- Email: [email protected]
Supply-side structural effects of air pollutant emissions in China: A comparative analysis
G Model
ARTICLE IN PRESS
STRECO-712; No. of Pages 7
Structural Change and Economic Dynamics xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
Structural Change and Economic Dynamics journal homepage: www.elsevier.com/locate/sced
Supply-side structural effects of air pollutant emissions in China: A comparative analysis Rui Xie a , Fangfang Wang a , Julien Chevallier b,c,∗ , Bangzhu Zhu d,e,∗ , Guomei Zhao a a
College of Economics and Trade, Hunan University, Hunan, PR China IPAG Business School, IPAG Lab, 184 Boulevard Saint-Germain, 75006 Paris, France Université Paris 8, LED, 2 avenue de la Liberté, 93526 Saint-Denis cedex, France d School of Management, Jinan University, Guangdong, PR China e Business School, Nanjing University of Information Science & Technology, Nanjing 210044, PR China b c
a r t i c l e
i n f o
Article history: Received 19 January 2017 Received in revised form 13 February 2018 Accepted 18 April 2018 Available online xxx JEL classification: N50 N70 N90 057 Keywords: Ghosh input-output model Air pollution Structural decomposition analysis Supply-side structure China International comparative analysis
a b s t r a c t China’s air quality has become a major issue affecting people’s livelihood and continues to deteriorate in recent times. It is an important issue of common concern for economists and policymakers to explore the drivers of the growth of air pollution emissions and the deteriorating environmental quality in China. From the perspective of supply-side structures, this paper adopts Ghosh input-output model to decompose the factors affecting the changes of air pollutant emissions into economic activities, economic structures, allocation structures and emission intensity. Using this model, we conduct a structural decomposition analysis of air pollutant emissions in China, India, USA, and Japan for 1995–2009. The results reveal that China’s economic structure initially promoted air pollutant emissions, but later played a role in reducing them. Further, whereas in Japan and particularly, China, allocation structures were found to be a key factor in increasing air pollutant emissions, in America and India, it played a critical role in reducing emissions. Our findings suggest that adjusting the distribution structure of intermediate products is crucial to reduce air pollution. © 2018 Elsevier B.V. All rights reserved.
1. Introduction China’s economy has rapidly grown since the reform and opening up policy, and since then, has surpassed that of Japan, becoming the second largest economy in the world in 2010. However, China economic development is largely dependent on the extensive input of factors, such as labor, capital, and resources, and industrial sectors with high energy consumption, and thus, is subject to severe environmental pollution and energy shortage (Yang et al., 2014; Xie, 2014). In particular, since 2013, the Beijing-TianjinHebei region, Pearl River Delta, and Yangtze River Delta have been frequently experiencing haze, indicating increasing threats to the climate, human health, and development of a harmonious society (He et al., 2013).
∗ Corresponding authors. E-mail addresses: [email protected] (J. Chevallier), [email protected] (B. Zhu).
Preliminary research on the causes of haze in China attributes its formation to two factors. First is the obstruction in the horizontal flow of air and the lack of diffusion given static weather conditions, an external cause which cannot be controlled by human behavior. Second, oxysulfide (SOx) and oxynitride (NOx) emissions from fire coal, motor vehicles, and industrial production, which convert into secondary particulate pollutants through chemical reactions in the air and further aggravate the formation of haze (He et al., 2013; He and Jiang et al., 2014). Therefore, a fundamental method to curb haze is to strictly control SOx and NOx emissions and prevent the formation of secondary particulate pollutants. According to the World Input-Output Database (WIOD), China’s total air pollutant emission (total emission including SOx and NOx) was 35.40 Mt in 1995. While this figure decreased to 32.87 Mt in 2001, it surged in 2009 to 65.77 Mt, increasing by more than 100%. Therefore, an in-depth exploration of the factors affecting China’s air pollutant emissions is needed to control the formation of haze and the severe deterioration of air quality. The method of decomposition is often used to analyze the factors influencing envi-
https://doi.org/10.1016/j.strueco.2018.04.005 0954-349X/© 2018 Elsevier B.V. All rights reserved.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model
ARTICLE IN PRESS
STRECO-712; No. of Pages 7
R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
2
ronmental pollution and classify various factors contributing to the total change of pollutant emission (Ang, 1995; Ang et al., 2003; Su et al., 2017; Meng et al., 2017). The two strains of the decomposition methods are the index decomposition analysis (IDA) and structural decomposition analysis (SDA). In analyzing the factors influencing energy use and pollutant emission, IDA further categorizes total effects as independent effects using aggregated sector data. Many scholars have adopted IDA to analyze changes in energy intensity, carbon emissions and pollutant emissions (Ang and Liu, 2001; Lyu et al., 2016; Meng et al., 2018). Specifically, IDA methods enjoy relatively lower requirements for data and are convenient to operate. But IDA method is unable to conduct an in-depth analysis of direct effects and indirect effects among economic activities, and not allowed for a detailed decomposition of economic structures. SDA methods can overcome some of the drawbacks of IDA method (Su and Ang, 2012, 2016), which examine direct and indirect effects using Leontief and Ghosh inverse matrices (Hoekstra and Van den Bergh, 2003). SDA method is applied in conjunction with the input-output model to examine factors underlying energy intensity and pollutant emissions (Rose and Casler, 1996; Nie et al., 2016; Wang et al., 2017; Mi et al., 2017). However, SDA method requires large amounts of data, as SDA method decomposes changes in various indicators using the input-output model and data from corresponding input-output tables, which distinguish between a range of technological and final demand effects. With improvements in input-output tables, a growing number of researchers are applying the SDA model to analyze factors influencing the environment. Therefore, this article also uses this method. Many studies have used the Leontief model to examine the factors affecting environment pollutants from a demand-side perspective. Weir (1998) and Rormose and Olsen (2005) used the SDA model to identify factors underlying Danish pollutant emissions including CO2 , SO2 , and NOx during 1966–1988 and 1980–2002. Guan et al. (2008, 2009) explored the factors influencing China’s carbon emission growth during 1981–2002 and 2002–2005. Wood (2009) investigated the causes of greenhouse gas emissions in Australia from 1974 to 2005. Using SDA, Zhang (2009) analyzed the historical change trend for China’s carbon intensity related to energy for 1992–2006. Li et al. (2014) and Xie (2014) determined the factors contributing to the growth in China’s energy consumption and use. Li and Wei (2015) determined factors affecting China’s carbon emission and Zhang et al. (2015) did similar research for China’s pollutant levels, including COD, NH3-N, SO2 , and NOx. Su and Ang (2017) used the SDA method to analysis the aggregate embodied energy and emission intensities. Based on spatial SDA, Meng et al. (2017) proposed an alternative input-output to elucidate inter-regional spillover effects in determining China’s regional CO2 emissions growth. Su et al. (2017) used the input-output (IO) method to analyze the city state’s carbon emissions from the demand perspective and used the SDA method to investigate the Table 1 Basic form of single regional input-output table. Output
Intermediate Input
Value Added
Total Input
Sector 1 . . Sector35
2.1. Data source and explanation We adopt single regional input-output table data for two developing countries, China and India, and two developed countries, the United States and Japan, from the World Input-Output Database (WIOD) for 1995–2009. Table 1 presents the basic form of the table. SOx and NOx are key precursor pollutants contributing to the formation of haze (He and Jiang et al., 2014). Thus, we retrieve emission data for SOx and NOx for 1995–2009 from environmental accounts by WIOD and combine the data for both by sector. We then use them as indicators to measure air pollutant emissions. In Table 1, Zis the matrix of intermediate inputs and zij denotes the input demand for sector i from sector j. V is a value-added row vector and vi represents the value added for sector i. Fis the final demand (used) column vector. Xis the output column vector and xi denotes the total output of sector i. 2.2. Theoretical model Consider the column identity of a typical input-output table in monetary terms: (1)
Then, Eq. (1) can be rewritten as
Sector 1 ... Sector 35
Final Demand Total Output
Z
F
V ... Vn V ROW X ...
