Cities, Vol. 17, No. 2, pp. 123–135, 2000 2000 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0264-2751/00 $- see front matter
Land market forces and government’s role in sprawl The case of China Tingwei Zhang* Urban Planning and Policy Program, College of Urban Planning and Public Affairs, University of Illinois at Chicago, Room 235, 412 S. Peoria Street, Chicago, IL 60607, USA
While urban sprawl is a controversial topic among researchers, the majority of planners have expressed concerns about sprawl’s environmental and social costs, and recommended policies for sprawl control. This research analyzes urban sprawl in China through a comparative approach, comparing the pattern and driving forces of sprawl in China to those of the US. It identifies main factors contributing to sprawl in China: the land market, local government’s willingness to lease land as the result of new tax revenue regulations, and the decentralization process after China’s economic reforms. Facing an enormous loss of cultivated land in the last decade, the central government has put a control policy in place in recent years. However, the policy has shown limited success. This paper finds that it is the presence of interest groups within the public sector that makes government consensus over sprawl hard to reach, which aggravates the sprawl problem. 2000 Elsevier Science Ltd. All rights reserved Keywords: Urban sprawl, Economic reforms, Interest groups
jurisdictions may have an impact on sprawl. It is widely accepted that sprawl occurs more often in fragmented metropolitan areas where numerous government jurisdictions have power over land use and zoning regulations (Richmond, 1995; Downs, 1994; Rusk, 1993). Second, public policy, especially federal policy on taxes, housing, transportation and infrastructure investment, is believed to have significant impact on the occurrence of urban sprawl (Moe, 1995; Parker, 1995; McDonald, 1993). In light of these research findings, this article intends to examine government’s role in sprawl in a different context — China — through a comparative approach on sprawl issues in China to that of the US. Since China’s economic reforms started in 1978, many scholars have noticed significant changes in the political economy and urban landscape of the country. Researchers found that these changes “altered the lines of authority and the flow of resources between the central government and municipalities”. As a result, “cities become economically more autonomous from Beijing and more embedded in their immediate
The phenomenon of disproportional expansion of urbanized areas into undeveloped land is generally called “urban sprawl”, but the definition of “urban sprawl” and whether sprawl is a “desirable” development form remain controversial. In recent years, the majority of urban planners have expressed concerns about sprawl’s possible consequences, especially its economic and social costs (Real Estate Research Corporation, 1974; Ewing, 1997; Urban Land Institute, 1998; Burchell, 1997; Freilich and Peshoff, 1997; Burchell et al., 1998). Researchers particularly focus on the relationship of government and sprawl: the organizational structure of government bodies, public policy at all levels, and their contribution to urban sprawl. This relationship in the American context has been examined from two perspectives. Researchers have found that, first, the organization of local government *Tel.: ⫹ 1-312-355-0303; e-mail: [email protected]
Land market forces and government’s role in sprawl: T Zhang
locale; the visual result was an urban-rural sprawl” (Davis et al., 1995). The concept of “urban sprawl in China” was first introduced early in the 1980s (Fung, 1981), and a number of researchers have made contributions on the topic. Naughton noticed the spillover of urban activities into suburban regions “to create new districts of undifferentiated urban sprawl”. For him, sprawl in China is more a result of changing internal economic forces that brought previously urban economic activities into peripheral areas. He also pointed out that government, rather than the marketplace, has a responsibility for sprawl: “district specialization is neither happenstance nor controlled by the market . . . rather, is based in both state priority for certain types of development or in a lack of state priority . . . [which] leaves land open for other uses” (Naughton, 1995). Shue’s research on Xinji, a small town in northern China, found that in addition to the expansion of urbanized area into farmland, sprawl in the Chinese context has another dimension: the sprawl of state power. She noticed the conflict between central government and local state bureaucracy. While decentralization reduced central government’s intervention in local life, “the local state’s presence now seems both more selective and more efficacious” (Shue, 1995). In analyzing the structure of Chinese cities, scholars noted the importance of China’s “work-unit” system (danwei in Chinese). As Wu indicated, there were many forms of conflicts between central and local government over land premiums, as well as between municipality and farmers, and state work-units and the public. Wu also pointed out a potential coalition for development consisting of municipal government, overseas capital, and “big builders” (large state workunits) (Wu, 1997). In summary, researchers have correctly attributed China’s urban sprawl to the land market force created after reform, and to the changes in power distribution over urban development. However, most researches tend to view the public sector (government at the central and local level) as a whole, although they may find differences between the central and local government over some issues. This study has found another key factor to sprawl: the presence of various interest groups within the public sector. It is the conflicts among these interest groups that aggravate sprawl and defeat efforts on sprawl control. To begin, this article paints an overall picture of China’s urbanization, and the Chinese version of urban sprawl in comparison to the American pattern, followed by an analysis of the government’s role from two perspectives, the organizational structure of government and the relationship among various government bureaus. The research examines land policies of three different groups of government departments: departments in charge of revenue, public spending, and land and land use administration, and highlights the contribution of conflicts between the 124
three interest groups to sprawl. Main findings are presented in the conclusion section. Data and case study information about China’s sprawl were collected from interviews in four Chinese cities: Beijing, Shanghai, Xiamen and Shenzhen in 1997 and 1998. A total of 32 senior government officers, city planners, and academic researchers provided insight on China’s urbanization and sprawl issues. The selection was based on interviewees’ involvement in land use research in recent years. Secondary data come from official publications, mainly Chinese statistical yearbooks of both the central and local statistical bureaus. Since the Chinese government has issued a new land management regulation (The Land Act of China, 1998), some new developments are discussed in the study as well.
