G

J

M

PN

a

ARRA

KUGUTB

I

i((telraicN‘‘t‘lc

0d

ARTICLE IN PRESSModel

LUP-936; No. of Pages 14

Land Use Policy xxx (2010) xxx–xxx

Contents lists available at ScienceDirect

Land Use Policy

journa l homepage: www.e lsev ier .com/ locate / landusepol

anaging urban growth in a transforming China: Evidence from Beijing

engjun Zhao ∗

ew Zealand Centre for Sustainable Cities, University of Otago, 23A Mein Street, 6242, New town, Wellington, New Zealand

r t i c l e i n f o

rticle history:eceived 14 October 2008eceived in revised form 12 May 2010ccepted 13 May 2010

eywords:rban containment strategiesrowth managementrban sprawl

a b s t r a c t

Managing urban growth in the current rapid urbanization process has become a key issue for land usepolicy in transformation China. This paper maps and assesses the performance of urban containmentstrategies in China, looking at the case of Beijing over a 19-year period (1990–2009). The analysis showsthat to a large extent containment strategies perform well in terms of concentrating urban growth inplanned suburban areas and promoting compact development. However, the unexpected growth in therural–urban fringe and the decreased compactness of the fringes of the mixed urban areas and plannedperipheral constellations caused by dispersed and illegal development suggest that the municipal con-tainment strategies are being challenged by local development activities. Most sprawling developments

ransformationeijing

at the local level are favoured by the new trend towards local autonomy and fiscal responsibility in thecurrent transformation process. The results reveal that municipal growth control might not be achievedby all local jurisdictions when local economic motivations are involved. Particularly, urban sprawl hasbeen fuelled by the development of urban real estate which can create significant revenues for local gov-ernment and private developers. In the interest of future policy development, the management capacityof current containment strategies should be enhanced to mitigate the negative effects of market-led

nt tra

ipmcsT1Mlvsa1tpl

development in the prese

ntroduction

After the 1960s, urban sprawl became a worldwide problemn relation to metropolitan growth, not only in North AmericaSquires, 2002), Western Europe (Couch et al., 2008) and JapanSorensen, 1999), but also in some large cities in developing coun-ries (Keiner et al., 2005). Many studies have shown the negativeffects of urban sprawl on sustainable development, among whichoss of farmland, loss of environmentally fragile lands and reducedegional open space are frequently reported (Johnson, 2001). Inddition, urban sprawl is one of the significant factors influenc-ng food production and security in rapidly growing developingountries such as China (Zhang et al., 2007). A recent Unitedations report stated that urban sprawl in China has turned some

mega-cities’ (a city with more than 10 million residents) intomega-regions’ which may stretch hundreds of kilometres acrosshe country and be home to more than 100 million people. This

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

endless’ sprawling development could cause more serious prob-ems for the environment if it is not managed properly in theurrent rapid urbanization process (UN-HABITAT, 2008).

∗ Tel.: +64 4 918 6718.E-mail address: pengjun.zhao@otago.ac.nz.

ig

oi1fe

264-8377/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.oi:10.1016/j.landusepol.2010.05.004

nsformation context.© 2010 Elsevier Ltd. All rights reserved.

Recognition of the costs of urban sprawl has prompted pol-cymakers worldwide to create urban containment policies thatrevent the outward expansion of the urban field and forceetropolitan areas to take a more compact form. National urban

ontainment strategies have been in place for many decades inome countries, such as the United Kingdom (Hall et al., 1973),he Netherlands (Dieleman et al., 1999) and Korea (Kim and Choe,997). In other countries, such as the United States (Nelson andoore, 1996), Japan (Nakai, 1988) and Canada (Stewart, 1996),

ocal urban containment strategies have been created by indi-idual municipalities. In some developing countries, containmenttrategies have also been applied to curb urban sprawl associ-ted with the rapid urban growth process. For example, since the980s, China has maintained development containment strategieso preserve agricultural land and promote efficient land use. Theseolicies mainly include limiting city size (both in terms of popu-

ation and the built-up area), tightly restricting new developmentn agricultural areas, encouraging intensive land development, andreenbelt management in some large cities.

However, the performance of urban containment strategies is

transforming China: Evidence from Beijing. Land Use Policy (2010),

ften criticized because their original objectives are frequentlynconsistent with the facts of local development (Longley et al.,992; Nelson, 1992; Pendall, 1999). Underlying the many reasonsor this are the political and institutional contexts that influ-nce policymaking and policy implementation. Particularly, the

ING

J

2 olicy x

ardi2ot(irmctnlafndtFcscTtiNtfat1

cCpigategpomategc

a‘1cslpoplsbb

ssgmsTtmtfboci

ucttotiIctseWfpctovmsta

U

e1cpwcCAtf

stroa

ARTICLEModel

LUP-936; No. of Pages 14

P. Zhao / Land Use P

cceptability of containment strategies to local jurisdictions shouldeceive more attention with respect to the impact of politicalecentralization and piecemeal decision-making concerning the

mplementation of metropolitan containment policies (Carruthers,002). The degree to which local governments implement nationalr metropolitan containment policies is one of the significant fac-ors determining the performance of such policies. As Pendall1999) pointed out, the ability of growth management, includ-ng containment strategies, to limit urban sprawl depends onesponsible regulations invoked by local jurisdictions within theetropolitan area. Some studies suggest there is a challenge

onfronting the implementation of containment policies: a par-icular policy might be good for one municipality but might notecessarily work in all local jurisdictions. Thus, in some ways

ocal growth management may assist in achieving metropolitannd national environmental goals but in other ways may detractrom this mission (Chinitz, 1990). In particular, when local eco-omic benefits are the primary concern of local government, localecision-making on growth – strengthened by political decen-ralization – may challenge municipal containment strategies.urthermore, some research has pointed out another challengeonfronting the implementation of containment policies: macro-cale containment strategies have disadvantages with respect toontrolling urban sprawl fuelled by market forces (Razin, 1998).here is empirical evidence to show that supply restrictions dueo growth control could have significant effects, increasing hous-ng prices or rental costs by reducing housing supply (Dawkins andelson, 2002). Hence, in an open market, housing development

ends to enhance housing availability and meet market demandor cheaper housing outside the range of growth control. However,s a result, sprawling development can occur in suburban jurisdic-ions where growth control is relaxed or absent (Moore and Nelson,994).

These two challenges confronting the implementation of urbanontainment policies in developed countries could well occur inhina. Since the 1978 reforms, China’s economy has been in therocess of transforming from state socialism to market capital-

sm. The transformation involves the interaction of three processes:lobalization, marketization and decentralization of administrativend fiscal powers (Chow, 2007). In the transformation con-ext, there was a clear tendency towards the decentralization ofconomic and fiscal administration and decision-making. Localovernments have now been given much more autonomy in theirursuit of growth. Moreover, the decentralization process has notnly involved changes in the relationship between national andunicipal governments, but also changes at the municipal, district

nd sub-district levels of government within a municipality, withhe local district and county governments becoming more pow-rful social and economic decision-makers. The role of municipalovernments in relation to growth is changing from ‘command-ontrol’ to ‘governance guidance’.

In the meantime, the relationship between local governmentnd enterprises or private developers has entered a phase of

local state corporatism’, as described by Oi (1992). Since the978 reforms, economic development has become an importantriterion in the performance review of local officials, which hasignificantly motivated the pursuit of economic growth at the localevel. The relationship between local government and the enter-rises within its administrative region is viewed as similar to thatf a large multilevel corporation, with local officials holding key

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

ositions and sitting on boards of directors. In some extreme cases,ocal leaders ‘often wear two hats with two different titles at theame time: one official, such as mayor or Party chief; the otherusinessman, such as chairman of the board of a local company orusiness conglomerate’ (Ma, 2002, p. 1552).

iohmp

PRESSxx (2010) xxx–xxx

Consequently, the challenges first faced by urban containmenttrategies in Western cities may now be confronting China. Manytudies have already reported that to maximize revenue gains localovernments have an urgent and large demand for local develop-ent within their own jurisdictions, which has led to serious urban

prawl, for example in Beijing (Deng and Huang, 2004; Wong andang, 2005) and Guangzhou (Wu and Yeh, 1999). Furthermore, inhe transformation context of rapid urban development and a dra-

atic trend towards decentralization in China, these challengeso urban containment strategies could be even greater than thoseaced in Western countries, suggesting that the problems facedy China could be more serious than those experienced in devel-ped countries. In particular, the rapid changes in the institutionalontext mean that urban master plans often lag behind reformsmplemented at national and local levels (Wei, 2005).

The debates introduced above concerning the effectiveness ofrban containment strategies reinforce the importance of empiri-al studies. As mentioned above, some researchers have examinedhe appearance of urban sprawl in China; however, little atten-ion has been paid to the effects of urban containment strategiesn urban sprawl. The research presented here is a first step inhis direction. Taking Beijing as a case study, this paper exam-nes the performance of urban containment strategies in China.n view of the ever-growing and significant role of China’s mega-ities in world urbanization, this empirical evaluation should be ofheoretical, practical and political importance. Theoretically, manytudies have already assessed urban containment strategies. How-ver, the existing literature is dominated by case studies fromestern countries, especially from the United States, while cases

rom developing and transitional countries are scarce. This studyrovides an opportunity to examine the challenges confrontingontainment strategies in a transformation context, with few coun-ries having experienced such dramatic growth and transformationver such a short period as China. In practice, this study can pro-ide actual case studies for policymakers working on urban growthanagement in fast-growing metropolitan areas. Politically, this

tudy provides information for policymakers concerning the wayso curb urban sprawl through containment strategies in transitionalnd developing countries, especially in China.

rban containment strategies in China

Since market reform started in 1978, China has experi-nced rapid urbanization and industrialization. During the period978–2009 China had an average urban growth ratio of 0.93 per-ent per annum and by the end of 2009 China’s overall urbanopulation had reached 622 million, nearly one-fifth (19%) of theorld’s total city population (CSB, various years). According to offi-

ial forecasting, in 2020 there will be 850 million people living inhina’s cities, making up 56 percent of the total population of China.s a result, there is a strong need to balance agricultural land pro-

ection and economic growth as China has a growing population toeed, requiring good-quality agricultural land.