2. Theoretical model and data
X = HT X + V T ,
Intermediate Demand
Input
drivers of emission changes, which was the first comprehensive analysis of Singapore’s emissions using the I-O framework. However, few articles have used Ghosh input-output model to analyze air pollution issues from a supply-side perspective (e.g. Zhang, 2010). China’s economic structure is undergoing major transformations, as a result of which its demand management faces limitations such as imbalances between domestic and external demands or investment and consumption. Against this background, supply policies can be used as a tool (Cai et al., 2008; China’s Economic Growth Report, 2010; Zhang, 2010) by the government to exercise direct control and reduce policy uncertainties and risks. Thus, this study examines the factors influencing pollutant emissions from a supply-side perspective and accordingly, offers policy suggestions to reduce emission and conserve energy. This study makes the following contributions: (i) we provide an analysis of the factors influencing air pollution, and (ii) we address the key issue of increased haze formation in China. We adopt the SDA approach using Ghosh input-output model from a supply-side perspective. Then, using WIOD’s input-output tables for 40 countries with 35 consolidated sectors for 1995–2009 and air pollutant data including SOx and NOx, we compare the different factors between developed and developing countries. It provides an international reference for China’s efforts to control haze. The remainder of this paper is organized as follows. Section 2 details the theoretical model and data. Sections 3 and 4 analyze air pollution emissions and the factors affecting them. Section 5 offers a summary and policy suggestions.
X T = V (I − H)−1 = VG,
(2)
X
In Eq. (2), we present Ghosh’s (1958) model, where H is the direct distribution matrix in which hij = zij ⁄xi . Here, hij represents the share of distributed output from the ith sector to the jth sector. G = (I − H)−1 is the Ghosh inverse matrix that indicates the dependence between the production sector and the sector that uses its products. Ghosh’s supply-driven input-output model takes primary inputs (sectoral value-add) as the starting point and accounts for both direct and indirect relationships between sectors.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model
ARTICLE IN PRESS
STRECO-712; No. of Pages 7
R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
Next, similar to Zhang (2010), we estimate air pollutant emissions across countries from a supply-side perspective using Eq. (2). P = X T E = VGE =
SGE,
p1 − p0 = (
(3)
where P is total air pollutant emission, Eis the emission intensity column vector, ei = pi ⁄xi , and is a scalar denoting the total volume of sectoral primary inputs (value added), that is, Chinese GDP, accounted for using an income approach. Sis the supply-side economic structure vector, whose ith entry is the share of value added for sector i in GDP. The air pollutants allocated to the sectors can be estimated as follows:
+
1
1 S1 (G1
−
3
0 )S0 G0 E0
− G0 )E0 +
diag(SG)E,
(4)
where p is the column vector for sectoral air pollutant emissions, whose element pi represents the ith air pollutant emissions. Structural decomposition is a popular descriptive tool and is used to identify the sources of historical changes (Dietzenbacher and Los, 1998; 2000). Using Eq. (3), changes in air pollutant emissions can be decomposed as p1 − p0 =
1 S1 G1 E1
−
0 S0 G0 E0 ,
(5)
where subscripts 1 and 0 represent time points. If the decomposition results in n variable factors, n first-order difference equations can be obtained. In this study, factor decomposition is conducted using Dietzenbacher and Los (1998) bipolar decomposition method. First, consider the following decomposition methods: p1 − p0 = ( +
0 S0 (G1
1
−
0 )S1 G1 E1
− G0 )E1 +
+
0 (S1
0 S0 G0 (E1
− S0 )G1 E1
− E0 ).
(6)
1 (S1
1 S1 G1 (E1
− S0 )G0 E0
− E0 ).
(7)
Next, take the average value for both methods: p1 − p0 =
1 ( 2
1
−
0 )(S1 G1 E1
+ S0 G0 E0 ) +
1 2
0 (S1
− S0 )G1 E1 +
Economic activity
+
1 [ 2
0 S0 (G1
− G0 )E1 +
1 S1 (G1
1 (S1
− S0 )G0 E0
Economic structure
− G0 )E0 ] +
Allocation structure
p = diag(X T )E = diag(VG)E =
+
1 ( 2
0 S0 G0
+
1 S1 G1 )(E1
− E0 )
(8)
Emissions intensity
The decomposition in Eq. (8) shows the sources of air pollutant emission can be further divided into economic activity, economic structure, allocation structure, and emissions intensity effects. Economic and allocation structure effects reflect the change in air pollutant emissions from a supply-side perspective. 3. Total air pollutant emissions analysis 3.1. Analysis of Air Pollutant Emissions in China China’s total air pollutant emissions experienced a slight decrease during 1995–2001, followed by a sharp increase between 2002 and 2009. According to Fig. 1, China’s total emissions decreased from 35.4 Mt in 1995 to 32.87 Mt in 2001 with a difference of 2.16 Mt, which accounts for 7.85% of total emissions change. However, after China’s accession to the WTO, in 2003, the total air pollutant emissions rapidly increased by 14% and by 2009, it grew by 29.62 Mt to 65.77 Mt, accounting for 107.85% of the rise in emission during 1995–2009.
Fig. 1. Total emissions and growth rate of air pollutants in China, 1995–2009.
Fig. 2. Emissions trend for major pollutant sectors in China, 1995–2009.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model STRECO-712; No. of Pages 7 4
ARTICLE IN PRESS R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
As shown in Fig. 1, China’s total air pollutant emissions have an obvious turning point in 2001, indicating a crucial change in China’s economic growth pattern after its accession to the WTO. Thus, in this study, we conduct an additional analysis of the factors affecting changes in air pollutant emissions during 1995–2001 and 2001–2009. Fig. 2 clearly shows that electricity, gas, and water supply; other non-metallic mineral and basic metals; and fabricated metal are the three main sectors contributing to the increase in China’s total air pollutant emissions. While electricity, gas, and, water supply account for 51% of total emission, other non-metallic mineral and basic metals and fabricated metal have a relatively small share of 10%. A further analysis of sectoral data reveals that electricity, gas, and water supply account for the largest share, especially following the first several years since China was admitted to the WTO. However, this share rapidly declined from 43% in 1995 to 33% in 2009. This is due to the fact that, on one hand, the output of electricity, gas and water supply dropped from $358375 million in 2003 to $185932 million in 2004 according to the input-output table provided by WIOD. Similarly, the China Statistical Yearbook in 2003 and 2004 also shows a downward trend of output in this sector. It means that the decline of output has led to a drop in the sector emissions. On the other hand, as of 2003, the sector of China’s electricity, gas, and water supply has undergone great transformation and expansion, forming an industrial chain, thereby increasing the comprehensive utilization of resources, to achieve a green development, and ultimately reduce environmental emissions. 3.2. International Comparative Analysis of Total Air Pollutant Emissions Fig. 3 illustrates the trend of air pollutant emissions in China, India, the United States, and Japan during 1995–2009. These four countries have a strong international presence and the highest
emission levels among other developing and developed countries. In 2001, China’s total pollutant emissions exceeded that of the United States, becoming the highest emitter of air pollutants. At present, the United States ranks second and continues to show a downward trend, followed by India and Japan; while India has a gradually increasing trend, that of Japan is relatively stable. According to Fig. 4, the intensity of air pollution emissions in developing countries such as China and India are much higher than those in developed countries such as the United States and Japan. Evidently, the drop in China’s air pollution emissions is more rapid than those in the other three countries. During the sample period, the intensity of China’s air pollution emissions steadily declined from 214.21 t/billion dollars in 1995 to 43.41 t/billion dollars in 2009, with an average annual decline of 5.32%. By 1999, India’s intensity of air pollutant emission was higher than that of China, thus recording the highest intensity level. Nevertheless, India appears to show a declining trend with a reduced rate of 3.26%. As for the United States and Japan, their air pollution intensities are far lower than those of China and India. The air pollution intensity in the United States reported an average annual decline rate of 4.41%, while that of Japan recorded an increase of 0.17%. 4. Contributing factors 4.1. Factors affecting China’s air pollution from a supply-side perspective Table 2 shows the SDA results for China’s air pollutants during 1995–2009. We see that economic activity contributed the most to China’s increased air pollutant emissions, while emission intensity played an important role in reducing them. On the other hand, allocation and economic structures, which are two types of supplyside factors, caused air pollutant emissions in China to significantly rise. China’s total air pollutant emissions increased by 27.47 Mt from 1995 to 2009. During the same period, economic activity
Fig. 3. Air pollutant emissions in four countries, 1995–2009.
