*The authors thank Dr. Feng Jian, Peking University, and Mr. David Chunyu, University at Albany, for their assistance in the preparation of this essay.
Suburbanization of Urban China: A Conceptual Framework
Suburbanization is treated as a universal phenomenon in Western social science. In Chinese cities a similar pattern appears to be developing. Major urban centers are beginning to hollow out at their core and shift both people and economic activity toward outer zones (Zhou and Ma, 2000). Large scale urban renewal and reconstruction in the central cities since the 1980s have relocated many city residents to suburbs (Wang and Zhou, 1999), and massive waves of migrants from other regions have been settled in peripheral areas. Suburbanization responds broadly to the rapid economic development of China since 1980, a process carried out through explicit public policies of re-engagement with the global economy and removal of barriers to urban growth. We might think of suburbanization as a natural and inevitable accompaniment to economic development. A more realistic view is that these two phenomena were a package deal, that economic and spatial restructuring of the metropolis was an intentional goal of the Chinese government as it sought to secure a new and stronger international position. Thus we have a fundamentally political understanding of suburbanization, in which at every level the economic processes that facilitate and shape suburban growth are grounded in policy choices.
Suburbanization responds to many different conditions, ranging from global shifts such as China’s reincorporation into the world economy and introduction of market mechanisms in its economy, to the specific details of land development and housing choices. We find it impossible simply to apply theoretical models that have been relied on in the West, such as distance-density gradients, the natural operation of the property market, or the impact of economic expansion. These forces play their part, but in a very specific historical and political context. We distinguish multiple levels of analysis, moving from the most general to the most specific. At each level we identify the main forces for change, describe their consequences, and show how these create the conditions within which processes at the next level need to be understood. From a theoretical perspective our contribution is to demonstrate the importance of such contextualization, and just as we draw on Western models to understand China, we believe that the peculiarities of the Chinese experience raise new questions about suburbanization in market societies.
In the following sections we begin by describing the timing and pace of suburban growth, emphasizing the case of Beijing that has been intensively studied by Zhou and his colleagues (Zhou and Ma 2000, Zhou and Meng 1998, Feng and Zhou 2003a). We then analyze the processes involved in this transformation.
The timing and pace of suburbanization
Before the 1980s, Chinese planners sought to decentralize the largest cities but with little effect. Certainly there was a potential for suburbanization, especially to meet the needs for housing. The population density in central cities reached very high levels. Per capita living space was only 3 to 4 square meters, so that typically three persons would share a bedroom that was only 10 feet by 10 feet. Population growth, even after fertility was limited by the one-child family policy and despite stringent restrictions on rural-urban migration, was greater than the rate of housing construction. Environmental quality was poor, especially where industrial plants were located in the inner city. In the 1960s and 1970s, some new industrial satellite towns were built around cities like Beijing and placed downwind of the city. The northwestern sector of the city (upwind) had already been designated for development of scientific research and education since the 1950s. But although many plants were constructed, their employees mostly remained in the city and commuted to the suburbs to work. Hence there was a potential demand for expansion on the periphery of the city, but it was not satisfied. Why not?
There is broad agreement that the poor performance of the national economy is one reason. Relatively underdeveloped anyway, China emerged from the 1940s with a decades-long legacy of foreign occupation and civil war. Central planning directed investments especially toward the heavy industries believed to be essential for national defense. Officials chose to repeat the Soviet Union’s strategy of redistributing resources to meet very basic needs for food and shelter but postponing “unproductive” investments in real estate and services. Failed experiments during the “Great Leap Forward” from 1958 to 1960, the “Black Years” from 1961 to 1963, and the Cultural Revolution further undermined development.
Beyond the overall lack of resources, the organization of the planned economy favored a continuing buildup of central cities. One factor is the refusal to treat land as a scarce resource. Urban land was allocated by the government at no cost to end users and without regard to the relative value of different locations. As a result, for example, large industrial zones continued to be built within the central cities. In Beijing, industrial plants were disproportionately located in the four central districts through the 1970s. Another is the centralization of fiscal resources in the hands of national ministries. Municipal officials did not have the means or authority to support large-scale construction of new residential districts, or of high-grade roadways to connect cities and satellite towns, or to create infrastructure that would bring standard public services to outlying areas. Consequently neither work unit leaders nor their employees perceived any benefit to relocating to suburbs and pressed instead to remain in central locations. A third factor is the treatment of housing as a welfare service. In urban areas most housing – aside from a leftover stock of private residences of generally low quality – was provided by work units or city housing authorities at rents that did not even cover basic maintenance costs. Consequently no investment in new housing could be self-supporting, and public housing in suburbs (because it required additional infrastructure investments and offered no savings in the cost of land) was especially costly.