China’s urbanization and the Chinese version of urban sprawl China’s urbanization and urban growth There have been a multitude of reports on urbanization in China, from both Chinese scholars and their foreign peers (to name some, including Dowall, 1993; Kirkby, 1985, 1996; World Bank, 1997; Wu, 1997; Song, 1992). Although methods of measuring China’s urbanization level, just as methods of measuring its GNP, have long been a controversial topic, by all measures, China is still at a low urbanization level. The World Bank reported that 30 percent of the Chinese population was urban in 1995, with 11 percent of total land as cropland and 43 percent as permanent pasture in 1994 (World Bank, 1998). The official Chinese statistical report indicates that 29.04 percent of the population was urban in 1995, and the proportion increased slightly to 29.92 percent in 1997 (State Statistical Bureau of People’s Republic of China, 1998). The State Statistical Bureau also reported that cultivated land was 94.97 million hectares, or 9.89 percent of total land in 1996 (State Statistical Bureau of People’s Republic of China, 1998). However, the most recent national land use survey found that China’s cultivated land was 134.4 million hectares in 1997, 41% more than the amount reported by the statistical bureau. (Zhao et al., 1998). The new figure has been accepted by international organizations (see FAO’s web page www.fao.org/WAICENT/faoinfo/). China’s city classifications divide cities into four types, plus towns — activity centers of townships and some counties. A city with a population of more than 1 million is a mega city, over 500,000 but less than one million is a large city, from 200,000 to 500,000 is a medium city, and less than 200,000 is a small city, defined by the City Planning Act of China, 1990. Table 1 shows the numbers of the four groups of cities, together with towns, in 1981 and 1990. Although the data shows the number of cities increasing, no official comprehensive data about the amount of China’s urbanized area have ever been published. Chinese researchers have estimated the
Land market forces and government’s role in sprawl: T Zhang Table 1
China’s cities and towns, 1981 and 1995
City Mega city Big city Medium city Small city Town Total
18 28 70 117 2678 2911
32 43 192 373 16992 17632
Source: Department of City Planning, Ministry of Construction of China, 1997.
distribution across the four groups of cities in 1990 (Ma, 1997. See Table 2). According to their estimate, mega cities had about one-third of the urbanized area and 41 percent of total urban population. From 1990 to 1995, the growth rates of medium size and mega cities have increased faster than that of the two other groups (Table 3). For instance, urbanized land in mega cities grew at an average rate of 9.8%, and the population of these super cities increased by 2.88% annually from 1990 to 1995 (Zhao et al., 1998). In the classification of non-farm use land, there are “urbanized areas” in cities and towns, and “non-farm use land” (residential, village owned manufacturing, and institutional uses) in rural areas in China. The amount of China’s urbanized area in cities and towns had increased to 42,400 square km (22,100 square km in cities and 20,300 square km for towns) in 1995, plus an additional 138,100 square km rural non-farm use land in the same year. The two land use categories accounted for 1.88 percent of China’s total territory (China Department of City Planning, 1997a). In that same year, the urban population increased to 180 million (State Statistical Bureau of People’s Republic of China, 1998). Table 2
While urban population in cities increased by 21.6% from 1990 to 1995, the urbanized area increased by 90.4% in the same period. This means that the growth rate of urbanized land was much faster than that of the urban population, as summarized in Table 4. The rapid growth of urbanized areas, particularly in mega cities, has caught the attention of the central government, especially some top leaders. The main concern is the loss of cultivated land. Statistics shows that China’s per capita cultivated land has decreased significantly since 1949. In 1949, per capita cultivated land was 1800 square meters and it decreased to 1133 square meters in 1995, a loss of over one third in 45 years (Ma, 1997; Li and Nie, 1996). Given China’s huge population base, “Will China be able to feed its population?” became a hot topic attracting worldwide attention. The government’s official status is that national land use policy should be built on the principle of food self-sufficiency, which means cropland preservation (Hu, 1998). By the end of 1996, the State Council issued an administrative order that no more cultivated land would be allowed to be converted to other uses for a whole year. The State Council also asked for comprehensive research on the causes of cultivated land losses, which resulted in the birth of the Land Act of 1998. In reporting the amount of China’s cultivated land and its loss, the State Statistical Bureau always warns in footnotes “underestimated, subject to further verification” (see notes of the Bureau’s statistical yearbooks from 1988 to 1995), thus making it difficult to calculate the accurate amount of loss. From the statistical yearbooks, the amount of cultivated land was 96.85 million hectares in 1985, reduced to 94.97 million in 1996. If these numbers were correct, the loss would be 1.88 million hectares, or 1.9% of the total cultivated land in a decade (Table 5).
China’s urbanized area in cities and towns (1990)
Urbanized area (sq. km) Percent of urban area Urban population (million) Percent of urban population
3655.4 31.5% 60.04 40.7%
1673.2 14.4% 20.90 14.2%
3003.5 25.9% 34.87 23.6%
Small city 3276.2 28.2% 31.71 21.5%
Total 11608.3 100.0% 147.52 100.0%
Source: Ma, 1997. Table 3
Annual growth rate of Chinese cities, 1990–1995
Number of cities
⬍ 200,000 200,000 to 500,000 500,000 to 1 million >1 million
5.1% 10.4% 9.0% 3.2%
7.2% 13.8% 11.4% 14.4%
4.7% 13.9% 6.4% 49.0%
Source: Based on Zhao et al., 1998, Table 1-1-1, p. 4.
Land market forces and government’s role in sprawl: T Zhang Table 4
Growth of urban population and urbanized area in China
Urban population Urbanized area
148 million 11,608 square km
180 million 22,100 square km
⫹ 21.6% ⫹ 90.4%
Source: Based on China Statistical Yearbook 1991 to 1996. Table 5
Cultivated land in China, 1985–1996
Amount of cultivated land (million ha.) As percent of China’s territory
Source: China Statistical Yearbook, 1986 to 1997.