Concomitant to the rapid urban expansion, urban containmenttrategies have been implemented since the 1980s. The key objec-ive of containment strategies has been to preserve farmland in theapid urbanization process and enhance the compactness of devel-pment. The containment strategies were implemented throughseries of policies relating to the strict protection of farmland,

transforming China: Evidence from Beijing. Land Use Policy (2010),

ntensive and high-density land use, the restriction of new devel-pment permission, control of the overall size of big cities andigh-density community development. These policies were imple-ented at all administrative levels, from the macro-level (national,

rovincial and municipal) to the micro-level (county, town, village

ARTICLE IN PRESSG Model

JLUP-936; No. of Pages 14

P. Zhao / Land Use Policy xxx (2010) xxx–xxx 3

ng (19

asa

dAtmToalfml‘pu

tiuewal(ahu

duaoNiNmoicmb

siiuithbuto

Fig. 1. Decentralized concentration strategy in Beiji

nd community). The regulations related to urban containmenttrategies were enacted in stages, beginning in 1986 and continuingt present.

The containment strategies place tough restrictions on newevelopments in the rural–urban fringe. The Land Managementct China (1986) prescribed the strict protection of farmland by

he government. In the 1999 Amendment to the Land Manage-ent Act, even stricter farmland protection policies were enacted.

he amendment states that the ‘efficient use and strict protectionf farmland is a primary national strategy’ (CPGPRC, 1999). Themendment added four detailed policies for the protection of farm-and: the ‘national management of land categories’, ‘the primaryarmland protection system’, ‘limiting the total area of develop-

ent’ and the ‘land development monitoring system’ which hased to the implementation of ‘land use change registration’ and theland survey and statistics system’. In order to implement theseolicies, the amendment clearly stipulated the adoption of the landse planning system.

Various land use planning decisions were made at all adminis-rative levels – national, provincial, county, city or town – as anmportant tool in implementing containment policies. The landse plans were authorized by higher jurisdictions and strictlynforced. In the land use planning system, containment strategiesere implemented in two ways; firstly, by ‘controlling the total

mount of land development’. For example, in 1997, the national

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

and management bureau introduced the national land use plan1997–2010), which stated that the area of new development ongricultural land was to be strictly limited to less than 2 millionectares until 2010 (CPGPRC, 1997). According to the national landse plan, local governments at different jurisdictional levels must

Plgro

91). Source: the author, edited from BIUPD (1991).

evelop a land use plan which is consistent with the national landse plan. The second implementation method involved ‘intensivend efficient urban land use’. For example, in Beijing, the total areaf urban land was to be limited to 61,000 hectares up to 2010.ew development land would be concentrated in the areas des-

gnated as key new towns and industrial zones, such as Tongzhouew Town, the CBD, Yizhuang Economic and Technology Develop-ent Zone and certain peripheral constellations (see Fig. 1). In the

ther areas, urban land development was severely limited. The res-dents of villages were encouraged to move to towns and factoriesoncentrated in industrial zones. Furthermore, urban redevelop-ent and infilling development were encouraged in the existing

uilt-up areas, especially in the city centre.The containment strategies applied within the urban planning

ystem in China include the strict control of the urban size (bothn terms of population size and the built-up size) of big cities (hav-ng non-agricultural residential populations over 500,000 in therban and nearby suburban areas), and a major effort to achieve

ntensive and high-density land use. The big cities have becomehe main places where urban sprawl has occurred, as these citiesave undergone a dramatic urbanization process (Ma, 2004). Theig cities accommodated 61 percent of the total number of newrban residents from 1990 to 2009 (CSB, various years). Hence, con-rolling urban size and urban expansion is one of the main policiesf national containment strategies. For example, in the first City

transforming China: Evidence from Beijing. Land Use Policy (2010),

lanning Act (1989), the urban size of China’s big cities was strictlyimited. Many policies have been implemented to limit the urbanrowth of the big cities. For example, new migrants moving fromural to urban areas are encouraged to live in small cities or townsutside the big cities. Additionally, the system of household res-

ING

J

4 olicy x

iZpbtcN1cIset

astocsacpAtlptonsftcaitl

ommPniTctsstarvslomltwhkiH

U

acagfsmf

bittrmaanormtjmjtbtpalg‘cgmpe1gtFrmntZ

toifuabhv

ARTICLEModel

LUP-936; No. of Pages 14

P. Zhao / Land Use P

dential permits is also used to control this migration (Chan andhang, 1999). Restricting land development permission is anotherolicy applied in big cities. As well as limiting the urban size ofig cities, ‘[i]ntensive and high-density land use’ was promoted inhese big cities through urban planning. In particular, density wasontrolled by the planning regulations. For instance, in 1990, theational Urban Land Use Classification and Planning Standard (GBJ37-90) prescribed that the residential population density in bigities was to be no less than an average of 140 persons per hectare.n addition, the city’s facilities (for example, schools, hospitals,hops) and utilities (for example, water, sewerage, electricity), andven some infrastructure (such as roads), were planned accordingo existing and future trends in residential population density.

In some big cities, a ‘decentralized concentration’ strategy waspplied to shape compact urban form at the city level. For example,ince 1983, Beijing has implemented a decentralized concentra-ion strategy, which officially encourages compact developmentn the urban fringe (BIUPD, 1983, 1991, 2004). This decentralizedoncentration strategy was greatly enhanced in the latest two ver-ions of the Beijing urban master plan (BIUPD, 1991, 2004). Theim of the decentralized concentration strategy is to encourageoncentrated development in planned suburban areas, with theopulation in the city centre being dispersed into the suburbs.ccording to this strategy, the city is classified into three areas:

he city centre, main urban area and suburban area. New towns areocated in the suburban area. The main urban area is zoned into fourarts: mixed urban areas, greenbelt, peripheral constellations, andhe rural–urban fringe (Fig. 1). There are three main componentsf the decentralized concentration strategy: development of selectew towns in the suburban area, development of peripheral con-tellations and greenbelt management. According to the strategy,urther developments occurring in the suburbs would be concen-rated in several planned new towns and ten planned peripheralonstellations. The greenbelt was located between the mixed urbanrea and the peripheral industrial areas to act as a buffer, restrict-ng further growth and development in the mixed urban areas andhe city centre. The development in the rural–urban area is strictlyimited by the municipal government.

Experience from Western countries shows that the objectivesf urban containment strategies are not successfully achieved inost cases, as macro-growth management is often not well imple-ented by local jurisdictions (Longley et al., 1992; Nelson, 1992;

endall, 1999). For example, Alterman (1997) conducted a cross-ational comparison of institutional, policy and contextual factors

nfluencing the implementation of urban containment strategies.he author came to the conclusion that the effectiveness of urbanontainment strategies was moderate and that stringent legal con-rols, for example, planning laws, are not enough to control urbanprawl. The criticisms of the effectiveness of urban containmenttrategies reveal that alongside the strategies themselves, institu-ional factors have a significant influence on how the strategiesre implemented and therefore their success. In particular, theelationship between state government, local government and pri-ate developers affects the degree to which the objectives of thetrategies can be achieved by actual developments at the localevel (Carruthers, 2002). In fact, objections to the implementationf urban containment strategies particularly arise in places wherearket forces operate freely with the support and assistance of the

ocal government (Frenkel, 2004). International experience revealshat the effectiveness of urban containment strategies could be

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

eak in China where political decentralization and marketizationave greatly increased the power of local government and the mar-et in local urban development. In fact, urban sprawl has occurredn big cities despite strict urban containment policies (Deng anduang, 2004; Wong and Tang, 2005).

ads2a

PRESSxx (2010) xxx–xxx

rban transformation of China

Since the reforms of the 1980s, China has been undergoingn obvious transformation, which has at least two fundamentalharacteristics. The first is that it involves three contemporarynd interrelated processes: decentralization, marketization andlobalization (Chow, 2007; Wei, 2001). The second is that the trans-ormation in China has been a gradual process in which the marketystem and the planning system coexist (Zhu, 2000). The transfor-ation process has created new institutional and social contexts

or urban development and land development management.In the decentralization process, the hierarchical relationship

etween the state and the city has been complicated by the twossues of local autonomy and fiscal responsibility. Before 1978,he power of the administrative system in China was highly cen-ralized and hierarchical, with the Party-state playing multipleoles in urban growth and management as an ultimate decision-aker, regulator and participant. After the early 1980s, there wasclear tendency towards decentralization in fiscal administration

nd decision-making. The retreat from an ideology of central plan-ing, the decentralization of decision-making and the injectionf foreign investment into local economies have contributed to aeorganization of the decision-making system, from a ‘top-down’odel in which the central Party-state determines the ‘rules of

he game’, to a negotiation and interactive system in which localurisdictions, firms and residents participate more in the decision-

aking process (Ma, 2002). The relationship between state, localurisdictions and enterprises has been rescaled dramatically inhe decentralization process (Shen, 2005). Local government hasecome increasingly important to the urban growth process sincehe introduction of decentralization and marketization. For exam-le, since the 1980s, local governments have emerged as majorgents of investment, while the central government now plays aess significant role in investment allocation (Wei, 2001). Municipalovernments are changing their roles from ‘government control’ to

governance guidance’ in growth management. Local district andounty governments are becoming more powerful and acquiringreater administrative control in the social and economic decision-aking process through new regulatory, taxation and licensing

owers (Wank, 1999, p. 250). The relationship between local gov-rnment and enterprises or firms has developed into what Oi (1992,998, 1999) has called ‘local state corporatism’, in which the localovernment treats enterprises within its administrative jurisdic-ion as one component of the local corporate whole (Zhu, 2000).urthermore, the state has also granted more decision-makingights to state-owned enterprises (SOEs). SOEs have thus becomeore autonomous and have been relieved of the burden of plan-

ing quotas to the extent that they can make decisions based onhe market demand for production (Hay, 1994; Naughton, 1995;hu, 2000).