Nevertheless suburbanization took hold in China’s major cities in the 1980s. The earliest studies identified this trend in Beijing ( Zhou, 1992; Zhou, 1995). From 1964 to 1982 the average annual population growth rate of Beijing was 1.1%, but only 0.2% for four central districts (Dongcheng, Xicheng, Xuanwu, and Chongwen). From 1982 to 1990, the core lost 82,000 population (-3.4%), the inner suburbs gained 1,149,000 (up 40.5%), and the outer suburbs gained 521,000 (up 13.1%). The trend intensified in the period from 1990 to 2000, as the core lost 222,000 (-9.5%), the inner suburbs grew by 2,400,000 (60.2%), and outer suburbs grew by 572,000 (up 12.7%). The inner suburbs became the main destination for movers from the central city, as well as the area where temporary residents from other regions were mostly concentrated. Both centripetal migrants and centrifugal migrants met in this zone. These shifts are mapped in Figures 1a and 1b.
Figures 1a and 1b about here
Similar results have been found other big cities, such as Guangzhou ( Zhou and Xu, 1996; Xie and Ning 2002), Shanghai ( Ning and Deng, 1996; Gao and Jiang 2002 ), Shenyang ( Zhou and Meng, 1997), Dalian ( Chai and Zhou, 2000), Hangzhou ( Zhou, M.1997; Feng and Zhou 2002), Suzhou, Wuxi, and Changzhou ( Zhang, Y. 1998). Our task is to understand why this happened.
The sources of suburbanization
Chinese suburbanization cannot be understood without reference to China’s market transition. Figure 2 reproduces a conceptual model that identifies several aspects of the process (drawn from Feng et al 2004). Feng depicts all of these as consequences of development of a market economy, but distinguishes processes “at the level of the economy” from processes “at the level of society.” Suburbanization itself is shown to include shifts in economic activity (industrial decentralization and development of large supermarkets, which represents the wider shift in retail services) as well as changes in residential locations (including both permanent and seasonal housing). The more “economic” sources of change involve urban land use transformation and the renovation of the inner city, on the one hand, and real estate investments (both domestic and foreign) and improvement of public transportation, on the other. Feng notes two types of social changes. One is in the class structure – the growing class inequalities that are manifested in spatial inequalities and represented in both practical and symbolic terms by growth in the use of private automobiles. The other is in residents’ attitudes toward the use of space, which provide support for a mix of high-class and moderate-income housing in suburbs.
Figure 2 about here
This illustration calls attention to the many components of suburbanization. Suburbanization can be indexed with population figures like the ones for Beijing presented above, but it is much more than a matter of numbers and locations. The issues are how and where people’s daily economic and social activities are carried out and how are these affected by China’s development of a market economy? In the spatial restructuring of the metropolis, what exactly does it mean to invoke market transition?
We analyze this question at three levels: the national economy and system of urban planning, the system of housing and real estate development, and in the composition of neighborhoods.
1. Processes at the national level
First, China has experienced two decades of rapid economic growth coinciding with the government’s decision to become more closely integrated with the global economy. Goods produced in China account for quickly growing shares of world trade, and by 2004 China had become the fourth most important country in the world in terms of volume of trade. China has also welcomed foreign investment, and investors have been especially active in real estate projects in major coastal cities. The average annual rate of increase in Gross Domestic Product (GDP) was 10.8% in 1981-85, 7.9% in 1986-90, 11.6% in 1991-95, and 8.3% in 1996-2000. In 2003 per capita GDP exceeded $1000 for the first time. It has exceeded $5,000 in some major cities in the Zhujiang Delta, and reached $4,000 in the Changjiang Delta and $3000 in Beijing. The residents of these cities have become consumers at an unprecedented scale for houses, cars, and education. In this context, there are obviously new possibilities for urban development, and the trend toward suburbanization has been one result.
