Habitat International 43 (2014) 274e282

Contents lists available at ScienceDirect

Habitat International
journal homepage: www.elsevier.com/locate/habitatint

Factors influencing the service lifespan of buildings: An improved

hedonic model
Guiwen Liu a, Kexi Xu a, c, Xiaoling Zhang b, *, Guomin Zhang c
a
Faculty of Construction Management and Real Estate, Chongqing University, Chongqing 400045, China
b
Department of Public Policy, College of Liberal Arts and Social Sciences, City University of Hong Kong, Hong Kong
c
School of Civil, Environmental and Chemical Engineering, RMIT University, Melbourne, VIC 3001, Australia

a r t i c l e i n f o a b s t r a c t

Article history: A large number of existing buildings were demolished indiscriminately during the urban renewal process
Available online 13 May 2014 in China, significantly shortening the average lifespan of buildings. This paper measures the average
service life of existing buildings based on the investigation of 1732 demolished buildings in seven
Keywords: communities of Jiangbei District in Chongqing from 2008 to 2010, and then explores the influencing
Building’s lifespan factors for a building’s lifespan with an improved Hedonic model. It is found that the average lifespan of
Sustainable
buildings is 34 years which is much shorter than the designed lifespan. The buildings with features such
Urban renewal
as small scale, near business centers, railway station, riverside and colleges, far away from the highway,
Influencing factors
Hedonic model
and with high speed development of economy and high level social investment in fixed assets are more
Case study likely to be demolished and hence generally short-lived. The external influencing factors are more
important than the internal influencing factors. And the internal factors, other than the floor area, are
less important than expected. At last, this paper concludes some suggestions for prolonging the build-
ing’s lifespan from the aspects of the demolition decision making system, maintenance and adaptive
reuse of the existing buildings.
Ó 2014 Elsevier Ltd. All rights reserved.

Introduction demolition in China annually constitutes about 40% of the total land
area acquired for construction. It is estimated that 4.2 million m2 of
With the implementation of the reform and opening-up policy housing have been demolished in the old district of Beijing from
in 1978, China has been experiencing rapid modernization and 1990 to 1998 (He & Wu, 2007). From 1995 to 2004, more than 745
urbanization process. By the end of 2011, China’s urbanization rate thousand households were relocated and over 33 million m2 of
had reached 51.3%, and nearly 60% of population will live in urban housing were demolished in Shanghai (see the report by SSB,
areas by 2020 (see the report by NPFPCC, 2012). The urbanization 2005). The scale of demolition in China is much larger than that
growth in conjunction with the pursuit of more comfortable living of the urban renewal period in US cities (He & Wu, 2007).
environment has resulted in great amount of new housing de- The demolition and reconstruction of existing buildings have
mands and therefore huge number of construction projects (Chang, always been considered as the phenomenon of urban renewal.
Ries, & Wang, 2013). The demands are commonly met through However, the large scale of demolition works is undertaken indis-
urban sprawl and urban renewal strategies. As urban sprawl has criminately in China, leading to the premature demise of existing
limits due to the strategy is at the expense of the loss of agricultural buildings. Qiu Baoxing, the Vice Minister of the Ministry of Housing
and ecological land, urban renewal strategy plays a major role in and Urban Construction of China commented that buildings in China
the process of urbanization (Shen, Yuan, & Kong, 2013). can only last for approximately 30 years on average which is much
China has been implementing huge plans of urban renewal in shorter than their designed service life Wang (2010). Furthermore, it
recent years. During the urban renewal process, massive demoli- is estimated by some Chinese experts that the lifespan of the urban
tion and reconstruction works have been undertaken. According to buildings constructed in the 1970s and 1980s is 30e40 years (Hu,
the report by Wang (2010), the land vacated from building Bergsdal, van der Voet, Huppes, & Müller, 2010). Buildings, as a
carrier of culture, could store the memory of city development. They
play an important role in urban cultural transmission and innova-
* Corresponding author. Tel.: þ852 34422402.
E-mail addresses: [email protected] (G. Liu), [email protected] (K. Xu), tion. The merit of architectural structures and its significant histor-
[email protected] (X. Zhang), [email protected] (G. Zhang). ical value will get lost after the demolition (He & Wu, 2007), and the

http://dx.doi.org/10.1016/j.habitatint.2014.04.009
0197-3975/Ó 2014 Elsevier Ltd. All rights reserved.
 G. Liu et al. / Habitat International 43 (2014) 274e282 275

