Ecological Indicators 50 (2015) 206–214

Short Communication

Assessment of sustainable development: A case study of Wuhan as apilot city in China

Xiaohong Chen, Xiang Liu *, Dongbin HuCollaborative Innovation Center of Resource-conserving & Environment-friendly Society and Ecological Civilization, School of Business, Central SouthUniversity, Changsha, China

A R T I C L E I N F O

Article history:Received 6 December 2013Received in revised form 29 October 2014Accepted 3 November 2014

Keywords:Sustainable developmentEnvironmental friendlyResource conservingVector angleEuclidean distanceChina

A B S T R A C T

Building resource-conserving and environmental-friendly society (referred to as “two-oriented society”,TOS) is an important way proposed by the Chinese government to achieve sustainable development. Inthis paper, a pilot city of constructing TOS in China-Wuhan is taken as a case to evaluate the performanceof TOS from 2005 to 2012. Treating the indicators of TOS as multi-dimensional vectors, this paperproposes a methodological framework by integrating the methods of vector angle and Euclidean distanceto measure the angle and distance between the vector of annual status of TOS and the vector of planningtarget of TOS. Based on this, the paper employs coordination (the angle between the two vectors) andeffectiveness (the distance between the two vectors) to describe the performance of TOS and itssubsystems, including economic development (ED), social and people's well-being (SW), resourceconsumption (RC), resource recycling (RR), environmental quality (EQ) and pollution control (PC).Moreover, grey relational analysis approach is used to analyse the core factors influencing TOSconstruction. Results reveal that the performance of TOS in Wuhan is continually improved in the studyperiod while the performance of its subsystems shows several differences: (1) both coordination andeffectiveness of TOS, ED and SW keep on improving, while those of EQ fluctuate; (2) the effectiveness ofboth RR and PC shows a downward trend while the coordination of them fluctuates; (3) the actual valueof RC indicators reaches the planning targets. Based on the analysis of determinants, we suggest thatstrengthening technological ability and adding investment are extremely important to improve theperformance of RC and PC. For the sake of improving RR, it is critical to provide more government publicexpenditure and encourage financial institutions to provide more loans to stimulate and support thebusinesses. Moreover, increasing the amount of R&D funding and maintaining steady external economicenvironment are proved to be effective to improve all subsystems of TOS.

ã 2014 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Ecological Indicators

journa l home page : www.e l sev ier .com/ loca te /eco l ind

1. Introduction

Sustainable development (SD) is a common goal across theglobe in the 21st century. However, large amount of energyconsumption, environmental pollution and carbon emissionsgenerated in China pose negative threats to Chinese SD (Zhangand Wen, 2008). Therefore, the Chinese government has beenpaying increasingly more attention to exploring ways forconserving resources and protecting the environment withoutdamaging economic growth. The Chinese government directlyproposed SD as part of the national strategies in 1994, and hasimplemented it by various economic models such as circular

* Corresponding author. Tel.: +86 18569517974.E-mail addresses: xiangliu1018@gmail.com, liuxiang@csu.edu.cn (X. Liu).

http://dx.doi.org/10.1016/j.ecolind.2014.11.0021470-160X/ã 2014 Elsevier Ltd. All rights reserved.

economy, low-carbon economy, as well as building two-orientedsociety (TOS).

Over the past decades, plenty of literature focus on proposingindicators and methods for evaluating the performance of SD(Rinne et al., 2013; Li et al., 2009; Böhringer and Jochem, 2007;Hopwood et al., 2005) and investigating its determinants (Zhanget al., 2012; Wei et al., 2012; Conroy and Berke, 2004). For example,Compendium of Sustainable Development Indicator Initiativesidentified about 600 sustainable indicators for SD performance(Parris and Kates, 2003). Böhringer and Jochem (2007) andHammond et al. (1995) reviewed the consistency and meaning-fulness of various indicator and composite index, and proposedmany principles and theoretical frameworks about SD assessment.Dalal-Clayton and Bass (2006) and Moldan et al. (2012) proved thatan aggregate SD index and indicators linked to targets are of greatimportance for policy makers because they can be unambiguouslyinterpreted and easily communicated to the general public.

X. Chen et al. / Ecological Indicators 50 (2015) 206–214 207

However, scientific rules elaborated by Ebert and Welsch (2004)have not been taken into account enough for aggregatingindicators towards composite indices because of lacking thegenerally accepted rules for normalization and weighting, andbeing short of the commensurability of various variables(Böhringer and Jochem, 2007). Moreover, it was still an openquestion to construct a versatile methodology (Singh et al., 2012).Even though various evaluation methods had been employed, suchas principal component analysis (PCA) (Yang et al., 2011), superefficiency data envelopment analysis (DEA) model (Wu et al.,2014), and super slack-based measure (SBM) model withundesirable outputs (Li et al., 2013). The evaluation results ofPCA and DEA models are greatly influenced by sample size andstructure, which means that different data and sample size oftenlead to changes in evaluation results. In addition, from the practicalview, planning is helpful for promoting SD for all countries (Conroyand Berke, 2004), especially for China which has a highlycentralized planning system to lead and guide the developmentof cities (Xie and Costa, 1993). Therefore, the objectives mentionedin the plan should be taken into account when evaluating theperformance of SD.