Fig. 4. Intensity of air pollutant emissions in four countries, 1995–200.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model
ARTICLE IN PRESS
STRECO-712; No. of Pages 7
R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
5
Table 2 SDA of air pollutant emission changes in China, 1995–2009 (Mt). Countries
Period
EconomicActivity
Economic Structure
Allocation Structure
Emission Intensity
Total
China
1995–2001 2001–2009 1995–2009
18.90 61.66 80.56
1.80 −0.84 0.95
0.55 19.01 19.56
−23.40 −50.20 −73.60
−2.16 29.62 27.47
Source: Author estimations. Table 3 SDA of air pollutant changes in major polluting industries in China, 1995–2009 (Mt). Industry
Period
Economic Activity
Economic Structure
Allocation Structure
Emission Intensity
Total
Electricity, Gas, and Water Supply
1995–2001 2001–2009 1995–2009 1995–2001 2001–2009 1995–2009 1995–2001 2001–2009 1995–2009
5.06 15.44 20.50 0.92 2.40 3.31 1.18 2.91 4.09
2.40 −0.24 2.15 −0.42 −0.63 −1.05 −0.43 0.89 0.46
−0.10 5.50 5.40 0.03 −0.01 0.02 0.04 0.32 0.36
−6.58 −19.82 −26.40 −0.92 −0.89 −1.81 −1.32 −3.47 −4.79
0.78 0.88 1.65 −0.39 0.87 0.48 −0.54 0.65 0.11
Other Non-Metallic Mineral
Basic Metals and Fabricated Metal
Source: Author estimations.
increased China’s air pollutant emissions by 80.56 Mt, accounting for 293.26% of total air pollutant emissions, whereas emissions intensity decreased them by 73.60 Mt, accounting for −267.93%. Allocation and economic structures also increased China’s air pollutant emissions by 19.56 Mt (71.02%) and 0.95 Mt (6.55%). From 1995–2001, China’s air pollutant emissions decreased by 2.16 Mt. While economic activity, economic structure, and allocation structure caused China’s air pollutant emissions to increase, emission intensity contributed towards reducing them. Compared with 1995–2009, the economic and allocation structure in 1995–2001 showed significant differences. In 1995–2001, economic structure increased China’s air pollutant emissions by 1.80 Mt (83.33%) and allocation structure brought about a smaller increase of 0.55 Mt. From the viewpoint of these two supply-side factors, China adopted an extensive growth model that promoted industries with high-intensity emissions. Nevertheless, China’s allocation structure of intermediate products was more reasonable and had little impact on its air pollutant emissions. During 2001–2009, China’s air pollutant emissions increased by 29.62 Mt, indicating a higher growth rate than that in 1995–2001. Economic activity and allocation structure caused China’s air pollutant emissions to increase, while emissions intensity and the economic structure caused them to decrease. Compared with 1995–2001, allocation and economic structures showed significant differences. Economic structure reduced air pollutant emissions by 0.84 Mt, whereas allocation structure significantly increased them by 19.01 Mt (64.18%). These results indicate that China’s economic structure moved towards reducing air pollutant emissions, although the effects are not particularly evident. However, the allocation structure of intermediate products was centered on high energy consumption and high emission industries, which led to China’s economic growth being further dominated by a growth extensive model. Thus, supply-side effects (allocation and economic structure) increased China’s air pollutant emissions to 2.35 Mt during 1995–2001 and 18.72 Mt in 2001–2009. Moreover, although the negative effects of economic structure decreased, those of allocation structure increased. This indicates that China’s intermediate products relied heavily on resources and energy and its economy followed a typical extensive model that led to high energy consumption and pollution levels. Table 3 presents the SDA results for China’s major polluting industries—electricity, gas and water supply; other nonmetallic mineral; and basic metals and fabricated metal—during
1995–2009. From this perspective, economic activity, economic structure, and allocation structure were the main factors contributing to the increased air pollutant emissions, whereas emissions intensity played an important role in decreasing them. In 1995–2009, economic activity led to an increase in air pollutant emissions by the electricity, gas, and water supply; other non-metallic mineral; and basic metals and fabricated metal industries by 20.50 Mt, 3.31 Mt, and 4.09 Mt, respectively. Emissions intensity, on the other hand, contributed towards decreasing these emission levels by 26.40 Mt, 1.81 Mt, and 4.79 Mt, respectively. Allocation structure caused an increase of 5.4 Mt, 0.02 Mt, and 0.36 Mt, respectively, accounting for 327.27%, 4.17%, and 327.27%. Economic structure contributed to an increase in emission levels in electricity, gas, and water supply and basic metals and fabricated metal by 2.15 Mt (130.30%) and 0.46 Mt (418.18%) and a decrease of 1.05 Mt (−218.75%.) in other non-metallic minerals. In 1995–2001, economic activity was the main factor contributing to increased air pollutant emissions by the three major polluting industries, whereas emissions intensity played a vital role in decreasing them. Allocation structure increased air pollutant emissions in other non-metallic mineral and basic metals and fabricated metal industries by 0.03 Mt (7.69%) and 0.04 Mt (7.41%) but decreased emissions in electricity, gas, and water supply by 0.10 Mt (−12.82%). By contrast, economic structure increased air pollutant emissions in electricity, gas, and water supply by 2.40 Mt (307.69%) and decreased them in other non-metallic mineral and basic metals and fabricated metals by 0.42 Mt (−107.69%) and 0.43 Mt (−79.63%). In sum, economic structure increased air pollutant emissions in other non-metallic mineral and basic metals and fabricated metal and substantially promoted air pollutant emissions in electricity, gas, and water supply. Allocation structure, on the other hand, had a small effect on air pollutant emissions in the three major polluting industries. In comparison with the decomposition results for 1995–2001, in 2001–2009, economic activity increased pollutant emissions by the major polluting industries, whereas emissions intensity decreased them. Economic and allocation structures also had significant impacts. China’s economic structure transformed from a factor that increased pollutant emissions from electricity, gas, and water supply to one that decreased them, while the reverse is true for basic and fabricated metals. Allocation structure changed by creating small increases in air pollutant emissions from electricity, gas, and water supply to large ones.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model
ARTICLE IN PRESS
STRECO-712; No. of Pages 7
R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
6
Table 4 Comparative SDA of air pollutant emission changes in four countries, 1995–2009 (Mt). Countries
Period
EconomicActivity
Economic Structure
Allocation Structure
Emission Intensity
Total
China
1995–2001 2001–2009 1995–2009 1995–2001 2001–2009 1995–2009 1995–2001 2001–2009 1995–2009 1995–2001 2001–2009 1995–2009
18.90 61.66 80.56 2.18 11.97 14.15 11.26 8.77 20.03 −1.23 0.91 −0.32
1.80 −0.84 0.95 −0.43 −0.89 −1.32 −2.24 1.48 −0.76 −0.25 0.10 −0.15
0.55 19.01 19.56 −0.34 −0.20 −0.97 5.30 −6.49 −1.19 0.07 0.75 0.82
−23.40 −50.20 −73.60 0.02 −3.87 −3.85 −19.16 −13.04 −32.20 1.17 −1.73 −0.56
−2.16 29.62 27.47 1.43 6.59 8.02 −4.83 −9.28 −14.11 −0.24 0.02 −0.22
India
USA
Japan
Source: Author estimations.