physical coherence of neighborhoods could be destroyed afterward The physical condition of a building is an index to estimate the
(Zhai & Ng, 2013). Followed by these, the potential social contra- potential for adaptive reuse of the existing building (Langston,
dictions are accompanied by the massive demolition activities. It is Wong, Hui, & Shen, 2008). So the variables of building condition
inevitable that the old community will be torn down to rebuild a are factors influencing the decision making of building demolition
new one and the original residents have to be relocated after having in the urban renewal process, such as the construction quality, and
built a relatively stable social network (Zhang, Hu, Skitmore, & the aging of building structure and components. The poor condi-
Leung, 2014). In this context, social conflicts occurred (He & Wu, tions of existing buildings, including the inadequate architecture
2007; Yau & Chan, 2008). According to Shan and Yai (2011), land design, poor construction quality and unconformity to building
expropriation and dismantling of existing buildings are among the standards and regulations, caused by the lack of construction su-
major causes of social conflicts in China. The reduced building’s pervision and jerry-build are generally considered to be the major
lifespan also causes tremendous waste of resources and the aggra- internal factors for the short life of buildings in China (Shen et al.,
vation of environment pollution (Bullen & Love, 2010; Ellison, Sayce, 2013). Moreover, the number of stories and the total floor area of
& Smith, 2007; Thormark, 2002). Buildings are responsible for more building reflect the efficiency of land use (Weber, Doussard, Bhatta,
than 40% of global energy use, 12% of water consumption, one third & McGrath, 2006). Therefore, the low floor-area ratio buildings are
of global greenhouse gas emissions (Conejos, Langston, & Smith, more likely be demolished as it is one kind of “spatial fix” that
2013), and account for approximately 136 million tons of waste, prepares land for conversion to higher density and better use
nearly half of which is from demolition (Fu, Pan, Ma, & Li, 2013). The (Harvey, 1989; Smith, 1996).
waste produces a lot of hazardous substances and a bad effect on the The previous studies reveal that buildings are demolished not
water and soil surface due to the lack of waste-disposal technologies only because they are old or in a bad condition but also because
(Huang, Shi, Tanikawa, Fei, & Han, 2013). Moreover, demolition and proprietors or developers want to demolish them (Hzoka, 1988).
rebuilding will increase the life cycle cost of buildings from the The external factors are over-riding reason for the decision making
perspective of overall costs (Dong, Kennedy, & Pressnail, 2005). of building demolition and determinants of building’s service life-
Every year, new building construction in China amounts to 2 billion span (Liu, Xu, & Zhang, 2012). Braid (2001), Brueckner (1980) and
m2 and consumes approximately 40% of the world’s cement and Wheaton (1982) found that the changes in a city’s population
steel, and the construction waste accounts for 30e40% of the total density and housing preferences directly drive the building de-
amount of the urban waste (Gilbert, 2012). DeSimone and Popoff molition. As a city seeks to house more people, the ground rents
(1998) appreciated that one way of extending resource productiv- rise and landlords or developers replace small buildings with larger
ity is by prolonging the service life of products, and prolonging the ones to get more profit (Hufbauer & Severn, 1973). However, Harvey
existing building’s lifespan is one of the effective approaches to the (1989) and Smith (1996) indicated that the rent gap is the major
energy saving (Power, 2008). The volume of construction and de- reason for building demolition. Economists have found widespread
molition waste from future construction in China depends largely on evidence that the value differential accounts for renovation activity.
the lifespan of the existing buildings (Hu, 2010). As the building’s Using data on single-family home sales in Vancouver, Rosenthal
lifespan is considered as a contributor to sustainable urban devel- and Helsley (1994), the evidence for the hypothesis that demoli-
opment (Laure & Gerda, 2007; Lee, Lim, & Kim, 2013), the issue that tion takes place when the sale price of a property that is purchased
concerns about the lifespan of existing buildings has attracted specifically for renovation is equal to the price of vacant land (plus
widespread attention and intense debate in China. However, it ap- the cost of demolition). According to Ravetz (2008), it is shown that
pears that little effort was made to investigate the determinants of much of the reinvestment in the existing buildings depends on
the building’s lifespan through analyzing a large number of exem- perception of value and quality at the neighborhood scale, cultural
plar buildings in previous studies. heritage and landscape are two determinants for value. Moreover,
As an effort contributing to sustainable urban development, the some researchers have hypothesized answers to the determinants
focus of this paper is to measure the service lifespan of buildings in of demolition and renovation of urban existing buildings in their
China based on the investigation of a large number of demolished case studies of gentrified neighborhoods, and the common char-
buildings, and then explores the key factors influencing the service acteristics of these buildings are easily identifiable (Clay, 1979;
lifespan of the existing buildings with a case study in Jiangbei Helms, 2003; Johnson, Drew, Keisler, & Turcotte, 2012; Zahirovic-
District, Chongqing. The influencing factors are considered from Herbert & Chatterjee, 2011), which are presented in Table 1.
five aspects which include building characteristics, location char- In spite of these results, this paper plans to test different hy-
acteristics, neighborhood characteristics, economic variables and potheses put forth in the literature based on a case study of
politic variables based on an improved Hedonic model. The structures in Jiangbei District, Chongqing. Empirically, the effects of
outcome of this research is considered as the scientific foundation five aspects of influencing factors (most of which are measured at
for developing the effective and efficient strategies for prolonging
the building’s lifespan. Table 1
Key factors influencing the building’s lifespan in previous studies.
Literature review
Groups Aspects Factors

The reasons for demolishing buildings in the process of urban Internal Physical condition Construction quality (building structure)
factors Number of stories
renewal have been investigated in previous studies (Bender, 1979; Floor area
Golton, 1997; O’Connor, 2004). Golton (1997) indicated the rea- External Location condition Distance to CBD
sons for the demolition of existing buildings are complex. Thus, the factors Convenient to mass transit
factors influencing building’s lifespan are multiplex. According to Neighborhood Population density
characteristics Distance to disamenities (public housing
the survey conducted by Japanese researchers, 47% of short-lived
project, highway)
buildings are caused by aging of the building, 7% are caused by Distance to pleasant views (park, lake,
facilities update and others are caused by the social and economic university campus)
factors in Japan (Song, 2004). All of the influencing factors for the Economic variables Land value
building’s lifespan can be grouped into internal and external fac- Politic variables Urban planning
Political jurisdiction
tors, as shown in Table 1.
276 G. Liu et al. / Habitat International 43 (2014) 274e282