In the marketization process, market rules have become impor-ant in determining the ‘rules of the game’. Marketization haspened much space for enterprises other than SOEs to becomenvolved in urban development, such as private, joint-venture ororeign enterprises. A multi-stakeholder arrangement (MSA) forrban development has been introduced. In China’s cities, amongll the marketization reforms, housing and land market reform maye two of the most important. The housing and land market reformsave led to two changes in property right: marketization and pri-atization. In the planned system in the pre-reform era, housing

transforming China: Evidence from Beijing. Land Use Policy (2010),

nd land were allocated through administrative orders that wereetermined according to the economic plan. Land was not con-idered a commodity and had no ‘value’ (or market price) (Ding,003). Housing was seen as a form of social welfare rather thancommodity and it was provided to all employees for free or at

ING

J

olicy x

ldrarpoattIa1ss2

t(thut2sa(swiob2bhTs

cctuTcdobba

drwtAtcatdccBla

dtTctCdmaZeicohkehsisesthotcs

caalmTtpieti

T

uml[ciuattmet

ARTICLEModel

LUP-936; No. of Pages 14

P. Zhao / Land Use P

ow rent (Ma, 2004; Wang et al., 2005). Housing ownership wasominated by public ownership. Since the 1980s, when urban landeform and housing reform were implemented, market rules medi-ted by price mechanisms have emerged in allocating land useights and housing ownership. In the urban land market reformrocess, land use right has been separated from state-owned landwnership. Land use rights can now be transferred according todetermined monetary value (Walker and Li, 1994). At the same

ime, non-government organizations have been given legal accesso a land use right transfer system (Xie et al., 2002; Zhang, 2000).n the housing reform process, housing was commoditized and as

result real estate markets have now emerged in China (Chen,998; Dowall, 1993). The housing reforms aimed to transform theocialist welfare-oriented housing system into a market-orientedystem through the privatization of housing ownership (Huang,004; Huang and Clark, 2002; Wang and Murie, 1999).

In the privatization process, occupiers have been encouraged toake full ownership of their homes through a ‘purchasing’ processHuang, 2005; Li, 2000; Li and Li, 2006). For example, in Beijing,he empirical findings show that workers who resided in publicousing provided by danwei (a state or collectively owned worknit) have gradually moved from renting their work-unit housingo owner occupation over the period since the 1980s (Li and Yi,007). Housing reform broke the public sector-dominated housingystem and diversified housing provision and investment (Wangnd Murie, 1996). According to research by Li and his colleaguesLi, 2000; Li and Tang, 1998), four types of commodity housingources can be identified: the open market, danwei (state-ownedork unit), housing bureau and resettlement housing. Public hous-

ng within a danwei declined although it was the main sourcef commodity housing (housing sold on the open market) at theeginning of the housing market reform period. For example, in005, commodity housing (the housing sold on the open market)ought privately amounted to 89.3 percent of the total commodityousing in Beijing (Beijing Real Estate Association, various years).he booming housing market has become one of the major factorstimulating suburban development in China’s cities.

Transformation in China is an ‘experimental and gradual’ pro-ess in which the old planned system and new market systemoexist. Compared with eastern European socialist countries, theransformation towards a market system in China has been a grad-al process which started in 1978 and continues to the present-day.he dualism resulting from the coexistence of marketization andentral planning creates a unique institutional context for urbanevelopment and growth management, as central planning hasften been largely retained, while some aspects of capitalism haveeen allowed to operate. The market system is strongly controlledy the forces of centralized government and its planning apparatus,lthough it is very apparent in some parts of the country.

There are four dimensions to this dualism in China’s urbanevelopment (Zhu, 2000). Firstly, the central planning system hasetained great power in the decision-making process (Hu, 2000),ith the notion of ‘managed openness’ being part of the tradi-

ional ideology relating to China’s reform process (Weiss, 1999).ccording to the conception of ‘managed openness’, marketiza-

ion is not only favoured by government but also monitored by theentral government. Secondly, the state-owned danweis still playn important role in most cities, although the scale and impor-ance of state-owned enterprises (SOEs) has declined in the pastecades (Wang et al., 2005). Thirdly, the emerging land market

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

oexists alongside land allocation occurring through a command-ontrol process (Yeh, 2005; Yeh and Wu, 1996). For example, ineijing, 9219.05 hectares, which is nearly 98 percent of the total

and allocated during 1992–2002, was transferred through negoti-tions between government and developers, despite the fact that

r

bro

PRESSxx (2010) xxx–xxx 5

ramatic market-oriented land reforms were carried out duringhe same period (Beijing Real Estate Association, various years).he coexistence of the land allocation system and the free marketreates the so-called ‘double-track system’ (Ding, 2003) or ‘dual-rack land system’ (Zhang, 1997). It indicates that the land market inhina is still far from mature (Wu, 2001; Xie et al., 2002; Zhu, 2004),espite many reforms relating to land development and manage-ent. One main reason for the dualism in the land market is the

mbiguous nature of land use rights in China (Lin and Ho, 2005;hu, 2005). The fourth aspect of the dualism occurs in the prop-rty market, where two kinds of property rights still coexist. Ones the full property right, available to commodity housing whichan be bought and sold on the open market at market prices. Thether is the limited property right, available to non-commodityousing which cannot be freely bought and sold on the open mar-et. The main reason is that housing reform has enabled statenterprises to withdraw from direct housing production; however,ousing consumption is still often subsidized and allocated throughtate danweis (Wang and Murie, 1996; Wu, 1996). Danweis are stillmportant purchasers and distributors of commodity housing anduch a traditional redistribution system, as suggested above, stillxerts a large influence on housing consumption, in particular inome less open cities, for example, Beijing and some other cities inhe remote west of China, where fewer market-oriented reformsave been conducted than in other cities (Li, 2000). In these lesspen cities, market reforms will usually not be implemented untilhey have been successfully tested in the more open cities, as it isonsidered more important to maintain political stability and pre-erve management systems than to encourage economic reforms.

In summary, the dramatic urban transformation in Beijing hasreated a unique institutional context for urban developmentnd land development management. Market forces are growingnd play an important role, affecting urban development andand development management. However, marketization does not

ean there is no intervention by the state and local governments.ransformation in China is characterized by a dualism in whichhe growing market coexists alongside the remaining centrallylanned system. In some areas the market plays a major role, but

n other areas command-control by the government continues toxert considerable influence. This dualism creates a unique insti-utional context for urban development and growth managementn China’s cities.

he case of Beijing and methodology

In this section, the paper will use Beijing as a case to evaluaterban containment strategies. The city of Beijing has existed forore than 3000 years and is China’s capital. In 2008, its total popu-

ation was 16.95 million, with a total land area of 16,410 km2 (BSBBeijing Statistic Bureau], various years). Its administrative areaomprises sixteen districts and two counties. The city is dividednto three zones: the city centre, the main urban area and the sub-rban area (see Fig. 2). This paper looks at the main urban areand the city centre, two zones which provided 79.1 percent ofhe employment and made up 68.3 percent of the population ofhe city in 2009 (BSB [Beijing Statistic Bureau], various years). The

ain urban area in particular is the primary zone in which urbanxpansion has occurred in Beijing since the 1990s and it consists ofhe mixed urban area, greenbelt, peripheral constellations and the

transforming China: Evidence from Beijing. Land Use Policy (2010),

ural–urban fringe, shown in Fig. 1.The study period for this paper is from 1990 to 2009, with 1990

eing selected principally because it marks the start of the mostapid process of urbanization in Beijing, which continued through-ut the 1990s. At the same time, several fundamental policies were

ARTICLE ING Model

JLUP-936; No. of Pages 14

6 P. Zhao / Land Use Policy x

dNnwi

usemltatwtsiwbmib

gsitcct(uigdop

jtwasaoo

muaadmupiittmfisdr

wadsitrmZoseTtuld

M

isceti

ohbever, the entropy index requires detailed data concerning land use

Fig. 2. The location of case study areas of Beijing.

eveloped in relation to urban growth management, such as theational City Planning Act, enacted in 1989, and, in particular, theew version of the City Master Plan of Beijing (1992–2010), inhich containment strategies were emphasized as one of the most

mportant principles of growth management in Beijing.The analysis units are the 139 sub-districts located in the main

rban area and city centre of Beijing. These sub-districts consist oftreet areas (Jiedao) and townships (Xiangzhen). There are three lev-ls of administrative management in Beijing. The first level is theunicipal government level, the second is the district or county

evel and the third is the sub-district level, the basic administra-ive unit in Beijing. There are two sorts of administrative unitst the sub-district level. One is the street area, the other is theownship area. Street areas are usually located in urban regionsith state land ownership. Townships are usually located within

he rural–urban fringes with predominantly collective land owner-hip. These two different administrative units are treated equallyn the analysis as they have a similar geographical scale. The reason

hy the sub-district is selected as the analysis unit in this paper isecause it is the basic administrative unit of Beijing. Most of theunicipal policies are implemented at this level. For research aim-

ng at policy evaluation, this geographical scale would thus be theest choice.