In recent years, some studies evaluate SD performance byregarding the indicators of SD as an n-dimensional vector. Forexample, Wei et al. (2012) defined the indicators describing China'sregional pollution control performance (PCP) as an n-dimensionalvector. They aggregated the best and worst performance of eachindicator in all regions of China and defined them as optimal vectorand threshold vector, respectively. Taking the optimal vector as abenchmark, all components of this vector can be viewed as thevalue requirements of the corresponding components of the vectorof China's regional PCP. Afterwards, Euclidean distance withweighting was employed to calculate the gap between the vector ofthe current regional PCP status and the optimal vector. The gap thatreflects the effectiveness of value requirements of all indicators canbe recognized as regional PCP. However, besides the valuerequirement of each indicator, the coordination requirement ofall components of a vector also should be taken into account. Thecoordination requirement means the proportional relations amongall components of a vector (the vector of regional PCP in Wei et al.(2012)) should be consistent with the proportional relationsamong the corresponding components of the other vector (theoptimal vector of PCP in Wei et al. (2012)), which indicates that allindicators of PCP should be developed simultaneously. Thecoordination requirement can be reflected by the directions ofvectors. The direction of vector can be calculated by vector anglewhich is widely used as a term of cosine similarity in various areasincluding information retrieval and decision making (Ye, 2011).Duwairi (2006) viewed several keywords in a student essay as avector and compared it with the vector of keywords in a modelessay to evaluate the performance of the student essay by cosinesimilarity. Cosine similarity treats each object (the keywords in astudent essay in Duwairi (2006)) as a vector, and then calculatesthe cosine of the angle between two vectors as a similarity measureof them.

In the case of SD, the planning targets of SD compiled by thegovernment can be regarded as the ideal objective of SDconstruction (like the model essay in Duwairi (2006)). Moreover,only when all indicators achieve targets, can the status of SD berecognized as reaching the ideal objective. Therefore, by taking theSD indicators as a multi-dimensional vector, the planning targets ofSD indicators can be viewed as an objective vector. The direction ofthe objective vector implies the coordination requirements of itscomponents. Each component of the objective vector indicates thevalue requirement of the indicator. Consequently, comparing thevector of the annual status of SD with the objective vector, both thecoordination (direction of the two vectors) and effectiveness

(distance of the two vectors) can be calculated to evaluate SDperformance.

This paper presents a detailed analysis of SD performance ofWuhan which is a pilot city of building TOS in China. The methodsof vector angle and Euclidean distance are used to calculate thecoordination and effectiveness of TOS. The coordination andeffectiveness are displayed in the same diagram to depict theperformance of Wuhan's TOS construction. Furthermore, thispaper analysed core factors influencing TOS performance ofWuhan by grey relational analysis.

The rest of this article is organized as follows: in Section 2 wegive a short introduction of Wuhan and introduce the methodologyin detail. Section 3 presents a summary of the findings of ouranalysis. Finally, we propose recommendations for the governmentto improve TOS more coordinately and effectively.

2. Materials and methods

2.1. Study area

Wuhan is the capital city of Hubei province, which is located incentral China (see Fig.1). As one of the pilot areas constructing TOS,Wuhan covers an area of 8.49 thousand km2 and has a populationof 10.22 million inhabitants in 2013.

Wuhan is an old industrial base collecting optic-electronic,automobile manufacturing, iron and steel manufacturing. Coalis the dominant energy source in the city. Environmentalpollution caused by rapid economic growth and unbalancedindustry structure is a great challenge to the SD of Wuhan.Therefore, Wuhan was approved to be a pilot city ofconstructing TOS by the Chinese government in 2007. In orderto promote and accelerate TOS construction, Wuhan has putforward a comprehensive plan called Synthetically ReformExperimental Implementing Scheme for TOS Construction ofWuhan depicting the vision, objectives and measures ofconstructing TOS. The plan works out as a basic plan guidingthe development of the city.

2.2. Evaluating the performance of TOS

2.2.1. Index system of TOSAs a practical way for SD in China, TOS refers to a system in

which resources are efficiently exploited and environment iseffectively preserved to acquire the highest economic and socialbenefits. As such, we can divide the TOS system into sixsubsystems, including economic development (ED), social andpeople's well-being (SW), resource consumption (RC), resourcerecycling (RR), environmental quality (EQ) and pollution control(PC). Specifically, RC and RR are essential contents of resourceconserving, and EQ and PC represent the core features ofenvironment-friendly.

The first three main features of good indicators are easy to beunderstood, sensitive to changes and relevant (OECD, 2008).Therefore, indicators for evaluating TOS should be: (1) describingthe situation of TOS scientifically, objectively and comprehen-sively. (2) Simple and data availability. (3) Constituted by statusindicators and process indicators. Status indicators reflect theresult of constructing TOS, while process indicators show thechange of it. For instance, environment-friendly indicators canbe divided into environmental quality and pollution control. Theformer describes the status while the latter indicates the resultsand procedures. Therefore, with reference to Synthetically ReformExperimental Implementing Scheme for TOS Construction ofWuhan and other plans, we can construct evaluation indicatorsof TOS performance as shown in Table 1.

Fig. 1. Location of Wuhan in Hubei province and China.

208 X. Chen et al. / Ecological Indicators 50 (2015) 206–214

2.2.2. Models and calculation proceduresThe indicators describing TOS are taken as an n-dimensional

vector Ynt , where n is defined as the number of evaluation

indicators for TOS, and t indicates the year. Hence, the performanceof TOS in t year can be characterized by a specific vector

Ynt ¼ ðy1t; y2t; . . . ; yntÞ; (1)

Table 1Evaluation indicators of TOS.

