growth City urban population growth City rural population growth GDP per capita, 2007 Growth of GDP per capita, 2007–2000 Human capital Consumer amenities (relative to national average) 8 100 150 7 90 140 80 130 6 70 120 5 60 110 4 50 100 3 40 90 30 80 2 20 70 1 10 60 0 0 50 Coastal Large Cities Coastal Small Cities Inland Large Cities Inland Small Cities Coastal Large Cities Coastal Small Cities Inland Large Cities Inland Small Cities % of college grads, 2000 Colleges, per capita, 2007 Average floor space, per capita, 2000 Public transport vehicles, per capita, 2007 College teachers, per capita, 2007 College students, per capita, 2007 Water consumption, per capita, 2007 Doctors, per capita, 2007 Sources: The 2000 population census, 2007 permanent population estimates from ECUST, and the 2007 Statistical Yearbook. 2 It is measured by the total number of urban population within 50km radius. 3 Similar evidence is found in Brazil during 1970-2000 where city growth was primarily driven by market potential for goods, inter-city transport costs, and the population concentration in nearly towns and villages (Da Mata, et al, 2007). China urbanization review: balancing urban transformation and spatial inclusion 4 International experience highlights that manufacturing initially concentrates in large cities of countries at early stages of economic development (India), then disperses evenly across the urban system (Brazil), and finally becomes specialized in small cities and rural areas of mature systems (USA). Innovation and financial services tend to concentrate in largest cities. In China, large “coastal� cities have a disproportionate concentration of manufacturing employment, but not financial services (Figure 5). It suggests that Chinese economic structure is still in the early stage of industrialization and will undergo significant structural changes in the near future. In addition, large cities either coastal or inland have strong concentration of scientific research activities—similar to the USA. Finally, agriculture activities are more specialized in inland small cities. Figure 5. Economic specialization of China: the location quotient 2.0 2.0 1.8 1.8 1.6 1.6 1.4 1.4 1.2 1.2 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 Coastal Large Cities Coastal Small Cities Inland Large Cities Inland Small Cities Coastal Large Cities Coastal Small Cities Inland Large Cities Inland Small Cities Location quotient of manufacturing emp. 2000 Location quotient of whole sale/retail trade emp. 2000 Science and research Agriculture Finance of insurance Real estate Sources: Authors’ own calculation using the 2000 population census. Going forward, Chinese cities can amplify the gains from economic transformation by accelerating the pace of product market integration. Fluidity in the markets for goods will facilitate economic specialization and enhance urban productivity. While China has made massive investments in infrastructure and particularly in transport networks of all modes, its domestic logistic costs rank 72nd out of 150 countries.4 2. Many poor people far from prosperity While overall poverty reduction has been impressive, the coastal-inland development gap and the rural-urban divide are the two major components of spatial inequality in China. Spatial transformations have stimulated progress in large coastal cities— however many poor people in rural areas and small hinterland cities still need to be connected with prosperity. The rural poor account for 91 percent of the total poor; and western provinces (north and south) account for over 50 percent of China’s poor, though the western region only represents around 20 percent of the national population. In addition to rural-urban and broad regional divisions, the urban underclass is concentrated in second-tier cities (non provincial capitals).5 Four largest mega provincial cities (Beijing, Shanghai, Chongqing and Tianjin) have the lowest urban disadvantaged rate of around 1 percent. More than 80 percent of the urban underclass live in prefectural or lower-level cities. Recent policy initiatives indicate an increasing commitment of the government to a broader poverty reduction, social protection, and human development agenda. In line with China’s 11th Five-Year Plan, new spending programs target the development of an inclusive countryside with the objective to reduce the existing sharp rural urban disparities in public service delivery6 . Policies now address some of the inadequacies, inequalities and inefficiencies in the allocation of public resources for the delivery of basic public services at the national level targeting the basic needs of the poor through the national poverty alleviation strategy. 4 Arvis et. al, 2007. 5 The urban underclass represents a group of relatively poor people under twice of the World Bank poverty line. 6 Brixi, 2009. China urbanization review: balancing urban transformation and spatial inclusion 5 Figure 6. distribution of poverty 2003, in percent 2003, in percent 100 100 90 90 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Population Poor people Population Poor people Coastal Inland Rural residents Urban migrants Urban hukou residents Source: World Bank (2009). Investments under the national poverty alleviation program have Figure 7. Migration patterns been supporting basic infrastructure, including safe water and 2000, in percent roads, in poor counties and villages across China. Public spending 100 on the 9-year basic education has also been rising, emphasizing the need to accomplish the objective of universal 9-year schooling 80 in the Western provinces and in rural areas across provinces. 