An evaluation model for urban carrying capacity: A case study of China's mega-cities

Habitat International 53 (2016) 87e96

Contents lists available at ScienceDirect

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

An evaluation model for urban carrying capacity: A case study of China's mega-cities Yigang Wei a, b, Cui Huang b, *, Jing Li c, Lingling Xie d a

School of Economics and Management, Beihang University, Beijing, China School of Public Policy and Management, Tsinghua University, Beijing, China c Department of Public Policy, City University of Hong Kong, Hong Kong, China d Institute of Management Science and Engineering, Guangxi University of Finance and Economics, Nanning, China b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 April 2015 Received in revised form 27 October 2015 Accepted 28 October 2015 Available online xxx

China experienced unprecedented urbanization development in the last two decades. During the rapid urbanization, cities have been attracting large population inflows from rural areas, and concentrating a wide range of social and economic activities. However, an over-concentration of population and human activities has lead to severe and diverse challenges for sustainable urban development, such as environmental degradation, poor infrastructure, and inadequate public services etc. Against this backdrop, concepts within urban carrying capacity (UCC) have received growing attention. It provides local government and urban planners key conceptual underpinnings to improve urban sustainability. However, there remain huge ambiguities in its definitions, implications, particularly measurable indicators, and analytic procedures. These deficiencies significantly hamper the effective implications of UCC concepts in routine urban management. Using the mean variance analysis method, this paper aims to establish an integrated UCC analytic framework to improve decision-making on sustainable urban land use and development. 30 representative indicators drawn from literature are selected to systematically evaluate the UCC conditions. 30 provincial capital cities and municipalities in China are selected as data sample. The results reveal several important findings. First, there exists a positive link between the city scale and UCC. Second, this exists a geographical pattern that costal cities have a high UCC than the central and western regions. Third, infrastructural and environmental factors are of salient weights in evaluating the UCC. Through the broad validations in China's mega-cities, this system has demonstrated capabilities of simplifying, appropriately quantifying, and evaluating the complex process of urban planning and management towards sustainability. © 2015 Published by Elsevier Ltd.

Keywords: Urban carrying capacity China's mega-cities Sustainability Urban planning and management Urban land use

1. Introduction Urbanization has been an important feature in the process of human development all throughout history. This trend is often associated with a sweeping population migrating from the countryside to the cities (McKinsey Global Institute, 2011). Onishi (1994) summarized three features of a city that can attract a large population in a densely developed area. First is the centrality of public administration and private decision-making. For example, the centrality of decisions in peripheral regions significantly reduces the communication costs. Second is security for urban residents'

* Corresponding author. E-mail address: [email protected] (C. Huang). http://dx.doi.org/10.1016/j.habitatint.2015.10.025 0197-3975/© 2015 Published by Elsevier Ltd.

daily livings and commercial opportunities. And the third is higher efficiency, due to a benefit of relatively easier cooperation and concentration of various factors of production. From a resident's perspective, these incentivizing features guarantee clear advantages for living and doing business, relative to rural or suburban areas. Therefore, attracted by the richer economic opportunities that cities can provide, people migrate from the rural areas to cities in search of better lives. Particularly in the last two centuries, cities with fast advancements in economy, technology, and transport, have contributed to unparalleled affluence and far better lives than the rural areas of many countries. Nowadays, the urbanization process has been increasing across the world. According to data from the United Nations, a new city with 1.3 million inhabitants will be built every week for the next four decades (Bentham, 2014). Meanwhile, rapid urbanization forms an important impetus for

88

Y. Wei et al. / Habitat International 53 (2016) 87e96

economic growth (Li & Yao, 2009). Therefore, urban areas are of significant importance for the society since a large population and substantial social and economic activities are concentrating there. China is currently at the stage of rapid urbanization. Heikkila and Xu (2013) have systematically analyzed the history, incentives, and means by Chinese government to promote the development of urbanization. They argued that the Chinese government holds a “pro-urbanization stance”, i.e., the government is proactively guiding and controlling this unprecedented urbanization process. Promoting urbanization is not an end in and of itself, but to serve the strategic goals of the government. Urbanization is a centrally important step in China's reform and opening up and socioeconomic development plans because the government treats it as a strategy for driving economic growth (Heikkila & Xu, 2013). China's present urbanization rate remains low and is not compatible with its per capita income level (World Bank, 2014). The Chinese government has thus committed to significantly promoting the urbanization in the next two to four decades. World Bank predicts that China's urbanization rate will increase consistently from the current 50%e70% by 2030 (World Bank, 2014). Backed by strong government willpower, China's urbanization has been encouraged to grow on a fast and unprecedented scale. In the past 35 years, China's urbanization increased rapidly from less than 20% in 1978 to 52% in 2012, much faster than that of the U.S. and U.K., although slightly slower than the rates of Japan and South Korea from the same development phrases (World Bank, 2014). Fig. 1 compares the urbanization process in China and U.S. The incremental population in China's urban areas will reach 425.53 million from 2000 to 2030, compared with 93.13 million new urban residents in U.S. in the same period, meaning that China's new city dwellers will far exceed the total U.S. population. With the fast-paced urbanization process, continuous congregation of larger population, urban services, production, consumption, and social wealth have been occurring in most cities around the world. However, these factors have made cities vulnerable in terms of achieving sustainable development and providing comfortable living standards for urban inhabitants (Chen, Tao, & Zhang, 2009). A host of urban symptoms induced by excessive population inflows and overdevelopment of the urban areas have been emerging and growing more severe (Abernethy, 2001; Oh, Jeong, Lee, Lee, & Choi, 2005). Due to the worsening living environments in urban areas, particularly in mega-cities, concerns related to the urban carrying capacity (UCC) concept have often been voiced when debating whether the current rate of urban development has exceeded inherent limit of the city (Wei, Huang, Lam, & Yuan, 2015). The issue of overladen urban carrying capacity has become a widespread challenge, despite the immensity and variety of global cities (Oh et al., 2005; Onishi, 1994). Currently, China has 288 cities categorized at the prefectural