This study will evaluate the performance of containment strate-ies by examining the gaps between the objectives of containmenttrategies and development facts. Containing further urban growthn planned areas and shaping a compact urban form are two ofhe important objectives of containment strategies. This paperonstructs several key indicators to measure urban growth andhanges in compactness and then evaluates to what extent the con-ainment goals have been achieved by actual local developmentssee Tables 1 and 2). Two types of indicators are constructed to eval-ate the urban containment strategies. The first is an urban growth

ndicator, which is represented by population growth, housing

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

rowth and the growth of industrial development. The populationata used in the study comes from the Beijing statistics yearbooksf 1990 and 2009. The population here refers to the ‘long-termopulation’ (chang zhu ren kou). According to a note in the Bei-

ctlj

PRESSxx (2010) xxx–xxx

ing Statistical Yearbook, the ‘long-term population’ is defined ashose residents who have resided in Beijing for at least one yearhen the statistics surveys were conducted. Temporary residents,

lso referred to as the ‘floating population’, are not included in thistudy because their numbers cannot be accurately measured by theuthorities. Housing growth is measured by changes in the numberf houses. Industrial growth is indicated by changes in the numberf registered enterprises.

The second indicator is related to urban form and is employed toeasure the changes in urban compactness. This paper examines

rban compactness over two aspects of urban form, the functionalnd the physical aspects (see Table 1). In relation to the functionalspect, density is measured using four indicators. Firstly, the grossensity of the population is the gross population of a sub-district,easured in persons per hectare. This measure is most commonly

sed in urban research because the data is easy to obtain from theopulation census (Galster et al., 2001; Tsai, 2005). However, the

ndicators cannot describe the actual urban development situationn one administrative area. Thus, in the second density indicator,he administrative land area is replaced by the built-up land areao show the ‘net’ density of the population in the urban develop-

ent area, omitting the non-built-up areas such as greenland andarmland. The third density indictor is the net employment densityn a built-up area, which can measure the degree of activity inten-ification in one area. The fourth density indicator is the householdensity per hectare in a built-up area. This indicator is used to detectesidential density and real estate development in one area.

Mixed land use can be measured by the degree of diversity,hich depends on the geographical scale of the analysis (Batty et

l., 2003). There is different diversity at different levels, such as city,istrict and neighbourhood. Galster et al. (2001) quantified diver-ity in a single square-mile grid scale using an exposure index. Thisndex could show land use diversity in a specific area by showinghe average density of residential land use in relation to other non-esidential land use. Song and Knaap (2004) measured the land useix with the concept of entropy at the TAZ level (Traffic Analysis

one). The entropy measure is used widely to examine the degreef mixed land use (Cervero, 1989; Frank and Pivo, 1995). The diver-ity of urban land use (in terms of commercial or business uses, forxample) can be reflected in the diversity of employment activities.herefore, this paper uses the diversity of employment activitieso measure the degree of mixed use in the primary administrativenits. The entropy indicator is weighted by the population of the

ocal unit. The higher the entropy in the urban area, the higher theegree of mixed use in the local unit.

ei =(−∑5

j=1(pj) ln(pj))/ ln(s)

pi

where Mei is the degree of mixed employment activities in unit; pj shows the proportion of each of the five employment types,uch as agricultural employment, industrial employment, commer-ial employment, state and municipal government employment,ducation and research employment and employment in other ter-iary sectors; pi presents the size of the population in unit i; and ss the number of employment types. In this case, s is 5.

The degree of mix in employment activities is just one aspectf mixed use. The other important aspect is the balance betweenouseholds and jobs at the local level. The entropy index can alsoe used to measure the jobs–housing balance (Frank, 1994). How-

transforming China: Evidence from Beijing. Land Use Policy (2010),

ategories. Cervero (1996) used an indicator of the ratio of jobso employed residents to measure the jobs–housing balance in thearge cities of the San Francisco Bay area. Peng (1997) employed theobs–housing ratio to determine the jobs–housing balance in Port-

ARTICLE IN PRESSG Model

JLUP-936; No. of Pages 14

P. Zhao / Land Use Policy xxx (2010) xxx–xxx 7

Table 1Selected key performances indicators and measures.

Indicator Explanation Measure Data source

Urban growth indicatorsPopulation growth Persons BSB, various years (1990–2009)Housing growth Number of housing BSB, various years (1990–2009)Industrial growth Number of registered enterprises BSB (1996) and BSB (2010)

Urban form indicators

Density

Gross density of population Persons per hectare BSB, various years and BLHB (1990, 2009)Net density of population Persons per hectare in built-up area BSB, various years and BLHB (1990, 2009)Net density of employment Employees per hectare in built-up area BSB, various years and BLHB (1990, 2009)Net density of households Households per hectare in built-up area BSB, various years and BLHB (1990, 2009)

Mixed land useThe mixed degree of employment Entropy indictor BSB (1996) and BSB (2010)The degree of distribution equality ofhouseholds and jobs

Standard deviation BSB (1996) and BSB (2010)

Concentration The concentration degree of patchesshape

Area–perimeter ratio of built-up patches BLHB (1990, 2009)

Continuity The continuous degree of the built-upareas

Local Moran’s I BLHB (1990, 2009)

Data source: BSB (various years), Beijing Statistical Yearbook (1990–2009); BSB (1990), Beijing 4th Census; BSB (1996), Beijing The First National Basic Unit Census; BSB(2010), Beijing The Second (2008) National Economic Census; BLHB (1990), Beijing Land Readjustment Data; BLHB (2009), Beijing Land Readjustment Data.

Table 2Urban growth in Beijing, 1990–2009.

Number of units Changes in populationon average (%)

Changes in the number ofhousing on average (%)

Changes in the number ofregistered enterprises onaverage (%)

City centre 28 −10.3 −6.6 44.5Mixed urban areas 49 66.5 73.9 129.2

luIsehb

dotaltp

icahiauwb

A

C

re

ipgaOiutftei

caiagco

oihmtlt

Planned peripheral constellations 28 137.2Green belt 18 93.7Rural–urban area 16 120.6Total 139 104.5

and, Oregon. The ratio method is the simplest and most frequentlysed measure; however, this method neglects the land use factor.

n this article, a jobs–housing balance indicator is measured by thetandard deviation between household density and job density inach local area. The higher the standard deviation between house-old density and job density, the more unbalanced the relationshipetween households and jobs within a local jurisdiction.

In the local development process, containment policies limit theevelopment of strip and scattered patches of land use. The shapef patch development is one important indicator when evaluatinghe performance of urban form at the local level. This paper uses anrea–perimeter ratio (APR) of one patch to measure the shape ofocal land development. The higher the APR of one patch, the higherhe degree of concentration. The average value of the total APR ofatches represents the degree of concentration in sub-areas.

Another measure is a continuity indicator used to measure phys-cal attributes in the polycentric city. The continuity of urban forman be measured by the Moran coefficient. The Moran coefficient isquantitative index used to measure the degree of continuity in theigh-density clusters (Tsai, 2005). It can also measure continuity

n a built-up area. A high Moran coefficient means that a built-uprea has good characteristics in relation to clustered and contin-ous development. This paper employs the local Moran’s index,hich measures the degree of spatial continuity at the local level,

ased on feature locations and attribute values.

ssessment of urban containment strategies

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

hanges in urban growth

Beijing has experienced rapid population growth followingapid urbanization and a booming economy since the 1990s. How-ver, the growth of the population has occurred at a different pace

Si2ao

125.4 173.588.2 135.1

158.3 167.7111.5 130.0

n different areas. Table 2 shows that the sub-districts located inlanned peripheral constellations have had the highest populationrowth and the highest growth in industrial development on aver-ge. The city centre experienced a dramatic decrease in population.f all the factors influencing the distribution of the population and

ndustrial sites, urban growth management is important. In partic-lar, the development of industrial parks is strongly managed byhe municipal government. These population and industrial growthacts suggest that urban containment strategies have achieved theirwo main aims: to encourage urban growth in the planned periph-ral constellations and new towns and to limit population growthn the city centre.

However, Table 2 shows that outside the planned peripheralonstellations, the sub-districts located in the rural–urban fringelso had a relatively high population growth and the highest hous-ng development. The results are inconsistent with one of the mainims of urban containment strategies – that further populationrowth and housing development in the suburbs should be con-entrated in the planned peripheral constellations and the growthccurring in the rural–urban area should be limited.

Compared with the municipal government-led industrial devel-pment, housing development in the suburbs is substantiallynfluenced by the market. Especially in relation to commercialousing development, private developers and the local govern-ent (district, county and town) play a more active role than

he municipal government. The main reason for this is that theocal government needs revenues from real estate developmento cover the high costs of providing local public social services.

transforming China: Evidence from Beijing. Land Use Policy (2010),

ince the 1990s, commercial housing prices have been increas-ng in Beijing, with the housing price soaring after 2001. Between001 and 2008 the commercial housing price increased on aver-ge by 216.5 percent (BMCHURD [Beijing Municipal Commissionf Housing and Urban-Rural Development], 2009). The booming

ARTICLE IN PRESSG Model

JLUP-936; No. of Pages 14

8 P. Zhao / Land Use Policy xxx (2010) xxx–xxx

Table 3The overall changes to urban compactness in Beijing, 1990–2009.

Indicators Mean Standard deviation Maximum Minimum

Gross density of population 1990 126 124 518 22009 151 116 896 5

Net density of population 1990 145 113 518 72009 179 109 896 11

Net density of employment 1990 85 66 289 32009 68 71 415 5

Net density of households 1990 44 38 163 22009 53 32 329 5

Mixed degree of employment 1990 0.7177 0.1181 0.8872 0.43802009 0.8102 0.1451 0.9103 0.4386

Jobs–household balance 1990 29.80 22.53 97.41 0.202009 19.76 13.85 78.26 0.83

Concentration degree of patches shape 1990 177.53 126.96 670.55 45.832009 274.85 133.86 709.92 71.11

85

rnrs

epdtrtTbopcocmgl

laMd

O

cbv(h

otsdiat

rr

8sridLldcFtiy

c(aetibb2Orttrimaecgc

Continuous degree of the built-up areas 1990 0.0012009 0.003

eal estate market provides local government with the opportu-ity to obtain increased revenue from housing development. Localevenue from land development is called ‘extra income outside thetate budget’ (yu suan wai shou ru).