Subsystems Indicators

Economic Development (ED,5) The amount of GDP (100 millioGDP per capita (yuan)

The added value of high-tech iTertiary industry proportion (%General budget revenue in loca

Social and People's Wellbeing (SW,3) Urbanization rate (%) (househoPer capita annual disposable inPer capita annual net income o

Resource Consumption (RC,2) Energy consumption of unit GDWater consumption per unit in

Resource Recycling (RR,2) Utilization rate of industrial waRatio of industrial solid wastes

Environmental Quality (EQ,3) Urban per capita public green sCoverage rate of green areas deUrban air quality rate (%)

Pollution Control (PC,5) Urban sewage centralized dispoPercentage of harmless treatmThe amount of COD emissions

The amount of SO2 emissions (CO2 emissions per unit of GDP

a The 12th five-year plan for national economic and social development of Wuhan (b The plan for Wuhan building people's happiness city.c Wuhan's circular economy development plan (2011–2015).d Wuhan's 12th five-year plan on environmental protection (2011–2015).e Wuhan's implementation plan on building low carbon city.

in which the components yit correspond to the yi indicator of TOSin t year. For each indicator yi, the planning target can be found ingovernmental official documents (see details in Table 1). Specially,the planning targets of TOS are defined as Yn

0 ¼ ðy10; y20; . . . ;yn0Þwhich can be interpreted as the ideal objective to which both thedirection and distance of Yn

t should be close.The raw data of indicators including positive indicators (such as

GDP per capita, etc.) and negative indicators (such as the amount of

The target value in 2015

n yuan) 11,244.8a

111,666b

ndustry as a share of GDP (%) 0.27a

) 52a

l finance (100 million yuan) 1396a

ld population) 0.77a

come of urban households (yuan) 41,000b

f rural households (yuan) 16,000b

P (TCE/10,000 yuan) 0.87c

dustrial added value (m3/10,000 yuan) 80.8c

ter for irrigation (%) 88a

utilized (%) 98a

pace (m2/person) 10b

veloped (%) 40d

85b

sal rate (%) 93b

ent for living carbage (%) 100b

(10,000 tonnes) 13.73c

10,000 tonnes) 10.9c

(tonne/10,000 yuan) 1.48e

2011–2015).

X. Chen et al. / Ecological Indicators 50 (2015) 206–214 209

COD emissions, etc.) which are normalized by

yi ¼ yi=Pn

i¼1 yi ; Positive indicatorsPni¼1 yi=yi; Negative indicators

((2)

Therefore, the cosine of the angle between the vector of currentstatus of TOS in t year Yn

t and the vector of the planning targets Yn0 is

cosut ¼ Yn0 � Yn

t

k Yn0 k �k Yn

t k ¼Pn

i¼1yi0 �yitffiffiffiffiffiffiffiffiffiffiffiffiffiffiPn

i¼1y2i0

q�

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiPn

i¼1y2it

q (3)

cosut is monotonically decreasing and 0 � cosut� 1. We define thecoordination degree of the current status of TOS in t year as

Ct ¼ 1 � cosut (4)

The smaller Ct is, the more coordinated the development ofvarious subsystems and indicators of TOS are.

Moreover, the distance between Ynt and Yn

0 can be calculated byEuclidean distance

Dt ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiXni¼1

ðyit � yi0Þ2vuut (5)

The smaller Dt is, the closer Ynt and Yn

0 is, which means the bettereffectiveness of TOS in pt year.

Specially, the values of Ct and Dt of Yn0 in 2015 are zero, which is

C0 ¼ 0D0 ¼ 0

�

Ct and Dt can be used to measure the coordination andeffectiveness of TOS in OA ¼ ða1; b1Þ year. As shown in Fig. 2, wesuppose that two-dimensional vectors OA ¼ ða1; b1Þ,OB ¼ ða2; b2Þ,OC ¼ ða3; b3Þ, and OD ¼ ða4; b4Þ represent TOS status in differentyears, and OH ¼ ða0; b0Þ is the objective vector of TOS. The vectorcomponents ai and bi (i = 0, 1, 2, 3, 4) represent the actual value of

Fig. 2. Comparison of the coordination (Ct, cos u) and the distance of (Dt).

TOS indicators and we have a3 > a0 > a4 > a2 > a1, b3 > b0 > b1 > b4 > b2.The distances and the cosine of the angle between OA and OH areD1 and cosu1. The distances and the cosine of the angle between OBand OH are D2 and cosu2. The distances and the cosine of the anglebetween OC and OH are D3 and cosu3. The distances and the cosineof the angle between OD and OH are D4 and cosu4. Fig. 2 shows thatD1 = D2 = D3 > D4, C2 = C4 = 0, C1, C3 > 0. Therefore, regarding OH as abenchmark, OD shows a better effectiveness than OB for D4< D2.However, the coordination of OD and OB are equal (C2 = C4 = 0),which means ai and bi of these two vectors are in balance (a2/b2 = a4/b4 = a0/b0).

Moreover, as shown in Fig. 2, although the actual value of OC isgreater than that of OA (a3 > a1, b3 > b1), their distances are equal(D1 = D3). In this case, in order to distinguish D1 and D3, we definethe distance as zero when the actual value of the indicator isgreater than the target value of this indicator. Therefore, theperformance of OA and OC distances (D1 > D0

3ðD3Þ ¼ 0) areconsistent with the performance of their actual value (a3 > a0 > a1and b3 > b0 > b1). Considering the purpose of this paper is toevaluate the performance of TOS in different years, the relative sizeof Dt is more important than the actual size of it.

Consequently, the performance of TOS can be clearly shown inthe coordinate axis constituted by Ct (the coordination of TOS) andDt (the effectiveness of TOS). In special, the origin of the coordinateaxis is (C0, D0) = (0, 0) which can be seen as the performance of TOSplanning targets.