levels or above. According to the CEIC database in 2013, there have been 31 cities with a population of 2e4 million, and 14 cities with a population over 4 million. According to the Green Book of Small and Medium-sized Cities released in 2010, cities with a population of 3e10 million are defined as mega-city in China. Since the megacities have been the highest concentrated areas of people and human activities, resources, and environmental pollution (Liu, 2012), they are thus more prone to the issues of overloaded UCC than small and medium cities. Thus, the mega-city is especially subjective to the occurrence of various “urban diseases”, reflected in a degrading environment, poor infrastructure, and insufficient public services, etc. Hence, to develop a reliable UCC evaluation model is of strategic importance to China's sustainable development. The government has understood the importance of promoting urban sustainability as a priority policy objective. The phrase “urban carrying capacity improvement”, which has been permeating official documents and regulations, has been fully institutionalized in national development planning and policies (see Table 1). Sustainable urban development may be defined as “a process of synergetic integration and co-evolution among the great subsystems making up a city (economic, social, physical and environmental), which guarantees the local population a non-decreasing level of wellbeing in the long term, without compromising the possibilities of development of surrounding areas and contributing by this towards reducing the harmful effects of development on the bio-sphere” (Camagni, 1998, p.4). Progressing sustainability is essential responsibility for urban planning and development. The UCC concept provides a useful theoretical foundation and methodological base for guiding sustainable urban development. According to the UCC concept, there is a certain inherent limit on a given urban area, beyond which will lead to irrecoverable changes, degradation or damages to the environment (Liu & Borthwick, 2011). Therefore, a UCC assessment can provide an indication on the maximum potential population, and also serve as an important guide to the service load of the region, which should be maintained above a specified/minimal/acceptable standard (Summers, 2004). UCC has become a popular term in the field of urban planning and management, environmental, and social studies. However, there remain huge ambiguities on its definitions, implications, and particularly, its measurable indicators and evaluative methods. The elusiveness surrounding UCC concept is mainly attributed to the integrative elements and properties associated with urban development. These problems become inhibitors for the effective implications of UCC concept in routine urban management and planning. This study aims to develop an effective UCC evaluation framework to fill the gap of previous studies. The evaluation model can systematically assess the present state of UCC and identify its deficient factors. The applicability of the model is then widely demonstrated in China's 30 mega-cites. This research is of

Fig. 1. Comparisons on urban-rural population in China and the U.S. Source: http://www.unhabitat.org/stats/Default.aspx (Accessed on 14 November 2013).

Y. Wei et al. / Habitat International 53 (2016) 87e96

89

Table 1a Milestones of UCC improvement in national policy and regulations. Year Policy and regulations

Contents

2010 The 12th Five-year Plan

The 12th Five-year Plan explicates the requirements for improving the comprehensive UCC, i.e. increasing population density, avoiding excessive urban sprawls, optimizing land use structures, remedying various “urban diseases”, improving the urban service and infrastructure, strengthening city management, promoting ecological and humanistic environment, etc. ” 2006 The 11th Five-year Plan The 11th Five-year Plan raised the detailed requirements for China's urbanization i.e. “to promote urban comprehensive carrying capacity”. The plan addresses that the scale and layout of urban development should be scientifically designed, consistent to natural carrying capacity (such as water and land resources, environmental endowments, geological conditions), economic development, employment potentials, and urban services and infrastructure.” 2005 Notifications on strengthening formulation, examining, In January 2005, the Ministry of Construction require the local government to improve the comprehensive and approval of urban master plan. UCC, by addressing the main tasks including resource conservation, ecological construction, and key infrastructure projects.

Table 1b Criteria for indicator selection. Items

Contents

i. ii. iii. iv. v. vi. vii. viii.

“Scientific accuracy, operability, hierarchy, completeness and dynamic” (Yu & Mao, 2002, p.181). Representative and sensitive to the state of present conditions. Direct link between human's impacts and their causing activities and events (Button, 2002). Rich policy implications for forecasting the trends of changes. Offering a meaningful ground for testing the relevant theories. Avoiding the repetitive information due to the inclusion of too many indicators. Reliably measurable and quantifiable (Button, 2002; Graymore et al., 2010). Ensuing the uniformity and consistency of indicators across different city prototypes (Button, 2002).

important theoretical and practical implications.

3. Relationships between UCC and urban sustainability

2. Development of the carrying capacity concept

There has been no consensus on the definitions of sustainability or sustainable development. The World Commission on Environment and Development defined sustainable development as “meeting the needs of current generations without compromising the ability of future generations to meet their own needs” (World Commission on Environment Development, 1987, p.8). Achieving sustainable urban development is an ultimate goal that planners, city managers, and residents seek. Thus, the task for addressing the present needs without compromising regenerative capacity to meet the demand raises a great challenge for planners and city managers (World Commission on Environment Development, 1987). Integrating sustainability concepts in the realms of environmental, social and economic concerns have played a centrally important role in the formulation of urban management decision and policies (Button, 2002). Organizations such as the Organization for Economic Cooperation and Development (OECD), United Nations Commission on Sustainable Development (UNCSD), the World Trade Organization (WTO) have proposed various indicators for assessing the status of sustainable development (Oh et al., 2005). These indicators, as suggested by various organizations, primarily focus on the natural environment such as air, water, forest, and biodiversity and have yet adequately considered other more comprehensive factors (Oh et al., 2005). In general, the concept of sustainability often leads to substantial confusions, and a need to understand how such a growth limit can be defined and identified. The concept of carrying capacity encompasses sustainability. “Sustainability is a necessary and sufficient condition for a population to be at or below carrying capacity” (Daily & Ehrlich, 1996, p992). For achieving sustainability, carrying capacity assessment is an important yardstick to gauge the level and state of urban sustainability (Sarma et al., 2012), and thus to better guide urban development. The carrying capacity concept also provides valuable evaluation methods and measurable indicators for assessing