According to the regulation governing the distribution of rev-nue arising from the transfer of the land use right from the state torivate ownership, 60 percent of the total revenue arising from landevelopment is collected by the state and 40 percent is retained byhe municipal governments. Each year, the municipal governmentsedistribute their share to the local governments (districts, coun-ies and towns) based on the revenue that each have contributed.his means that increased land development and higher contri-utions will bring greater local revenue return, although it is alsoften the case that local governments retain revenue that should beaid to the state and municipal governments (Chen, 2003). In someases, revenue from land development amounts to over one-thirdf total local revenue (Jia and Le, 2010). This trend has not obviouslyhanged, despite the fact that the state and the municipal govern-ents have introduced several strict land policies requiring local

overnments to control their land permission grants and againstand hoarding and real estate speculators since 2003.

The above results suggest that actual urban growth at theocal sub-district level in the rural–urban fringe conflicts with theims of macro-urban growth management at the municipal level.arket-led real estate development particularly influences land

evelopment in the rural–urban fringe.

verall changes in urban compactness

From the indicators of urban compactness, a general conclusionan be drawn that the overall functional aspects of urban form haveecome more compact since the 1990s, except with regard to indi-idual aspects such as a decrease in the density of employmentsee Table 3). In relation to physical form, local development had aigher compactness in 2009 than in 1990.

The density of the gross population increased from an averagef 126 persons per hectare to 151 persons per hectare. In addi-ion, gross population density shows a more even trend, with the

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

tandard deviation of gross population density in the sub-areasecreasing from 124 to 116 since 1990. The two other density

ndicators, the net population density and net household density,lso show obvious growth. The increased density indicators implyhat to a large extent containment strategies performed well with

olrem

0.0050 0.0177 −0.04000.0061 0.0283 −0.0041

espect to the objective of enhancing urban compactness in theapid urban growth process.

However, the average net density of employees declined from5 persons per hectare to 68 persons per hectare, while most den-ity indicators increased. The decreases in employment densityeveal one challenge faced by current municipal containment pol-cy: the conflict between municipal land control and huge localemand for land development in the local industrialization process.ike most big cities in China, Beijing aims to achieve a successfulevel of industrialization and become an economically indepen-ent industrialized city. Local industrial developments outside theity centre play a significant role in this industrialization process.or example, the gross output value of industry in suburban dis-ricts accounted for 67.3 percent of the total gross output value ofndustry in Beijing in 2009 (BSB (Beijing Statistic Bureau), variousears).

Among the factors influencing the local industrialization pro-ess, the need to enhance local revenue is one of the most importantWu, 1990a, 1990b). Since the 1990s, greater fiscal responsibilitynd competition between local jurisdictions have led local gov-rnments to stress the need for rapid local industrialization. Manyown or village enterprises (xiangzhen qiye) have been developedn various local industrial development zones which are managedy the local district or county government. In Beijing, the num-er of industrial development zones reached 470 by the end of003, and they were located across all suburban areas (Li, 2007).f these industrial development zones, 94 percent were autho-

ized by local governments (district, county or town), while athe state and municipal levels they were seen as illegal. Rela-ively rapid land development in local industrial zones is one maineason for a decrease in the density of employment. When thesellegal industrial development zones were closed by the state and

unicipal governments in 2003, 467 km2 of land which had beenssigned to be developed was saved. The decrease in the density ofmployment alongside rapid local industrialization suggests thaturrent municipal containment strategies are being challenged byrowth-oriented development. Clearly, there is at least one policyhallenge confronting containment strategies in the growing city

transforming China: Evidence from Beijing. Land Use Policy (2010),

f Beijing: the objectives of municipal containment strategies areikely to conflict with the facts of local development in the cur-ent transformation process as a result of local governments beingncouraged to make their own decisions on economic develop-ent.

ARTICLE IN PRESSG Model

JLUP-936; No. of Pages 14

P. Zhao / Land Use Policy xxx (2010) xxx–xxx 9

eholds

iosmcntddso

uaiccaaif

C

st2d(ddtctctAp

cdteppaitilcTts

C

icastiTuip

hsstt

Fig. 3. The changes in net density of hous

The analysis shows that the average degree of mixed usencreased during the period 1990–2009. The increasing indicatorf ‘mix entropy’, which measures the diversity of employment,hows that the degree of mixed use has been upgraded at theunicipal level. In a similar way, the jobs–housing balance indi-

ator, which is measured by the standard deviation between theet density of households and employment, also verifies the facthat the jobs–housing balance has improved. The average standardeviation between the net density of households and employmenteclined distinctly, from 29.80 in 1990 to 19.76 in 2009. The resultsuggest that containment strategies perform well in promotingverall mixed land use.

The results also show that the shape of areas developed forrban land use has greater continuity in 2009 than in 1990. Theverage value of the local Moran’s I indicator increased from 0.0018n 1990 to 0.0035 in 2009. This means that to some extent theontainment strategies achieved their goal of enhancing overallompact urban form by encouraging new development locateddjacent to the existing built-up area. The APR indicator for built-upreas increased from 177.53 in 1990 to 274.85 in 2009. This resultndicates that urban expansion in Beijing tends to take a compactorm when containment strategies are implemented.

hanges in the density of Beijing

The results of the comparison of the figures for 1990 and 2009how that the sub-districts located in the city centre had nega-ive growth in population, household and employment density. In009, the population and household density of these sub-districtsecreased by 13.7 and 10.9 percent on average, respectivelysee Fig. 3). The employment density of these sub-districts alsoecreased by 8.6 percent on average. The main reason for the lesserecrease in employment is that while the decentralized concentra-ion policy tends to disperse population and households outside theity centre with the aim of reducing the congestion in the city cen-

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

re, the local government tends to replace old communities in theity centre with high-density office buildings for commercial func-ions in order to increase local revenue (Fang and Zhang, 2003).s a result, employment density declined relatively slowly whileopulation and household density decreased rapidly.

ddHdp

and employment in Beijing, 1990–2009.

Despite this trend, population density in some sub-districtslose to the centre has increased. In particular, because of theecentralized concentration strategy, ten peripheral constella-ions became the ‘growth poles’ where most population andmployment became concentrated. The sub-districts located in thelanned peripheral constellations had a growth of 11.4 percent inopulation density and 18.5 percent in employment density onverage. However, the results show that some sub-districts locatedn the rural–urban fringe experienced obvious decreases in popula-ion density. The density of employment and households changedn the same way as population density. The result indicates thatand development on the fringes of urban areas in Beijing pro-eeded at a faster rate than population and employment growth.his suggests that the municipal environmental goal to increasehe intensification of land use was not being achieved by someub-districts on the urban fringe.

hanges in mixed land use in Beijing

In 1990, the city centre had the highest value for the mixed usendicator. However, by 2009, the degree of mixed use in the cityentre had declined, while most sub-districts in the mixed urbanrea and the greenbelt had become more mixed (see Fig. 4). A rea-on for an increase in mixed land use in these areas might be thathe new suburban industrial areas, encouraged by the decentral-zed concentration strategy, attracted greater numbers of residents.he decreased mixed use in the centre owes much to market-ledrban development, which favours non-residential development

n urban central areas and has been pursued by local officials andrivate developers for economic benefits.

Most sub-districts in the peripheral constellations experiencedigh growth in relation to the jobs–housing balance. The main rea-on is that the decentralized concentration strategy encourageself-contained development, with a high level of mixed land use inhe peripheral constellations. The result indicates that the decen-ralized concentration strategy performed well to a large extent,

transforming China: Evidence from Beijing. Land Use Policy (2010),

espite there being a citywide trend towards revenue-enhancingevelopment as a result of the current marketization process.owever, some sub-districts in the rural–urban fringe have hadramatic decreases in the degree of mixed land use. For exam-le, Zhongguancun, which is known as ‘China’s Silicon Valley’, is

ARTICLE IN PRESSG Model

JLUP-936; No. of Pages 14

10 P. Zhao / Land Use Policy xxx (2010) xxx–xxx

the job

lshtmhAnhto

C

oc

cftfflaga

to

Fig. 4. The changes in mixed land use and

ocated between Xiyuan, one of the ten planned peripheral con-tellations (see Fig. 1), and the mixed urban area. Zhongguancunas shown a considerable decline in mixed land use. This was dueo revenue-enhancing development which addressed the require-

ent for rapid commercial and retail development while inhibitingousing development, in particular, social housing development.ccording to the municipal ‘decentralized concentration strategy’,ew developments in the planned peripheral constellations shouldave a high level of mixed land use. The results, however, show thathe actual local developments are inconsistent with the objectivesf macro-development management at the municipal level.

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

hanges in the concentration and continuity of Beijing

The analysis shows that in the period from 1990 to 2009 devel-pment had taken a more compact urban form at the local levellose to the city centre (see Fig. 5). Peripheral constellations, espe-

iaddl

Fig. 5. Changes in concentration and

s–housing balance in Beijing, 1990–2009.

ially, achieved a more compact development shape as these areasell under the control of strict municipal containment strategies. Athe same time, some sub-districts on the edge of the rural–urbanringe show decreases in the degree of local concentration. Thisact suggests that new developments in the local towns and vil-ages on the urban fringe are likely to have taken an irregular formnd would challenge containment strategies which have the sin-ular objective of achieving concentrated development rather thanllowing urban sprawl in urban fringe areas.

The value of the local Moran’s I indicator shows that since 1990,he city centre has had a relatively higher value and that the devel-pments had high continuity. In fact, most of the development

transforming China: Evidence from Beijing. Land Use Policy (2010),

n the city centre occurred as infilling redevelopment encour-ged by containment strategies. Outside the city centre, some newevelopments in the surrounding greenbelt occurred in a moreiverse manner. The main reason for this was that some individual

ocal jurisdictions with a strong desire for local growth permit-

continuity from 1990 to 2009.