60 Similarly in health, new initiatives in the form of investments in public health infrastructure, New Rural Cooperative System in 40 rural areas, and the urban basic health insurance for unemployed urban residents, have been launched and backed by higher 20 public spending with the objective to enable all Chinese citizens to benefit from public health programs and essential care. But 0 Coastal Large Cities Coastal Small Cities Inland Large Cities Inland Small Cities China’s journey in bridging spatial inequalities in service delivery Migration from other provinces Migration from same province is not complete with the largest cities having consistently better Migration within the same city indicators for social and basic service. Source: The 2000 population census. Beyond access to services, the hukou system is the main barrier limiting the extent to which people can seek better economic opportunities. In general, it is more difficult to obtain a local hukou in large city such as Beijing, Shanghai, and Guangzhou. Big cities specialize in high-end and high value-added products demanding skilled workers. They also provide consumer amenities and public services which attract skilled workers. Urban migrants gravitate towards centers of economic opportunities—large cities offering higher wages and better public services. Limiting entry into these cities lowers the extent to which urbanization can be spatially and socially inclusive. 3. Fluid factor markets and differentiated public policies: key to efficient and inclusive urbanization 3.1. Fluid land, labor and product markets as a foundation Strategies that balance spatial efficiency and equity will be the main challenge for next round of urbanization in China. While objectives of spatial efficiency may seem at odds with spatial equity, evidence from countries such as the United States shows that enhancing fluidity of factor markets along with progressive policies can turn the apparent tradeoff into a win-win. In the United States factor markets at the national level integrated at different rates for goods, capital, and labor.7 Goods market integration was realized in the latter half of the nineteenth century, capital markets by the early twentieth century, and finally labor markets by the mid-twentieth century (Figure 8). Only after all three markets of goods, capital, and labor 7 Kim and Margo (2004). China urbanization review: balancing urban transformation and spatial inclusion 6 were nationally integrated, local wages and earnings became Figure 8: Factor market integration and the convergence of regional income per capita in the less tied to place, and regional economic structure and regional United States8 incomes finally converged significantly. In sum, the dissipation of regional disparities was realized only after goods, capital, and Later 19 century: th Early 20 century: th Mid 20 century: th Goods market integration Capital market integration Labor market integration labor markets were fully integrated. 250 200 China has been very successful in integrating goods market thanks to massive construction of trunk transport infrastructure. 150 However, from the USA experience, goods market integration by 100 itself cannot induce convergence in regional and local incomes. Other factor markets, such as labor and capital, need to be fluid 50 and integrated into the national market. Fluid factor markets are the key for inclusive urbanization. In addition to fluid labor market, 0 1840 1860 1880 1900 1920 1940 1960 1980 1990 what are priorities for public policy and investment? ne enc sa wsc pc ma wnc esc mt Source: Kim and Margo, 2004. 3.2. A typology for policy prioritization Note: ne refers to the New England region; ma the Middle Atlantic region; enc the East North Central region; wnc the West North Central region; sa the South Atlantic region; esc the East South Central region; wsc the West South Central region; mt the Mountain region; and pc the Pacific region. To be most effective, urban policies of each city need to be tailored to meet specific urbanization challenges, which are interlinked with its position in the urban hierarchy (systems of cities) and economic geography endowments. The following typology can be proposed to differentiate priorities across different places in China’s urban portfolio: Figure 9. Typology for prioritizing urban policies in China City size China's urban system Big Small Coastal Group I Group II Economic geography Inland Group III Group IV • Group I: Big cities in coastal areas, such as Beijing, Shanghai, and Guangzhou. Those cities have grown rapidly serving as a main gateway to access international markets. They have served as engines of Chinese economic growth. • Group II: Rapidly expanding suburban cities near big coastal cities. • Group III: Large inland cities. These cities are excluded from the current development cluster, but possess high-quality human capital and good amenities, which will serve as a foundation for economic development. • Group IV: small hinterland cities and rural towns. These cities constitute the bottom tier of the Chinese urban hierarchy, and have strong linkage to agricultural activities in neighboring rural areas. 8 The chart plots regional income per capita, 1840-1990 (U.S. = 100). It is quoted from Kim and Margo (2004), and originally from Easterlin (1960, 1961), Statistical Abstract of the U.S., various years. China urbanization review: balancing urban transformation and spatial inclusion 7 References 1. Arvis, J-F., M. Mustra, J. Panzer, L. Ojala, and T. Naula (2007), Connecting to Compete: Trade Logistics in the Global Economy – The Logistics Performance Index and Its Indicators, World Bank, Washington DC. 2. Brixi, H. (2009), “China : Urban services and governance,� The World Bank Policy Research Working Paper Series, No. 5030. 3. Da Mata, D., U. Deichmann, V. Henderson, S. Lall, and H. G. Wang (2007), “Determinant of City Growth in Brazil,� Journal of Urban Economics, 62, 252-272. 4. Kim, S. and R. Margo (2004), “Historical Perspectives on U.S. Economic Geography,� in: J.V. Henderson, J.F. Thisse (Eds.), Handbook of Regional and Urban Economics, vol. 4, North-Holland. 5. Schneider A., M. Friedl, and D. Potere (2009), “A new map of global urban extent from MODIS satellite data,� Environmental Research Letters, 4. 6. World Bank (2009), “From Poor Areas to Poor People - China’s Evolving Poverty Reduction Agenda,� World Bank Report, No. 47349-CN. An Eye on East Asia and Pacific is a series of short notes related to the East Asia and Pacific Region prepared by the region’s economists. These notes are not peer reviewed or edited. They do not represent the official position of the World Bank Group, its directors, or management. Contact the authors for discussion on the individual notes, or Ivailo Izvorski (iizvorski@worldbank. org) about the overall series or to have a note considered. www.worldbank.org/eapeye China urbanization review: balancing urban transformation and spatial inclusion 8