Carrying capacity is conventionally expressed as the number of individuals that a standard area of land can support over a long period of time, in ecological studies. According to the logistic growth model, animal population growth can be constrained to an upper asymptote, i.e., the carrying capability. It suggests that there exists a finite carrying capability for a given species, either the optimum or maximum, and being close to this optimum level of population density is secure for their future survival and proliferation (Campbell, 1998). An excessive population growth exceeding this limit leads to dramatic negative impacts, manifested as overcrowding and shortages of foods (Campbell, 1998). Therefore, the carrying capacity concept provides guidance for a sustainable size in population relative to the supporting ecosystem. Likewise, there may exist a maximum or optimum level of population size for the human society (Campbell, 1998). Thomas Malthus firstly proposes human carrying capacity concept. It is defined as the total size of human population that the earth or a region can sustain without destroying the “natural, cultural, and social environment” and damaging the perpetuity of future carrying capacity (Abernethy, 2001, p9; Wei et al., 2015). Natural endowments are also vitally important for human carrying capacity. For example, economic activities rely on the natural capitals including the ecological services and natural resources. The excessive use of natural capital beyond the regenerative capacity will lead to the depletion in natural capital stock. Sustainability necessitates humanity development within the world's biosphere regenerative capacity. Ecological carrying capacity only considers natural resources as a primary dimension for understanding carrying capacity. However, human carrying capacity concept consists of more mad-made and complex factors such as social, economic, cultural etc. aspects, making it different from the ecological carrying capacity.

90

Y. Wei et al. / Habitat International 53 (2016) 87e96

sustainability of a region (Baldwin, 1985; Liu & Borthwick, 2011). Graymore, Sipe, and Rickson (2010) support the recognition of carrying capacity as a “sustainable threshold”. If the population exceeds this threshold of carrying capacity, it leads to negative impacts on the integrity, function, productivity, and resilience of the social, economic, and ecological supporting systems (Graymore et al., 2010; Yue, Tian, Liu, & Fan, 2008), and the damage could be irreversible and even calamitous. UCC helps determine the optimal level of population growth and urban development that the environment in a specified locality can support without degrading longterm sustainability. It also reminds the society to what extent natural and man-made resources, in terms of quantity and quality, should be adequately maintained above a specified standard. Fig. 2. The components of UCC.

4. Determining components of UCC The UCC concept derives from the above-mentioned carrying capacity theories, but with a special focus on the urban setting. UCC is of practical significance for urban sustainable development. For example, Onishi (1994) argued that “a city cannot expand infinitely, but has limits on population and economic activities, alongside with which the citizens can utilize urban facilities and services comfortably” (p. 40). Intuitively, a city with a higher UCC leads to “its residents love to live in and take pride in being part of it”, where the environment makes its residents physically, spiritually, and culturally devoted to the city (Wang, 2013, p.13). Some researchers have attempted to provide clear definitions of UCC. Onishi (1994), for example, defined UCC as human activities engaged in the city at a comfortable level. Since comfort is difficult to be judged objectively, UCC can be more precisely defined as an adequate supply of man-made and natural resources for demands of the public (Onishi, 1994). Likewise, Joardar (1998) and Oh et al. (2005) mainly pay attention to the physical or infrastructural factors of a city by assessing them against acceptable norms and standards. Sustainable development necessitates harmonious and balanced relationship among natural resources, bio-ecology, demographic growth, and human socioeconomic activities. As a yardstick for measuring urban sustainability, UCC is also a multidimensional assessment (Jin, Xu, & Yang, 2009). However, previous definitions, either focusing on infrastructure or urban ecology, seem fragmented and incomplete (Sarma et al., 2012; Summers, 2004; Tan, Shi, & Sun, 2008). This study adopts the definition of UCC as the limits of sustainable urban development from the perspective of five determining components: infrastructure and urban services, environmental impacts and natural resources, public perception,1 institutional setting, and society supporting capacity2 (Wei et al., 2015) (See Fig. 2). The measurable criteria for each dimension respectively are sufficient and well-maintained infrastructure and urban service, green environment and endurable resource uses, perceptual (both psychological and visual) satisfaction, institutional viability, economic affluence.

5. Limitation of existing literature Through an extensive literature review, the limitation of the previous study is summarized as follows:

1 Public perception refers to the behavioral psychology perceived by the urban residents, such as senses, attitudes, anticipations etc. towards the overall improvements of the urban settings. 2 Society supporting capacity is defined as the economic, fiscal and technological capacity of a society to improve its UCC by means of proactive investment.


Yue et al. (2008) proposed a general definition of “carrying capacity” as the maximum population of human, livestock, or wild animals that can be supported indefinitely without generating permanent damage to the earth. Presently, based on different underlying theories and emphasis, carrying capacity studies have been conducted in five research strands: tourism/recreational carrying capacity,3 safety or disaster carrying capacity4 (e.g. Chen et al., 2009), ecological carrying capacity,5 human carrying capacity6 (Graymore et al., 2010), and UCC (Li et al., 2009; Liu, 2012; Oh et al., 2005; Onishi, 1994; Sarma et al., 2012). An integration of these above analytical dimensions is necessarily important to establish a complete UCC evaluation framework for city managers. However, this has yet been appropriately addressed.
Current UCC related studies have been conducted alongside two strands (Liu, 2012), either concentrating on the single factor carrying capacity of a limited resource such as water and land, or focusing on the comprehensive carrying capacity by encompassing the economic, ecological, and social aspects of human activities. Currently, researchers have paid more attentions to the physical factors during UCC assessment, particularly focusing on infrastructure, pollution, and resources availability, but leaving socio-economic and institutional factors out of the analysis. Therefore, single carrying capacity can only provide partial understandings of urban sustainability. A comprehensive perspective is adopted in this study. Comprehensive UCC should completely cover all aspects of economy, environment and society (Liu, 2012). How to coordinate the relationships between various UCC elements while ensuring their consistent improvements is an important issue.
Traditional economic disciplines are limited in scope for integrating the environmental components and ecological significance into the economy (Pillet & Odum, 1984). The environmental components, unlike their economic counterparts

3 Tourism carrying capacity focuses on the negative impacts of tourisms on the destinations from ecological, physical, and experiential aspects. 4 With the fast urbanization pace, continuous congregation of population, urban services, production and wealth have been occurring in most cities of the world, and these factors make those cities vulnerable when sudden disasters happen (Chen et al., 2009). Disaster carrying capability refers to the capacity of a city or region to predict, prevent, rescue, or recover from disasters and accidents (Guo & Liu, 2003). These disasters and accidents broadly include “natural disasters, industrial accidents, and public health and social safety incidents” (Chen et al., 2009, p.50). 5 Based on biometric perspective, ecological carrying capacity speculated on the probable maximum number of species a specific region could indefinitely support. 6 Human carrying capacity refers to the maximum scale of human's consumptions of renewable resources, which can be indefinitely supported without causing irreversible damage to a defined region.