ING

J

olicy x

tditmgtri2(

jMiHodi

D

igmiCdceitllcwta

tBssTldmj1w

tiigtumtjs

o

umdssspsiamfioa

rdHuadksoroldol‘dfs

io1rrtcC(alaemcbHnra

jmls

ARTICLEModel

LUP-936; No. of Pages 14

P. Zhao / Land Use P

ed developments based on their individual benefits, although newevelopment in the greenbelt was strictly limited by the munic-

pal government. The results reveal that in the greenbelt areashe conflict between local demands for land development and the

unicipal controls on new development is becoming fierce. Thereenbelt, one of the most important tools of the municipal con-ainment strategy, faces challenges from local developments. Thisesult is consistent with the findings of previous studies which havenvestigated failures in Beijing’s greenbelt strategy (Jun and Zhou,007). Similar findings have also been reported in Western citiesMillward, 2006; Munton, 1983).

The analysis shows that outside the greenbelt areas, some localurisdictions in the rural–urban fringe have an increased local

oran’s I value. This reveals that the containment strategy of lim-ting discontinuous development performed well in these areas.owever, some sub-districts in the rural–urban fringe show anbvious decrease in the local Moran’s I value, indicating that urbanevelopment in some sub-districts is still occurring in an increas-

ngly discontinuous and scattered manner.

iscussion

Similarly to the experiences of Western countries, rapid urban-zation has greatly contributed to the promotion of economicrowth and social wellbeing in China. Meanwhile, the environ-ental effects of urban expansion, particularly urban sprawl, are

ncreasingly attracting the attention of politicians and researchers.onsidering the ever-growing urban population and the hugeemand for land for development within developing countries,ontainment strategies designed to curb urban sprawl and enhanceffective land use are vital aspects of land use policies. However,n practice the experiences of developed countries suggest thathere are conflicts between macro-growth management and theocal requirement for development. These conflicts tend to chal-enge the implementation of urban containment strategies. Thishallenge becomes far more serious in a transformation countryhere the power influencing urban development has fundamen-

ally changed, with an increase in the authority of local governmentnd the power of the market.

The analysis undertaken in this paper provides evidence onhe performance of urban containment strategies in the case ofeijing. The analysis found that the objectives of containmenttrategies to encourage urban growth in planned peripheral con-tellations and to increase urban compactness have been achieved.he sub-districts located inside the planned peripheral constel-ations experienced higher growth in population numbers andensity. The overall average density of the population, employ-ent and households increased, and mixed land use and the

obs–housing balance also improved. The results suggest that since990 to a large extent the containment strategies have performedell in Beijing.

However, this study also found three trends that are contraryo the municipal containment strategy objective of encourag-ng compact urban development. Firstly, the sub-districts locatedn the rural–urban fringe had dramatic unexpected populationrowth and housing development. Secondly, some sub-districts onhe edge of rural–urban fringe experienced obvious decreases inrban compactness, as indicated by residential density, employ-ent density, the concentration of land use and its pattern, and

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

he continuity of land development. Thirdly, mixed use and theobs–housing balance greatly decreased in the centre and someub-districts in the peripheral constellations.

These contrary trends indicate that urban sprawl has stillccurred on the urban fringe of Beijing, although relatively strict

e‘s‘a

PRESSxx (2010) xxx–xxx 11

rban containment strategies have been implemented at theunicipal level. There are many reasons for this. However, the

eep institutional context is the most important. In Western cities,prawl is characterized by low-density housing and commercialtrip development which is a result of and dependent upon exten-ive automobile use. Essentially, urban sprawl in Western cities isromoted by lifestyle changes (Squires, 2002). In China, as well asocial factors based on lifestyle changes, urban sprawl is stronglynfluenced by political decentralization and marketization. Firstly,s mentioned above, local fiscal responsibility has led local govern-ents to promote local industrialization as a means of enhancing

scal income. However, since this local industrial development isften dominated by scattered town-village enterprises in suburbanreas, fiscal reform has ultimately stimulated urban sprawl.

Secondly, the local authorities’ need for local revenue from landental and the competition for investment are the main factorsetermining dispersed development in suburban areas (Deng anduang, 2004). Since the 1980s, urban land reform, in which a landse right can be transferred as a component of the ‘price’, hasllowed local governments to obtain large fiscal benefits from landevelopment. For example, the state allows local governments toeep up to 60 percent of land revenues to encourage land use rightales. There are two kinds of fiscal benefit arising from land devel-pment. The first is the direct land development income from landental and revenue arising from land use right transfer. The sec-nd kind is indirect land development benefits. For example, theocal government takes their provision of land as a substitute forirect capital investment when they enter into partnerships withther foreign investors (Zhu, 1999). In some extreme situations,and can be provided free to investors by local government as apreferential policy’ to attract investment. As a result, urgent, highemand for land development, combined with severe competitionor local economic growth among local jurisdictions, has led to thecattered, even illegal, sprawling development around Beijing.

Thirdly, land development is a more fragmented process due tomperfect land market reform in China. Many studies have pointedut that a land market has emerged in China (Chen, 1998; Dowall,993). In this land marketization process, the market mechanismeplaces central planning controls and the land use right is sepa-ated from land ownership. Land users have been allowed to let,ransfer, rent and mortgage their land use right. However, in theurrent, more gradual transformation process, the land market inhina is not ‘perfect’, as it still allows unusual types of transfersKeng, 1996; Zhu, 2005). For example, Yeh and Wu (1996, p. 336)rgued that ‘because of the introduction of the paid transfer ofand use rights, a dual land use system composed of market-basedllocation and non-market administrative allocation has begun tomerge in China’. Some land can be bought on the market at thearket price through tender and auction, while other land is allo-

ated administratively without a price or at a price negotiatedetween developers and local government (Zhu, 2000). Therefore,aila (2007, p. 7) has argued that ‘a formalized market model doesot exist in China’. The fragmented mechanism of the land useight is one main reason for the loss of control of land marketsnd scattered land development in China (Ding, 2003).

In addition, land ownership also influences urban sprawl in Bei-ing. As mentioned above, there are three levels of administrative

anagement in Beijing. The first level is the municipal governmentevel, the second is the district or county level, and the third is thetreet or township level. There are also two different land own-

transforming China: Evidence from Beijing. Land Use Policy (2010),

rship systems and related land management systems. The termscounty’ and ‘township’ refer to aspects of the collective ownershipystem in the rural counties or towns, while the terms ‘district’ andstreet area’ refer to elements of state land ownership in the urbanreas. Most of the land development management at the municipal

ING

J

1 olicy x

lcvtsTlrHbaa

mgouhbmherl

gtcfwow1cstwjiuv1itfoermom

C

itasttcpm

ordCfiotortamgac

rnfemIiaoiemNo

csswarhrgaaidttssrsmis

R

A

B

ARTICLEModel

LUP-936; No. of Pages 14

2 P. Zhao / Land Use P

evel covers both collective and state ownership systems. In someases, collective land ownership is managed by individual suburbanillages on the urban fringe. Generally speaking, the implementa-ion of land development management in the urban districts andtreet areas is stricter than in the rural counties and townships.he relaxed land development management of collectively ownedand in suburban counties and townships is believed to be one maineason for urban sprawl on the urban fringe of Beijing (Deng anduang, 2004). In particular, rapid real estate development favouredy the growing housing market in the suburban areas tends toggravate urban sprawl, as indicated by the findings in the abovenalysis.

Fourthly, the increase in revenue-enhancing development is oneain reason for the decline in mixed land use. Against the back-

round of fiscal responsibility, local government tends to take anypportunity to gain as much local revenue as possible. In the actualrban development process, the developments which can bringigher local revenue are given priority, such as office buildings,usiness parks and high-quality apartment blocks, while develop-ents that have a low-revenue contribution, for example, social

ousing and public facilities, are rarely considered by local gov-rnment. Consequently, these micro-development activities haveesulted in a decrease in the degree of mixed land use at the regionalevel.

The three contrary trends mentioned above actually verify aeneralizable challenge confronting the implementation of con-ainment strategies, which has already been reported in developedountries. This is that a particular policy which might be goodor one municipality might not necessarily work for all. In someays, local growth management will assist in the achievement

f metropolitan and national environmental goals but in otherays it will detract from this mission (Carruthers, 2002; Chinitz,

990). In particular, when local economic benefits are the mainoncern of the local government, local decision-making on growth,trengthened by political decentralization, may create challengeso municipal containment policies. This challenge could be greaterhen market factors are involved. The empirical findings from Bei-

ing presented above show that the booming real estate marketn China is a major factor causing a decline in the effectiveness ofrban containment strategies. The findings are consistent with pre-ious findings reported in Western countries (Moore and Nelson,994). As mentioned in the section ‘Urban containment strategies

n China’, China’s big cities have been undergoing a rapid urbaniza-ion process for over 30 years (1978–2009). However, huge demandor housing deriving from the current rapid urbanization process isnly one reason for urban expansion and urban sprawl. Local gov-rnment need for local revenue from the real estate market oftenesults in a relaxation of growth control. As a result, the housingarket promoted by local government activities has become one

f the major factors challenging the effectiveness of urban contain-ent strategies.

onclusions

Urban containment strategies should play an important rolen managing land development towards greater sustainability inhe context of transformation and development in China, as theyre one of the best known land use policies for managing urbanprawl. However, the empirical evidence from this study suggests

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

hat although containment strategies generally perform well, inhe current transformation context they nevertheless face serioushallenges from local development fuelled by marketization andolitical decentralization. With respect to future policy develop-ent, closer attention should be paid to local solutions. The ability

B

B

PRESSxx (2010) xxx–xxx

f containment strategies to limit urban sprawl depends on theesponsible application of the regulations employed by local juris-ictions. In this respect, the municipal containment strategies inhina should examine the decentralization trend and local demand

or growth. The key issue for containment strategies at this stagen Beijing would not only be to make the controls over local devel-pment more strict, as is the case in developed countries, but alsoo emphasize the need for guidance at the local level so that devel-pment proceeds in a more sustainable way and uncoordinatedapid urban growth is avoided. Local benefits and the equity ofhese benefits, in particular economic benefits, should be taken intoccount by municipal governments. Furthermore, they should putore effort into harmonizing the relationship between municipal

rowth control and local growth, using ‘market’ strategies suchs fiscal arrangements and tax redistribution rather than mereommand-control.