2.3. Impact factors assessment

2.3.1. Impact factorsEconomic, political culture, institutional and intergovernmen-

tal management variables are key factors influencing effectiveimplementation of local sustainability plan (Zeemering, 2012). Weidentified a comprehensive set of factors affecting TOS construc-tion as shown in Table 2.

Firstly, investment-driven has been regarded as a key develop-ment strategy for China (Qin et al., 2006). The growth of industrialeconomy, industrial transformation and upgrading, as well asenergy saving and emission reductions, are inseparable fromvarious capital investments. Therefore, two indicators namelyinvestment in fixed assets and funds for pollution treatment aretaken as proxy indicators.

Secondly, government public finance expenditure in scienceand technology, and environmental protection are important. Theformer indicator contains funds provided by the government tosupport science and technology activities. The latter one includesexpenditure of environmental and resources monitoring andsurveillance, pollution reduction, supporting new energy andrenewable energy industries, etc.

Thirdly, with the capital support provided by financial markets,businesses are stimulated and encouraged to promote economicdevelopment and industrial transformation and upgrading. InWuhan, bank loans are the most important ways of financial support.

Fourthly, technologies are key factors in developing a sustain-able society (Rennings, 2000; Fleiter and Plötz, 2013). In China,improving technological abilities plays a key role in improvingproductivity and environmental performance (Ang, 2009; Fisher-Vanden and Sue Wing, 2008). Therefore, the portion of R&D(Research and Development) expenditure in GDP and the numberof granted patents are taken as proxy indicators.

Last but not the least, a stabilized macroeconomic environmentwill help the government to better foster development of theinfrastructure and institutions, which are necessary for sustainablegrowth (Huang et al., 2006). As in Wuhan, GDP growth rates ofChina and Hubei province are the first two important externaleconomic environment factors.

Table 3The performance of TOS in Wuhan (2005–2012).

2005 2006 2007 2008 2009 2010 2011 2012

TOS Ct (%) 16.79 14.59 12.21 9.81 8.46 6.66 3.44 2.79Dt (%) 4.35 4.13 3.87 3.53 3.29 2.97 2.31 1.81

ED Ct (%) 23.90 19.99 15.91 11.47 9.28 6.83 1.65 0.75Dt (%) 3.52 3.38 3.22 3.00 2.83 2.58 1.98 1.55

SW Ct (%) 12.64 10.34 7.67 5.07 3.59 2.08 1.09 0.52Dt (%) 2.06 1.97 1.85 1.68 1.56 1.38 1.14 0.91

RC Ct (%) 3.83 2.12 0.72 0.02 0.01 0.06 0.51 0.28Dt (%) 1.06 0.95 0.79 0.55 0.41 0.29 0.19 0.00

RR Ct (%) 0.01 0.03 0.00 0.00 0.03 0.03 0.08 0.04Dt (%) 0.21 0.19 0.16 0.13 0.10 0.00 0.09 0.09

EQ Ct (%) 0.05 0.05 0.02 0.00 0.02 0.02 0.02 0.05Dt(%) 0.17 0.16 0.16 0.12 0.11 0.17 0.08 0.05

PC Ct (%) 2.32 1.69 1.05 0.72 0.32 0.37 0.25 0.34Dt (%) 1.05 0.89 0.71 0.57 0.40 0.36 0.28 0.21

Note: the value of Ave is calculated by the mean of two indicators in the aspect.Taking investment (INS) as an example, Ave of INS is calculated by the mean ofX1 and X2.

Table 2Factors affect TOS construction in Wuhan.

Theme Indicators

Investment (INV,2) Investment in fixed assets (10,000 yuan) (X1)Funds for pollution treatment (10,000 yuan) (X2)

Government Public Expenditure (GPE,2) Government expenditure for science and technology (10,000 yuan) (X3)Government expenditure for environmental protection (10,000 yuan) (X4)

Financial Support (FIS,2) Short-term loans (10,000 yuan) (X5)Medium and long term loans (10,000 yuan) (X6)

Technological Ability (TEA,2) The portion of R&D expenditure in GDP (%) (X7)The number of granted patents (X8)

External Economic Environment (EEE,2) GDP growth rate of China (%) (X9)GDP growth rate of Hubei Province (%) (X10)

210 X. Chen et al. / Ecological Indicators 50 (2015) 206–214

2.3.2. Method of grey relational analysisGrey relational analysis (GRA) was applied to further discover

the factors affecting TOS constructing. GRA was proposed byProfessor Deng (1989) to analyse the correlation between areference sequence and comparable sequences. The more similarthese two sequences’ curves are, the closer relationship betweenthem is (Deng, 2010). Yin (2013) reviewed the theoretical andpractical trends of GRA, and demonstrated the application of GRAshowing a continuous increase in various areas includingengineering, business economics, environmental sciences ecology,etc. According to Deng's definition (Deng, 1989, 2010), the first stepof GRA is to calculate the grey relational coefficient

jiðkÞ ¼ miniminkjx0ðkÞ � xiðkÞj þ rmaximaxkjx0ðkÞ � xiðkÞjjx0ðkÞ � xiðkÞj þ rmaximaxkjx0ðkÞ � xiðkÞj

: (6)

ji(k) is the grey relational coefficient of comparable sequence xi(xi = xi(k) = {xi(1), xi(2), . . . , xi(n)} , k = 1, 2, . . . , n, i = 1, 2, . . . , m),and reference sequence x0 (x0 = x0(k) = {x0(1), x0(2), . . . , x0(n)} ,k = 1, 2, . . . , n) at time k. xi and x0 are used after dimensionlessprocessing and are expressed as normalized values. The distin-guishing coefficient r reflects the degree to which the minimumscores are emphasized relative to the maximum scores. r isgenerally assumed to be 0.5.

