As environmental pollution and resource crises intensified, policymakers are increasingly concerned about the impact of economic activities on the ecological environment. Urban agglomerations are not only the core areas leading economic growth but also the fronts facing severe resource and environmental challenges. The spatial differentiation of economic development and of ecological environment exist simultaneously within urban agglomerations. To promote green and coordinated development, it is urgent to improve the performance appraisal system of regional development and to optimize the interregional cooperation mechanism. Eco-efficiency, which takes into account factors such as resources, environment, and economy, can reflect green development performance comprehensively, and therefore, providing an important reference for measuring the long-term development advantages of a region and formulating sustainable development policies. Undoubtedly, the calculation of urban eco-efficiency and the identification of its influencing factors are prerequisites for achieving sustainable urban development [1
The concept of eco-efficiency was first introduced by Schaltegger and Sturm [2
] and has attracted much attention since [3
]. The World Business Council for Sustainable Development (WBCSD) defines eco-efficiency as providing cost-effective services and products to meet human needs and high-quality life; at the same time, these services and products can reduce environmental impacts and resource consumption to levels that match the Earth’s carrying capacity [4
]. According to Kuosmanen [5
], eco-efficiency emphasizes acquiring economic output with minimal natural resources consumption and environmental degradation. Since eco-efficiency essentially reflects the effectiveness of input and output, it is suitable for the application of frontier analysis methods, which mainly including the Data Envelopment Analysis (DEA) method [6
] and the Stochastic Frontier Analysis (SFA) method. Compared with the SFA model, the DEA model provides a more straightforward and more flexible estimation method, as there is no need for knowledge of the functional relationship between input and output, let alone assigning input and output weights [7
], which is why it has gained more popularity.
Various models have been developed to evaluate eco-efficiency. For instance, Korhonen [8
] estimated the eco-efficiency of 24 European power plants using an extended DEA model that treats pollutants as the inputs. Sarkis [9
] provided six DEA models and some extensions to assess the ecological efficiency of power plants. Zhang et al. [10
] took six kinds of environmental pollutant emissions together with water resources, raw mining resources, and energy as the inputs of DEA to measure ecological efficiency. However, as shown in many studies, conventional DEA models have some drawbacks. An eco-efficiency derived from a contemporaneous production possibility set (PPS) may encounters problems of misleading technical regress and also suffers non-circularity and linear programming infeasibility when measuring cross-period directional distance functions (DDF), which are referred to as “discriminating power problem” and “technical regress” [11
]. To overcome these problems, scholars have developed many improved approaches. Tone [13
] put forward the super-efficiency model based on the non-angular, non-radial Slacks-based Measure (SBM) model to address the “discriminating power problem”. Additionally, Pastor and Lovell [15
] proposed a new model based on global benchmark technology. Oh [16
] further extended it with sequential technology, which is circular and can overcome “technical regress”, and thus, has been widely used in the estimation of eco-efficiency in recent years [17
]. For example, using an improved SBM model combines global benchmark technology, directional distance function and a bootstrapping method with the Global Malmquist Luenberger (GML) index, Yang and Zhang [18
] analyzed the dynamic trends of eco-efficiency of 30 sample provinces in mainland China from 2003 to 2014. Adopting the SBM model, Zheng et al. [19
] measured the eco-efficiency of 31 Chinese regions from 2000 to 2015. Furthermore, based on a Meta-US-SBM model (meta-frontier, undesirable outputs, and super-efficiency SBM), Huang et al. [20
] estimated the composite eco-efficiency using in China’s provincial data from 2001 to 2014.
Some studies have explored the factors that influence eco-efficiency. For example, Zhang et al. [21
] examined the factors affecting industrial eco-efficiency based on the three-stage DEA model using Chinese provincial data between 2005 and 2013, and their results showed that China’s regional industrial eco-efficiency was majorly affected by the factors of the environmental regulations, technological innovations, as well as level of economic development and industrial structure. Bai et al. [22
] quantitatively investigated the relationship between urbanization and urban eco-efficiency using the data of 281 prefecture-level cities in China from 2006 to 2013. Their results showed that apparent spatial disparities exist among different cities, and there was an N-shaped relationship between urbanization and urban eco-efficiency. Ren et al. [23
] analyzed the effects of three types of environmental regulations, namely, command-and-control, market-based, and voluntary regulation, on the eco-efficiency of China’s three regions based on the Stochastic Impacts by Regression on Population, Affluence and Technology (STIRPAT) model. Their findings revealed that the effects of different types of environmental regulation on eco-efficiency vary from region to region. Huang et al. [24
] investigated the impacts and mechanisms of the urban cluster on urban eco-efficiency. They find that the improvement of the urban cluster is conducive to enhancing urban eco-efficiency, and there is a “core-periphery” spatial structure in the process of urban cluster development. Li et al. [25
] studied the relationship between government transparency and eco-efficiency utilizing the data of 262 cities in China from 2005 to 2012. Their results suggested that the overall eco-efficiency of Chinese cities was low, and a nonlinear relationship exists between government transparency and eco-efficiency performance.