The Chinese statistical bureau also released annual cultivated land change from 1978 to 1995. According to the statistical yearbook, from 1985 to 1994, a period when China experienced rapid urbanization, cumulative loss of cultivated land was 7.82 million hectares. The cumulative loss is about 8% of China’s total cultivated land in 1985 (State Statistical Bureau of People’s Republic of China, 1998). Chinese researchers estimated that the average loss of cultivated land was 580,000 hectares per year from 1986 to 1993 (Li and Nie, 1996; Ma, 1997). If this number is accepted, the amount of cultivated land loss was 5.8 million hectares in the last ten years (1986 to 1996), meaning that about 6 percent of China’s cultivated land had been converted to non-farm uses over the decade. The most accurate data may come from an internal report of the Department of City Planning (DCP) of the Ministry of Construction. The 1997 report showed that the cumulative loss of cultivated land was 68,442 square km, or 6.8 million hectares from 1986 to 1995 (China Department of City Planning, 1997b, see Table 6). What contributes to cultivated land loss? Official reports indicate four main categories of loss from 1990 to 1995. Structural adjustment of the agricultural sector was the number one contributor (66.84 percent), mainly the conversion of cultivated land to other non-urban uses such as orchard, meadow and timberland. The second contributor was urban construction activity (18.52 percent) including residential, industrial, infrastructure development and other capital improvement uses. Third, rural construction activiTable 6
ties (10.62 percent) included rural industrial developments and public uses in rural areas. The remainder was due to residential construction by and for farmers (4.04 percent) (State Statistical Bureau of People’s Republic of China, 1998, see Fig. 1). Market forces had a significant impact on land use conversion in rural areas. Since the increase of the market price of food crops (rice and wheat mainly) was slower than that of other farm products such as meat, fruit and wood after the reform, it is understandable to see a very large amount of cultivated land being converted to orchard, meadow and timberland. Urban construction activity has contributed 18.52% to the loss of cultivated land, which means about onefifth of the loss was the result of urban growth. Further evidence of rapid urban growth, an obvious phenomenon in all Chinese cities, could be found at the local level. Scholars have reported the growth of “built-up area” (in most cases, it means the central area) in Chinese cities. Table 7 shows the change of built-up area of the four city categories from 1990 to 1995. According to the report, mega cities such as Chongqing and Shanghai have demonstrated impressive expansion, and their growth rates were faster than that of other city categories. Fig. 2 illustrates the urban growth of Shanghai, China’s largest city. The urbanized area grew rapidly in the reform period; at the same time, there have been many plots of vacant farmland in the urbanized area, especially in the four new urban districts: Pudong, Jingshan, Baoshan and Jiading. In 1949, the city’s jurisdiction was 618 square km, and the urban district (“Shi-Qu” in Chinese, meaning
Cultivated land loss in China, 1986–1995 (in square km)
Source: China Department of City Planning, 1997b.
Land market forces and government’s role in sprawl: T Zhang
Figure 1 Causes of cultivated land loss
Land expansion of built-up area in Chinese cities, 1990–1995
City Size Mega city (population > 2 million)
Growth Rate from 1990 to 1995 (%) Beijing Tianjing Shanghai Guangzhou Chongqing
Mega city (1 million to 2 million) Medium city (500,000 to 1 million) Small city (200,000 to 500,000)
20% 7.3% 56.2% 38.3% 112.7% 13.5% 18.3% 23.9%
Source: Based on Zhao et al., 1998, Table 1-1-3, p. 5.
Figure 2 The growth of Shanghai
Land market forces and government’s role in sprawl: T Zhang
the built-up urban area) was 23 square km with a population of 4.19 million. By 1982, when China started economic reforms in cities, the municipality managed a jurisdiction of 6185 square km and the urban district was 232 square km with a population of 6.27 million. In 1995, although the jurisdiction remained almost the same (6340 square km), the urban district had increased to 2057 square km (including the Pudong New District and four previously rural counties: Jingshan, Baoshan, Jiading and Shanghai County, now Minghong District). The population in this area increased to 9.57 million. Table 8 shows the growth of the urban district of Shanghai from 1949 to 1995. While the population in the urban district increased by 37 percent from 1985 to 1995, the land of the district grew 486 percent in the same period of time. It may be argued that there are other factors contributing to the urban district growth, such as the adjustment of the district boundary and the creation of new urban districts in recent years. Pudong New District alone has added 523 square km to the urban district after reassessing boundaries, but the area has not been fully built-up. However, even without the addition of all these “new” urban districts (most land in these new districts had already been urbanized), the “old” urban district itself had expanded to about 400 square km in 1995; it is still 123% as large as in 1985. (In August 1999, the central government has approved Shanghai’s new master plan. The new plan predicts that the urban district will be 585 square km in 2010.) Not only did mega cities like Shanghai grow rapidly, medium cities and small towns had similar growth patterns. Xiamen, a city of 327,200 urban residents in 1980, grew to 552,800 in 1996. Its urbanized area increased from 14.9 square km to 62.5 square km in the same period. With a population increase of 68.9 percent, urbanized land increased by 320 percent. Although urbanization does need land, land conversion to urban uses is much faster than the increase of urban population. Moreover, urban growth often takes cultivated land of the best quality in terms of agricultural productivity and accessibility to the market-fringe areas of cities and towns. Table 8
The Chinese version of urban sprawl It is important to distinguish urban sprawl from “rational” or “normal” urbanization. Although defining urban sprawl is still a controversial topic in the US, most researchers tend to characterize urban sprawl with the following characteristics: (1) leapfrog or scattered development, (2) commercial strip development, and (3) large expanses of low-density or single-use development (Ewing, 1997; Moe, 1995). For a developing country with a low urbanization level like China, normal urbanized area growth with increasing rural–urban migration is a common process of modernization. The Chinese version of urban sprawl is the disproportionate expansion of the urbanized area. Although the Chinese context is quite different from the US, there are similarities in urban sprawl patterns of the two countries. The first is the disproportionate conversion of farmland to urban uses. The population of the Chicago metropolitan area increased 4 percent in the last two decades, but its urbanized area increased by 45 percent in the same period ((Openlands Projects, 1998) McDonald et al., 1992). The urbanized area surrounding New York City expanded by 65 percent, while the population grew by only 8 percent between 1960 and 1985 (Richmond, 1995). As mentioned above, the population of the urban district (Shi-Qu) of Shanghai increased by 37 percent, but the urbanized area grew by 486 percent from 1985 to 1995. The second character of the Chinese version of sprawl is scattered development in the urban fringe. The US version takes a form of development located on the edges of existing communities or leapfrogging into previously undeveloped areas. The Chinese version has some variations but retains the main feature of the dispersed development pattern. New development districts (kaifa qu in Chinese) of various types such as “High-tech industry park”, “Tariff-free district”, and “Special economic development zone” are often located in areas that were previously farmland. Those development zones increased so rapidly that in only one year (1991 to 1992) the number of development districts increased from 117 to 8,700 and occupied 16,000 square km of land, although fewer new
The growth of Shanghai (1949–95)
Total population of municipality (million) Total land of municipality (sq. km) Urbanized area (sq. km) Population in central area (million)
Change 1985–95 154 (2%) 0.84 (7%) 1706 (486%) 2.59 (37%)
a Including the Pudong New District and four new urban districts (previously rural counties). Without the five new urban districts, the urbanized area was 390.2 square km in 1995. However, a big proportion of land in these new urban districts is for urban uses.Source: Shanghai Statistical Yearbook, 1986 to 1996.