Y. Wei et al. / Habitat International 53 (2016) 87e96

such as human labors and capital investment, are usually not valued their true worth by the economy and the market (Campbell, 1998). Since the neglect of real value by the market, ecological systems that contribute to the creation of products and services are vulnerable to depletion without supplementary or remedial provisions being made for their consequent replacement and rehabilitation (Campbell, 1998; Repetto, 1992). However, as summarized by Oh et al. (2005), current related studies primarily focus on environmental dimension, rather than a holistic perspective integrating socioeconomic, environmental, and institutional lenses.
Current UCC assessment studies seem rather subjective and rudimentary, while the complexities involved require a more scientific and objective means. The existing literature still lacks of quantitative-based framework for UCC assessment. There are three key reasons for the scarcity of empirical studies. First, carrying capacity assessment is both a quantitative and qualitative study (Summers, 2004). The highly subjective attributes makes it particularly impossible to quantify and calculate in any accurate way (Sarma et al., 2012). Second, assessment on the UCC is complicated by the numerous factors, large varieties of natural and man-made resources, etc. Third, UCC assessment should ideally address the variability in technology, institutions, and human lifestyle (Sarma et al., 2012).

91

From a different classification perspective, Beijing, Shanghai, and Guangzhou are first-tier cities. They are the most economically developed cities in China. The rest are second-tier cities, which are of significant importance to the associated regions and provinces in terms of economic, social, political aspects. This study uses the data of 2011, which is sourced from published consensus, including the China City Statistical Yearbook 2012, China Statistical Yearbook For Regional Economy 2012, and the China Urban Construction Statistical Yearbook 2011.

6.2. Research process Fig. 4 illustrates the procedure for the development of UCC

6. Data and research methodology 6.1. Data and investigated cities This study evaluates the UCC conditions of 30 provincial capital cities and municipalities in China. Fig. 3 provides the geographic positions of the cities investigated. The four municipalities consist of Beijing, Tianjin, Shanghai, and Chongqing. They are directly governed by the central government in the administrative order.

Fig. 3. Map of Chinese cities investigated.

Fig. 4. Research procedure.

92

Y. Wei et al. / Habitat International 53 (2016) 87e96

evaluation model. It mainly consists of three steps i.e. indicators selection, evaluation method determination, and data processing. 6.3. Indicator selection In general, various estimate techniques have demonstrated in a general ways that some determining factors are involved. For a critical analysis, several rules for variables selection is determined through literature review (See Table 1). Based on the selection principles, 56 indicators were originally selected. Then, in reference to expert consulting and a correlation test, a final 30 indicators were developed into an evaluation system. The system consists of five subcategories of variables, i.e., economic, resources, environmental, infrastructural, and transportation. Due to the important role of transportation for a mega-city, this study separates transportation from infrastructure as an individually focused analytical dimension. 6.4. Research methodology

Table 3 The major methods for determining indicator weights. Subjective-based

Objective-based

A$J$Klee method Delphi Analytical Hieratical Process (AHP)

Deviation method Mean-Variance Analysis Principle Components Analysis (PCA) BP Neutral Network

methods, such as Analytical Hieratical (AHP), Principle Components Analysis (PCA), etc. (See Table 3). These methods can be classified into either subjective or objective-based approach. The main advantages and limitations of each quantitative method are compared in Table 4. Generally, the results by objective-based methods will not be affected by the whim or subjective opinions of the investigated individuals, and therefore leads to objective estimation results. To select the appropriate evaluation method, the rationales and features of sustainable urban development should be considered. The methodology should be effective in identifying the determining limitation of urban sustainability, and be useful to

Related studies have provided a wide range of evaluation Table 2 Indicator system. Sector

Indicators & unit

Attributes

Data source

Economic

X1-Urban registered unemployment rate (%) X2-Per capita disposable income of urban households X3-Per capita Fiscal income (Yuan) (X3 ¼ Fiscal income/population) X4-Per capita GDP (Yuan) X5-Annual GDP growth rate X6- Per capita water supply (ton): (X6 ¼ Total water supply of urban districts/total urban population) X7-Per capita daily domestic water consumption (liter) X8-Per capita constructive land (m2) (X8 ¼ urban constructive land/total urban population) X9-Per capita gas supply (m3) (X9 ¼ Total gas supply in urban districts/total urban population) X10-Per capita domestic electricity consumption (kwh) (X10 ¼ Total domestic electricity consumption in urban districts/total urban population) X11- industrial wastewater discharged per 10,000 Yuan GRP (ton) (X11 ¼ Total volume of industrial wastewater discharged*10,000/GRP) X12- industrial CO2 emissions per 10,000 Yuan GDP (kg) (X12 ¼ Total volume of industrial CO2 emissions*10,000/GRP) X13-The ratio of industrial solid waste which is comprehensively utilized X14- The ratio of sewage treated (%) X15-Living garbage treatment rate X16-The number of days with air quality above Grade-2 standard per years X17-Per capita green area (m2) (X17 ¼ Green areas of urban district/urban population) X18-Green coverage rate of urban built-up areas (%) X19-Number of hospital beds per 10,000 persons: (X19 ¼ Total hospital beds*10,000/total population) X20-Per capita floor space of urban residents (m2) X21- The density of drainage pipe in urban built-up areas (km/km2) X22-water access rate (%) X23-gas access rate (%) X24-Number of Internet per 10,000 persons (user) (X24 ¼ Urban Internet users*10,000/total population) X25- Number of mobile phone users per 10,000 persons (user) (X25 ¼ Urban mobile phone*10,000/total population) X26-Number of fixed telephone users per 10,000 persons (user): X26 ¼ Urban fixed telephone users*10,000/total population X27-Number of bus per 10,000 persons (unit) X28-Number of private cars per 10,000 persons (unit): X28 ¼ Number of private cars*10,000/total population X29-Per-capita urban road areas (m2) X30- Highway density (km/km2): X30 ¼ Length of highway/land area