At the same time, the management capacity of the cur-ent containment strategies should be enhanced to mitigate theegative effects of market-led development in the current trans-

ormation context. In relation to this context, it is important tostablish a clear mechanism for determining accountability as aeans of achieving the goal of increasing administrative capacity.

n the transformation context, urban development is character-zed by unpredictable situations because of the lack of claritybout the accountability of different sectors. The determinationf accountability would provide clear and stable rules concern-ng relative powers, functions and resources, and as a result wouldnhance the administrative capacity of different levels of govern-ent, enabling them to coordinate the various sectors – private,GOs and CBOs – to achieve the municipal containment strategybjectives.

In relation to future research directions, further assessment ofontainment strategies in developing and transforming countrieshould be undertaken. The empirical evaluation of containmenttrategies is attracting attention in land use research. However,ith the realization that containment policies are widely viewed

s important tools in curbing urban sprawl, at least two advancesemain to be made in the research. Firstly, few empirical studiesave been devoted to developing countries and as a result theesearch on containment strategies does not represent a generallobal view. The empirical evaluation of containment strategies haslready been widely investigated in North America and Europe,nd it is now necessary that this becomes the focus of debaten developing countries, given the emergence of sprawling urbanevelopment in these countries. Secondly, current research intohe verification of the effects of containment strategies limits itselfo the topics of housing development, environmental effects andocial effects. Little attention has been paid to the acceptability ofuch strategies to local jurisdictions, and there is often no explicitecognition of the institutional context. The complex institutionalituation in a developing and transformation country such as Chinaay provide a good opportunity to study the role and influence of

nstitutional factors in the implementation of urban containmenttrategies.

eferences

lterman, R., 1997. The challenge of farmland preservation: lessons from a six-nation comparison. Journal of the American Planning Association 63 (2),220–241.

atty, M., Besussi, E., Maat, K., Harts, J.J., 2003. Representing Multifunctional Cities:

transforming China: Evidence from Beijing. Land Use Policy (2010),

Density and Diversity in Space. University College, London.eijing Real Estate Association, various years. Beijing Real Estate Statistical Year-

book (in Chinese: Beijing shi fang di can tong ji nian jian). China Statistics Press,Beijing.

IUPD (Beijing Institute Urban Planning Design), 1991. Beijing City Master Plan(1991–2010). BIUPD (Beijing Institute Urban Planning and Design), Beijing.

ING

J

olicy x

B

B

B

B

B

B

B

B

B

C

C

C

C

C

C

C

C

C

C

C

C

D

D

D

D

D

F

F

F

F

G

H

H

H

H

H

H

H

J

J

J

K

K

KL

L

L

L

L

L

L

M

M

M

M

M

N

N

N

N

O

O

O

P

P

R

S

ARTICLEModel

LUP-936; No. of Pages 14

P. Zhao / Land Use P

IUPD (Beijing Institute Urban Planning Design), 1983. Beijing City ConstructionMaster Plan (1982–2000). BIUPD (Beijing Institute Urban Planning and Design),Beijing.

IUPD (Beijing Institute Urban Planning Design), 2004. Beijing City Master Plan(2004–2020). BIUPD (Beijing Institute Urban Planning and Design), Beijing.

LHB (Beijing National Land Resource Housing Management Bureau), 1990. BeijingLand Readjustment Data. Beijing National Land Resource and Housing Manage-ment Bureau, Beijing.

LHB (Beijing National Land Resource Housing Management Bureau), 2009. BeijingLand Readjustment Data. Beijing National Land Resource and Housing Manage-ment Bureau, Beijing.

MCHURD (Beijing Municipal Commission of Housing Urban-Rural Development),2009. Beijing Real Estate Statistical Yearbook (2009). Zhong guo ji liang Press,Beijing.

SB (Beijing Statistic Bureau), 1990. Beijing the Fourth (1990) Census. Beijing Statis-tic Bureau, Beijing.

SB (Beijing Statistic Bureau), 1996. Beijing First (1996) National Basic Unit Census.Beijing Statistic Bureau, Beijing.

SB (Beijing Statistic Bureau), 2010. Beijing the Second (2008) National EconomicCensus. Beijing Statistic Bureau, Beijing.

SB (Beijing Statistic Bureau), various years. Beijing Statistic Yearbook. China Sta-tistical Publishing House, Beijing.

arruthers, J.I., 2002. Evaluating the effectiveness of regulatory growth managementprograms. Journal of Planning Education and Research 21 (4), 391–405.

ervero, R., 1989. America’s Suburban Centers: The Land Use-transportation Link.Unwin-Hyman, Boston.

ervero, R., 1996. Jobs–housing balance revisited. Journal of the American PlanningAssociation 62 (4), 492–511.

han, K., Zhang, L., 1999. The hukou system and rural–urban migration in China:processes and changes. The China Quarterly 160, 818–855.

hen, A., 1998. China’s urban housing market development. Problems and prospects.Journal of Contemporary China 7 (17), 43–61.

hen, F., 2003. Market ‘black hole’ encroaching land revenue Governments andDevelopers encroaching land revenue. China Industrial and Commercial Times(in Chinese: zhong hua gong shang shi bao) (June).

hinitz, B., 1990. Growth management. Journal of the American Planning Association56 (1), 3–8.

how, G.C., 2007. China’s Economic Transformation, 2nd ed. Blackwell PublishingLtd., London.

ouch, C., Petschel-Held, G., Leontidou, L. (Eds.), 2008. Urban Sprawl in Europe:Landscape, Land-Use Change & Policy. Wiley-Blackwell, London.

PGPRC, 1997. National Land Use Plan (1997–2010) (in Chinese: quan guo tu di liyong zong ti gui hua). The Central People’s Government of the People’s Republicof China, Beijing (in Chinese: zhong hua ren min gong he guo zhong yang renmin zheng fu).

PGPRC, 1999. Land Management Act (Amendment) (in Chinese: tu di guan li fa (xiuding ban)). The Central People’s Government of the People’s Republic of China,Beijing (in Chinese: zhong hua ren min gong he guo zhong yang ren min zhengfu).

SB, various years. China Statistic Yearbook (in Chinese: zhong guo tong ji nian jian).China Statistic Bureau (in Chinese: zhong guo tong ji ju).

awkins, C.J., Nelson, A.C., 2002. Urban containment policies and housing prices: aninternational comparison with implications for future research. Land Use Policy19 (1), 1–12.

eng, F.F., Huang, Y., 2004. Uneven land reform and urban sprawl in China: the caseof Beijing. Progress in Planning 61 (3), 211–236.

ieleman, F.M., Dijst, M.J., Spit, T., 1999. Planning the compact city: the RandstadHolland experience. European Planning Studies 7 (5), 605–621.

ing, C., 2003. Land policy reform in China: assessment and prospects. Land UsePolicy 20 (2), 109–120.

owall, D., 1993. Establishing urban land markets in the Peoples Republic of China.Journal of the American Planning Association 59 (12), 182–192.

ang, K., Zhang, Y., 2003. Plan and market mismatch: urban redevelopment in Beijingduring a period of transition. Asia Pacific Viewpoint 44 (2), 149–162.

rank, L.D., 1994. An Analysis of Relationships Between Urban Form (Density, Mix,and Jobs: Housing Balance) and Travel Behavior (Mode Choice, Trip Generation,Trip Length, and Travel Time). Washington State Department of Transportation,Olympia, WA.

rank, L.D., Pivo, G., 1995. Impacts of mixed use and density on utilization of threemodes of travel: single-occupant vehicle, transit, and walking. TransportationResearch Record 1466, 44–52.

renkel, A., 2004. The potential effect of national growth-management policy onurban sprawl and the depletion of open spaces and farmland. Land Use Policy21 (4), 357–369.

alster, G., Hanson, R., Ratcliffe, M.R., Wolman, H., Coleman, S., Freihage, J., 2001.Wrestling sprawl to the ground: defining and measuring an elusive concept.Housing Policy Debate 12 (4), 681–717.

aila, A., 2007. The market as the new Emperor. International Journal of Urban andRegional Research 31 (1), 3–20.

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

all, P., Gracey, H., Drewett, R., Thomas, R., 1973. The Containment of Urban England,vol. II. Sage Publications, Inc., Beverly Hills.

ay, D., 1994. Economic Reform and State-Owned Enterprises in China, 1979–1987.Oxford University Press, Oxford.

u, X., 2000. The state, enterprises and society in post-Deng China. Asian Survey 40(4), 641–657.