Land market forces and government’s role in sprawl: T Zhang
zones were added in recent years (Zhao et al., 1998). Most districts were located away from the often crowded and fully developed existing urban centers to attract foreign investment. However, in many cases, local government designated new development districts but either had no financial ability to complete (or even to start) the project or no foreigners to invest in it. The result is hundreds of hectares of idle land surrounding one or two incomplete high-rise office buildings on the far edges of these cities. (This is especially true in some small and medium cities, such as Beihai in Guangxi Province.) A 1995 survey found that only 2% of the development zones had been completed by the end of 1992 (Zhao et al., 1998). Low density and commercial strip developments are not characteristics of the Chinese version of sprawl, in contrast to the American version of urban sprawl. Also, central city decline resulting from negative impacts of urban sprawl is not a Chinese phenomenon. Central cities are still booming in China, although statistics show that since 1982 they have been slowly losing their population (Zhou and Meng, 1998). The main cause of population loss in central districts is the high price of housing. Not only are ordinary city residents forced to move out from central areas, but average enterprises are often forced to relocate to municipal fringes due to problems of affordable land in central areas. New housing projects in fringes have provided ordinary urban residents better living conditions, at the cost of longer commuting time, and the loss of cultivated land in peripheral areas. While “richer” people in general prefer to live in suburban areas in the US, it is the “poorer” who have to move to fringe areas in Chinese cities. This is one of the most important differences between the Chinese and the US versions of sprawl. Causes of urban sprawl in China In discussing causes of sprawl, American researchers have listed market forces, problematic public policy, and the American culture (“the American character”), as main contributors. Government policy on taxation, public spending, and infrastructure provision have especially been questioned (Richmond, 1995; Katz and Bernstein, 1998). In China, we found factors which are similar to the American case. Disproportional urban land growth was almost nonexistent before the reform. Land was public property; land transfer and changing land uses were all administrative decisions, not economic actions. No market forces were involved because no land market existed. When a factory needed land for a new workshop, it applied for and gained land from local government with token reimbursement to farmers. Since land was “free of charge”, no one could make any profit from land development or land transference. The real estate industry had been totally removed from the Chinese economy by the beginning of the 1956 “Socialization Movement”. Also, since more than 80 percent of the urban population lived in public housing, no concept
of a “housing market” existed in urban areas. Property exchange and transference did exist in rural areas where 99% of the residents owned their homes, but housing construction was for self-consumption rather than for profit. Because land was also owned by the public collectively in rural areas, it was still almost free of charge. Moreover, from the demand side, as a developing country, China had extremely limited capital to invest in urban development projects, except on “productive projects” such as factories and power stations, so land demand was limited. The Chinese government adopted a strict urban population control policy that eliminated the possibility of free rural-to-urban migration, thus significantly reducing the demand for new housing developments in urban areas. The consequence of these policies was low levels of urbanization, low demand for land, and limited urban growth. Since the introduction of the market economy after reform, the economic value of land has been fully recognized in China. All land-related policies are based on the understanding that land is a primary economic element and should generate revenue for the government. The 1982 Constitution recognized two kinds of land ownership: urban land is the property of the state and rural land is collectively owned by villagers. With additional amendments, the 1988 Constitution was the first national document that legalized the separation of land ownership from land use rights; it also allowed the transference of land use rights. The State Council passed the Regulation of Urban Land Use Rights in 1988 and enacted the regulation in 1990 (China Academy of Social Science, 1992). The 1988 Constitution and the Regulation laid the foundation for land policy in the post-reform era. Although there is still no privately-owned land in China today, a land market has been created and land use rights entered the market, under the policy of separation of use rights from ownership (Dowall, 1993; Li, 1997; Yeh and Wu, 1996). Land became a main source of profits for real estate companies, villages, individual farmers, and government of all levels as well. These interest groups thus emerged with the rebirth of land as a market force in China. From the demand side, while wealthy foreign investors seek land for their projects, urban residents try their best to improve their living conditions, and rural–urban migrants look for basic housing in cities. Under the pressure of high demand, land prices have skyrocketed. For instance, in Shanghai, acquisition cost of one Mu (a Chinese land measurement unit, one Mu equals 0.16 acre) of land was about US$90 in the early 1970s, and then increased to US$ 2,700 in 1988 and US$183,342 in the Special Economic Zone in 1995 (Shanghai Municipal Statistics Bureau, 1988, and from interviews). In the decision-making structure, the reforms have decentralized a considerable amount of decision power to localities and enterprises, which has significant impacts on land disposition, and ultimately on 129
Land market forces and government’s role in sprawl: T Zhang