In this paper, the two reference sequences are Ct and Dt, thecomparable sequences are the determinants mentioned in Table 2.Generally, these determinants have positive effects on TOSconstruction. Hence, the change direction of the two referencesequences and comparable sequences are opposite. Therefore, ji(k)is calculated based on the reciprocal of the determinants.

Therefore, the average value of ji(k) can be calculated torepresent the grey relational grade (GRG)

g i ¼1n

Xnk¼1

jiðkÞ: (7)

gi denotes importance of the factors, the greater gi is, the moreimportant it is.

2.4. Data sources

The socio-economic data are mainly collected from WuhanStatistical Yearbook. The resource and environment data areoriginally obtained from China City Statistical Yearbook, WaterConservancy Annals of Hubei province and Report on the State ofthe Environment in Wuhan. The sources of the target values in2015 are introduced in detail in Table 1. All indicators are availablein the period 2005–2012.

3. Results and discussion

3.1. Performance of TOS

According to formulas (3), (4) and (5), the coordination degreeCt and distance Dt of TOS in t year in Wuhan during 2005–2012 canbe calculated as shown in Table 3.

The performance of TOS in t year can be seen as points (Ct, Dt)and presented in coordinate axis. We can depict constructingstatus of TOS in Wuhan from 2005 to 2012 in Fig. 3.

It can be concluded from Fig. 3 that Wuhan's TOS performancecontinues to close to the planning targets in 2015 from 2005 to2012. On the one hand, the coordination degree (Ct) greatlydecreased from 16.79% in 2005 to 2.79% in 2012, which means allsubsystems of TOS including ED, SW, RC, RR, EQ and PCexperienced similar development tendency that gets close tothe planning targets. On the other hand, the distance (Dt)continues to drop from 4.35 in 2005 to 1.81 in 2012, whichdecreased by 58.4%. Moreover, the performance of TOS in differentyears is various. After 2007 when the city was approved as a pilotcity, TOS construction achieved sound and fast development. In2009, the growth rate of TOS dropped significantly for the globalfinancial crisis. The analysed result resonates that the Chinesegovernment pays more attention to ensuring China's rapideconomic growth, and the efforts of environment protectionrelegated to secondary position for the global financial crisis (Songet al., 2011). In addition, the performance of TOS improved mostfrom 2010 to 2011, which Ct declined by 48.31% (from 6.66% in 2010

Fig. 3. The performance of TOS (2005–2012).

X. Chen et al. / Ecological Indicators 50 (2015) 206–214 211

to 3.44% in 2011) and Dt declined by 22.22% (from 2.97 in 2010 to2.31 in 2011).

3.2. Performance of subsystems

Fig. 4 shows the performance of six subsystems of TOS inWuhan during 2005–2012.

The performance of ED greatly improved from 2005 to 2012.Both Ct and Dt of ED are close to the planning targets. Specifically,the value of Ct sharply decreased from 23.9% in 2005 to 0.75% in

Fig. 4. The performance of TOS'

2012. The result indicates that all indicators of ED are coordinateddevelopment.

In SW subsystem, the growth rate of Ct continuously rises from18.16% in 2006 to 52.27% in 2012. It shows that indicators of SW areimproved in balance during the period. However, compared withthe other two indicators of SW, the actual value of urbanizationrate witnesses the greatest difference from the planning target.Therefore, the government should pay more attention to promot-ing urbanization development.

The performance of RC has greatly improved during the period2005–2009. Ct of RC decreases from 3.83% in 2005 to 0.01% in 2009.However, it goes up to 0.51% in 2011 because of the varyingperformance of two indicators of RC in that year. Moreover, theactual values of these two indicators in 2012 are greater than thoseof the planning targets in 2015.

As for RR, its performance appears highly volatile. Both Ct andDt have been decreasing over the 2005–2010 period. Theperformance of two indicators of RR in 2010 is better than theirplanning targets in 2015. However, they declined in 2011 and resultin Ct and Dt increase.

The performance of EQ fluctuates in the period of 2005–2012.Both Ct and Dt of EQ see continuous improvement from 2005 to2008. However, the actual values of the three indicators of EQdeclined from 2009 to 2010. The pressure for the government topromote economic growth has greatly increased after 2008 due tofinancial crisis. Therefore, the government has to pay moreattention to promoting economic development but neglect thedevelopment of EQ. The reason is that GDP growth, which is closelyrelative to officer's promotion (Li and Zhou, 2005), is much moreimportant than environmental protection for government leadersin China.

All indicators of PC have been continuously improved from2005 to 2012 except urban sewage centralized disposal rate, which

s subsystems (2005–2012).

212 X. Chen et al. / Ecological Indicators 50 (2015) 206–214

decreased from 94.96% in 2010 to 88.78% in 2012. The bestperformance can be seen in the indicator of CO2 emissions per unitof GDP which roughly declines from 3.31 in 2005 to 1.39 in 2012.The reason for this phenomenon is that Hubei is a pilot area toconstruct low carbon province. As the capital city of Hubei, Wuhanmakes great efforts to promote carbon emissions reduction, andthe value of CO2 emissions per unit of GDP in 2012 is smaller thanthe planning target.

Comparing the performance of various TOS subsystems, it canbe concluded that ED is the best, and subsystems about resourceconserving (RC, RR) are better than subsystems about environmentfriendly (EQ, PC). The government should be focused more on theconsistency and coordination of different TOS subsystems, such aspaying more attention to the quality of economy than the quantityof it, to recycling of resources rather than consumption of it, and toprotecting the environment rather than controlling of it.

3.3. Analysis of influent factors

Table 4 shows the results of grey relational grades (GRG) amongfactors influencing the performance of TOS.