the parcel or block level), including the building characteristics, likely to be correlated (Dye & McMillen, 2007). The standard he-
location characteristics, neighborhood characteristics, economic donic price function for the sales of building is given by the
variables and political variables, are definitively established by following equations (Sirmans & Macpherson, 2005):
analyzing a set of micro-data that documents the demolition ac-
tivities among the buildings in Chongqing and the macro-data P ¼ f ðXi ; Ym ; Zn ; u1 Þ; (1)
which affects the demolition activities over the years from 2008
to 2010. where P represents the sales price of an existing building, Xi rep-
resents the building characteristic at number i, Ym represents
Methods location characteristic of building at number m, Zn represents the
neighborhood characteristic at number n, u1 represents other var-
Sample description iables that have an influence on the sales price and f (*) denotes the
idiographic function.
The survey was conducted in Jiangbei District, Chongqing. The exposition of this paper starts with this standard approach
Chongqing is considered more active than other cities, in terms of and then extends it. Let Y represent the lifespan of the existing
the absolute number of urban renewal projects and demolished building. And this paper puts forward an assumption that the
buildings. The urban renewal has been implemented in large scale building characteristics (X1i), location characteristics (X2i), neigh-
in Chongqing since 2001, and the urban renewal process was borhood characteristics (X3i), economic variables (X4i), political
further promoted in 2008 when the policy “Guideline to Speeding jurisdiction (X5i) are the determinants of building’s lifespan, i de-
up Renovation of Old and Dilapidated Buildings in Metropolitan notes the number of the variable in the context of each aspect and
Districts in Chongqing” was issued by Chongqing Municipal Gov- u2 represents other variables that have an influence on the service
ernment, with the plan of an area of 7,860,000 m2 to be renewed in life of building. The building’s lifespan equation is given by
three years. As a major urban renewal region in Chongqing, Jiangbei
Y ¼ f ðX1i ; X2i ; X3i ; X4i ; X5i ; u2 Þ; (2)
District has a territory of 221 km2 with twelve communities. In the
three years from 2008 to 2010, the floor area of demolished Different kinds of equations of the Hedonic model are available,
buildings in Jiangbei District is 1,150,000 m2, accounting for more and there is always a debate on the selection of the function
than 15% of the total amount of the demolition area in Chongqing. (Cropper, Deck, & McConnell, 1988). Follain and Malpezzi (1980)
The data about the basic structural characteristics of these demol- indicated that the Semi-log linear function has three advantages
ished buildings were collected with the assistance of the Municipal as a kind of equation for the Hedonic model. Firstly, this model
Bureau of Land Resources and Housing of Chongqing. And these takes into account the difference between the units of independent
characteristics include the year built, year demolished, buildings’ variables. Secondly, it is easy to explain the meanings of the co-
addresses, structures, functions, floor areas and the number of efficients of the model. Thirdly, this kind of equation helps to
stories. The raw data set includes 2127 demolished buildings. In the minimize the heteroscedasticity of the model. Therefore, the Semi-
data set, two types of observations were deleted. One type is the log linear equation is selected in this paper. Denoting the building’s
observations with missing or inconsistent values for important lifespan by Y, independent variables by Xni (1  n  5), the build-
variables. Another type is observations with addresses that could ing’s lifespan function is given by
not be geo-coded or consistent with the addresses in the demoli- X X X
tion data set. Thus, the final sample consists of 1732 buildings with Y ¼ B0 þ B1i ln X1i þ B2i ln X2i þ B3i ln X3i
specified addresses and fundamental information, which are pre- X X
þ B4i ln X4i þ B5i ln X5i þ E (3)
sented in Table 2. The sample buildings were located in seven
communities in Jiangbei District, as shown in Fig. 1.
where B0 denotes the constant term, Bni (1  n  5) denotes the
parameters to be estimated and E denotes random error term for
An improved hedonic model for building’s service life
the model.
Developers often look for old, small existing buildings on large
lots to replace with new buildings built to the limits of local Data and variable descriptions
building codes and zoning regulations. Well-maintained, high-
priced houses are less likely to be demolished than buildings whose The service lifespan of existing building is defined as the
poor condition has led to relatively low sales prices. Thus, factors dependent variable. The theoretical discussion implies that the
determining sales prices also influence the probability that a lifespan of existing building depends primarily on the utility pro-
building is demolished. And the lifespan of existing buildings de- vided by the physical condition and its proximity to amenities,
pends on the probability of demolition. Hedonic price function neighborhood identity, cultural value, political factors, and other
estimates will be subject to selection bias because the errors of the external influence factors. Due to the lack of heritage buildings and
sales price and the service life of demolished building equations are historical blocks in the sample area, the cultural variables are not
considered as the influencing factors of building’s lifespan in this
paper. As such, the explanatory variables can be grouped into five
Table 2
Demolished buildings by community. broad categories that might influence building’s service life, i.e.,
building characteristics, location characteristics, neighborhood
Community name Floor area (m2) Sample capacity Average lifespan (year)
characteristics, economic variables and political variables. These
Shima he 109,830.51 206 37 characteristics alter the demolition probability and hence indirectly
Dashi ba 249,510.50 250 36
decide the demolition year and the building’s lifespan. Using a
Guanyin qiao 261,191.12 196 27
Huaxin jie 29,247.75 61 31 geographic information system (GIS), the buildings’ addresses were
Wuli dian 82,289.05 255 21 geocoded, and several more variables were defined based on the
Cun tan 25,406.00 197 42 previous studies and local conditions. The means and standard
Tieshan pin 82,351.64 567 38 deviations of variables by taking the natural logarithm used in the
Total 839,826.57 1732 34
analysis are presented in Table 3.
 G. Liu et al. / Habitat International 43 (2014) 274e282 277

Fig. 1. The location of the sample buildings in seven communities in Jiangbei District, Chongqing.

Building characteristics space and the number of stories) and condition. Increased floor
These characteristics are the building’s inherent attributes, space or improved condition increases a building’s housing service
including the structure, number of stories and floor area, which are level (Smith, 1996). If the demolition decision is based on the gap
not affected by any external factors. The total “housing services” between the value of a property as presently constituted and its
provided by a building are a function of its size (measured by floor value after significant transformation, any existing physical trait
that makes the property less desirable will enlarge this gap and
make demolition more likely to happen. Thus, this paper assumes
Table 3 that the STRUCTURE (measured by the designed lifespan of different
Descriptive statistics of the dependent variables. kinds of structure, as shown in Table 4), FLOORAREA and STORIES are
Variables Full sample Expectation significant predictors of building’s lifespan and the coefficients are
(N ¼ 1732) expected to be positive. The data were provided by database of the
Mean Std. basic structural characteristics of samples.
dev.

Building
Location characteristics
STORIESdstories of building (X1) 1.824 1.446 þ Virtually every case study has shown that gentrified neighbor-
FLOORAREAdfloor area of a building (X2) 4.990 1.417 þ hoods are likely to be proximate to the city’s central business dis-
STRUCTUREdthe designed service life of 4.125 0.633 þ trict. Clay (1979) found that most are within two miles of the CBD,
structures in years (X3)
half (49%) within one mile, and more than one third (38%) within
Location
half a mile. Sites that are closer to downtown workplaces and city-
DISTCBDddistance to central business 7.653 1.170 þ center amenities are more attractive than outlying neighborhoods.
district in miles (X4) To account for the impact of this factor on building’s lifespan, the
DISTRAILWAYddistance to railway 8.780 0.346 þ distance in miles from the demolished building to nearest business
station in miles (X5)
center (Jiangbeizui and other business centers in Jiangbei District as
DISTSUBWAYddistance to nearest subway 8.002 0.818 þ
station in miles (X6) shown in Fig. 1) was measured. According to (Weber et al., 2006),
DISTHIGHWAYddistance to highway in 8.066 0.538 þ proximity to CBD should increase the probability of demolition.
miles (X7) Therefore, the coefficient of DISTCBD is expected to be positive.
Besides, the traffic condition (the distance from the existing
Neighborhood
building to the nearest public transportation station) also in-
DISTRIVERddistance to nearest riverside 6.094 0.792 þ
in miles (X8)
fluences the building’s lifespan (Weber et al., 2006). In this paper,
DISTPARKddistance to nearest park in 7.127 0.551 þ the distance to closest public transit stop (distance to railway sta-
miles (X9) tion in miles, distance to nearest subway station in miles and dis-
DISTCOLLEGEddistance to nearest college 7.587 1.802 þ tance to highway in miles) is measured. Then this paper assumes
in miles (X10)
that proximity to this infrastructure is more likely to increase the
ISTSCHOOLddistance to nearest key high- 7.721 0.836 þ
school in miles (X11) housing values and courage the renovation activity. Thus, the co-
DISTHOSPITALddistance to nearest hospital 6.921 0.814 þ efficients of DISTRAILWAY, DISTSUBWAY and DISTHIGHWAY are
in miles (X12)