However, despite all the fruitful results and substantial advances, there are still some limitations in the previous literature. Firstly, compared with numerous studies that focused on the provincial and sectoral level [21
], much less attention has been paid to the eco-efficiency of cities, especially that of the emerging cities in developing countries [26
]. Considering that urban areas make a tremendous contribution to resource consumption and pollution emissions in the developing world, it is of great significance to carry out in-depth research on their green development. Another essential drawback is that most studies ignored the spatial correlations between cities when discussing the influencing factors of urban eco-efficiency. However, apparent neglect of the spatial spillover effects, which tend to become significant with the intensification of interregional economic connections and the more frequent flow of factors of production, could result in biased estimations [1
To address these limitations of extant studies, this paper extends research at the provincial and sectoral level to the urban level utilizing a prefecture-level panel dataset in the Yangtze River Delta Urban Agglomeration (YRDUA), China, during the period of 2003–2016, and constructs a model that incorporated the super-efficiency DEA model with slacks-based-measure and global-frontier technology (SSBM-GF) to estimate total-factor eco-efficiency (TFEE). Moreover, to take into account spatial spillover effects, a spatial panel Tobit model is constructed to analyze the influencing factors of urban eco-efficiency. The results are supportive of understanding the spatial difference and driving mechanism of urban eco-efficiency in the YRDUA, providing a scientific basis for governments to formulate policies to promote the development of green and sustainable urbanization.
The remainder of the paper is organized as follows. Section 2
introduces the methodology and data sources. Section 3
presents the empirical results. Section 4
provides some discussion and implications. Section 5
provides a conclusion.
4. Discussion and Implications
This study shows that there were significant regional disparities of TFEE in the YRDUA. The cities with high eco-efficiency were majorly located in the coastal areas, while the cities with low eco-efficiency were mostly situated in the inland parts throughout the study period. This finding is similar to Xing et al. [56
], which suggests that these spatial disparities are primarily attributable to geographical and economic differences among areas. Compared with the cities in the inland areas, the cities in the coastal areas are more developed. They have better infrastructure, more advanced technologies and more stringent environmental regulation, which contribute to improving resource utilization and reducing pollutant emissions and thus promoting urban eco-efficiency.
TFEE in the YRDUA and its four regions showed a similar trend of “decline first and then rise with fluctuation” during the period 2003–2016. The temporal evolution of TFEE in the YRDUA generally reflected the transformation of China’s economic development model during this period. After joining the WTO in 2001, China started a process of rapid industrialization. As the frontier for opening up to the outside world, the Yangtze River Delta quickly developed a thriving manufacturing industry during this period, and consequently put enormous pressure on resources and environment. Therefore, TFEE in the YRDUA showed a downward trend during 2003–2005. In response to the deteriorating ecological environment, the Chinese central government introduced a series of regulatory policies and increased environmental governance. For instance, Export Control of High-energy-consuming, High-polluting and Resource-intensive Products (2005), Assessment of Corporate Environmental Behavior (2005), Government Procurement List of Energy-saving Products (2006), Emissions Trading Scheme (2006), Environmental Liability Insurance System (2007), Project for Reducing Major Pollutants Emission (2007), Energy Conservation Law (2008), etc. However, stimulated by the blooming external demand before the global financial crisis, China’s economy was heavily reliant on resources, presenting a characteristic of high pollution and high growth. Thus, economic growth at this stage was still not coordinated with environmental improvement in the YRDUA; thus, TFEE in this area showed a fluctuating state during the period 2005–2012. In the post-financial crisis era, China gradually strengthened ecological protection and environmental governance and proposed an “ecological civilization construction” strategy. As the most developed region in China, the Yangtze River Delta has taken the lead in ecological construction by accelerating the elimination of low-end industries and encouraging innovation, which has significantly reduced pollution and improved resource utilization efficiency. Accordingly, a steady increase of TFEE in the YRDUA has been observed during the period 2012–2016.
Therefore, some policy implications for achieving green and sustainable development in urban areas can be put forward based on this research. Firstly, the pace of integrated development of the Yangtze River Delta Urban Agglomeration should be accelerated. There was an apparent disparity of ecological efficiency among cities in the Yangtze River Delta, and the gap between inland and coastal cities has been expanding, which is not conducive to the construction of regional ecological civilization. To achieve coordinated and shared development, it is urgent to strengthen guidance and support for the backward areas and give full play to positive spillover effects. While strengthening the responsibility for ecological protection, it is also indispensable to promoting the free flow of factors of production and encouraging the dispersion of advanced technologies and industries to backward areas, thereby gradually eliminate regional inequality.