 þ þ

a a b

þ þ þ

b b b

 þ

c b

þ

b



b



b



b

þ þ þ þ þ

b c c b b

þ þ

b b

þ þ þ þ þ

a c c c b

þ

b

þ

b

þ 

d a

þ þ

b a

Resources

Environmental

Infrastructural

Transport

Note: a refers to data sourced from China Statistical Yearbook for Regional Economy 2012; b refers to data sourced from China City Statistical Yearbook 2012; c refers to data sourced from the China Urban Construction Statistical Yearbook 2011; d refers to data sourced from (Liu, 2012); þ indicates benefit indicator that is the bigger the better; indicates cost indicator that is the bigger the worse.

Y. Wei et al. / Habitat International 53 (2016) 87e96

93

Table 4 The pros and cons of qualitative-based approaches. Approach

Strength

Delphi

Expert's experience and opinions are solicited by means of brainstorming.

AHP Principle Components Analysis (PAC) BP Neutral Network Mean-Variance Analysis

Limitations

The estimation results are highly subjective to evaluators' judgments. The method combines the qualitative and quantitative merits and provides a multi-dimensional Expert's opinions may subjectively affect the analysis. results. The statistical methods can effectively summarize the multiple variables to a limited number of The evaluation relies on exigent data quality. synthesized indices, and avoids the correlation among these indices. The method leads to a mutual evolution process with relatively objective results reflecting the real state. This method leads to high-accuracy estimation results and the underling rationales are easy for understandings.

A large number of training sample data are essential for appropriate evaluation. The estimation results are sensitive to the quality of historical data.

Note: A$J$Klee method is a derivative from AHP approach. The characteristics of A$J$Klee method are in reference to AHP.

improve the UCC. Table 4 summarizes the cons and pros of each method. This study determines the weight of each individual indicator based on two basic principles regarding conceptual framework and data quality: i) prefer objective-based approach to subjective-based approach; ii) prefer a wider range of indicators/ indicator system to a few representative variables. Principle i) excludes the methods of Delphi and AHP, and Principle ii) excludes the methods of PAC and BP Neutral Network. The Mean-Variance Analysis method is thus chosen for its accessibility for a general stakeholder of urban development and high-accuracy estimation results. 6.5. Data processing



 yij ¼ xij  xjmin = xjmax  xjmin

i ¼ 1; 2; 3; …; n; j

¼ 1; 2; 3; …; m

(1)



 yij ¼ xjmax  xij = xjmax  xjmin ¼ 1; 2; 3; …; m Note: xjmin and xjmin

i ¼ 1; 2; 3; …; n; j (2)

respectively refers to the minimal value and maximal

value of Ij .

After the dimensionless processing, the conformity and consistency of data across different indicator units is ensured. 6.5.2. Mean variance analysis This study uses the method of Mean Variance Analysis to determine the relative weights of each individual indicator. The analysis procedure consists of three steps (Equations (3)e(5)).
Sample mean:


Mean square error of Ij:

vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u n 
 uX
2 yij  E Ij s Ij ¼ t

(4)

i¼1

, n
 X
 u4 ¼ s Ij s Ij

(5)

j¼1

The aggregate weights of each hieratical levels are derived by adding the weight coefficients of subcategory indicators (see Table 5). The estimation results are derived by means of multiobjective linear weighting function as Equation (6):

RA ðwÞ ¼

m X

yij ui

ði ¼ 1; 2; 3……::; nÞ

(6)

i¼1

7. Results and discussion

For the group of “cost indicators”, data is transformed by equation (2):

bÞ

(3)


Weight coefficient of Ij:

6.5.1. Dimensionless standardization The indicators are grouped into two types, i.e. “benefit indicators” and “cost indicators” (see Table 2). The former refers to the ones that result in improving carrying capacity with their values increasing. “Cost indicator” is on behalf of deteriorating carrying capacity with their values increasing, such as CO2 emission, wastewater discharged, etc. The first step for data analysis is to conduct the dimensionless standardization to remove the data contamination issues due to different units indicators among indicators. For the group of “benefit indicators”, data is transformed by equation (1):

aÞ

n

 1X E Ij ¼ E Ij ¼ y n i¼1 ij

Table 5 shows the weights of the evaluative indicator system. Among the five key UCC subcategories, infrastructural and environmental aspectsdwith a statistical weight of 0.260 and 0.259 respectivelydare the most important determining factors, followed by resources and economic factors. Transportation has lowest contribution to urban carrying capacity in our model. 7.1. Urban infrastructure Urban infrastructure, such as utilities, communication, healthcare, amenity facilities, are essentially important for sustainable urban development and comfortable resident's living. It is the basic responsibility of the local government to provide versatile, adequate, and well-maintained infrastructural facilities. Due to the data limitation, this study only investigates healthcare (determined by the number of hospital beds), housing, public utility, and communication factors. Various age groups may have different preferences regarding their infrastructure needs: the elderly may

94

Y. Wei et al. / Habitat International 53 (2016) 87e96

Table 5 The weight of indicator system. Sector

Evaluative areas

Indicators

Weight

Economic (0.177)

Employment (0.034) Affluence (0.07)