Economic Table 4
LANDVALUEdthe assessed value of the land 5.338 0.402  Designed service life of different typed of structure according to the standard.
in Yuan (X13)
Structure type Designed lifespan Rate of depreciation
PCVAIdpercent change in value added of 1.449 1.436 
of structure (year) (per year)
industry 2008e2010 (X14)
PCARSdpercent change in annual retail 1.003 0.296  Reinforced concrete 90 0.0125e0.010
sales 2008e2010 (X15) Brick and concrete 75 0.020e0.0125
Brick and timber 47.5 0.040e0.014
Political Brick and stone 137.5 0.008e0.070
SIFAdthe social investment in fixed assets 12.73 0.331 
Source: standard for the service life of structure according to the Ministry of Housing
in 2008e2010 (X16)
and Urban Construction of China.
278 G. Liu et al. / Habitat International 43 (2014) 274e282

expected to be positive. The data about the distances were

measured by the geographic information system (GIS).

Neighborhood characteristics
Helms (2003) observed that neighborhood amenities are likely
to attract renovation and stimulate demolition activities. Natural
amenities, such as high elevations, attractive views, and proximity
to water, are characteristics of many gentrified neighborhoods. This
paper calculates the distance in miles to Riverside (Yangtze River or
Jialing River), which is an important amenity in Chongqing. The
Yangtze River and Jialing River are scenic and the riverside is highly
valued by the people in Chongqing, so buildings near the riverside
are particularly desirable. Therefore, the coefficient of DISTRIVER
(distance to nearest riverside in miles) is expected to be positive.
Man-made amenities that are attractive to renovators include
parks, landmarks, university campus and other atmospheric factors
that enhance a neighborhood’s character. Thus, the DISTPARK (dis-
tance to nearest park in miles), DISTCOLLEGE (distance to nearest
college in miles), DISTSCHOOL (distance to nearest prestigious high-
school in miles) and DISTHOSPITAL (distance to nearest hospital in
miles) are calculated in this paper. And all of these dependent
variables are expected to have a positive coefficient. The distances Fig. 2. The descriptive statistics of demolished building's lifespan.
were measured by the geographic information system (GIS). As the
population of sample area remained unchanged from 2008 to 2010, Results
the changing of population density is not taken into account as a
neighborhood characteristic in this paper though it is an important The average service lifespan of demolished buildings
factor affecting the building’s lifespan in previous studies.
Weber et al. (2006) indicated that it is possible that demolitions Fig. 2 presents the descriptive statistics of building’s lifespan.
are more likely in areas where the property values were initially The mean value is 34.36, the medium value is 34, the mode value is
low and where majority of buildings cannot reflect increasing 59, the standard deviation is 15.466, the skewness is 0.196 and the
ground rents. Therefore, it is expected that there is a negative kurtosis is 0.995. According to the eigenvalues cited above, the
relationship between the land value and the building’s lifespan. lifespan of demolished buildings exhibits a skewed distribution. In
Besides, PCVA (percent change in value added of industry) and order to the avoid the influence of a high degree of scatter of the
PCARS (percent change in annual retail sales) are opted as the samples and the maximum and minimum value, this study chooses
variables to reflect economy development level. The high speed of the median value, 34 years, as the average service lifespan of the
the economy development may increase the probability of demo- demolished buildings (Sheldon, 2009). As the Code for Design of
lition. Thus, this paper assumes that the coefficients of PCVA and Civil Buildings in China (GB 50068-2001) stipulates that the
PCARS are negative. These data were selected from the “Guidance of designed service lifespan of magnitude building and the main
the Land Value of Chongqing (2009)” and the “Statistical Yearbook structure of high-rise building is 100 years, and the designed ser-
of Jiangbei (2011)”. vice lifespan of general building is 50e100 years. By comparing
these standards of building service life to the lifespan of demol-
Political variables ished buildings affected in the process of urban renewal in
The urban policies and political jurisdiction have influence on Chongqing, it can be found that the lifespan of these demolished
the probability of demolition (Weber et al., 2006). As a variable of buildings is far shorter than the design standards.
the urban policy, the SIFA (the social investment in fixed assets) is
measured (Chen & Galbraith, 2011). In China, government officials Distribution of the demolished buildings by year built
wield a substantial amount of power over the decision-making of
urban planning and building demolition. Since the GDP has become As presented in Table 5, a majority of the demolished buildings
the only standard for the evaluation of government cadres’ per- were built between 1970 and 1989, which accounts for 45.55% of
formance and achievements in their official career in China, they the total amount of samples. Since the reform and opening-up
prefer to demolish the existing buildings and to build the new ones policy in 1970s, the high speed of economic growth drove devel-
rather than rehabilitate and reuse the old buildings. Therefore, it is opment of the construction industry. Thus, plenty of buildings and
expected that the SIFA has a negative influence on the building’s infrastructures were built during 1970s and 1980s. However, these
lifespan. The data were provided by the “Statistical Yearbook of
Jiangbei (2011)”. Besides, the precious studies adopted the influ-
Table 5
ence of the political preferences of aldermen by adding dummy Demolished buildings by year built.
variables that could identify the parcel’s ward, each of which has an
Built year Frequency Percent (%) Accumulative percent (%)
elected alderman (Weber et al., 2006). The sample in this paper fell
into seven communities with different sub-district offices. How- Before 1940 1 0.06 0.06
1940e1949 254 14.67 14.72
ever, the government of Jiangbei District plays a major role in the
1950e1959 127 7.33 22.06
decision-making of demolition and reconstruction rather than the 1960e1969 262 15.13 37.18
sub-district offices. In this regard, the dummy variable such as 1970e1979 353 20.38 57.56
factors reflecting the contexts of local community area are not 1980e1989 436 25.17 82.74
considered as an effective variable of the political jurisdiction in 1990e1999 261 15.07 97.81
After 2000 38 2.19 100.00
this paper.
 G. Liu et al. / Habitat International 43 (2014) 274e282 279