urban sprawl. City and town governments and enterprises have been given power over local land use regulation and financial decisions. Supervision over land use from superior administrations has largely been removed. Land disposition is no longer an administrative decision but rather an economic action. All parties involved in land development and transference manage land simply based on its economic value with little consideration of long-term consequences of their actions, except economic considerations. At the same time, a tax sharing arrangement between the central government and provinces was adopted. Before the reform, under the planned economy, the central government provided funding to almost all local development projects. The new revenue distribution regulation entitles local government to share tax revenue, but it cuts most funds that previously came from the central government. Similar arrangements have been made within local government entities. A local municipal government now has to find its own funding for its infrastructure projects, schools, and most of its welfare programs. The pressure on municipal government is tremendous. Since land is the most valuable “commodity” under the control of a municipal government, generating revenue from leasing land use rights and charging land use fees has become the most popular practice for local government. Today, there are three types of revenue from land in China: land tax, charges on land leasing, and a variety of land use fees. Land tax has existed since 1949, but it was eliminated in urban areas in 1956, as a result of the Socialization Movement that nationalized all urban economic activities. Because all enterprises, institutions, housing stocks and urban land were owned by the public sector, a land tax on an enterprise or an institution would become a charge on government itself. After the reform, however, investors received land use rights for up to 70 years and thus became the real “owners” during the leasing period. A land tax is charged on users now. Land tax has two forms: urban land use tax and cultivated land occupation tax. The former refers to tax on urban land currently in nonfarm uses, and the latter is the tax on farmland acquisition. According to the current Chinese tax regulation, the two kinds of land tax are to be shared by the central and local governments. As a result, only 50 percent of cities collect this tax today. Many municipalities prefer not to charge land tax on local enterprises so as to keep revenue from being lost to the central government, especially in recent years when local enterprises have been facing financial problems. Land leasing charge refers to the charge on leasing land use rights. Because the state owns all urban land, as defined by the Constitution, revenue from leasing land use rights is to be shared by the central and local governments. However, local governments often do not think it fair to give “a big piece of the pie” to the central government, because it is local government 130
that invests “local money” on land development before the land can be leased. This is still a problem to be solved. For local governments, the land leasing charge is the most important revenue from land. The leasing charge alone annually brings municipalities more than 10 billion yuan (8.3 yuan equals US$1, 1998) to China’s east coast, reported the People’s Daily (the Chinese Communist Party’s official newspaper). From 1990 to 1995, Shanghai’s Pudong New District, a newly developed area of the city, had collected US$190 million directly from land leasing (People’s Daily, August 23, 1997). The Shanghai municipal government had collected $8.51 billion plus 10.3 billion RMB (about $1.3 billion) from land leasing in the period of 1988 to 1996 (Zhao et al., 1998). Experts estimated that, on average, 15 to 20 percent of local revenue came from land leasing in recent years. Most of the revenue is spent on infrastructure improvement and extension projects, which makes urban sprawl financially possible. Land use fees are charges paid to local government for its provision of infrastructure services to users. These fees are fully under the control of local government, and represent another main revenue source for the municipality. Many municipal governments prefer to charge use fees rather than collecting land tax. Income from use fees is used to support infrastructure extension projects. Some municipal governments (e.g. Shanghai) also charge fees for the “vegetable production foundation”. The foundation collects money to support the re-creation of farmland for the land lost in urban development. The amount of land generated revenue has increased since the urban construction boom. Table 9 shows the increase of land generated revenue from 1987 to 1990 in Shanghai, and the land income became even more important from 1990 to 1996. Both “push” and “pull” forces drive local government on urban growth policy. Growth of urban population, an enormous demand for housing, and, especially, increasing construction activity push municipal governments to acquire more land. While more farmland is being converted to urban uses, more investment is needed to develop infrastructure in new areas. When more money is needed, the central government cuts most development funds under the policy of decentralization and with the philosophy of “share decision power, share responsibility”, forcing local governments to seek their own sources. All these factors push local governments to acquire and lease more land so as to collect more revenue from land development. With decentralization, local governments now grant land use regulation power, which had previously been under the control of higher government (the central and provincial government) before the reform. Peasants in edge areas surrounding central city long to become “urban residents”, county and village leaders would like to “sell” their land and to annex their municipalities to the central city. This decentralized
Land market forces and government’s role in sprawl: T Zhang Table 9
Land revenue of Shanghai, 1987–1991 (in million RMB)
Cultivated land occupation tax Vegetable foundation Land use fees Urban land use tax Land leasing charge Total
37 103 6 – – 146
89 106 11 – 104 310
57 65 9 236 31 398
39 25 11 206 134 415
88 40 17 230 149 523
Source: Based on Zhao et al., 1998, Table 2-4-2, p. 121.