X1-Urban registered unemployment rate (%) X2-Per capita disposable income of urban households X3-Per capita Fiscal income (Yuan) X4-Per capita GDP (Yuan) X5-Annual GDP growth rate X6- Per capita water supply (ton) X7-Per capita daily domestic water consumption (liter) X8-Per capita constructive land (m2) X9-Per capita gas supply (m3) X10-Per capita domestic electricity consumption (kwh) X11- industrial wastewater discharged per 10,000 Yuan GDP (ton) X12- industrial CO2 emissions per 10,000 Yuan GDP (kg) X13-The ratio of industrial solid waste which is comprehensively utilized X14- The ratio of sewage treated (%) X15-Living garbage treatment rate X16-The number of days with air quality above Grade-2 standard per years X17-Per capita green area (m2) X18-Green coverage rate of urban built-up areas (%) X19-Number of hospital beds per 10,000 persons X20-Per capita floor space of urban residents (m2) X21- The density of drainage pipe in urban built-up areas (km/km2) X22-water access rate (%) X23-gas access rate (%) X24-Number of Internet per 10,000 persons (user) X25- Number of mobile phone users per 10,000 persons (user) X26-Number of fixed telephone users per 10,000 persons (user) X27-Number of bus per 10,000 persons (unit) X28-Number of private cars per 10,000 persons (unit) X29-Per-capita urban road areas (m2) X30- Highway density (km/km2)

0.034 0.037 0.033 0.038 0.034 0.030 0.044 0.037 0.031 0.037 0.039 0.029 0.040 0.034 0.029 0.030 0.025 0.032 0.029 0.034 0.029 0.037 0.038 0.030 0.027 0.035 0.031 0.031 0.036 0.029

Resources (0.180)

Economic Scale (0.038) Growth (0.034) Water (0.074) Land (0.037) Energy (0.068)

Environmental (0.259)

Pollution (0.068) Treatment (0.133)

Green (0.057) Infrastructural (0.260)

Healthcare (0.029) Housing (0.034) Utility (0.104)

Communication (0.092)

Transport (0.126)

consider availability to healthcare services as most important. The working population may prioritize housing conditions and public utilities. Younger generations may be more dependent on communication access than other age groups. 7.2. Environment Environment is another important dimension for UCC monitoring and evaluation. The environment is an envelope around human activities (Abernethy, 2001). The environment supplies essential inputs to economic production and consumption, and also has to incorporate the waste generated. This study mainly focuses on the pollution discharged and associated treatment effect by man-made phenomenon. Pollution indicators include the industrial wastewater and CO2 emission. Pollution treatment variables mainly consist of the treatment rates of various wastes. Green areas, recognized as an important environmental asset, are included in the evaluation. 7.3. Resource Resource is also treated as a key subcategory in the UCC indicator evaluation system. Resources should not be exploited faster than they are regenerated or produced. Resources in this study refer to both the natural and man-made resources, including land, water, energy, all crucial to the city development and the lives of urban residents. Introducing the important concept of “appropriated carrying capacity” is necessary. “Appropriated carrying capacity” refers to import carrying capacity (i.e., resources) from remote places to sustain the urban development of the destination city. In this study, “appropriated carrying capacity” concept is considered. For example, water, gas, and electricity supply are often partially outsourced through domestic or even international trade. Beijing, as a water-deficient city, has to divert substantial water resources for nearby regions. Therefore, to ensure sustained resources supply and efficient utilization are two indispensable factors for

sustainable resource consumption. 7.4. Economics The concept of UCC cannot be understood in isolation from economic dimension of the urban environment. Economic vitality and diversity is an essential feature of urban sustainability. The key mission of the effective daily function of a city is to promote economic well-being, which ensures a high quality of life and is fundamental for the capability of urban services and facilities provision. Economic conditions in this study is systematically represented by several areas of variables, such as employment, affluence of the citizens, economic scale, and growth rate. 7.5. Transportation Transportation is also investigated in the study. In terms of data availability, buses represent public transportation; the conditions of road and private cars are also taken into account. 8. Discussion and conclusions Table 6 shows each city's rank in terms of their current UCC conditions. Both the comprehensive UCC and individual UCC are presented. Major findings are as follows: Firstly, Beijing, Guangzhou, Nanjing and Shanghai are the highest-ranking cities. Their high UCC is substantially supported by their strong economic and infrastructural strength. For example, although Beijing suffers from the poor transportation and a lack of water resources, it performs best in terms of economic development. Strong economic capacity is fundamental for government to proactively enhance UCC by means of direct investment and imported carrying capacity. In addition, economic affluence, as shown by low unemployment rates and high disposable income of urban household, can ensure the high material standards of living of the

Y. Wei et al. / Habitat International 53 (2016) 87e96

95

Table 6 The UCC Ranks of mega-city in China. Rank

UCC

Economic

Resources

Environmental

Infrastructural

Transport

Beijing Guangzhou Nanjing Shanghai Wuhan Changsha Shenyang Urumqi Jinan Hefei Hangzhou Tianjin Haikou Chengdu Changchun Fuzhou Nanchang Taiyuan Xian Yinchan Hohhot Kunming Shijiazhuang Zhengzhou Nanning Harbin Chongqing Guiyang Lanzhou Xining

0.617 0.598 0.597 0.589 0.553 0.549 0.546 0.545 0.545 0.543 0.542 0.541 0.536 0.514 0.512 0.507 0.489 0.481 0.472 0.470 0.467 0.462 0.454 0.444 0.392 0.379 0.372 0.363 0.355 0.321

0.121 0.118 0.097 0.100 0.065 0.092 0.075 0.054 0.063 0.057 0.104 0.108 0.062 0.068 0.049 0.072 0.049 0.042 0.054 0.043 0.073 0.066 0.033 0.074 0.036 0.044 0.052 0.055 0.044 0.028

0.098 0.050 0.074 0.078 0.087 0.050 0.081 0.126 0.084 0.078 0.050 0.087 0.060 0.066 0.096 0.048 0.072 0.098 0.057 0.128 0.114 0.100 0.094 0.080 0.054 0.079 0.073 0.043 0.097 0.081

0.177 0.176 0.193 0.181 0.171 0.199 0.196 0.154 0.186 0.192 0.158 0.158 0.196 0.185 0.182 0.187 0.183 0.152 0.163 0.115 0.123 0.138 0.125 0.151 0.151 0.116 0.136 0.159 0.105 0.089