Table 6 lifespan of commercial building and office building are shorter than
Distribution of the building’s lifespan by functions. the average service life of all the demolished buildings.
Function of building Frequency Percent (%) Median of the lifespan
(year) Hedonic model based description of the factors influencing building
Residential building 1010 58.31% 35 service life
Commercial building 131 7.56% 32
Office building 62 3.58% 29 Stepwise regression analysis was performed to identify the de-
Industrial building 366 21.13% 30
Other buildings 163 9.41% 42
terminants for the dependent variable with the software of
SPSS18.0 based on the improved Hedonic model proposed previ-
ously in this paper. The stepwise regression analysis can eliminate
buildings are generally in poor physical condition, such as poor the insignificant independent variables. Also, it helps to ensure the
construction quality, inadequate architecture design and uncon- model is highly significant overall and each of the identified inde-
formity to building standards, caused by the backward construction pendent variables has a significant influence on the dependent
technologies, unenlightened architectural design concepts and variable (Cevik, 2007). Table 7 represents the results of regression
one-sided pursuit of materials saving and construction speed. analysis of the model for building’s lifespan. The adjusted R2 of this
Moreover, the buildings which were built after 1990s can meet the model is 0.277. Despite the relatively low goodness-of-fit measure,
need of modern life due to the improvement of the construction the model is highly significant overall. The p value of F-test (indi-
quality, building function and neighborhood circumstance. So these cated by italic in the table) which is less than 0.05 confirms that the
buildings which were built between 1970s and 1980s with func- probability is virtually zero that all of the coefficients are equal to
tional obsoleteness are the main objects of demolition in the urban zero. And all the VIF statistics in the regression analysis as reported
renewal process. below are less than 10 which are safely remote from threshold
values that would signal problem of multicollinearity (Fox, 1997).
According to Table 7, it is found that there are eleven de-
Distribution of the building’s lifespan by functions terminants for the building’s lifespan, including two internal fac-
tors and nine external factors. It suggests that both the internal and
The data in Table 6 present the distribution of lifespan of the external factors contribute to the building’s lifespan, and the
surveyed demolished buildings with different types of functions, external influencing factors are relatively more important than the
including residential building, commercial building, office building, internal factors to the building’s lifespan.
industrial building and other buildings (including schools, hospi-
tals, auditoriums, mess halls). The lifespan of residential building is Discussion
the closest to the average service life of all the demolished build-
ings. That’s because the amount of the residential building occupies What’s new by comparing with previous studies?
more than half of the samples. Furthermore, there is a big differ-
ence between the lifespan of different types of buildings. Especially, As expected, the coefficient of FLOORAREA is positive, which
the lifespan of other buildings is 13 years longer than the lifespan of suggests that the existing buildings with small floor area are more
office building. The reason for this phenomenon is that the com- likely short-lived. Since urban renewal in China is largely grounded
mercial land and official land are usually updated faster than other upon real estate development (He & Wu, 2007), many small-scale
kinds of land in order to get more economic benefits. So, both the buildings with poor design in structure and function are usually

Table 7
Estimation results of the model for building’s lifespan.

Model Un-std. coef. Std. coef. T Sig. Collinearity statistics

B Std. error Beta Tolerance VIF

(Constant) 81.654 29.452 2.772 0.006

Independent variables
Building
STORIES (X11) 1.861*** 0.298 0.174 6.238 0.000 0.574 1.742
FLOORAREA (X12) 0.694** 0.324 0.064 2.142 0.032 0.508 1.968
Location
DISTCBD (X21) 4.292*** 0.702 0.325 6.113 0.000 0.158 6.313
DISTRAILWAY (X22) 21.928*** 2.378 0.492 9.234 0.000 0.158 6.343
DISTSUBWAY (X23) 6.280*** 1.149 0.333 5.466 0.000 0.121 8.275
DISTHIGHWAY (X24) 2.040*** 0.718 0.071 2.842 0.005 0.718 1.393
Neighborhood
DISTRIVER (X31) 1.651*** 0.590 0.085 2.797 0.005 0.489 2.043
DISTCOLLEGE (X33) 1.877*** 0.280 0.219 6.700 0.000 0.420 2.384
Economic
PCVAI (X42) 2.182*** 0.427 0.203 5.109 0.000 0.284 3.523
PCARS (X43) 19.285*** 1.622 0.368 11.889 0.000 0.466 2.147
Political
SIFA (X51) 3.533*** 1.291 0.076 2.737 0.006 0.585 1.710
R2 0.232
Adjusted R2 0.227
F-test 47.214*** P ¼ 0.000 < 0.05

Dependent variable: lifespan of demolished building.

***Significance at the 1% level.
**Significance at the 5% level.
280 G. Liu et al. / Habitat International 43 (2014) 274e282