Secondly, the improvement of ecological efficiency should be set as one of the core factors in the performance assessment system for local government. In a decade (2003–2012), the ecological efficiency of the Yangtze River Delta urban agglomeration has been stagnated. The fundamental reason is that the old development model places too much emphasis on GDP, thus neglecting resources, environment, and ecology. In recent years, especially after 2012, the reform of the performance assessment system for local government has had a profound impact on the resources and environment. The central government, whose incentives play a vital role in shaping regional plans [57
], has firmly strengthened the supervision of the ecological environment, placing more focus on the preservation of natural resources and the improvement of environmental quality. Accordingly, the urban eco-efficiency of the Yangtze River Delta region has achieved steady growth since 2012. However, Chinese cities still have a lot of room to improve their eco-efficiency, which depends to a great extent on further reforming the performance evaluation mechanisms to improve green development.
Thirdly, a timely enhancement of environmental regulations is critical for developing economies to achieve high-quality development. As industrialization goes to a certain stage, developing economies should raise their environmental standards promptly, and put a limit on the energy-intensive and high-pollution industries, and therefore avoid becoming the “pollution heaven” for FDI. For the Yangtze River Delta Urban Agglomeration, more stringent environmental regulations are needed in the future to formulate and implement regional integrated environmental policies.
Furthermore, steady investing in innovation is indispensable to promoting eco-efficiency. Innovation is vital to the development of emerging industries and is the fundamental driving force for achieving sustainable development. Therefore, the government must continuously invest financial resources to support scientific research, technology development and business innovation in multiple dimensions, and create an efficient innovation system. Furthermore, the increase in the proportion of the tertiary industry can reduce resource consumption and pollution emissions, thereby improving ecological efficiency. Hence, the government should emphasize the transformation and upgrading of industries in the backward cities and encourage the development of service industries.
Finally, population concentration should be well-guided to make full use of scale effects and agglomeration effects. There was a weak inverted U-type relationship between population density and eco-efficiency in the YRDUA; thus, differentiated urbanization policies should be formulated for cities of different sizes. For small and medium-sized cities, population concentration should be further enhanced to make full use of scale effects and agglomeration effects to improve economic efficiency and environmental efficiency. However, for some big cities such as Shanghai, Hangzhou, and Nanjing, properly control of the population density is essential to prevent “big city diseases” from threatening sustainable development.
It is of great significance to carry out in-depth research on the dynamics and driving mechanisms of eco-efficiency of urban areas in developing countries since they make a tremendous contribution to resource consumption and pollution emissions in the developing world. This paper extends research at the provincial and sectoral level to the urban level utilizing a prefecture-level panel dataset in the Yangtze River Delta Urban Agglomeration, China, between 2003 and 2016, and proposes an SSBM-GF model that incorporated the super-efficiency DEA model with slack-based-measure as well as global-frontier technology to estimate total-factor eco-efficiency. Moreover, to take into account spatial spillover effects, a spatial lag Tobit model is constructed to analyze the factors influencing urban eco-efficiency.
Our measure revealed that there were great regional disparities of TFEE in the YRDUA; the cities with high eco-efficiency were majorly located in the coastal areas, while the cities with low eco-efficiency were mostly situated in the inland areas throughout the study period. TFEE in the YRDUA and its four regions demonstrated a similar trend of “decline first and then rise with fluctuation” during the period 2003–2016. The regression analysis shows that there was a noticeable positive spatial spillover effect of eco-efficiency across cities in the YRDUA. An increase in the proportion of tertiary industry could promote the improvement of eco-efficiency. Moreover, the impacts of environmental regulation and innovation on eco-efficiency both were significantly positive in the YRDUA. Notably, the inflow of FDI was not conducive to the growth in eco-efficiency in this region. The relationship between population intensity and eco-efficiency is an inverted U-shaped curve. The results are supportive of understanding the spatial difference and driving mechanism of urban eco-efficiency in the YRDUA, providing a scientific basis for governments to formulate policies to promote the development of green and sustainable urbanization.
This study inevitably has some limitations, which in turn point to directions for future research. Firstly, this paper used an SSBM-GF model based on the DEA method to estimate the total-factor eco-efficiency. Although DEA has some advantages, it is a non-parametric mathematical programming approach that does not consider statistical noise, which might lead to biased measures to a certain extent. Moreover, our analysis merely focused on the Yangtze River Delta region; therefore, the findings should not be taken as an accurate depiction of the overall picture of the urban development across China. Nevertheless, this approach can be extended to more parts of China and other countries without difficulty.