0.177 0.180 0.152 0.167 0.147 0.138 0.137 0.182 0.134 0.115 0.168 0.134 0.161 0.126 0.112 0.130 0.110 0.146 0.139 0.118 0.106 0.104 0.114 0.090 0.095 0.091 0.065 0.066 0.056 0.097

0.044 0.073 0.081 0.062 0.083 0.070 0.056 0.029 0.078 0.100 0.061 0.054 0.057 0.069 0.074 0.070 0.075 0.044 0.059 0.066 0.051 0.053 0.089 0.048 0.056 0.049 0.045 0.040 0.054 0.026

urban residents, even though it may also indicate higher work pressure and longer working hours. Secondly, cities in eastern regions generally have high UCC rankings. Most of the low-ranked cities are located in central and western parts of China. Eastern cities generally perform well in terms of economic and infrastructural development, since they are major beneficiaries of reform and opening policies that were initiated from the eastern coastal regions. In the strand of literature on regional disparity, it is demonstrated that China's economic inequality is attributed to various factors, i.e., favorable government policy for coastal regions, unequal infrastructure development, labor market distortion, and imbalanced migration pattern. In recent years, such disparity seems to be lessened due to the Chinese government's “Western Development Strategy” (Fan & Sun, 2008). However, our estimation suggests that UCC still bears a pattern where coastal regions are more attractive to inland regions. Whether the “One Belt (Silk Road Economic Belt) and One Road (21st Century Maritime Silk Road)” projects will have significant impact on regional disparity in economic growth and infrastructure development remains debatable, but labor markets and migration patterns cannot be altered easily. Hence large cities located in the eastern coastal regions would be first choices for university graduates, rural and inland migrants to reside. And ironically, megacity problems such as air pollution, traffic jams, and expensive housing may persist in high-ranked UCC cities for the foreseeable future. Thirdly, the study shows a positive link between the city scale and UCC. By definition, UCC indicates how much a city can hold its population, under a variety of economic, social and environmental constraints. Although there is no causal relationship between city scale and UCC by definition, our results suggest that the largest cities, including Beijing, Guangzhou, Shanghai and Wuhan (China's four most densely populated cities, each with over 10 million people) also rank in the top five of the UCC evaluation. It is conjectured that a larger population scale leads to the more economic

and efficient human activities patterns: Larger cities tend to have urban agglomeration effect, which increases labor demand for high-tech or high-skilled workers. Better urban public infrastructure facilitates business activities and improves labor productivity too, which contributes to the efficient operation of cities (Eberts & McMillen, 1999). It is therefore not surprising that larger cities also rank top in UCC, although the hazard of urban sprawl cannot be ignored. Fourth, environmental capacity has a salient impact factor in evaluating a lively and attractive mega-city. According to Chen et al. (2013), life expectancies in northern China (considered north of the Huai River) are about 5.5 years lower than in southern China (considered south of the Huai River), due to an increased incidence of cardiorespiratory mortality caused by free provision of coal for boilers for winter heating. Our model concurs that in terms of environmental capacity, cities in southern China has 23% higher average rate than cities in northern China. Indeed, industrial carbon emission per 10,000 yuan GRP (ton) is in the north is on average 11% higher than in the south, thus exerting more negative impact on environmental capacity. The number of days with air quality above Grade 2 standard per years in the south is on average 27% higher than that in the north. In view of the individual subcategory rankings, several other findings appear as follows. First, economic performance is an important determining factor for UCC conditions. In general, economic strength shows a positive relationship with UCC performance. Beijing shows the strong economic strength. It ranks the lowest in unemployment rate, the third highest in average household disposable income, and the most abundant in fiscal income per capita, although the GRP growth rate is the lowest. Second, in terms of resource carrying capacity, most of the topranked UCC cities do not have consistently high levels of resource endowment. It suggests that the safe and adequate supply of

96

Y. Wei et al. / Habitat International 53 (2016) 87e96

resources is a short board for these high ranked mega cities. Interestingly, some of the low-ranked cites shows a high level of resource endowment. For example, Yinchuan and Hohhot, ranked 20th and 21st, have the highest and third highest resource capacity. The imbalanced distribution between resource endowment and UCC suggests low efficiency in the allocation between natural capital and human capital, a prevailing problem underlying China's regional inequality. Although the central government has endeavored to attract capital flows from coastal to inland regions, most of the money has poured into the real estate sector rather than industrial and manufacturing sectors. This capital flow trend generates a growing property bubble, which is featured by China's “ghost cities” (marked by extremely high residential vacancy rates), particularly in inland and western provinces. Since these inland cities lack prosperous economic conditions and lag in UCC rankings, it is difficult to attract prospective enterprises but easy to attract speculators. This further lowers the efficiency of capital allocation, resulting in a chain of actions and reactions in which larger cities become even larger while smaller cities become less competitive. Third, in terms of transportation, Hefei performs best, which is followed by Shijiazhuang and Wuhan. Beijing is ranked the forth lowest. Although Beijing has the largest number of bus per pita and a relatively high road density, the highest per capita private car ownership leads to a deficiency in road space and traffic congestions. Considering sustainable transportation strategy for a smart city, the government could control the private car ownership and encourage green and high-efficiency carpooling at first glance, and propose different measures which fit different cities: For middle and small sized cities with lower work pace, reducing vehicle miles travelled by car, promoting shorter distances and encouraging all modes of transport such as walk and bicycling are feasible solutions. For large cities with an intense work pace, the technological transition to low carbon emission transport system would not only enhance the efficiency of traffic regulation but also reduce air pollutants. The concept of sustainable urban development is receiving wide recognition. The findings in this study may assist city managers in identifying problems and then in finding means to make improvements. Meager government investment and distribution of resources should be more fairly invested in the sectors mentioned above to improve the urban sustainability. This research still remains limited. The analytic dimensions mainly consist of the economic and physical aspects of UCC, due to the limits of available data. Other important factors such as technological, institutional, and perceptual factors have been rarely integrated. Future studies should conduct wider investigations on the perceptual and institutional aspects of UCC when given an improvement in data availability. Acknowledgment The first author would like to thank the China Postdoctoral Science Foundation (Grant No.: 2014M550755) and the Chinese Academy of Engineering (Grant No.: 2013-XZ-25) for the financial support. The fourth author would like to thank Key Laboratory of Carrying Capacity Assessment for Resource and Environment, the Ministry of Land and Resources P.R.C for the financial support (Grant No.: CCA2015.06). References Abernethy, V. D. (2001). Carrying capacity: the tradition and policy implications of limits. Ethics in Science and Environmental Politics, 23, 9e18. Baldwin, J. H. (1985). Environmental planning and management. Boulder: Westview Press.