Besides sheer growth, the metropolitan economy has undergone a sectoral transformation. In Beijing, for example, during the period 1982 to 2000 the proportion of employment in primary industry decreased from 29% to 13%, and that in secondary sectors decreased from 39% to 29%. As the location of the national government, service sectors were traditionally stronger in Beijing than in most cities, but service employment is now dominant, having grown from 32% to 58%. Among service activities, banking, insurance, real estate and social services increased especially rapidly from 2% to 14%, and transportation, storage, postal and telecommunication services, and wholesale, retail trade and catering services more than doubled from 11% to 23%. This change does not have obvious implications for spatial shifts. Older heavy industries were good prospects for relocation to suburbs on environmental grounds, but if industry was not the main component economic expansion this might not have contributed much to suburban growth. The new service activities – especially business services – could have been concentrated in central cities, restoring to cities the “central place” functions that they would naturally have had earlier in the 20th Century. But they could as well have been shifted to outer zones.
In the era of the socialist planned economy the actual direction of urban development was governed by urban planning. The guiding ideology in the planned economy was strong controls by government and limitation of population growth, although specific sorts of development were expected and carried out. In the late 1950s, it was decided in the city’s master plan that Beijing should be developed into a modernized base of manufacturing. The city’s built-up area increased from 109 sq. km. in 1949 to 221 sq. km. in 1959. Emphasis was placed on construction in the industrial quarter in the northwestern and eastern suburbs. Later, the industrial base of the Capital Iron and Steel (Shougang), the culture-education quarter in the northwest and the embassy quarter in the central core were constructed one after another. Some satellite towns in the outer suburbs were also listed in the development plan at the time, but this aspect of the plan was not much implemented.
Although government has maintained weaker planning controls in the period of market transition, formal plans have continued to affect development. In the master planning completed in 1982, officials decided to reinterpret Beijing’s economic function, from an industrial center to a center of politics and culture (Beijing’s Planning Committee, 1982). This anticipated the restructuring of the labor market. In the 1993 master plan, more specific steps were taken to promote spatial restructuring: it was decided to promote dispersed clusters of growth (Fensan Jituanshi Buju) throughout the region, to invest in transportation systems to link the city with towns in the outer suburbs, to coordinate the development of new districts and the renewal of the old city, to protect and improve the urban physical environment, and to shift from heavy industry to new high-tech industry (Beijing’s Planning Committee, 1993). Without doubt these policies have facilitated suburbanization.
Finally perhaps the most important policy shift at the national level has been the accommodation of government to the reality of massive rural-urban migration. At one time the full array of state powers had been directed against migration, with strict enforcement of a household registration system and procedures for food rationing that made it difficult for unauthorized migrants to live or work in cities. The registration system has been in flux in China since the 1980s. For example, the external population (wailai renkou) in Beijing increased from 170,000 in 1982 to 600,000 in 1990, a growth rate of 255%. It then boomed in the 1990s, rising to 2,570,000, for a growth rate of 326%. As a result, in 2000, concentrated areas of external population have become a specific form of suburban settlement.
2. Reform of housing and urban land utilization system
The emergence of real estate markets, including a partial privatization of housing, has significantly altered the urban development process (Wu, 1995).
For the half century from 1949 to 1998, housing provision and allocation had been considered to be a kind of welfare, with the majority of housing owned by work units. Housing was constructed or purchased by work units near their premises and rented at a very low rate to their workers, thus, residential location was governed by the location of housing provided by the work units, often within or near them (Yeh et el., 1995). Some powerful work units, such as government departments and large enterprises, had the ability to construct large concentrative residential quarters for their cadres and workers, creating whole neighborhoods of relative privilege.
In the early years of market reform up to 1998, housing reform had led to some significant changes in the structure of housing provision. Only after 1998 did the role of working units in allocating housing change greatly. At that time the traditional system of housing allocation was cancelled, and a new housing allocation system, with subsidy by the nation and the working units, was established.
Residential mobility before the mid-1980s was low and insignificant in shaping urban social spatial structure, because of a housing shortage and a housing allocation system based on the construction and allocation of housing by working units.
With the abolishment of housing allocation by working units beginning in 1998, the working units are taking measures to encourage employees to buy their private housing through an accumulation fund of housing, loans for private housing, housing subsidies by the nation and by the working units, etc. As a result, the freedom of purchasing housing and the flexibility of selecting housing location are strengthened. It is not necessary for the employees in the same working units to live together, and for the new housing purchasers to live near their working units. Thus, the spatial separation between residence and employment appears, with more and more commuters in Chinese cities. The new housing system had played an important role in restructuring the social space in urban China,
As a result of innovation in the housing allocation system, the structure of construction of each kind of housing changed since 1998. The numbers of housing units built by the central government agencies or local units were decreasing, while the number of housing built by real estate enterprises was increasing. In 2000, the proportion of the housing built by the central units was only 18.6%, while in 1997 it had been 25.4%. The proportion of the housing built by local units in the total of city was 13.8% in 2000, compared to 26.6% in 1997. The proportion of housing built by real estate enterprises increased from 48.0% in 1997 to 67.7% in 2000.