4.2. International Comparison of Factors Affecting Air Pollutant Emissions from a Supply-Side Perspective Table 4 presents the SDA results for changes in air pollutant emissions for two developing countries (China and India) and two developed ones (the United States and Japan) for 1995–2009. In 1995–2009, economic activity and emissions intensity had the highest influence on air pollutant emissions across all four countries, whereas economic and allocation structure had varying impacts. Overall, economic and allocation structures decreased air pollutant emissions in India and the United States, but increased emission levels in China. In Japan, while economic structure decreased air pollutant emissions, allocation structure increased them. This suggests that there is significant potential to reduce supply-side air pollutant emissions in China. In particular, economic structure decreased air pollutant emissions in India, the United States, and Japan, respectively, by 1.32 Mt, 0.76 Mt, and 0.15 M, accounting for −16.46%, −5.39%, and −68.18%. In China, however, economic structure increased air pollutant emissions by 0.95 Mt (3.46%). Allocation structure decreased air pollutant emissions by 0.97 Mt (−12.09%) and 1.19 Mt (−8.43%) in India and the United States and significantly so in China (19.56 Mt; 71.20%) and Japan (0.82 Mt; 372.73%). During 1995–2001, economic structure decreased air pollutant emissions in India, the United States, and Japan, respectively, by 0.43 Mt, 2.24 Mt, and 0.25 Mt, accounting for −30.07%, −46.38%, and −104.17%, but increased them in China by 1.80 Mt (83.33%). Allocation structure decreased air pollutant emissions in India by 0.34 Mt (−23.78%), but increased them in China, the United States, and Japan, respectively, by 0.55 Mt, 5.30 Mt, and 0.07 Mt, thus accounting for 25.46%, 109.73%, and 29.17% of total emissions. In 2001–2009, economic structure caused air pollutant emissions in China and India to decrease by 0.84 Mt (2.84%) and 0.89 Mt (13.51%), but significantly increased it in the United States and Japan by 1.48 Mt (15.95%) and 0.10 Mt (500.00%). allocation structure led to an increase of 19.01 Mt and 0.75 Mt in China and Japan, accounting for 64.18% and 3750.00%, but a decrease in India and the United States of 0.2 Mt and 6.49 Mt, accounting for −3.03% and −69.94%. A comparison of 1995–2001 and 2001–2009 reveals that China’s economic structure is transforming to one that promotes the reduction of air pollutant emissions, while the opposite holds true for the United States and Japan. As for India, its economic structure has always been conducive to the reduction of air pollutant emissions. Allocation structure, however, seems unfavorable to the reduction of air pollutant emissions in China and Japan and a critical factor in increasing air pollutant emissions in the United States. However, it has always been conducive to reducing air pollutant emissions in India. In sum, economic and allocation structures depict opposing trends and have been conducive to air pollutant emission reduc-
tion in China. Nevertheless, to improve the distribution structure such that it promotes the reduction of air pollutant emission, highly pollutant intermediate inputs should be replaced by less polluting ones. 5. Conclusions and Policy Suggestions Since 2013, China has been experiencing frequent haze formation in its urban economic belts, particularly in the BeijingTianjin-Hebei region, Pearl River Delta, and Yangtze River Delta. While several studies have been conducted on factors contributing to environmental pollutants using the Leontief input-output model, they have largely done so from a demand perspective, which may be subject to limitations. Ghosh (1958) input-output model, which uses data from WIOD and those on SOx and NOx, decomposes factors influencing air pollutant emissions into economic activity, economic structure, allocation structure, and emissions intensity. Using his model, we identify the main factors but from a supplyside perspective. In addition, under the unified framework of SDA, we compare these factors between China and the United States, Japan, and India. The main results can be summarized as follows. First, the overall air pollutant emissions trend in China showed a slight decline with a substantial increase between 1995 and 2009. Economic and allocation structure, two supply-side factors, had varying impacts in 1995–2001 and 2001–2009. During 1995–2009, economic structure caused air pollutant emissions in China to substantially increase, whereas, in 2001–2009, it helped reduce them. This indicates that, overall, economic structure positively affected China’s air pollutant emissions. On the other hand, China’s allocation structure marginally increased air pollutant emissions from 1995 to 2009, but substantially so between 1995 and 2009. This indicates that China’s allocation structure of intermediate products was based on pollution-intensive industries and was deteriorating over time. Second, during 1995–2009, electricity, gas, and water supply; other non-metallic mineral and basic metals; and fabricated metal were the three main sectors contributing to increased total air pollutant emissions in China. The SDA results for these major polluting industries reveal that in 1995–2001, economic structure increased air pollutant emissions in electricity, gas, and water supply and decreased those in other non-metallic mineral and basic metals and fabricated metal. In 2001–2009, while economic structure decreased air pollutant emissions in electricity, gas, and water supply, it increased them in basic and fabricated metal. As for allocation structure, it transformed from a weak impact on air pollutant emissions in the former period to substantially increasing them in the latter through the electricity, gas, and water supply and basic metals and fabricated metal industries. Third, the international comparison revealed that during 1995–2009, economic structure decreased air pollutants in the
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model STRECO-712; No. of Pages 7
ARTICLE IN PRESS R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
United States, Japan, and India but increased them in China. A comparison between 1995–2001 and 2001–2009 showed that China’s economic structure has transformed to one that promotes the reduction of air pollutant emissions, while those of the United States and Japan seem to contribute towards increasing them. India, by contrast, has always had an economic structure that is conducive to reducing air pollutant emissions. In 1995–2009, allocation structure reduced air pollutant emissions in the United States and India but increased them in China and Japan. In other words, allocation structure is transforming from being unfavorable to reduce air pollutant emissions in China and Japan but is becoming a major contributing factor to increased levels in the United States. India’s allocation structure has always been conducive to the reduction of air pollutant emissions in India. The main policy recommendations can be formulated as follows. First, although allocation structure promoted increased air pollutant emissions in China, it has been beneficial to reducing them in the United States and India. Therefore, China should adopt various measures to reduce the promoting impact of its distribution structure on air pollutant emissions by upgrading its high-polluting industries—i.e. electricity, gas, and water supply; other non-metallic mineral; and basic metals and fabricated metal—and decreasing or optimizing the distribution proportion of intermediate products to these high-polluting industries. Second, economic structure has been critical in reducing China’s air pollutant emissions. Thus, China should aim to further optimize its industrial structure, with its growth-stimulating industries switching from high-polluting industries to less-polluting ones, such as the information industry and financial intermediation, thus decoupling economic growth from air pollution. Finally, it is imperative that China undertakes to strengthen measures to reduce air pollution from both the demand and supply side. For the demand side, the production of pollution-intensive products should be reduced by encouraging consumers to choose greener products (Zhang et al., 2015). Compared to the end-of-pipe measures on the demand side, the supply side should curb primary sources of air pollution that aggravate the formation of haze (Zhang, 2010). These changes can improve the production, distribution, circulation, and consumption of products and lead to reduced air pollutant emissions. Acknowledgments Our heartfelt thanks should be given to the Editor, Dr Coffman, as well as anonymous reviewers, and Zhifu Mi for their helpful comments, which have allowed us to improve the manuscript. We thank the National Natural Science Foundation of China (NSFC) (71673083,71303076, 71203062 and 71303174), and Hunan Science and Technology Key Program of China (Grant no. 2015zk2002) for funding support. References Ang, B.W., 1995. Decomposition methodology in industrial energy demand analysis. Energy 20 (11), 1081–1095. Ang, B.W., Liu, F.L., 2001. ‘A new energy decomposition method: perfect in decomposition and consistent in aggregation’. Energy 26 (No. 6), 537–548.