transformed into high floor-area ratio buildings with better phys- The DISTRIVER and DISTCOLLEGE are statistically significant at
ical conditions which are more popular in the real estate market. the neighborhood level, while the ISTSCHOOL, DISTHOSPITAL and
However, the coefficient of STORIES is negative which is contrary to DISTPARK are insignificant in the model. The results confirm the
expectation. This result implies that the variable of STORIES is hypothesis that the neighborhood amenities are significant as in-
insignificant in the model for this sign is counterintuitive. The ur- fluence factors for building’s lifespan. The neighborhoods near
ban existing buildings have experienced a process of construction water bodies which have good natural view are generally well-
from low stories to high stories, as shown in Table 8. So the developed, so buildings located in riverside are particularly desir-
demolished buildings with higher stories in the samples were able (Jim & Chen, 2007). As shown in Fig. 1, the Yangtze River and
usually built later leading to a short lifespan of these buildings Jialing River flow into Jiangbei District separate from Shima he and
compared to others. Moreover, the STRUCTURE is not the determi- Jiangbeicheng, and then convergence at the Jiangbeizui (CBD of
nant of the building’s lifespan as it is insignificant in the model, Jiangbei). The riverside regions in Jiangbei are not only the impor-
though it was expected to have a positive coefficient. tant areas of livable urban construction, but also the city image
The results consistently demonstrate that the location charac- display areas of Jiangbei. However, the majority of the existing
teristics DISTCBD, DISTRAILWAY, DISTSUBWAY, and DISTHIGHWAY are buildings which are located in the riverside regions in Jiangbei are
significant predictors of building’s lifespan. DISTCBD is an impor- industrial buildings and old residential buildings with poor condi-
tant determinant of the independent variable and has a positive tions, such as the Tianyuan Factory and Paper Mill. These existing
influence on the building’s lifespan. That’s because there is a strong buildings cannot match the needs of public recreation and pedes-
relationship between access to the CBD and land values (Alonso, trian accessibility which are required in the riverside regions. So the
1964). The prosperous and land value decrease monotonically buildings near the riverside are more likely to be demolished and
with distance from the CBD. So proximity to the CBD should in- replaced by new structures with better physical conditions and
crease the probability of demolition and upgrade of the existing higher value. Moreover, as a main evaluating indicator for the social,
buildings which the values are initially low to reflect increasing ecological and land benefits and the input amount of materialized
ground rents. Besides, the business centers in Jiangbei are usually labor, public service facilities enhance a neighborhood’s character
located in the city center. According to the urban housing filtering (Jim & Chen, 2007). A college as the a man-made amenity helps to
theory (Dong, 2001), along with the urban sprawl and urban ameliorate the neighborhood’s landscape and education condition,
renewal, the existing residential buildings located in the city center stimulate the increase of the housing price and land value, and then
are replaced by the new economic buildings and official buildings attract building regeneration and demolition activities. Therefore,
while the new residential buildings are built in the neighborhood the existing buildings near both the natural amenities and man-
which is far away from the city center. Thus, the buildings which are made amenities are more likely to be demolished and updated in
located near the CBD (city center) are more likely to be demolished order to match the neighborhood circumstance.
and short-lived. As expected, the coefficients of PCVAI and PCARS are negative.
In the model, the positive coefficient of DISTRAILWAY is the most The coefficient of LANDVALUE is also expected to negative, but
strongly significant of all, which supports finding of other empirical insignificant in the model. The results support the expectation that
studies that buildings near a large transport hub are more likely to the economic variables have negative influence on the building’s
be demolished and short-lived. The Jiangbei Longtoushi railway lifespan. Moreover, the coefficient of PCARS is the second strongly
station is an important distribution center of passengers and goods significant of all variables explored in the model and nearly ten
of Chongqing. The agglomeration effect of the neighborhoods times of the coefficient of PCVAI. There is a relationship between
around the station is improved (Martínez & Araya, 2000). So the this result and the changing of the industrial structure of Jiangbei
demolition of the existing buildings and reconstruction are more District as the result of high-speed development of economy in
likely to be implemented in these neighborhoods to get more living recently years. Transformation of industrial structure brings the
area and activity space. Though DISTSUBWAY and DISTHIGHWAY large-scale conversion of urban land and the functional decline of
were expected to have positive influence on the building’s lifespan, existing urban buildings. As the financial center and business
the coefficients are negative in the model. The results imply that center of Chongqing, the tertiary-industry has replaced the
the convenience value of living near a highway may be outweighed secondary-industry and becomes the leading industry of Jiangbei
by the accompanying noise, pollution, and traffic congestion District. Tertiary-industry concentrated in the old city while the
(Rietveld & Wagtendonk, 2004). The buildings near the highway industrial center shifts to the edge of the city due to the disad-
are less likely to be demolished and have long lifespan. Moreover, vantages of land competition (Guo, Jiang, Lu, Sun, & Wang, 2002).
the No. 3 subway line in Jiangbei District was operated in 2011 Thus, a large number of the vacant industrial buildings which still
while the demolition years of these buildings were from 2008 to have safe structure and historical value are demolished in order to
2010. It leads to the convenience value of the subway station cannot get more construction land and development space for the tertiary-
be well reflected among the samples. Thus, the traffic conditions industry.
have bidirectional effects on the lifespan of existing buildings. The negative coefficient of SIFA suggests that the urban policy is
an important influence factor for the building’s lifespan. The SIFA
has a positive influence on the neighborhood’s combined effect.
Table 8
The SIFA can improve the physical conditions of the neighborhood,
Distribution of the demolished buildings by stories.
including the infrastructure, public service facilities and environ-
Built year 1e3 Stories 4e6 Stories >6 Stories ment. Then the productivity of labor in this neighborhood is
Before 1940 0.06% 0.00% 0.00% improved. At the same time, it promotes the industrial agglomer-
1940e1949 16.11% 2.24% 0.00% ation of the neighborhood, and helps to take full advantage of the
1950e1959 7.51% 7.46% 0.00%
combined effect caused by the economy of scale (Guo et al., 2002).
1960e1969 15.98% 9.70% 0.00%
1970e1979 19.58% 33.58% 7.50% The changing of the distribution and improvement of the combined
1980e1989 23.88% 36.57% 37.50% effect causes the adjustment of the urban space and the structure of
1990e1999 14.44% 10.45% 55.00% land use. Thereby, the existing buildings in this neighborhood must
After 2000 2.44% 0.00% 0.00% be updated faster in order to match the need of the new and
Total 100.00% 100.00% 100.00%
modern urban function.
 G. Liu et al. / Habitat International 43 (2014) 274e282 281