land use power thus helps urban sprawl. In addition, since all municipalities are competing with each other for investment to “speed up economic development”, no single county or village can stop sprawl. Investors can easily make a deal with a neighboring county or village, similar to what happened in the US. Government’s role in sprawl and sprawl control Clearly, under the pressure of the central/superior government’s fund cutting and increasing land demand, the main driving force of urban sprawl in China is local government’s willingness to lease more land in responding to market forces after the reform. Decentralization of land regulation power gives local governments a tool to realize their objective. High land prices stimulate local government’s willingness to “sell” more land. The combination of the marketplace and government’s willingness results in the loss of cultivated land and causes urban sprawl. As the loss of cultivated land has become a serious problem not only economically but also politically among senior government leaders, the central government has initiated a new policy to control over-expansion. The motivation of sprawl control in China is to preserve cultivated land and ensure food provision, which is different from the motivation of sprawl control in the US and European countries. In developed countries, the main concerns of sprawl control are basically environmental protection and social justice: to protect open space and reduce deterioration of central cities (Hu, 1998; Pucher, 1995a, b; Freilich and Peshoff, 1997; Burchell, 1997). China’s sprawl control policy China’s sprawl control policy is a top–down approach; it attempts to contain sprawl from both the supply and the demand side. From the supply side, the central government issued an administration order that no farmland should be allowed to convert to nonagricultural uses from April 1996 to April 1997, allowing the government a “break” to redesign land use policy. Second, starting from 1996, anyone who converts farmland to non-agricultural uses must recreate the same amount of land as farmland. Sources of land to be developed into new farmland include vacant land previously utilized for urban uses and idle land in rural areas. Third, in a land quota system, each
municipality is allowed a certain amount of farmland to convert to non-agricultural uses, but no cultivated land can be leased if the municipality has used all its quota. The central government decides land quotas and provincial government manages the quotas. In Spring, 1998, the revised Land Act was formally approved by the National People’s Congress. The Act eliminates the existing “five administration-level” system of land management, and reclaims power on farmland conversion to the central government. The Act also uses economic leverage to reduce cultivated land loss by increasing farmland acquisition reimbursements to farmers, so land is more expensive now. Starting from January 1, 1999 with the implementation of the 1998 Land Act, any new conversion of farmland is up to the central government’s approval. The re-centralization of land management power reflects the central government’s efforts on sprawl control. Last, each municipality is asked to establish a “permanent cultivated land system” in which farmland will never be allowed for non-agricultural uses, a strategy similar to the concept of “permanent open space” in some developed nations. From the demand side, the central government asked local governments to tighten control over building permits, especially permits for factory construction and certain new commercial developments. The tightening policy is a “one stone–two birds” strategy reflecting the central government’s concern over sprawl, and factory and commercial space overbuilding, which worsens competition among factories and real estate companies, given the huge amount of idle productive equipments and vacant buildings. The central government also requires review and revision of land use plans of all municipalities, the purpose being to examine land use patterns and save more land in the revised plan. The Department of City Planning of China, a central government body, will supervise the revision, but the State Land Bureau now grants the final decision on any new farmland acquisition, a power given by the new Land Act of 1988. Government’s role in sprawl and sprawl control Although sprawl control represents the public interest and therefore should be supported and implemented by the public sector, the public sector itself is not homogeneous. From a land management perspective 131
Land market forces and government’s role in sprawl: T Zhang
in the Chinese context, three sets of government departments act in the public sector—revenue collecting, public spending, and land administration. Departments in charge of revenue collection include the Central Internal Revenues and local taxation bureaus. Public spending departments have a wide range, from budget bureau at both central and local level, to all public services such as transportation, public health and education. The third interest group within the public sector is land administration and land management institutions which include planning bureaus and land bureaus. Each of the three groups has its own motivations and attitudes toward urban sprawl and sprawl control. In addition, the attitude of bureaus in the central government is different from that of local government. Because revenue from land leasing and land tax is becoming more important to government budgets, it is not surprising to see that among the three groups of government bureaus, two of them, revenue collection and public spending, both at the central and local level, actually desire more land leasing in general (see Table 10). It is logical that revenue collection departments in both the central and local government would like to draw more income from land development, so they Table 10
are passive in following top leaders’ decisions on sprawl control. Facing increasing budget deficits at both levels in recent years, the voice for generating more revenue from all sources including land tax and land leasing charges is increasingly louder. The only difference between revenue collection departments in the central and local government is over the distribution of land related revenues. The central Internal Revenues asks local government to convert all forms of land use fees (a local revenue) to land tax (a national revenue) so that the central government can share in the income. Although several new revenue regulations have been issued, local government is still reluctant to follow the central government’s regulations (cf. Davis et al., 1995). There is a difference between attitudes of the central and local departments in charge of budget and public spending. Because the central government manages the national budget that includes spending on food import programs, it recognizes the importance of farmland preservation. The State Statistical Bureau reported that from 1986 to 1995, China imported 41.2 million tons of grain (Zhao et al., 1998). So, central public expenditure departments in charge of those programs tend to support sprawl control policy. The control policy also receives supports
Attitude of government bureaus on sprawl and sprawl control
Functions of Government Bureaus
Revenue collection (Internal Revenues, local taxation bureau)
Seeks more revenue from land tax and land leasing charges, therefore supporting the increase of land leasing. Asks local government to convert land use fees to land tax to be shared with the central government. Passively follows top-leaders’ decision on sprawl control.
Prefers more land use fees rather than land tax. Strongly supports leasing more land, although cannot say so openly. Still pushes local leaders to acquire more farmland for leasing, and against the central government’s control policy.
Public spending (budget bureau, public works, transportation bureau, other public services including public health and education)
A mixed attitude on sprawl: increasing revenue for a bigger budget is welcome, but spending more on agriculture related expenditures, especially on importing food due to cultivated land losses is more serious.
Supports leasing more land since spending on agricultural-related and food supply programs are less a local responsibility, while land revenue is a main revenue source for funding public projects, especially on infrastructure extension. Capital shortage in infrastructure construction due to over-expansion is a real problem, forcing bureau of public works to reduce the provision of services in new areas. Public transportation and public schools are the most needed in new areas.
Debate on who should provide infrastructure within a metro area: now it is a local responsibility, but local governments ask for central subsidy since land tax is shared with the central government. As in the central government, these bureaus have to take all consequences of sprawl into consideration, which makes them relatively supportive on sprawl control policy. Land administration/land use management (planning bureau, land bureau)
Initiates and supports sprawl control strategies, concerned about the negative consequences of sprawl. A slight difference between the planning and the land bureau: planners intend to solve the sprawl problem from the land use plan approach and argues that some expansion is needed in urbanization; land bureau wants to put land use permits under its control and supports stricter control regulations.
Local planning bureaus do not support strict sprawl and enjoy power over land regulation after decentralization. Local land bureaus face a dilemma of either following local decision makers for leasing more land, or supporting the central policy of controlling sprawl more strictly. Competition exists on land use power between the two bureaus at the local level too.