The development of private real estate began in urban China in the 1990s. The government has adopted a series of policies, for example to raise rents, to increase the proportion of housing accumulation fund, to permit the purchased houses or apartments to come into the market, to give employees cash allowance to purchase house or rehabilitation move, to provide low-interest loan for house purchase by banks, to establish house quality system, to push living community administration, and so on. Most important, housing is becoming commodified, and people are increasingly choosing suburban location as consumers because land prices are lower there.
The first wave of suburbanization in the 1980s was driven by relocation of persons displaced by urban redevelopment projects, in housing built and financed by government, and by relocation of industry. Since then, as Zhou and Ma (2000) point out, the most powerful force for change has been the creation of a system requiring land users to pay for development rights based on land value and other market factors. In 1985????, payment for use rights of urban land was established, and this caused a fundamental shift of land use in the central city core – from industrial and administrative to commercial and other tertiary uses. Survey data shows that in Beijing’s central section the profit per square meter of space for industry is only one tenth – in some cases even one fiftieth – as high as that for business. This disparity has begun to force industry out of the central city, starting with enterprises that pollute the environment, or needed space to develop, or needed money to improve technology.
With the commercialization of housing, residents have begun to select their residential location independently of the housing compounds made available by their employers. As housing reform has progressed, a higher share of suburban housing has been developed through market mechanisms.
This is seen first in the growing proportion of individual purchases of housing. Individual purchases account for above 70% and in some case up to 90% of transactions in coastal cities. A survey of residents in new dwelling units from 1990 to 1995 in Beijing found that at that time only 14% purchased their own unit, and for those moving from the central city to inner suburbs the rate was only 8%. At that time Guangzhou had a more advanced private housing market, but still only 38% of movers had purchased homes as individuals (Zhou et al 2000).
The real estate developers extensively market suburban locations, advertising the higher living quality and more comfortable space in suburbs. They design and sell housing for all segments, ranging from apartments for people with modest incomes, to townhouses and single-family villas. These tend to be located in different parts of the metropolis, adding to spatial differentiation. In Beijing, for example, 70% of new housing affordable to low income people is found in inner suburbs, 20% in outer suburbs, and only 10% in the central city. In contrast 60% of villas for the wealthy are located in outer suburbs, more than 30% in inner suburbs, and almost none in the city itself (Feng et al 2004).
Processes at the neighborhood level
Our concern in this section is with the kinds of neighborhoods that are being produced in the reform period, and more specifically with the characteristics of suburban neighborhoods. Analyses of local area census data from 1982 and 2000 suggest that there have been major changes of urban social spatial structure. These studies are based on the methods of factorial ecology, which are used to identify the main sources of variation in the social composition of neighborhoods and to show where different kinds of neighborhoods are located. In most market societies social areas mainly reflect differences in socioeconomic status, family life cycle (young families live in different areas than single adults or the elderly), and race or ethnicity. Figures 3a and 3b summarize the results of such analyses for Beijing (a more complete presentation is provided in Feng and Zhou 2003b).
Figures 3a and 3b about here
The 1982 social areas map is relatively simple. The majority of territory in the Beijing metropolis was still primarily agricultural, and there was a very strong distinction between low density agricultural villages and other zones. There were also mining areas located mainly in the western suburbs. The typical neighborhoods in the urban core were distinguished from these agricultural and mining zones by their high population density and urban working class occupations. These “high density, worker areas” dominated the four oldest districts in the heart of the city as well as newer industrial zones in the northwest and to the east.
An interesting feature of the 1982 map is that it also identifies two sorts of neighborhoods that are set apart by their more elite social composition. Intellectual areas (standing out for high average education levels) are found in the northwest adjacent to the urban core, where planners had located Beijing’s most important universities and research centers. Cadre areas, the site of housing constructed for government employees, are found along the inner city’s eastern edge. Hence in the era of the socialist planned economy, the types of social areas mainly reflected the location of different types of jobs. We also see that early “suburbanization” had made little impact on the composition of the periphery, which appears as a vast and undifferentiated zone.