Previous studies have shown that the political jurisdiction has China is 34 years which is much shorter than the planning lifespan.
a significant impact on teardown probabilities of existing buildings And it is suggested that the small scale buildings near business
(Weber et al., 2006). A It is therefore considered that the political centers, railway station, riverside and colleges, far away from the
jurisdiction is one of the key determinants of the building’s lifespan highway, with high speed development of economy and high level
in China. It is also found that local authority in China played a more of social investment in fixed assets are generally short-lived.
influential role in the urban renewal process than their Western However, the structure, stories, the neighborhood amenities of
counter-parts (Hin & Xin, 2011). Meanwhile, the local government park, high school and hospital, land value, and subway are insig-
seldom seeks for public opinions in the current practices of building nificant as determinants of building’s lifespan or less important
demolition in China. In fact, the participation activities of the cur- than expectation. Furthermore, the political variables are consid-
rent citizens involve ‘Informing’ and ‘Consultation of Tokenism’ ered as one of the key factors that affect building’s lifespan by
phases. In this context, most of them are lack of the capability to comparing the findings in this paper with previous studies. In line
express their willingness (Shan & Yai, 2011). Due to the non- with the findings, two strategies for prolonging building’s service
transparent decision-making system, the existing buildings are lifespan are proposed. It is necessary to build a scientific decision
usually demolished without professional evaluation, leading to the making system of building demolition that could help to avoid the
premature demise of these buildings. Besides, the exclusion of large scale demolition being undertaken indiscriminately. The
political variables is considered as an important reason for the low maintenance and adaptive reuse of the existing buildings is
value of R-square (0.227). considered as an efficient strategy to help achieve the sustainable
urban development in the long run.
Suggestions for prolonging the building’s lifespan Due to the limited access to the micro-data of demolished
buildings, this paper only covers the data from the past three years.
As there are different factors that affect the building’s lifespan Accordingly, the effects of some independent variables cannot be
which is much shorter than the designed lifespan, strategies should well reflected in the sample, such as the DITSUBWY. Besides, there is
be investigated not only based on only physical conditions but on a lack of the political independent variables in this paper as it is
different criteria to prolong the service lifespan of existing building difficult to assess the effect of the political factors. Furthermore,
in China. Referring to previous studies and the case study in this other important variables that significantly affect the building’s
paper, some strategies are suggested as follows: lifespan, such as the urban planning, are not included in this study
To build a scientific decision making system of building demolition. as the sample region is not big enough. Therefore, further research
It is necessary to develop a decision making index system which is needed to acquire more micro-data from different districts and
helps to estimate the existing building’s condition from the aspects cities. It is thus anticipated that more political relevant variables
of functional variables, environmental variables, cultural variables should be taken into account in the model in the planned research
and economic variables. The procedures of decision making on the design.
demolition should be more transparent as the public opinions were
rarely consulted in line with the current practices of building de- Acknowledgments
molition in China. Moreover, a feasibility study should be con-
ducted to include full consultation with affected residents before The work presented here is funded by the National Social
the launching of the demolition of existing buildings. It is suggested Science Foundation of China (grant number 12AZD064), National
that the interests of all groups in the affected communities should Natural Science Foundation of China (Project No. 71303203,
be taken into account. Project No. 7200376), College Research Fund at City University of
To encourage the maintenance and the adaptive reuse of the Hong Kong (Project No: 9610282) and Fundamental Research
existing buildings; Whether to demolish or refurbish an existing Funds for the Central Universities (grant number CDJKXB12002).
building is often determined during the urban renewal process. The authors are grateful for the support of NSSFC and FRFCU. And
Rebuilding a new higher-valued structure on the vacated land from the authors also thanks for Dr. Carl M. Egan’s assistance in editing
the demolition of existing building in an area can quickly increase this paper.
the value of a property. However, it is unsustainable for the urban
development as it ignores the residual utility and value of the
existing buildings. Adequate maintenance can reduce the extent of References
dilapidation of building while regularly maintenance can contribute
Alonso, W. (1964). Location and land use: Toward a general theory of land rent.
to good performance of building. Also, adaptive reuse is considered Cambridge, MA: Harvard University Press.
as another effective strategy for prolonging the service lifespan of Bender, B. (1979). The determinants of housing demolition and abandonment.
existing buildings. This strategy generally involves a process to Southern Economic Journal, 31, 131e144.
Braid, R. M. (2001). Spatial growth and redevelopment with perfect foresight and
ameliorate the financial, environmental and social performance of durable housing. Journal of Urban Economics, 49, 425e452.
buildings (Bullen & Love, 2010; Langston et al., 2008). It has been Brueckner, J. (1980). Residential succession and land-use dynamics in a vintage
attempted sporadically in some urban renewal practices in China, model of urban housing. Regional Science and Urban Economics, 10, 225e
240.
such as the Kuan & Zhai Alley (Kuanzhai xiangzi) in Chengdu and the Bullen, P. A., & Love, P. E. D. (2010). The rhetoric of adaptive reuse or reality of
Nanluogu xiang in Beijing. And all these successful cases can be used demolition: views from the field. Cities, 27, 215e224.
for reference to the adaptive reuse of existing buildings during the Cevik, A. (2007). Unified formulation for web crippling strength of cold-formed
steel sheeting using stepwise regression. Journal of Constructional Steel
urban renewal process. Research, 63, 1305e1316.
Chang, Y., Ries, R. J., & Wang, Y. (2013). Life-cycle energy of residential buildings in
Conclusion China. Energy Policy, 62, 656e664.
Chen, J., & Galbraith, J. K. (2011). Institutional structures and policies in an envi-
ronment of increasingly scarce and expensive resources: a fixed cost perspec-
By now, the existing studies on the influencing factors of the tive. Journal of Economic Issues, 45, 301e308.
service life of existing buildings in China has yielded inconclusive Clay, P. L. (1979). Neighborhood renewal: Middle-class resettlement and incumbent
results, this paper fills the gap in the literature. Moreover, the detail upgrading in American neighborhoods. Lexington, MA: Lexington Books.
Conejos, S., Langston, C., & Smith, J. (2013). AdaptSTAR model: a climate-friendly
of the micro and macro-data used in these estimations are evi- strategy to promote built environment sustainability. Habitat International, 37,
denced from the results. The average service life of the building in 95e103.
282 G. Liu et al. / Habitat International 43 (2014) 274e282