7
Ang, B.W., Liu, F.L., Chew, E.P., 2003. Perfect decomposition techniques in energy and environmental analysis. Energy Policy 31 (No. 14), 1561–1566. Dietzenbacher, E., Los, B., 1998. Structural decomposition techniques: sense and sensitivity. Econ. Syst. Res. 10 (4), 307–324. Dietzenbacher, E., Los, B., 2000. Structural decomposition analyses with dependent determinants. Econ. Syst. Res. 12 (No. 4), 497–514. Ghosh, A., 1958. Input-output approach in an allocation system. Economica 25 (No. 97), 58–64. Guan, D.B., Hubacek, K., Weber, C.L., Reiner, D.M., 2008. The driver of Chinese CO2 emissions from 1980 to 2030. Global Environ. Change 18 (No. 4), 626–634. Guan, D.B., Peters, G.P., Weber, C.L., 2009. Journey to world top emitter: an analysis of the driving forces of China’s recent CO2 emissions surge. Geophys. Res. Lett. 36 (4), 1–5. He, H., Jiang, G.B., 2014. Scientific and reason explain haze problem of China. Qiushi (No. 6), 48–50, In Chinese. He, H., Wang, X.M., Wang, Y.S., Wang, Z.F., Liu, J.G., Chen, Y.F., 2013. The tracking and control of atmospheric haze. Bull. Chin. Acad. Sci. (No. 7), 344–352, In Chinese. Hoekstra, R., Van den Bergh, J.C.J.M., 2003. Comparing structural and index decomposition analysis. Energy Econ. (No. 1), pp. 39– 64. Li, F.Y., Song, Z.Y., Liu, W.D., 2014. ‘China’s energy consumption under the global economic crisis: decomposition and sectoral analysis’. Energy Policy 64, 193–202. Li, H.N., Wei, Y.M., 2015. Is it possible for China to reduce its total CO2 emissions? Energy 83, 438–446. Lyu, W., Li, Y., Guan, D., Zhao, H., Zhang, Q., Liu, Z., 2016. Driving forces of Chinese primary air pollution emissions: an index decomposition analysis. J. Clean. Prod. 133, 136–144. Meng, L., Crijns-Graus, W.H., Worrell, E., Huang, B., 2018. Impacts of booming economic growth and urbanization on carbon dioxide emissions in Chinese megalopolises over 1985–2010: an index decomposition analysis. Energy Efficiency 11 (1), 203–223. Meng, B., Wang, J., Andrew, R., Xiao, H., Xue, J., Peters, G.P., 2017. Spatial spillover effects in determining China’s regional CO2 emissions growth: 2007–2010. Energy Econ. 63, 161–173. Mi, Z., Meng, J., Guan, D., Shan, Y., Liu, Z., Wang, Y., Wei, Y.M., 2017. Pattern changes in determinants of Chinese emissions. Environ. Res. Lett. 12 (7), 074003. Nie, H., Kemp, R., Vivanco, D.F., Vasseur, V., 2016. Structural decomposition analysis of energy-related CO2 emissions in China from 1997 to 2010. Energy Efficiency 9 (6), 1351–1367. Rormose, P., Olsen, T., 2005. Structural decomposition analysis of air emissions in Denmark 1980-2002 Beijing, China. 15th International Input-Output Conference. Rose, A., Casler, S., 1996. Input-output structural decomposition analysis: a critical appraisal. Econ. Syst. Res. 8 (No. 1), 33–62. Su, B., Ang, B.W., 2012. Structural decomposition analysis applied to energy and emissions: some methodological developments. Energy Econ. 34 (1), 177–188. Su, B., Ang, B.W., 2016. Multi-region comparisons of emission performance: the structural decomposition analysis approach’. Ecol. Indic. 67, 78–87. Su, B., Ang, B.W., 2017. Multiplicative structural decomposition analysis of aggregate embodied energy and emission intensities’. Energy Econ. (No. 65), 137–147. Su, B., Ang, B.W., Li, Y., 2017. Input-output and structural decomposition analysis of Singapore’s carbon emissions. Energy Policy 105, 484–492. Wang, H., Ang, B.W., Su, B., 2017. A multi-region structural decomposition analysis of global CO2 emission intensity. Ecol. Econ. 142, 163–176. Weir, M., 1998. Sources of changes in emissions from energy: a structural decomposition analysis. Econ. Syst. Res. 10, 99–112. Wood, R., 2009. ‘Structural decomposition analysis of Australia’s greenhouse gas emissions’. Energy Policy 37 (No. 11), 4943–4948. Xie, S.C., 2014. ‘The driving forces of China’s energy use from 1992 to 2010: an empirical study of input-output and structural decomposition analysis’. Energy Policy 73, 401–415. Yang, C.J., Zhou, Y., Jackson, R.B., 2014. ‘China’s fuel gas sector: history, current status, and future prospects’. Utilities Policy 28, 12–21. Zhang, Y.G., 2009. Structural decomposition analysis of sources of decarbonizing economic development in China: 1992–2006. Ecol. Econ. 68 (No. 8), 2399–2405. Zhang, Y.G., 2010. Supply-side structural effect on carbon emissions in China. Energy Econ. 32 (No. 1), 186–193.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
ARTICLE IN PRESS
STRECO-712; No. of Pages 7
Structural Change and Economic Dynamics xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
Structural Change and Economic Dynamics journal homepage: www.elsevier.com/locate/sced
Supply-side structural effects of air pollutant emissions in China: A comparative analysis Rui Xie a , Fangfang Wang a , Julien Chevallier b,c,∗ , Bangzhu Zhu d,e,∗ , Guomei Zhao a a
College of Economics and Trade, Hunan University, Hunan, PR China IPAG Business School, IPAG Lab, 184 Boulevard Saint-Germain, 75006 Paris, France Université Paris 8, LED, 2 avenue de la Liberté, 93526 Saint-Denis cedex, France d School of Management, Jinan University, Guangdong, PR China e Business School, Nanjing University of Information Science & Technology, Nanjing 210044, PR China b c
a r t i c l e
i n f o
Article history: Received 19 January 2017 Received in revised form 13 February 2018 Accepted 18 April 2018 Available online xxx JEL classification: N50 N70 N90 057 Keywords: Ghosh input-output model Air pollution Structural decomposition analysis Supply-side structure China International comparative analysis
a b s t r a c t China’s air quality has become a major issue affecting people’s livelihood and continues to deteriorate in recent times. It is an important issue of common concern for economists and policymakers to explore the drivers of the growth of air pollution emissions and the deteriorating environmental quality in China. From the perspective of supply-side structures, this paper adopts Ghosh input-output model to decompose the factors affecting the changes of air pollutant emissions into economic activities, economic structures, allocation structures and emission intensity. Using this model, we conduct a structural decomposition analysis of air pollutant emissions in China, India, USA, and Japan for 1995–2009. The results reveal that China’s economic structure initially promoted air pollutant emissions, but later played a role in reducing them. Further, whereas in Japan and particularly, China, allocation structures were found to be a key factor in increasing air pollutant emissions, in America and India, it played a critical role in reducing emissions. Our findings suggest that adjusting the distribution structure of intermediate products is crucial to reduce air pollution. © 2018 Elsevier B.V. All rights reserved.