Land market forces and government’s role in sprawl: T Zhang
from the agriculture department whose basic responsibility is to ensure sufficient food provision; cultivated land preservation is thus tied directly to its own interest. While worrying about expenditures on food imports, public expenditure departments in the central government at the same time also want to gain more from land revenue for other national programs and projects, especially major infrastructure projects such as railroad and highway construction. However, there is a debate on the provision of infrastructure in the metropolitan region. It has been a local responsibility so far, but local government has asked for central subsidies since local land tax is shared with the central government. On the other hand, because agricultural and food import programs are less a responsibility of local municipalities, both local revenue collection and public spending bureaus desire leasing more land in order for funding local programs. In addition, under the pressure of fund cutting from the central government, local public spending departments rely more on local budgets, which “forces” local decision makers to support the two groups in drawing more money from land development. In the public sector, besides the agriculture department, only land administration departments (land bureaus and planning bureaus) are truly concerned with the negative consequences of urban sprawl. As the administrative body in charge of long term development, the department of planning pursues a comprehensive sustainable growth strategy. The main responsibility of land bureaus is to prevent cultivated land loss. The two departments are playing key roles in the fight against sprawl. However, even within this group, debate about sprawl continues. The central planning department supports sprawl control, but it argues that a certain degree of urban expansion is necessary for Chinese cities. Planners have listed several reasons. First, almost all Chinese cities are overcrowded. The density in central areas is so high that basic open space is very limited, not to mention heavy air pollution in cities. Second, there are millions of “liudong renkou” (“floating population”) meaning temporary rural– urban migrants in cities after the economic reforms. Shanghai alone has more than two million “floating population”, in addition to its nine million members of the “official urban population” in 1995 (from interview, 1997). To house and provide basic services to those “temporary urban residents” requires land. Third, infrastructure systems in Chinese cities are so under-invested and out-of-date they can no longer bear loads. Funding for infrastructure improvement projects depends heavily upon land revenue. In addition, these projects themselves need land. Last and most importantly, China’s economic development needs foreign investment, and land availability is a must in attracting foreign investors. Some Chinese scholars believe that China’s urban expansion is a
“healthy urban growth” different from sprawl in the US, because urban expansion does not cause central city decline in China (Zhou and Meng, 1998). Since decentralization, local planning bureaus have enjoyed power over land use regulation. They would not like to see control over land use, such as a land quota system, imposed again by a higher government. To them, the re-centralization of land use power entails a large loss. Moreover, local municipal governments, for whom the planners work, desire more land revenue in general. These decision makers often support land growth projects even in the face of planners’ opposition to sprawl. Land bureaus of the central and local governments, together with the agriculture departments, are the backbone of any control policy. For land bureaus, a hidden motivation in designing a new land use policy (such as the Land Act of 1988) is to take urban land use control power from planning bureaus at various levels. Here a conflict exists between two institutions within the same interest group. Also, local land administrators work under local governments; just as local planners, they have little to do if the local government does not want to stop growth. Because the control policy was initiated and implemented only recently, there is no solid data to examine the performance of the policy. Most planners and officials interviewed believed that the policy has some positive impacts on cultivated land loss, especially in showing the central government’s determination. However, local implementation of the policy is problematic. Scholars point out that local government and local planning bureaus are still leasing land to firms, especially foreign investors who are looking for land. One example is Shenzhen. The city has developed from a village of 1,200 residents to a mega city of 3.5 million in 19 years. Although the whole city is new, several district governments are actively engaging in “new area” development projects. Some district officers said that the existing downtown is “too old to be attractive to investors”, so “new development projects in ‘newer downtown’ and edge areas are necessary in speeding up the city’s economic development”; the implication leads to acquiring more land (based on the author’s interview, June 1998). Another example is a city government in Hainan province. Although the 1998 Land Act asks for higher acquisition charges to reduce land demand, the city government of Haikou, capital of the Hainan province, announced in August, 1998, that it would reduce land acquisition costs to investors in the city. The purpose is again “to attract more investment and speed up economic development” (People’s Daily, August 31, 1998).
Conclusion As stated at the beginning of the study, the purpose of the research is to examine government’s contri133
Land market forces and government’s role in sprawl: T Zhang
bution to sprawl. We have found that it is the combination of market forces and government’s reaction (especially at the local level) to the marketplace that cause urban sprawl in China. We may be able to generalize some important lessons from the Chinese case which shows that besides private interest groups outside the government, the public sector itself consists of and represents various “public” interest groups. Since these groups may have different interests in farmland conversion, it is difficult to reach a consensus among government bureaus. The lack of consensus within the public sector exists at two levels: the differing attitudes toward sprawl among government departments, and between the central and local government. Without consensus, controlling sprawl cannot be adopted as a public policy. Moreover, even if the policy were to be adopted by the central government, it would still be difficult to get implemented at the local level, because land administration is basically a local authority responsibility. The presence of various interest groups in the public sector thus significantly weakens any control efforts. This looks like a common problem in sprawl control throughout the world, regardless of the context. In analyzing the public sector’s attitude to sprawl, we may even add a third dimension to the issue: the possible interest conflict between government institutions as a whole, and individual officials in the institution. In the case of China, this means corruption. Officials in charge of land acquisition and land leasing may gain “profit” from urban sprawl through “deal making”, given the hot land market in recent years. Thus, regardless of whatever policy is adopted, sprawl control is still a distant goal. Many Chinese officials and planners told stories about corruption in planning and land bureaus during interviews. Wu’s finding of a “pro development coalition” by municipal officials, large state work-units and overseas capital may provide answers to the situation (Wu, 1998). The various players in the private sector, the interest groups in the public sector, and the complex relations between the private and the public sector at both organizational and individual levels, all make public policy hard to succeed, even a good policy such as sprawl control.