The 2000 map shows much greater spatial diversity, and it also reflects the emergence of some new kinds of neighborhoods. Let us review the changes one at a time, starting with the center and moving out to the periphery:
1. First, the districts in the very center of the region continue to be identified by their very high density, and there are now some similar areas not only in the northwest inner suburbs, but also farther to the west.
2. Cadre areas have disappeared, suggesting that there has been a dispersal of government employees to a wider range of neighborhoods. Intellectual areas are still found in a band around the west and north of the inner core. But they have added an ethnic component to their makeup. Noticeable shares of residents are members of Chinese minority groups, many of whom have come to Beijing to study or drawn to jobs requiring a higher level of education. There are also some migrant villages mixed into this part of the city, such as the Hui nationality settlement in Xinjiang Village.
3. Turning to the suburbs, the territory identified by the predominance of agricultural employment – still important at the edges of the metropolis – has diminished, and no areas are typified by mining employment. It is not that mining has disappeared, but rather than population growth with other characteristics has become more important in those areas.
4. A considerable portion of former agricultural zones are now characterized by their low population density and larger housing units, but now predominantly settled by people with urban occupations who previously lived in the central city. This stratum includes many inner suburbs on the south, east and north, as well as an extended suburban belt on the west of the city.
5. Another new and critically important kind of suburban neighborhood houses migrants from outside the region. These are Beijing’s floating population, people who do not have the same rights to rent public housing or purchase on favorable terms from their employers. They often rent in the private housing market provided by villagers on the edge of the city and constitute enclaves of persons who congregate based on their province of origin and the sector of the economy in which they work.
6. Finally some suburban neighborhoods are described as “areas of urban residents in the outer suburb.” These are defined in part by the low occupational and educational level of their residents, who nonetheless do not tend to have agricultural jobs. Several of these neighborhoods are county seats in outlying subdistricts.
Research on spatial inequalities in China is handicapped by the lack of data on household incomes for small areas. It is widely expected that the increasing inequality in earnings that characterizes the class structure will be mapped into space, but the tools to measure this phenomenon are not yet available. Still, we see much evidence that as suburbs have grown, they have also become more clearly differentiated.
Chinese suburbanization in a comparative context
We will use North American suburbanization as a specific point of reference, recognizing that this is only one of the versions in a market society. One of the defining characteristics of suburbs in the United States is their political autonomy – suburbia in many respects starts at the city limits, and some of the most important issues for American suburbs stem from this fact. To be “suburban” means to have an independent school system, to be reliant on your own tax base, and to control local development. This is generally not the case in major Chinese metropolitan regions. The largest cities cover extensive geographic areas that include much rural, undeveloped territory. The central city municipal government is likely to be responsible for services in many of the new “suburban” areas. Surrounding rural areas may be under the jurisdiction of other units of local government, and some rural villages that have been enveloped by leap-frog development may retain control of their local services, development controls, and collectively held land. But the general trend is for the central city government to extend its reach, and suburban development may be accompanied by political reorganization similar to annexation in the U.S. context.
Although the Chinese metropolis is not politically fragmented, it is highly differentiated spatially. There are large differences in land use and composition of the population between city and suburban zones and even within the suburban zone. We sketch here the typical American patterns and compare them to the Chinese case.
Suburbanization of Employment
Since the late l950s, the bulk of new manufacturing and trade employment in the U.S. metropolis has been located in small and middle-sized cities in the suburban ring (Berry and Kasarda 1977, Chapter 13). Downtown department stores compete with new suburban shopping malls. The highly developed expressway network around central cities frees manufacturing plants to take advantage of the lower land prices and taxes and the superior access to the skilled workforce offered by the suburbs. For the period of 1963 to 1977, in the largest twenty-five metropolitan areas, total manufacturing employment in central cities declined by about 700,000 (19 percent), while their suburbs gained 1.1 million (36 percent). At the same time, total central city retail and wholesale employment was stagnant (dropping by 100,000). Trade employment in the suburbs increased by 1.8 million (or 110 percent) in this period. Thus, total employment growth in the suburbs outpaced the growth of population (Logan and Golden 1986). This is the heart of the phenomenon popularized by Garreau (1991) as the creation of “Edge City.”