1. Introduction China’s economy has rapidly grown since the reform and opening up policy, and since then, has surpassed that of Japan, becoming the second largest economy in the world in 2010. However, China economic development is largely dependent on the extensive input of factors, such as labor, capital, and resources, and industrial sectors with high energy consumption, and thus, is subject to severe environmental pollution and energy shortage (Yang et al., 2014; Xie, 2014). In particular, since 2013, the Beijing-TianjinHebei region, Pearl River Delta, and Yangtze River Delta have been frequently experiencing haze, indicating increasing threats to the climate, human health, and development of a harmonious society (He et al., 2013).
∗ Corresponding authors. E-mail addresses: [email protected] (J. Chevallier), [email protected] (B. Zhu).
Preliminary research on the causes of haze in China attributes its formation to two factors. First is the obstruction in the horizontal flow of air and the lack of diffusion given static weather conditions, an external cause which cannot be controlled by human behavior. Second, oxysulfide (SOx) and oxynitride (NOx) emissions from fire coal, motor vehicles, and industrial production, which convert into secondary particulate pollutants through chemical reactions in the air and further aggravate the formation of haze (He et al., 2013; He and Jiang et al., 2014). Therefore, a fundamental method to curb haze is to strictly control SOx and NOx emissions and prevent the formation of secondary particulate pollutants. According to the World Input-Output Database (WIOD), China’s total air pollutant emission (total emission including SOx and NOx) was 35.40 Mt in 1995. While this figure decreased to 32.87 Mt in 2001, it surged in 2009 to 65.77 Mt, increasing by more than 100%. Therefore, an in-depth exploration of the factors affecting China’s air pollutant emissions is needed to control the formation of haze and the severe deterioration of air quality. The method of decomposition is often used to analyze the factors influencing envi-
https://doi.org/10.1016/j.strueco.2018.04.005 0954-349X/© 2018 Elsevier B.V. All rights reserved.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model
ARTICLE IN PRESS
STRECO-712; No. of Pages 7
R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
2
ronmental pollution and classify various factors contributing to the total change of pollutant emission (Ang, 1995; Ang et al., 2003; Su et al., 2017; Meng et al., 2017). The two strains of the decomposition methods are the index decomposition analysis (IDA) and structural decomposition analysis (SDA). In analyzing the factors influencing energy use and pollutant emission, IDA further categorizes total effects as independent effects using aggregated sector data. Many scholars have adopted IDA to analyze changes in energy intensity, carbon emissions and pollutant emissions (Ang and Liu, 2001; Lyu et al., 2016; Meng et al., 2018). Specifically, IDA methods enjoy relatively lower requirements for data and are convenient to operate. But IDA method is unable to conduct an in-depth analysis of direct effects and indirect effects among economic activities, and not allowed for a detailed decomposition of economic structures. SDA methods can overcome some of the drawbacks of IDA method (Su and Ang, 2012, 2016), which examine direct and indirect effects using Leontief and Ghosh inverse matrices (Hoekstra and Van den Bergh, 2003). SDA method is applied in conjunction with the input-output model to examine factors underlying energy intensity and pollutant emissions (Rose and Casler, 1996; Nie et al., 2016; Wang et al., 2017; Mi et al., 2017). However, SDA method requires large amounts of data, as SDA method decomposes changes in various indicators using the input-output model and data from corresponding input-output tables, which distinguish between a range of technological and final demand effects. With improvements in input-output tables, a growing number of researchers are applying the SDA model to analyze factors influencing the environment. Therefore, this article also uses this method. Many studies have used the Leontief model to examine the factors affecting environment pollutants from a demand-side perspective. Weir (1998) and Rormose and Olsen (2005) used the SDA model to identify factors underlying Danish pollutant emissions including CO2 , SO2 , and NOx during 1966–1988 and 1980–2002. Guan et al. (2008, 2009) explored the factors influencing China’s carbon emission growth during 1981–2002 and 2002–2005. Wood (2009) investigated the causes of greenhouse gas emissions in Australia from 1974 to 2005. Using SDA, Zhang (2009) analyzed the historical change trend for China’s carbon intensity related to energy for 1992–2006. Li et al. (2014) and Xie (2014) determined the factors contributing to the growth in China’s energy consumption and use. Li and Wei (2015) determined factors affecting China’s carbon emission and Zhang et al. (2015) did similar research for China’s pollutant levels, including COD, NH3-N, SO2 , and NOx. Su and Ang (2017) used the SDA method to analysis the aggregate embodied energy and emission intensities. Based on spatial SDA, Meng et al. (2017) proposed an alternative input-output to elucidate inter-regional spillover effects in determining China’s regional CO2 emissions growth. Su et al. (2017) used the input-output (IO) method to analyze the city state’s carbon emissions from the demand perspective and used the SDA method to investigate the Table 1 Basic form of single regional input-output table. Output
Intermediate Input
Value Added
Total Input
Sector 1 . . Sector35
2.1. Data source and explanation We adopt single regional input-output table data for two developing countries, China and India, and two developed countries, the United States and Japan, from the World Input-Output Database (WIOD) for 1995–2009. Table 1 presents the basic form of the table. SOx and NOx are key precursor pollutants contributing to the formation of haze (He and Jiang et al., 2014). Thus, we retrieve emission data for SOx and NOx for 1995–2009 from environmental accounts by WIOD and combine the data for both by sector. We then use them as indicators to measure air pollutant emissions. In Table 1, Zis the matrix of intermediate inputs and zij denotes the input demand for sector i from sector j. V is a value-added row vector and vi represents the value added for sector i. Fis the final demand (used) column vector. Xis the output column vector and xi denotes the total output of sector i. 2.2. Theoretical model Consider the column identity of a typical input-output table in monetary terms: (1)
Then, Eq. (1) can be rewritten as
Sector 1 ... Sector 35
Final Demand Total Output
Z
F
V ... Vn V ROW X ...