References Burchell, R (1997) Economic and fiscal costs (and benefits) of sprawl. The Urban Lawyer 29(2), 151–181. Burchell, R, Listokin, D, Shad, N, Phillips, H, Downs, A, Seskin, S, Davis, J, Moore, T, Helton D and Gall, M (1998) Costs of Sprawl Revised: the Evidence of Sprawl’s Negative and Positive Impacts. Transit Cooperative Research Program (TCRP) H-10. China Academy of Social Science and Institute of Public Administration USA (1992) Urban Land Use and Management in China (Final Report). Economics Science Press, Beijing, China. China Department of City Planning (1997a) Report on land use in Chinese cities (in Chinese). City Planning Review 21(2), 36–39. China Department of City Planning and China Academy of City Planning and Research (1997b) An Analysis on City and City
Development in China and in the World (Internal Report, in Chinese). Beijing, China. Davis, D, Kraus, R, Naughton, B and Perry, E (1995) Urban Spaces in Contemporary China: The Potential for Autonomy and Community in Post-Mao China. Woodrow Wilson Center Press and Cambridge University Press, Washington, DC. Dowall, D (1993) Establishing urban land market in the People’s Republic of China. Journal of the American Planning Association 59, 182–192. Downs, A (1994) New Visions for Metropolitan America. The Brookings Institution and Lincoln Institute of Land Policy, Washington, DC. Ewing, R (1997) Counterpoint: Is Los Angeles-style sprawl desirable? Journal of the American Planning Association 63(1), 107–126. Freilich, R and Peshoff, B (1997) The social costs of sprawl. The Urban Lawyer 29(2), 183–198. Fung, K I (1981) Urban sprawl in China: some causative factors. In Urban Development in Modern China, eds L J C Ma and E W Hanten. Westview Press, Boulder. Hu, W (1998) Letter to the Editor: Issues for sustaining Chinese agriculture. Land Use Policy 15(2), 167–170. Katz, B and Bernstein, S (1998) The new metropolitan agenda: Connecting cities and suburbs. Brookings Review, Fall, 4–7. Kirkby, R (1985) Urbanization in China: town and country in a reaction in the Pearl River Delta Region. Croom Helm, London; Columbia University Press, New York. Kirkby, R (1996) The Environmental constraints on China’s rural urbanization strategy: an essay of the past, present and future. Asian Geographer 15(1), 123–141. Li, L H (1997) The political economy of the privatization of the land market in Shanghai. Urban Studies 34(2), 321–335. Li, X and Nie, X (1996) Will Famine Hit China Again? (in Chinese). Reform Press, Beijing. Ma, W (1997) Urbanization and agricultural modernization: The risk and prospects of urbanization in China (in Chinese). City Planning Review 21(6), 26–29. McDonald, J (1993) Local property tax differences and business real estate values. Journal of Real Estate Finance and Economics 6, 277–287. McDonald, J, Orlebeke, C, Sen, A and Wiewel, W (1992) Real Estate Development and Property Taxes in DuPage County: Final Report. University of Illinois at Chicago, Center for Urban Economic Development. Moe, R (1995) In Alternative to Sprawl, D Young, ed. Lincoln Institute of Land Policy, Cambridge, MA. Openlands Projects (1998) Losing Ground: Land Consumption in the Chicago Region, 1990–1998. Chicago, Openlands Report. Parker, A (1995) Patterns of federal urban spending: Central cities and their suburbs, 1983–1992. Urban Affairs Review 31(2), 184–205. Pucher, J (1995a) Passenger transport in the United States and Europe: A comparative analysis of public policies, Part 1. Transport Reviews 15(2), 99–117. Pucher, J (1995b) Passenger transport in the United States and Europe: A comparative analysis of public policies, Part 2. Transport Reviews 15(3), 211–227. Real Estate Research Corporation (1974) The Costs of Sprawl: Environmental and Economic Costs of Alternative Residential Development Patterns at the Urban Fringe. Government Printing Office, Washington, DC. Richmond, H (1995) Regionalism: Chicago as an American Region. The MacArthur Foundation, Chicago. Rusk, D (1993) Cities Without Suburbs. Woodrow Center Press, Washington, DC. State Statistical Bureau of People’s Republic of China, 1998. China Statistical Yearbooks, 1988, 1990, 1995, 1996, 1997, 1998 (in Chinese). China Statistical Publishing House, Beijing, China. Shanghai Municipal Statistics Bureau, 1988. Shanghai Statistical Yearbooks, 1988, 1990, 1995, 1996 (in Chinese). China Statistical Publishing House, Beijing, China. Song, Q L (1992) Modern Urban Land Use Planning, (in Chinese,
Land market forces and government’s role in sprawl: T Zhang Xiandai chengshi tudi liyoung guihau). China Architectural Engineering Press, Beijing. Urban Land Institute (1998) ULI on the Future: Smart Growth. The Urban Land Institute, Washington, DC. World Bank (1997) China 2020: Development Challenges in the New Century. The World Bank Press, Washington, DC. World Bank (1998) World Development Report 1998. The World Bank Press, Washington, DC. Wu, F (December 1997) Urban restructuring in China’s emerging market economy: towards a framework for analysis. International Journal of Urban and Regional Research 21(4), 640–663. Wu, F (1998) The structure of building provision and the trans-
formation of the urban landscape in metropolitan Guangzhou, China. Urban Studies 35(2), 259–283. Yeh, A G O and Wu, F (1996) The new land development process and urban development in Chinese cities. International Journal of Urban and Regional Research 20(2), 330–354. Zhao, M, Guilang, B and Li, H (1998) Reform of Land Use Regulations and Urban and Rural Development (in Chinese: Tudi shiyong zhidu gaige yu chengxiang fazhang) Shanghai, Tongji University Press. Zhou, Y and Meng, C (1998) The tendency of suburbanization of big cities in China (in Chinese). Urban Planning Forum 3, 22–27.