Firstly, both Ct and Dt of TOS are highly influenced by theexternal economic environment (EEE). The GRG between EEE andCt is 0.898, and that between EEE and Dt is 0.845. Specifically, thetwo indicators of EEE which are X9 and X10 are the first twoindicators influencing Ct of TOS, the GRG between them is0.925 and 0.87, respectively. This finding implies the consistentrelationship between sustainable development and economicgrowth. Moreover, compared with other indicators, the portion ofR&D expenditure in GDP (X7) impacts Dt of TOS most, and GDPgrowth rate of Hubei province (X10) is followed. Besides, theamount of funds for pollution treatment (X2) also shows closerelationships with Ct and Dt of TOS, the GRG are 0.846 and 0.852,respectively.

Secondly, technological ability (TEA) and investment (INV) havethe most significant impacts on ED. The GRG between TEA and Ct is0.845 and the GRG between INV and Ct is 0.839. Specifically,investment in fixed assets (X1) (0.867) sees the most essential,followed by the number of granted patents (X8) (0.865) and theportion of R&D expenditure in GDP (X7) (0.825). These results

Table 4Results of grey relational analysis for factors affecting TOS construction.

TOS ED SW R

Ct Dt Ct Dt Ct Dt C

INV X1 0.652 0.716 0.867 0.641 0.843 0.645 0X2 0.846 0.852 0.812 0.856 0.772 0.789 0Ave 0.749 0.784 0.839 0.749 0.807 0.717 0

GPE X3 0.767 0.802 0.792 0.855 0.746 0.653 0X4 0.791 0.800 0.795 0.889 0.750 0.673 0Ave 0.779 0.801 0.794 0.872 0.748 0.663 0

FIS X5 0.797 0.811 0.797 0.862 0.754 0.675 0X6 0.797 0.754 0.790 0.854 0.745 0.651 0Ave 0.797 0.783 0.794 0.858 0.749 0.663 0

TEA X7 0.843 0.920 0.825 0.889 0.786 0.897 0X8 0.631 0.688 0.865 0.618 0.854 0.614 0Ave 0.737 0.804 0.845 0.753 0.820 0.755 0

EEE X9 0.925 0.830 0.808 0.818 0.769 0.750 0X10 0.870 0.860 0.815 0.889 0.777 0.803 0Ave 0.898 0.845 0.812 0.853 0.773 0.776 0

Note: the value of Ave is calculated by the mean of two indicators in the aspect. Taking i

confirm the viewpoint that China is an investment-driven country.Moreover, GDP growth rate of Hubei province (X10) (0.889) and theportion of R&D expenditure in GDP (X7) (0.889) are the first twoindicators affecting Dt of ED.

Thirdly, the two factors of external economic environment (X9and X10) are of vital importance for Dt of SW, and the GRG are0.803 and 0.75. Furthermore, the two technological ability factorsX8 and X7, and investment in fixed assets (X1) have significantimpacts on Ct of SW, and the GRG are 0.854, 0.786 and 0.843,respectively.

Fourthly, technological ability (TEA) and investment (INV)impact RC the most. The GRG between TEA and Ct is 0.843, and thatbetween INV and Ct is 0.842. The two indicators of TEA (X7 and X8)reflect the attitude and ability of government and businesses tosupport the development of technologies and sciences. The twoindicators of INV (X1 and X2) imply investment in technologicalupgrading and industrial equipment to treat pollution. Both thesefactors can significantly improve resource efficiency and result indecreases of the amount of resource consumption. Moreover,capital support including the government public expenditure(GPE) and financial support (FIS) has a significant impact on Ct ofRR. Dt of RR is mostly affected by the GDP growth rate of Hubeiprovince (X10) (0.845) and the portion of R&D expenditure in GDP(X7) (0.82). The results mean that providing capital support for thebusinesses to invest in improving technology ability is vital topromote RR development.

Lastly, the portion of R&D expenditure in GDP (X7) (0.869) iscritically important to improve Ct of EQ, followed by the growthrate of China (X9) (0.867) which also shows the closest relationshipwith Dt of EQ (0.935). The GRG between EEE and Ct, and thatbetween EEE and Dt of EQ are the greatest, which indicates thateconomic growth is the foundation of EQ improvement. Inaddition, like that of RC, technological ability and investmentare the first two factors influencing Ct of PC. The external economicenvironment (EEE), financial support (FIS) and government publicexpenditure (GPE) impact Dt of PC most. On one hand, it is muchmore possible for the government to shift their attention from thequantity of economics to the quality of that when economy runssmoothly. On the other hand, it is necessary for the companies andother pollution makers to take effective measures, such as

C RR EQ PC

t Dt Ct Dt Ct Dt Ct Dt

.844 0.638 0.757 0.757 0.626 0.625 0.722 0.658

.840 0.815 0.865 0.809 0.851 0.901 0.679 0.770

.842 0.727 0.811 0.783 0.738 0.763 0.701 0.714

.839 0.734 0.868 0.773 0.821 0.825 0.656 0.715

.839 0.759 0.866 0.775 0.853 0.815 0.650 0.737

.839 0.746 0.867 0.774 0.837 0.820 0.653 0.726

.839 0.741 0.878 0.803 0.829 0.846 0.665 0.713

.839 0.769 0.870 0.745 0.841 0.855 0.632 0.721

.839 0.755 0.874 0.774 0.835 0.851 0.648 0.717

.841 0.851 0.839 0.820 0.869 0.846 0.686 0.786

.845 0.610 0.742 0.727 0.603 0.604 0.732 0.619

.843 0.731 0.790 0.773 0.736 0.725 0.709 0.702

.840 0.764 0.860 0.796 0.867 0.935 0.675 0.730

.840 0.800 0.866 0.845 0.861 0.861 0.682 0.751

.840 0.782 0.863 0.820 0.864 0.898 0.679 0.740

nvestment (INS) as an example, Ave of INS is calculated by the mean of X1 and X2.