Bentham, J. (2014). The scenario approach to possible futures for oil and natural gas. Energy Policy, 64, 87e92. Button, K. (2002). City management and urban environmental indicators. Ecological Economics, 40(2), 217e233. Camagni, R. (1998). Sustainable urban development: definition and reasons for a research programme. International Journal of Environment and Pollution, 10(1), 6e27. Campbell, D. E. (1998). Emergy analysis of human carrying capacity and regional sustainability- an example using the state of Maine. Environmental Monitoring and Assessment, 51, 531e569. Chen, Y., et al. (2013). Evidence on the impact of sustained exposure to air pollution on life expectancy from China's Huai River policy. Proceedings of the National Academy of Sciences, 110(32), 12936e12941. Chen, G., Tao, L., & Zhang, H. (2009). Study on the methodology for evaluating urban and regional disaster carrying capacity and its application. Safety Science, 47(1), 50e58. Daily, G. G., & Ehrlich, P. R. (1996). Socioeconomic equity, sustainability, and earth's carrying capacity. Ecological Applications, 6(4), 991e1001. Eberts, R. W., & McMillen, D. P. (1999). Agglomeration economies and urban public infrastructure. Handbook of Regional and Urban Economics, 3, 1455e1495. Fan, C. C., & Sun, M. (2008). Regional inequality in China, 1978-2006. Eurasian Geography and Economics, 49(1), 1e18. Graymore, M. L. M., Sipe, N. G., & Rickson, R. E. (2010). Sustaining human carrying capacity: a tool for regional sustainability assessment. Ecological Economics, 69(3), 459e468. Guo, Z. L., & Liu, M. G. (2003). Secondary fuzzy comprehensive evaluation about the ability of a city's undertaken disaster by AHM. Journal of Hebei Institute of Architectural Science and Technology, 20(2), 5e8 (in Chinese). Heikkila, E., & Xu, Y. (2013). Seven prototypical chinese cities. Urban Studies, 1e21. Jin, W., Xu, L., & Yang, Z. (2009). Modeling a policy making framework for urban sustainability: incorporating system dynamics into the ecological footprint. Ecological Economics, 68(12), 2938e2949. Joardar, S. D. (1998). Carrying capacities and standards as bases towards urban infrastructure planning in India: a case of urban water supply and sanitation. Habitat International, 22(3), 327e337. Li, F., Liu, X. S., Hu, D., Wang, R. S., Yang, W. R., & Li, D. (2009). Measurement indicators and an evaluation approach for assessing urban sustainable development: a case study for China's Jining City. Landscape and Urban Planning, 90(3e4), 134e142. Liu, H. (2012). Comprehensive carrying capacity of the urban agglomeration in the Yangtze River Delta, China. Habitat International, 36(4), 462e470. Liu, R. Z., & Borthwick, A. G. L. (2011). Measurement and assessment of carrying capacity of the environment in Ningbo, China. Journal of Environmental Management, 92(8), 2047e2053. Li, B., & Yao, R. (2009). Urbanisation and its impact on building energy consumption and efficiency in China. Renewable Energy, 34(9), 1994e1998. McKinsey Global Institute. (2011). Urban World: Mapping the economic power of cities. Oh, K., Jeong, Y., Lee, D., Lee, W., & Choi, J. (2005). Determining development density using the urban carrying capacity assessment system. Landscape and Urban Planning, 73(1), 1e15. Onishi, T. (1994). A capacity approach for sustainable urban development: an empirical study. Regional Studies, 28(1), 39e51. Pillet, G., & Odum, H. (1984). Energy externality and the economy of Switzerland. Swiss Journal of Economics and Statistics, 120, 409e435. Repetto, R. (1992). Accounting for environmental assets. Scientific American, 94e100. June. Sarma, A. K., Mahanta, C., Bhattacharya, R., Dutta, S., Kartha, S., Kumar, B., et al. (2012). Urban carrying capacity: Concept and calculation (p. 23) (Guwahati, Assam,India). Summers, P. (2004). Population carrying capacity in Noosa Shire. Tewantin: Noosa Council. Tewantin: Noosa Council. Retrieved from http://scholar.google.com/ scholar?q¼populationþcarryingþcapabilityþinþnoosaþshire&btnG¼&hl¼zhCN&as_sdt¼0%2C5#2. Tan, W. K., Shi, Y. S., & Sun, L. (2008). Some theoretical issues on urban carrying capacity. China Population Resources and Environment, 18(1), 40e44 (in Chinese). The World Bank. (2014). China: To promote efficient, compatible, and sustainable urbanization. Beijing. Retrieved from http://www.worldbank.org/content/dam/ Worldbank/document/EAP/China/urban-china-overview-cn.pdf. Wang, F. (2013). Beijing as a globally fluent city. Beijing. Retrieved from http://www. brookings.edu/~/media/research/files/papers/2013/10/14 beijing as a globally fluent city/beijing as a globally fluent city.pdf. Wei, Y. G., Huang, C., Lam, P. T. I., & Yuan, Z. Y. (2015). Sustainable urban development: a review on urban carrying capacity assessment. Habitat International, 46, 64e71. World Commission on Environment Development. (1987). Our common future. Oxford: Oxford University Press. Yue, T. X., Tian, Y. Z., Liu, J. Y., & Fan, Z. M. (2008). Surface modeling of human carrying capacity of terrestrial ecosystems in China. Ecological Modelling, 214(2e4), 168e180. Yu, D. L., & Mao, H. Y. (2002). Regional carrying capacity: case studies of Bohai Rim area. Journal of Geographical Sciences, 12(2), 177e185.