Suburban development also has an important component of employment growth in China, but here housing and job development are taking place simultaneously. China never experienced the creation of dormitory suburbs whose residents mostly commuted to central city jobs.
Socioeconomic Differences between Central Cities and Suburbs
A common pattern in the U.S., though not in many other parts of the world, is for central cities in most metropolitan regions to have a less affluent residential population than their surrounding suburbs. There is much debate, however, whether this class segregation between central cities and their suburbs is a natural sorting out of social classes through the private market or whether its causes are political and institutional (Danielson 1976). Suburban housing stock is newer, suburban land is less expensive, and suburbs are more accessible to automobile and truck traffic. In the American context these differences draw more affluent residents to suburbs where they can buy larger houses as well as enjoy superior public services. The process works differently in China. Here there is no tradition of suburban single family homes, and no local political voice to object to high-density development. Low land prices, then, can facilitate apartment developments for working class people. In addition the municipal government itself is a major land developer, under increasing pressure to re-house people who have been displaced by redevelopment projects in the central city. Finally, because Chinese suburbs are not politically independent and have no fiscal advantages over city neighborhoods, their public services are often much less developed than in the central city.
The suburbanization process in the U.S. also increasingly involves minorities and immigrants, and the incorporation of these groups into suburban areas has become an important issue. As Massey and Denton (1987) document, the rate of growth of nonwhites and Hispanics in metropolitan areas is far outstripping the rate of growth of non-Hispanic whites. Much of this growth is occurring in suburbs. During the 1970s, for example, the number of blacks in the non-central-city parts of metropolitan areas increased by 70 percent, compared to just 16 percent in central cities; and the number of other nonwhites in them shot up by 150 percent, compared to approximately 70 percent in central cities. One reason for the rapidly increasing racial and ethnic diversity of suburbs may be that some new immigrant groups are bypassing central cities and settling directly in suburbs. Equally important is the increasing suburbanization of older racial and ethnic minorities, such as blacks (Frey and Speare 1988).
This phenomenon has encouraged researchers to study suburbanization as a mirror on the social mobility of minorities. Consistent with classical ecological theory, suburbanization has often been portrayed broadly as a step toward assimilation into the mainstream society and as a sign of the erosion of social boundaries. For European immigrant groups after the turn of the century, residential decentralization appears to have been part of the general process of assimilation (Guest 1980). In contrast the suburbanization process for blacks appears largely to be one of continued ghettoization (Farley 1970), as indicated by high and in some regions increasing levels of segregation and by the concentration of suburban blacks in communities with a high incidence of social problems (e.g., high crime rates), high taxes, and underfunded social services (Logan and Schneider 1984, Reardon 1997, Alba, Logan and Bellair 1994). These findings regarding black suburbanization have been interpreted in terms of processes that impede the free mobility of racial minorities: steering by realtors, unequal access to mortgage credit, exclusionary zoning, and neighbor hostility (Foley 1973).
China has no comparable “minority” population, but as we have seen it does have a rapidly growing population of people who have migrated across provinces or from rural to metropolitan areas, and they are disproportionately housed in the suburban periphery. Their placement certainly represents upward mobility for them in comparison with their place of origin, but at the same time it marginalizes them within the metropolis. They live in suburbs not because it offers better housing or neighborhoods than the central city, but because they are not allowed to live in most central city zones.
Looking to the future
Many of the forces that drove suburbanization in the 1980s, especially rapid urban population growth and the pent up demand for more and better housing, are still operative. China is in the midst of the transition from a planned economy to a socialistic market-oriented economy. It is also undergoing this change at a relatively early phase of economic development. If we compare China’s economy at the starting point of suburbanization ( a 1985 per capita GNP of $289 and an urbanization level of 24%) to that of the U.S. just prior to its post-World War II suburban boom (per capita GNP of $934 and urbanization at 57% in 1941), the gap is large indeed. One consequence is that the Chinese middle class is much weaker than that of the U.S. middle class, and suburbanization in China could not be based primarily on lifestyle choices by more affluent people. Another is that the role of the private car has been very different. The automobile was a necessary structural support for large-scale suburbanization in the U.S. (Muller, 1981), but was nearly invisible in China (Zhou and Ma 2000). Even in Beijing in 1985 only 1107 cars were owned by households in Beijing; most people traveled on bicycles and buses.