2. Theoretical model and data
X = HT X + V T ,
Intermediate Demand
Input
drivers of emission changes, which was the first comprehensive analysis of Singapore’s emissions using the I-O framework. However, few articles have used Ghosh input-output model to analyze air pollution issues from a supply-side perspective (e.g. Zhang, 2010). China’s economic structure is undergoing major transformations, as a result of which its demand management faces limitations such as imbalances between domestic and external demands or investment and consumption. Against this background, supply policies can be used as a tool (Cai et al., 2008; China’s Economic Growth Report, 2010; Zhang, 2010) by the government to exercise direct control and reduce policy uncertainties and risks. Thus, this study examines the factors influencing pollutant emissions from a supply-side perspective and accordingly, offers policy suggestions to reduce emission and conserve energy. This study makes the following contributions: (i) we provide an analysis of the factors influencing air pollution, and (ii) we address the key issue of increased haze formation in China. We adopt the SDA approach using Ghosh input-output model from a supply-side perspective. Then, using WIOD’s input-output tables for 40 countries with 35 consolidated sectors for 1995–2009 and air pollutant data including SOx and NOx, we compare the different factors between developed and developing countries. It provides an international reference for China’s efforts to control haze. The remainder of this paper is organized as follows. Section 2 details the theoretical model and data. Sections 3 and 4 analyze air pollution emissions and the factors affecting them. Section 5 offers a summary and policy suggestions.
X T = V (I − H)−1 = VG,
(2)
X
In Eq. (2), we present Ghosh’s (1958) model, where H is the direct distribution matrix in which hij = zij ⁄xi . Here, hij represents the share of distributed output from the ith sector to the jth sector. G = (I − H)−1 is the Ghosh inverse matrix that indicates the dependence between the production sector and the sector that uses its products. Ghosh’s supply-driven input-output model takes primary inputs (sectoral value-add) as the starting point and accounts for both direct and indirect relationships between sectors.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model
ARTICLE IN PRESS
STRECO-712; No. of Pages 7
R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
Next, similar to Zhang (2010), we estimate air pollutant emissions across countries from a supply-side perspective using Eq. (2). P = X T E = VGE =
SGE,
p1 − p0 = (
(3)
where P is total air pollutant emission, Eis the emission intensity column vector, ei = pi ⁄xi , and is a scalar denoting the total volume of sectoral primary inputs (value added), that is, Chinese GDP, accounted for using an income approach. Sis the supply-side economic structure vector, whose ith entry is the share of value added for sector i in GDP. The air pollutants allocated to the sectors can be estimated as follows:
+
1
1 S1 (G1
−
3
0 )S0 G0 E0
− G0 )E0 +
diag(SG)E,
(4)
where p is the column vector for sectoral air pollutant emissions, whose element pi represents the ith air pollutant emissions. Structural decomposition is a popular descriptive tool and is used to identify the sources of historical changes (Dietzenbacher and Los, 1998; 2000). Using Eq. (3), changes in air pollutant emissions can be decomposed as p1 − p0 =
1 S1 G1 E1
−
0 S0 G0 E0 ,
(5)
where subscripts 1 and 0 represent time points. If the decomposition results in n variable factors, n first-order difference equations can be obtained. In this study, factor decomposition is conducted using Dietzenbacher and Los (1998) bipolar decomposition method. First, consider the following decomposition methods: p1 − p0 = ( +
0 S0 (G1
1
−
0 )S1 G1 E1
− G0 )E1 +
+
0 (S1
0 S0 G0 (E1
− S0 )G1 E1
− E0 ).
(6)
1 (S1
1 S1 G1 (E1
− S0 )G0 E0
− E0 ).
(7)
Next, take the average value for both methods: p1 − p0 =
1 ( 2
1
−
0 )(S1 G1 E1
+ S0 G0 E0 ) +
1 2
0 (S1
− S0 )G1 E1 +
Economic activity
+
1 [ 2
0 S0 (G1
− G0 )E1 +
1 S1 (G1
1 (S1
− S0 )G0 E0
Economic structure
− G0 )E0 ] +
Allocation structure
p = diag(X T )E = diag(VG)E =
+
1 ( 2
0 S0 G0
+
1 S1 G1 )(E1
− E0 )
(8)
Emissions intensity
The decomposition in Eq. (8) shows the sources of air pollutant emission can be further divided into economic activity, economic structure, allocation structure, and emissions intensity effects. Economic and allocation structure effects reflect the change in air pollutant emissions from a supply-side perspective. 3. Total air pollutant emissions analysis 3.1. Analysis of Air Pollutant Emissions in China China’s total air pollutant emissions experienced a slight decrease during 1995–2001, followed by a sharp increase between 2002 and 2009. According to Fig. 1, China’s total emissions decreased from 35.4 Mt in 1995 to 32.87 Mt in 2001 with a difference of 2.16 Mt, which accounts for 7.85% of total emissions change. However, after China’s accession to the WTO, in 2003, the total air pollutant emissions rapidly increased by 14% and by 2009, it grew by 29.62 Mt to 65.77 Mt, accounting for 107.85% of the rise in emission during 1995–2009.
Fig. 1. Total emissions and growth rate of air pollutants in China, 1995–2009.
Fig. 2. Emissions trend for major pollutant sectors in China, 1995–2009.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model STRECO-712; No. of Pages 7 4
ARTICLE IN PRESS R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
As shown in Fig. 1, China’s total air pollutant emissions have an obvious turning point in 2001, indicating a crucial change in China’s economic growth pattern after its accession to the WTO. Thus, in this study, we conduct an additional analysis of the factors affecting changes in air pollutant emissions during 1995–2001 and 2001–2009. Fig. 2 clearly shows that electricity, gas, and water supply; other non-metallic mineral and basic metals; and fabricated metal are the three main sectors contributing to the increase in China’s total air pollutant emissions. While electricity, gas, and, water supply account for 51% of total emission, other non-metallic mineral and basic metals and fabricated metal have a relatively small share of 10%. A further analysis of sectoral data reveals that electricity, gas, and water supply account for the largest share, especially following the first several years since China was admitted to the WTO. However, this share rapidly declined from 43% in 1995 to 33% in 2009. This is due to the fact that, on one hand, the output of electricity, gas and water supply dropped from $358375 million in 2003 to $185932 million in 2004 according to the input-output table provided by WIOD. Similarly, the China Statistical Yearbook in 2003 and 2004 also shows a downward trend of output in this sector. It means that the decline of output has led to a drop in the sector emissions. On the other hand, as of 2003, the sector of China’s electricity, gas, and water supply has undergone great transformation and expansion, forming an industrial chain, thereby increasing the comprehensive utilization of resources, to achieve a green development, and ultimately reduce environmental emissions. 3.2. International Comparative Analysis of Total Air Pollutant Emissions Fig. 3 illustrates the trend of air pollutant emissions in China, India, the United States, and Japan during 1995–2009. These four countries have a strong international presence and the highest
emission levels among other developing and developed countries. In 2001, China’s total pollutant emissions exceeded that of the United States, becoming the highest emitter of air pollutants. At present, the United States ranks second and continues to show a downward trend, followed by India and Japan; while India has a gradually increasing trend, that of Japan is relatively stable. According to Fig. 4, the intensity of air pollution emissions in developing countries such as China and India are much higher than those in developed countries such as the United States and Japan. Evidently, the drop in China’s air pollution emissions is more rapid than those in the other three countries. During the sample period, the intensity of China’s air pollution emissions steadily declined from 214.21 t/billion dollars in 1995 to 43.41 t/billion dollars in 2009, with an average annual decline of 5.32%. By 1999, India’s intensity of air pollutant emission was higher than that of China, thus recording the highest intensity level. Nevertheless, India appears to show a declining trend with a reduced rate of 3.26%. As for the United States and Japan, their air pollution intensities are far lower than those of China and India. The air pollution intensity in the United States reported an average annual decline rate of 4.41%, while that of Japan recorded an increase of 0.17%. 4. Contributing factors 4.1. Factors affecting China’s air pollution from a supply-side perspective Table 2 shows the SDA results for China’s air pollutants during 1995–2009. We see that economic activity contributed the most to China’s increased air pollutant emissions, while emission intensity played an important role in reducing them. On the other hand, allocation and economic structures, which are two types of supplyside factors, caused air pollutant emissions in China to significantly rise. China’s total air pollutant emissions increased by 27.47 Mt from 1995 to 2009. During the same period, economic activity
Fig. 3. Air pollutant emissions in four countries, 1995–2009.