S

S

S

PRESSxx (2010) xxx–xxx 13

uang, Y., 2004. The road to homeownership: a longitudinal analysis of tenure tran-sition in urban China (1949–1993). International Journal of Urban and RegionalResearch 28 (4), 774–795.

uang, Y., 2005. From work-unit compounds to gated communities: housinginequality and residential segregation in transitional Beijing. In: Ma, L., Wu, F.(Eds.), Restructuring the Chinese Cities: Changing Society, Economy and Space.Routledge, London/New York, pp. 192–221.

uang, Y., Clark, W.A.V., 2002. Housing tenure choice in transitional urban China: amultilevel analysis. Urban Studies 39 (1), 7–32.

ia, H., Le, Y., 2010. Exploring 200 billion revenue income of Beijing: increase ordeclines of land based financial situation? 21 Century Economic Report (in Chi-nese: 12 shi ji jing jig bao dao) (January).

ohnson, M.P., 2001. Environmental impacts of urban sprawl: a survey of the lit-erature and proposed research agenda. Environment and Planning A 33 (4),717–735.

un, Y., Zhou, J., 2007. The failure and success of greenbelt program in Beijing. UrbanForestry and Urban Greening 6 (4), 287–296.

einer, M., Koll-Schretzenmayr, M., Schmid, W.A., 2005. Managing Urban Futures:Sustainability and Urban Growth in Developing Countries. Ashgate, Aldershot.

eng, K.C.W., 1996. China’s land administration system. Journal of ContemporaryChina 5 (13), 325–346.

im, J., Choe, S.-C., 1997. Seoul: The Making of a Metropolis. Wiley, New York.i, M., 2007. Shift from attracting investments to selecting investments and adjust

of Beijing industrial development zones (in Chinese: you yin zi dao xuan zi chanbian, Beijing kai fa qu qing li men hu). The Economic Observer Newspaper (inChinese: jing ji guan cha bao).

i, S.-M., 2000. Housing consumption in urban China: a comparative study of Beijingand Guangzhou. Environment and Planning A 32 (6), 1115–1134.

i, S.-M., Li, L., 2006. Life course and housing tenure change in urban China: a studyof Guangzhou. Housing Studies 21 (5), 653–670.

i, S.-M., Tang, Y.H., 1998. Market segmentation and consumer characteristics ofcommodity housing in Guangzhou: an analysis of household survey data. AsianGeographer 17 (1–2), 68–69.

i, S.-M., Yi, Z., 2007. Financing home purchase in China, with special reference toGuangzhou. Housing Studies 22 (3), 409–425.

in, G.C.S., Ho, S.P.S., 2005. The state, land system, and land development processesin contemporary China. Annals of the Association of American Geographers 95(2), 411–436.

ongley, P., Batty, M., Shepherd, J., Sadler, G., 1992. Do green belts change the shapeof urban areas? A preliminary analysis of the settlements geography of southeastEngland. Regional Studies 26 (5), 437–452.

a, L.J.C., 2002. Urban transformation in China: 1949–2000: a review and researchagenda. Environment and Planning A 34 (9), 1545–1569.

a, L.J.C., 2004. Economic reforms, urban spatial restructuring, and planning inChina. Progress in Planning 61 (3), 237–260.

illward, H., 2006. Urban containment strategies: a case-study appraisal of plansand policies in Japanese, British, and Canadian cities. Land Use Policy 23 (4),473–485.

oore, T., Nelson, A.C., 1994. Lessons for effective urban-containment and resource-land preservation policy. Journal of Urban Planning and Development 120 (4),157–171.

unton, R., 1983. London’s Greenbelt: Containment in Practice. Allen and Unwin,London.

akai, N., 1988. Urbanization promotion and control in metropolitan Japan. PlanningPerspectives 3, 197–216.

aughton, B., 1995. Growing Out of the Plan: Chinese Economic Reform, 1978–1993.Cambridge University Press, Cambridge, MA.

elson, A.C., 1992. Preserving prime farmland in the face of urbanization: lessonsfrom Oregon. Journal of the American Planning Association 85 (4), 467–487.

elson, A.C., Moore, T., 1996. Assessing growth management policy implementa-tion: case study of the United States’ leading growth management state. LandUse Policy 13 (4), 241–259.

i, J.C., 1992. Fiscal reform and the economic foundations of local state corporatismin China. World Politics 45 (1), 99–126.

i, J.C., 1998. The evolution of local state corporatism. In: Walder, A.G. (Ed.), Zoupingin Transition: The Process of Reform in Rural North China. Harvard UniversityPress, Cambridge, MA, pp. 35–61.

i, J.C., 1999. Rural China Takes off: Institutional Foundations of Economic Reform.University of California Press, Berkeley, CA.

endall, R., 1999. Do land use controls cause sprawl? Environment and Planning B26 (4), 555–571.

eng, Z., 1997. The jobs–housing balance and urban commuting. Urban Studies 34(8), 1215–1235.

azin, E., 1998. Policies to control urban sprawl: planning regulations or changes inthe ‘rules of the game’? Urban Studies 35 (2), 321–340.

hen, J., 2005. Space, scale and state: reorganizing urban space in China. In: Ma,L.J.C., Wu, F. (Eds.), Restructuring the Chinese City: Changing Society, Economyand Space. Routledge, New York, pp. 39–58.

transforming China: Evidence from Beijing. Land Use Policy (2010),

ong, Y., Knaap, G.-J., 2004. Measuring the effects of mixed land uses on housingvalues. Regional Science and Urban Economics 34 (6), 663–680.

orensen, A., 1999. Land readjustment, urban planning and urban sprawl in theTokyo metropolitan area. Urban Studies 36 (13), 2333–2360.

quires, G.D. (Ed.), 2002. Urban Sprawl: Causes, Consequences, and PolicyResponses. The Urban Institute Press, Washington, DC.

ING

J

1 olicy x

S

T

U

W

W

W

W

W

W

W

W

W

W

W

W

W

W

X

Y

Y

Z

ZZ

Z

Zhu, J., 2000. Urban physical development in transition to market. The case of China

ARTICLEModel

LUP-936; No. of Pages 14

4 P. Zhao / Land Use P

tewart, C., 1996. Managing growth: a regional planning perspective. In: Filion, P.,Bunting, T., Curtis, K. (Eds.), The Dynamics of the Dispersed City: Geographic andPlanning Perspectives on Waterloo Region. Department of Geography, Univer-sity of Waterloo, Waterloo, pp. 113–136.

sai, Y.-H., 2005. Qualifying Urban form: compactness versus “sprawl”. Urban Stud-ies 42 (1), 141–161.

N-HABITAT, 2008. State of the World’s Cities 2008/2009—Harmonious Cities. UN-HABITAT, London.

alker, A., Li, L.H., 1994. Land use right reform and real estate market in China.Journal of Real Estate Literature 2 (2), 199–211.

ang, Y., Murie, A., 1996. The process of commercialisation of urban housing inChina. Urban Studies 33 (6), 971–989.

ang, Y., Murie, A., 1999. Commercial Housing Development in Urban China. UrbanStudies 36 (9), 1475–1494.

ang, Y., Wang, Y., Bramley, G., 2005. Chinese housing reform in state-ownedenterprises and its impacts on different social groups. Urban Studies 42 (10),1859–1878.

ank, D.L., 1999. Producing property rights: strategies, networks, and efficiency inurban China’s nonstate firms. In: Oi, J.C., Walder, A.G. (Eds.), Property Rights andEconomic Reform in China. Stanford University Press, Stanford, CA, pp. 248–274.

ei, Y.D., 2001. Dencentralization, marketization and globalization: the triple pro-cess underlying regional development in China. Asian Geographer 20 (1/2),7–23.

ei, Y.D., 2005. Planning Chinese cities: the limits of transitional institutions. UrbanGeography 26 (3), 200–221.

eiss, L., 1999. Managed openness: beyond neo-liberal global. New Left Review225, 126–140.

ong, S., Tang, B., 2005. Challenges to the sustainability of “development zones”: a

Please cite this article in press as: Zhao, P., Managing urban growth in adoi:10.1016/j.landusepol.2010.05.004

case study of Guangzhou Development District, China. Cities 22 (4), 303–316.u, F., 1996. Changes in the structure of public housing provision in urban China.

Urban Studies 33 (9), 1601–1627.u, F., 2001. China’s recent urban development in the process of land and hous-

ing marketisation and economic globalization. Habitat International 25 (3),273–289.

Z

Z

PRESSxx (2010) xxx–xxx

u, F., Yeh, A.G.O., 1999. Urban spatial structure in a transitional economy: thecase of Guangzhou, China. Journal of the American Planning Association 65 (4),377–394.

u, J., 1990a. Impact of industry decentralisation on Beijing’s rural and urban devel-opment (in Chinese). Beijing City Planning and Construction Review 1, 18–20.

u, J., 1990b. Impact of industry decentralisation on Beijing’s rural and urban devel-opment (in Chinese). Beijing City Planning and Construction Review 2, 20–22.

ie, Q., Parsa, A.R.G., Redding, B., 2002. The emergence of the urban land market inChina: evolution, structure, constraints and perspectives. Urban Studies 39 (8),1375–1398.

eh, A.G.O., 2005. Dual land market and internal spatial structure of Chinese cities.In: Ma, L.J.C., Wu, F. (Eds.), Restructuring the Chinese City: Changing Society,Economy and Space. Routledge, London, pp. 59–79.

eh, A.G.O., Wu, F., 1996. The new land development process and urban develop-ment in Chinese cities. International Journal of Urban and Regional Research 20(2), 330–353.

hang, T., 2000. Land market forces and government’s role in sprawl: the case ofChina. Cities 17 (2), 123–135.

hang, X., 1997. Urban land reform in China. Land Use Policy 14 (3), 187–199.hang, X., Chen, J., Tan, M., Sun, Y., 2007. Assessing the impact of urban sprawl on

soil resources of Nanjing city using satellite images and digital soil databases.CATENA 69 (1), 16–30.

hu, J., 1999. Local growth coalition: the context and implications of China’s gradu-alist urban land reforms. International Journal of Urban and Regional Research25 (3), 534–548.

transforming China: Evidence from Beijing. Land Use Policy (2010),

as a transitional economy. Urban Affairs Review 36 (2), 178–196.hu, J., 2004. From land use right to land development right: institutional change in

China’s urban development. Urban Studies 41 (7), 1249–1267.hu, J., 2005. A transitional institution for the emerging land market in urban China.

Urban Studies 42 (8), 1369–1390.