Seen in this light we could try to explain Chinese suburbanization in terms of lagged development – beginning slowly due to obstacles placed by state central economic planning and disproportionately controlled by decisions of work unit managers and urban redevelopers in its early phase, but destined to catch up in the future. Indeed there are some signs that this is happening. In many respects – rapid economic development and improvement of standard of living, commodification of space and housing, and residents’ perceptions of their housing choices – the Chinese situation is becoming more similar to Western countries.
One key to the future is a change in people’s preferences about where to live. In the past a common saying was “preferring a bed in the central city to a house in suburbs”(Wang and Zhou, 1999), reflecting a sense that the city offered a better location even at the cost of less space. Early suburban development was largely decided by the state and by work units rather than by the residents. A survey in Shenyang found that forced moves accounted for 64% of all city-to-suburb mobility (Zhou and Meng, 1997). In Dalian the share was 67% (Chai and Zhou, 2000). In every city many people move out of the central city while still keeping their household registration there in order to retain urban residence rights (Zhou and Meng, 1997). Yet this sentiment is changing, and people are giving more weight to housing prices and the living environment. The pursuit of a comfortable environment for a large share of families is becoming more important than simple demand for living space.
In addition the automobile has arrived in China. By 2000, the total number of private automobiles in Beijing had amounted to over a million, with 12% of all households owning cars (Feng et al. 2004). And there are other ways that prosperity is promoting suburban growth. There are signs of development of a market for second houses in suburbia, and suburban shopping malls and supermarkets are transforming the spatial distribution of urban retail trade – increasing at an average annual rate of 68% between 1998 and 2001 (Wang and Zhou 2002).
We point to these phenomena because they have begun to play a stronger role in the reshaping of the metropolis. If one wants to find signs of suburbanization in China that are reminiscent of North American models, they can be found. But it would be risky and misleading to make too much of these parallels. If we begin to see that moving to suburbs reflects people’s hopes for better housing, more space, and a cleaner environment, this is not yet the reality encountered by many of China’s new suburbanites. More generally, though we have emphasized ways in which China is an economy in transition, it is far from a society dominated by private capital and a free market. The future depends as much on the continuing role of central government policy and direction as on the market trends made possible and stimulated by public policy.
References: Alba, Richard, John R. Logan and Paul Bellair. 1994. "Living with Crime: The Implications of Racial and Ethnic Differences in Suburban Location" Social Forces 73: 395-434.
Beijing’s Planning Committee. 1982. City Comprehensive Planning of Beijing (1981-2000) ( in Chinese).
Beijing’s Planning Committee. 1993. City Comprehensive Planning of Beijing (1992-2010) (in Chinese).
Berry, Brian, and John Kasarda. 1977. ContemporaryUrban Ecology. New York: Macmillan.
Chai, Y.W. and Zhou, Y. X. 2000. “The Characteristics, mechanisms and tendency of suburbanization in Dalian City.” Scientia Geographica Sinica. 20 (2), 127-132 (in Chinese).
Danielson, Michael. 1976. The Politics of Exclusion. New York: Columbia University Press.
Farley, Reynolds. 1970. “The Changing Distribution of Negroes Within Metropolitan Areas: The Emergence of Black Suburbs." American Journal of Sociology 75:512-529.
Feng, J. and Zhou Y. X. 2002. “ The spatial redistribution and suburbanization of urban population in Hangzhou City.” City Planning Review. 26(1), 58-65 (in Chinese)
Feng, J. and Zhou, Y. X. 2003a. “ The latest development in demographic spatial distribution in Beijing in the 1990s.” City Planning Review. 27(5), 55-63 (in Chinese)
Feng, J. and Zhou, Y. X. 2003b. “ The social spatial structure of Beijing metropolitan area and its evolution: 1982-2000.” Geographical Research. 22(4), 465-483 (in Chinese).
Feng, J. et al. 2004. “The development of suburbanization of Beijing and its countermeasures in the 1990.” City Planning Review. 28(3), 13-29 (in Chinese).
Foley, Donald. 1973. "Institutional and Contextual Factors Affecting the Housing Choices of Minority Residents." In Amos Hawley and Vincent Rock, eds., Segregation in Residential Areas. Washington, D.C.: National Academy of Sciences.
Frey, William, and Alden Speare. 1988. Regional andMetropolitan Growth and Decline in the United States. New York: Russell Sage Foundation.
Gao, X. D. and Jiang, Q. Z. 2002. “The redistribution of the population and suburbanization in Shanghai municipality. “ City Planning Review. 26(1), 66-69,89 (in Chinese).