Cropper, M. L., Deck, L. B., & McConnell, K. E. (1988). On the choice of functional Laure, I., & Gerda, K. (2007). Comparing environmental impacts of renovated
form for hedonic price functions. Review of Economics and Statistics, 70, 668e housing stock with new construction. Building Research and Information, 35.
675. Lee, Y., Lim, S., & Kim, G. (2013). Improvement characteristics shown in holistic
DeSimone, L. D., & Popoff, F. P. (1998). Eco-efficiency: The business link to sustainable regeneration of Ballymun toward sustainable community. Building and Envi-
development. MIT Press. ronment, 59, 654e661.
Dong, B., Kennedy, C., & Pressnail, K. (2005). Comparing life cycle implications of Liu, G. W., Xu, K. X., & Zhang, M. L. (2012). A study on the life-span of demolished
building retrofit and replacement options. Canadian Journal of Civil Engineering, buildings: based on the investigation of demolished buildings in Chongqing.
32, 1051e1063. Urban Studies, 19, 109e112.
Dong, X. (2001). The analysis on the location and impact factors of urban housing. Martínez, F., & Araya, C. (2000). Transport and land-use benefits under location
City Planning Review, 25, 33e39. externalities. Environment and Planning A, 32, 1611e1624.
Dye, R. F., & McMillen, D. P. (2007). Teardowns and land values in the Chicago NPFPCC. (2012). Report on China’s migrant population development 2012. Beijing:
metropolitan area. Journal of Urban Economics, 61, 45e63. China Population Publishing House.
Ellison, L., Sayce, S., & Smith, J. (2007). Socially responsible property investment: O’Connor, J. (2004). Survey on actual service lives for North American buildings. In
quantifying the relationship between sustainability and investment property Woodframe housing durability and disaster issues conference (pp. 1e9). Las Vegas.
worth. Journal of Property Research, 24, 191e219. Power, A. (2008). Does demolition or refurbishment of old and inefficient homes
Follain, J. R., & Malpezzi, S. (1980). Dissecting housing value and rent: Estimates of help to increase our environmental, social and economic viability. Energy Policy,
hedonic indexes for thirty-nine large SMSAs. Washington, DC: The Urban 36, 4487e4501.
Institute. Ravetz, J. (2008). State of the stockdwhat do we know about existing buildings and
Fox, J. (1997). Applied regression analysis, linear models and related models. Thousand their future prospects? Energy Policy, 36, 4462e4470.
Oaks, CA: SAGE Publications Inc. Rietveld, P., & Wagtendonk, A. J. (2004). The location of new residential areas and
Fu, F., Pan, L., Ma, L., & Li, Z. (2013). A simplified method to estimate the energy- the preservation of open space: experiences in the Netherlands. Environment
saving potentials of frequent construction and demolition process in China. and Planning A, 36, 2047e2063.
Energy, 49, 316e322. Rosenthal, S., & Helsley, R. (1994). Redevelopment and the urban land price
Gilbert, V. K. (2012). Toxic capitalism: The orgy of consumerism and waste: are we the gradient. Urban Economics, 35, 182e200.
last generation on earth? Authorhouse. Shan, C., & Yai, T. (2011). Public involvement requirements for infrastructure
Golton, B. L. (1997). Building obsolescence and the sustainability agenda. In CIB planning in China. Habitat International, 35, 158e166.
(Conseil International du Bâtiment) second international conference (pp. 87e94). Sheldon, M. R. (2009). Introduction to probability and statistics for engineers and
Paris, France. scientists. Academic Press Inc.
Guo, H. M., Jiang, M. Q., Lu, J., Sun, J., & Wang, J. (2002). Urban spatial economics. Shen, L. Y., Yuan, H. P., & Kong, X. F. (2013). Paradoxical phenomenon in urban
Beijing: Economic Science Press. renewal practices: promotion of sustainable construction versus buildings’
Harvey, D. (1989). The urban experience. Baltimore: Johns Hopkins University Press. short lifespan. International Journal of Strategic Property Management, 17, 377e
He, S., & Wu, F. (2007). Socio-spatial impacts of property-led redevelopment on 389.
China’s urban neighbourhoods. Cities, 24, 194e208. Sirmans, G. S., & Macpherson, D. A. (2005). The composition of hedonic pricing
Helms, A. C. (2003). Understanding gentrification: an empirical analysis of the models. Journal of Real Estate Literature, 13, 3e41.
determinants of urban housing renovation. Journal of Urban Economics, 54, 474e Smith, N. (1996). The new urban frontier: Gentrification and the revanchist city. New
498. York: Routledge.
Hin, L. L., & Xin, L. (2011). Redevelopment of urban villages in Shenzhen, China e an Song, C. H. (2004). Whole life and highgrade quality-stick to the idea of giving first
analysis of power relations and urban coalitions. Habitat International, 35, 426e consideration for the people and implement housing performance certification.
434. Urban Development, 9, 4e8.
Hu, M. M. (2010). Dynamic material flow analysis to support sustainable built envi- SSB. (2005). Shanghai statistical yearbook 2005. Beijing: China Statistical Publishing
ronment development, with case studies on Chinese housing stock dynamics. House.
Netherlands: Department of Industrial Ecology, Institute of Environmental Thormark, C. (2002). A low energy building in a life cycle e its embodied energy,
Sciences (CML), Faculty of Science, Leiden University. energy need for operation and recycling potential. Building and Environment, 37,
Hu, M. M., Bergsdal, H., van der Voet, E., Huppes, G., & Müller, D. B. (2010). Dynamics 429e435.
of urban and rural housing stocks in China. Building Research & Information, 38, Wang, Q. (2010). Short-lived buildings create huge waste. http://www.chinadaily.
301e317. com.cn/china/2010-04/06/content_9687545.htm Accessed 06.04.10.
Huang, T., Shi, F., Tanikawa, H., Fei, J., & Han, J. (2013). Materials demand and Weber, R., Doussard, M., Bhatta, S. D., & McGrath, D. (2006). Tearing the city down:
environmental impact of buildings construction and demolition in China based understanding demolition activity in gentrifying neighborhoods. Journal of
on dynamic material flow analysis. Resources, Conservation and Recycling, 72, Urban Affairs, 28, 19e41.
91e101. Wheaton, W. (1982). Urban residential growth under perfect foresight. Urban
Hufbauer, G. C., & Severn, B. W. (1973). The economic demolition of old buildings. Economics, 12, 1e21.
Urban Studies, 11, 349e351. Yau, Y., & Chan, H. L. (2008). To rehabilitate or redevelop? A study of the decision
Hzoka, H. (1988). A statistical study on life time of bridges. Japan Society of Civil criteria for urban regeneration projects. Journal of Place Management and
Engineers, 5, 51e60. Development, 1, 272e291.
Jim, C. Y., & Chen, W. Y. (2007). Consumption preferences and environmental ex- Zahirovic-Herbert, V., & Chatterjee, S. (2011). Historic preservation and residential
ternalities: a hedonic analysis of the housing market in Guangzhou. Geoforum, property values: evidence from quantile regression. Urban Studies, 49, 369e
38, 414e431. 382.
Johnson, M. P., Drew, R. B., Keisler, J. M., & Turcotte, D. A. (2012). What is a strategic Zhai, B., & Ng, M. K. (2013). Urban regeneration and social capital in China: a case
acquisition? Decision modeling in support of foreclosed housing redevelop- study of the Drum Tower Muslim District in Xi’an. Cities, 35, 14e25.
ment. Socio-Economic Planning Sciences, 46, 194e204. Zhang, X., Hu, J., Skitmore, M, & Leung, B. Y. P. (2014). Inner-city urban redevel-
Langston, C., Wong, F., Hui, E., & Shen, L. Y. (2008). Strategic assessment of building opment in China metropolises and the emergence of gentrification: the case of
adaptive reuse opportunities in Hong Kong. Building and Environment, 43, Yuexiu, Guangzhou. Journal of Urban Planning and Development, 05014004.
1709e1718. http://dx.doi.org/10.1061/(ASCE)UP.1943-5444.0000169.