X. Chen et al. / Ecological Indicators 50 (2015) 206–214 213

promoting equipment and technology improved, as well asreducing pollution to promote PC development. However, theyneed capital support from governments and financial institutions.

4. Conclusions and recommendations

4.1. Conclusions and policy implications

In this study, we presented integrate methods of vector angleand Euclidean distance to evaluate the coordination and effective-ness of TOS in the national pilot city Wuhan. Based on theperformance and determinants discussed above, we can drawconclusions and policy recommendations as follows.

Firstly, the performance of TOS, ED and SW is continuouslyimproved while the performance of the subsystems aboutenvironmental protection (EQ and PC) and resources conserving(RC and RR) fluctuate, because they are greatly affected by thefluctuated external economic environment. The results indicatethat when the external economic environment run smoothly, thesubsystems about environmental protection and resources con-serving will show a positive performance. The reason is thatChinese government officials prefer the quantity of GDP ratherthan the quality of economics because the quantity of GDP isclosely related to their professional promotion (Li and Zhou, 2005).Therefore, the aspects about the quality of economic developmentsuch as pollutants and emissions are considered less importantcompared to those about the quantity of economics growth.Consequently, it is necessary to reform the performance evaluationsystem of government officials to considering equally important oneconomic growth, environmental protection and resources con-serving. As such, government officials can be encouraged topromote both pollutants and emissions reduction and GDP growthsimultaneously.

Secondly, the government should balance the development ofvarious aspects of TOS to promote its improvement. From thispaper, we can see that the performance of coordination andeffectiveness of TOS and its subsystems are different. Both Ct andDt of TOS, ED and SW are continually improved. The values of themcontinuously close to the planning targets. The best performance ofCt and Dt can be seen in RC. The two indicators of RC realize theplanning target. Moreover, Dt of RR shows a downward trend whileCt of RR fluctuates in the study period except the year of 2010. Theinconsistent performance of the two indicators of RR contribute tothis difference, for utilization rate of industrial water for irrigationdropped by 13.1% from 2010 to 2012 while the ratio of industrialsolid wastes utilized decreased by 3.58% at the same time.Although Dt of PC shows a downward trend, Ct of PC expanded in2012. The reason is that indicators of PC show various perform-ances. For example, CO2 emissions per unit GDP realize theplanning target while urban sewage centralized disposal rate is faraway from the target. In addition, the inconsistent performance ofindicators in EQ result in fluctuated performance of its Ct and Dt.These results indicate that in order to promote the development ofTOS, instead of focusing on one aspect of TOS, such as developinglow-carbon city to promote carbon abatement or building water-conserving city to promote water saving, the government shouldpay attention to the balanced development of various aspects ofTOS, such as considering conserving water and energy, promotingcarbon, COD and SO2 abatement at the same time. Furthermore,the government should not only focus on the existing pollution andemissions, but also control the incremental quantity of those torealize stable improvements of these indicators.

Finally, government should take different strategies to deal withthe balanced development of TOS because the determinantsimpact TOS and its subsystems are various. In terms of RC and PC,technological abilities and investment are the most important

factors affecting their performance. Therefore, governments andenterprises should increase investment to promote upgradingequipment and technologies. As for RR, it is necessary to promotebusinesses recycling and reusing the wasted resources such aswater, etc. Therefore, the government should provide moregovernment public expenditure and encourage financial institu-tions such as commercial banks to provide more loans to stimulateand support the enterprises. Moreover, increasing the amount ofR&D funding and maintaining steady external economic environ-ment will be effective to improve all aspects of TOS.

4.2. Research limitations

This paper proposed a methodological framework by integrat-ing the methods of vector angle and Euclidean distance to evaluatethe performance of TOS from the perspectives of the coordinationand effectiveness. Based on this method, the performance of TOS inWuhan and the core factors influencing the performance arediscussed in-depth. However, because of the restriction of dataavailability, it is hard to choose indicators cover all aspects of theTOS. For example, indicators used in RC and RR are mainly relatedto industry production while only a few of them are related tohuman lives. Moreover, since the key points of developingeconomics in different years are different, indicators withweighting can be considered to use in this model in futureresearch.

Acknowledgements

Thanks to two anonymous reviewers for their constructivecomments. This work has been funded by the Foundation forInnovative Research Groups of the National Natural ScienceFoundation of China (No. 71221061), the National Natural ScienceFoundation of China (No. 71431006, No. 71271216), and KeyProjects of Philosophy and Social Sciences Research, Ministry ofEducation China (No. 13JZD0016).

References

Hammond, A., Adriaanse, A., Rodenburg, E., Bryant, D., Woodward, R., 1995.Environmental Indicators: A Systematic Approach to Measuring and Reportingon Environmental Policy Performance in the Context of SustainableDevelopment. World Resources Institute, Washington, DC, pp. 43.

Ang, J.B., 2009. CO2 emissions, research and technology transfer in China. Ecol. Econ.68 (10), 2658–2665.

Böhringer, C., Jochem, P.E., 2007. Measuring the immeasurable – a survey ofsustainability indices. Ecol. Econ. 63 (1), 1–8.

Conroy, M.M., Berke, P.R., 2004. What makes a good sustainable development plan?An analysis of factors that influence principles of sustainable development.Environ. Plan. A 36 (8), 1381–1396.