Garreau, Joel. 1991. Edge City: Life on the New Frontier. New York: Doubleday.
Guest, Avery M. 1980. "The Suburbanization of Ethnic Groups." Sociology and Social Research 64:497-513.
Logan, John R., and Reid Golden. 1986. "Suburbs and Satellites: Two Decades of Change." American Sociological Review 51:430-437.
Logan, John R. and Mark Schneider. 1984. "Racial Segregation and Racial Change in American Suburbs: 1970-1980." American Journalof Sociology 89:874-888.
Massey, Douglas, and Nancy Denton. 1987. "Trends in the Residential Segregation of Blacks, Hispanics, and Asians: 1970-1980." American Sociological Review 52:802-825.
Muller, P. O. (1981). Contemporary Suburban America. Englewood Cliffs, NJ: Prentice Hall.
Ning, Y. M. and Deng, Y. C. 1996. “A study of Shanghai’s suburbanization”. In S. M. Li et al, editors, Aspects of China’s Regional and Economic Development. Taipei and Hong Kong: Population Research Center of National Taiwan University and Hong Kong Baptist University. 129-153 (In Chinese)
Reardon, Kenneth M. 1997. “State and Local Revitalization Efforts in East St. Louis, Illinois.” Annals of the American Academy of Political and Social Science 551:235-247.
Wang, D. and Zhou, Y. 2002. “Study on Shanghai consumers’ choices for great supermarket.” Urban Planning Forum. No. 4 , 46－50 (in Chinese).
Wang, F. H. and Zhou, Y. X. 1999. “Modeling urban population densities in Beijing 1982-1990: suburbanization and causes”. Urban Studies. 36 (2), 263-279
Wu, F.1995. “The changing urban process in the face of China's transition to a socialist economy”. Environment and Planning C: Government and Policy. 13. 159-177
Xie, S. H. and Ning, Y.M. 2002. “Urbanization and suburbanization: the dual engines to spatial change of Chinese metropolis under transitional era: a case study of Guangzhou.” City Planning Review 27(11), 24-29,38 (in Chinese).
Yeh, A. G. O., Xu, X. Q. and Hu, H. Y. 1995. “The social space of Guangzhou City, China.” Urban Geography. 16 (7). 595-621
Zhang, Y. 1998. “A study of the suburbanization of population in Suzhou, Wuxi and Changzhou”. Economic Geography. 18(2). 35-40 (in Chinese).
Zhou, C. S. and Xu, X.Q. 1996. “A study of the spatial characteristics of population change in Guangzhou City”. Economic Geography. 16（2）. 25-30 (in Chinese).
Zhou, M. 1997. “A preliminary analysis of the question of Hangzhou's suburbanization.” Economic Geography. 17( 2). 85-88 (in Chinese).
Zhou, Y. X. 1992. “On the guiding ideology of urban planning in Beijing”. Beijing City Planning & Construction Review. No 4.14-16 (in Chinese).
Zhou, Y. X. 1995. “Suburbanization of Beijing and settlements problems in metropolitan core area”. in Settlements Problems in Metropolitan Core Area : Knowing the Facts and Finding Solution, Proceeding of the International Seminar on Human Settlements Policy '95, edited by National Land Agency of Japan and United Nations Center for Regional Development. 299-322.
Zhou, Y. X. and L. J. C. Ma. 2000. “Economic restructuring and suburbanization in China.” Urban Geography, Vol. 21. 205-236.
Zhou, Y. X. and Meng Y. C. 1997. “Suburbanization in China--A Case Study of Shengyang ” China City Planning Review 13 (2): 50-57.
Zhou, Y. X. and Meng Y. C. 1998. “The tendency of suburbanization of big cities in China.” Urban Planning Forum, No. 3,22-27 (in Chinese).
Zhou, Y. X. et al. 2000. “Report on the survey of a thousand relocation households in Beijing.” Planners, 16 (3), 86-89, 95 (in Chinese).
Figure 1a. The spatial distribution of population growth rate in Bejing metropolis by sub-district, 1982 to 1990.
Figure 1b The spatial distribution of population growth rate in Bejing metropolis by sub-district, 1990 to 2000
Figure 2. The process model of suburbanization in China since the 1990s.
Source: Feng, J. et al. 2004.
Figure 3a Social areas in Beijing metropolis in 1982
Figure 3b Social areas in Beijing metropolis in 2000