Dalal-Clayton, B., Bass, S., 2006. A Review of Monitoring Mechanisms for NationalSustainable Development Strategies. IIED, London Report prepared for theOECD.

Deng, J.L., 1989. Introduction to grey system theory. J. Grey Syst. 1 (1), 1–24.Deng, J.L., 2010. Introduction to Grey Mathematical Resource Science. Huangzhong

University of Science & Technology Press, Wuhan, China.Duwairi, R.M., 2006. A framework for the computerized assessment of university

student essays. Comput. Hum. Behav. 22 (3), 381–388.Ebert, U., Welsch, H., 2004. Meaningful environmental indices: a social choice

approach. J. Environ. Econ. Manage. 47 (2), 270–283.Fisher-Vanden, K., Sue Wing, I., 2008. Accounting for quality: issues with modeling

the impact of R&D on economic growth and carbon emissions in developingeconomies. Energy Econ. 30 (6), 2771–2784.

Fleiter, T., Plötz, P., 2013. Diffusion of energy-efficient technologies. In: Shogren, J.F.(Ed.), Encyclopedia of Energy, Natural Resource, and Environmental Economics.Elsevier, Waltham, pp. 63–73.

Hopwood, B., Mellor, M., O’Brien, G., 2005. Sustainable development: mappingdifferent approaches. Sustain. Dev. 13 (1), 38–52.

Huang, J., Yang, J., Rozelle, S., 2006. China's rapid economic growth and itsimplications for agriculture and food security in China and rest of world. Part 1,Rapid Growth of Selected Asian Economies: Lessons and Implications forAgricultural and Food Security – China and India, Policy Assistance Series 1/2.

214 X. Chen et al. / Ecological Indicators 50 (2015) 206–214

United Nations Food and Agriculture Organization (FAO) Regional Office for Asiaand the Pacific, Bangkok.

Li, F., Liu, X., Hu, D., Wang, R., Yang, W., Li, D., Zhao, D., 2009. Measurementindicators and an evaluation approach for assessing urban sustainabledevelopment: a case study for China's Jining City. Landsc. Urb. Plan. 90 (3),134–142.

Li, H., Fang, K., Yang, W., Wang, D., Hong, X., 2013. Regional environmental efficiencyevaluation in China: analysis based on the Super-SBM model with undesirableoutputs. Math. Comput. Model. 58 (5), 1018–1031.

Li, H., Zhou, L.A., 2005. Political turnover and economic performance: the incentiverole of personnel control in China. J. Public Econ. 89 (9), 1743–1762.

Moldan, B., Janoušková, S., Hák, T., 2012. How to understand and measureenvironmental sustainability: indicators and targets. Ecol. Indic. 17, 4–13.

OECD, 2008. Handbook on Constructing Composite Indicators: Methodology andUser Guide. OECD Publishing.

Parris, T.M., Kates, R.W., 2003. Characterizing and measuring sustainabledevelopment. Annu. Rev. Environ. Resour. 28 (1), 559–586.

Qin, D., Cagas, M.A., Quising, P., He, X.H., 2006. How much does investment driveeconomic growth in China? J. Policy Model. 28 (7), 751–774.

Rennings, K., 2000. Redefining innovation – eco-innovation research and thecontribution from ecological economics. Ecol. Econ. 32 (2), 319–332.

Rinne, J., Lyytimäki, J., Kautto, P., 2013. From sustainability to well-being: lessonslearned from the use of sustainable development indicators at national and EUlevel. Ecol. Indic. 35, 35–42.

Singh, R.K., Murty, H.R., Gupta, S.K., Dikshit, A.K., 2012. An overview of sustainabilityassessment methodologies. Ecol. Indic. 15 (1), 281–299.

Song, M., Wang, S., Yu, H., Yang, L., Wu, J., 2011. To reduce energy consumption and tomaintain rapid economic growth: analysis of the condition in China based onexpended IPAT model. Renew. Sustain. Energy Rev. 15 (9), 5129–5134.

Wei, J., Jia, R., Marinova, D., Zhao, D., 2012. Modeling pollution control andperformance in China's provinces. J. Environ. Manage. 113, 263–270.

Wu, H.Q., Shi, Y., Xia, Q., Zhu, W.D., 2014. Effectiveness of the policy of circulareconomy in China: a DEA-based analysis for the period of 11th five-year-plan.Resour. Conserv. Recycl. 83, 163–175.

Xie, Y., Costa, F.J., 1993. Urban planning in socialist China: theory and practice. Cities10 (2), 103–114.

Yang, Q., Gao, Q.Q., Chen, M.Y., 2011. Study and integrative evaluation on thedevelopment of circular economy of Shaanxi province. Energy Proc. 5,1568–1578.

Ye, J., 2011. Cosine similarity measures for intuitionistic fuzzy sets and theirapplications. Math. Comput. Model. 53 (1), 91–97.

Yin, M.S., 2013. Fifteen years of grey system theory research: a historical review andbibliometric analysis. Expert Syst. Appl. 40 (7), 2767–2775.

Zeemering, E., 2012. Recognising interdependence and defining multi-levelgovernance in city sustainability plans. Local Environ. 17 (4), 409–424.

Zhang, K.M., Wen, Z.G., 2008. Review and challenges of policies of environmentalprotection and sustainable development in China. J. Environ. Manage. 88 (4),1249–1261.

Zhang, X.H., Hu, H., Han, J., Deng, S.H., Xiao, H., Peng, H., Qi, H., 2012. Several novelindicators being applied to analyze the relationships between Chineseeconomic growth, energy consumption and its impact of emissions. Ecol. Indic.15 (1), 52–62.