Fig. 4. Intensity of air pollutant emissions in four countries, 1995–200.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model
ARTICLE IN PRESS
STRECO-712; No. of Pages 7
R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
5
Table 2 SDA of air pollutant emission changes in China, 1995–2009 (Mt). Countries
Period
EconomicActivity
Economic Structure
Allocation Structure
Emission Intensity
Total
China
1995–2001 2001–2009 1995–2009
18.90 61.66 80.56
1.80 −0.84 0.95
0.55 19.01 19.56
−23.40 −50.20 −73.60
−2.16 29.62 27.47
Source: Author estimations. Table 3 SDA of air pollutant changes in major polluting industries in China, 1995–2009 (Mt). Industry
Period
Economic Activity
Economic Structure
Allocation Structure
Emission Intensity
Total
Electricity, Gas, and Water Supply
1995–2001 2001–2009 1995–2009 1995–2001 2001–2009 1995–2009 1995–2001 2001–2009 1995–2009
5.06 15.44 20.50 0.92 2.40 3.31 1.18 2.91 4.09
2.40 −0.24 2.15 −0.42 −0.63 −1.05 −0.43 0.89 0.46
−0.10 5.50 5.40 0.03 −0.01 0.02 0.04 0.32 0.36
−6.58 −19.82 −26.40 −0.92 −0.89 −1.81 −1.32 −3.47 −4.79
0.78 0.88 1.65 −0.39 0.87 0.48 −0.54 0.65 0.11
Other Non-Metallic Mineral
Basic Metals and Fabricated Metal
Source: Author estimations.
increased China’s air pollutant emissions by 80.56 Mt, accounting for 293.26% of total air pollutant emissions, whereas emissions intensity decreased them by 73.60 Mt, accounting for −267.93%. Allocation and economic structures also increased China’s air pollutant emissions by 19.56 Mt (71.02%) and 0.95 Mt (6.55%). From 1995–2001, China’s air pollutant emissions decreased by 2.16 Mt. While economic activity, economic structure, and allocation structure caused China’s air pollutant emissions to increase, emission intensity contributed towards reducing them. Compared with 1995–2009, the economic and allocation structure in 1995–2001 showed significant differences. In 1995–2001, economic structure increased China’s air pollutant emissions by 1.80 Mt (83.33%) and allocation structure brought about a smaller increase of 0.55 Mt. From the viewpoint of these two supply-side factors, China adopted an extensive growth model that promoted industries with high-intensity emissions. Nevertheless, China’s allocation structure of intermediate products was more reasonable and had little impact on its air pollutant emissions. During 2001–2009, China’s air pollutant emissions increased by 29.62 Mt, indicating a higher growth rate than that in 1995–2001. Economic activity and allocation structure caused China’s air pollutant emissions to increase, while emissions intensity and the economic structure caused them to decrease. Compared with 1995–2001, allocation and economic structures showed significant differences. Economic structure reduced air pollutant emissions by 0.84 Mt, whereas allocation structure significantly increased them by 19.01 Mt (64.18%). These results indicate that China’s economic structure moved towards reducing air pollutant emissions, although the effects are not particularly evident. However, the allocation structure of intermediate products was centered on high energy consumption and high emission industries, which led to China’s economic growth being further dominated by a growth extensive model. Thus, supply-side effects (allocation and economic structure) increased China’s air pollutant emissions to 2.35 Mt during 1995–2001 and 18.72 Mt in 2001–2009. Moreover, although the negative effects of economic structure decreased, those of allocation structure increased. This indicates that China’s intermediate products relied heavily on resources and energy and its economy followed a typical extensive model that led to high energy consumption and pollution levels. Table 3 presents the SDA results for China’s major polluting industries—electricity, gas and water supply; other nonmetallic mineral; and basic metals and fabricated metal—during
1995–2009. From this perspective, economic activity, economic structure, and allocation structure were the main factors contributing to the increased air pollutant emissions, whereas emissions intensity played an important role in decreasing them. In 1995–2009, economic activity led to an increase in air pollutant emissions by the electricity, gas, and water supply; other non-metallic mineral; and basic metals and fabricated metal industries by 20.50 Mt, 3.31 Mt, and 4.09 Mt, respectively. Emissions intensity, on the other hand, contributed towards decreasing these emission levels by 26.40 Mt, 1.81 Mt, and 4.79 Mt, respectively. Allocation structure caused an increase of 5.4 Mt, 0.02 Mt, and 0.36 Mt, respectively, accounting for 327.27%, 4.17%, and 327.27%. Economic structure contributed to an increase in emission levels in electricity, gas, and water supply and basic metals and fabricated metal by 2.15 Mt (130.30%) and 0.46 Mt (418.18%) and a decrease of 1.05 Mt (−218.75%.) in other non-metallic minerals. In 1995–2001, economic activity was the main factor contributing to increased air pollutant emissions by the three major polluting industries, whereas emissions intensity played a vital role in decreasing them. Allocation structure increased air pollutant emissions in other non-metallic mineral and basic metals and fabricated metal industries by 0.03 Mt (7.69%) and 0.04 Mt (7.41%) but decreased emissions in electricity, gas, and water supply by 0.10 Mt (−12.82%). By contrast, economic structure increased air pollutant emissions in electricity, gas, and water supply by 2.40 Mt (307.69%) and decreased them in other non-metallic mineral and basic metals and fabricated metals by 0.42 Mt (−107.69%) and 0.43 Mt (−79.63%). In sum, economic structure increased air pollutant emissions in other non-metallic mineral and basic metals and fabricated metal and substantially promoted air pollutant emissions in electricity, gas, and water supply. Allocation structure, on the other hand, had a small effect on air pollutant emissions in the three major polluting industries. In comparison with the decomposition results for 1995–2001, in 2001–2009, economic activity increased pollutant emissions by the major polluting industries, whereas emissions intensity decreased them. Economic and allocation structures also had significant impacts. China’s economic structure transformed from a factor that increased pollutant emissions from electricity, gas, and water supply to one that decreased them, while the reverse is true for basic and fabricated metals. Allocation structure changed by creating small increases in air pollutant emissions from electricity, gas, and water supply to large ones.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model
ARTICLE IN PRESS
STRECO-712; No. of Pages 7
R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
6
Table 4 Comparative SDA of air pollutant emission changes in four countries, 1995–2009 (Mt). Countries
Period
EconomicActivity
Economic Structure
Allocation Structure
Emission Intensity
Total
China
1995–2001 2001–2009 1995–2009 1995–2001 2001–2009 1995–2009 1995–2001 2001–2009 1995–2009 1995–2001 2001–2009 1995–2009
18.90 61.66 80.56 2.18 11.97 14.15 11.26 8.77 20.03 −1.23 0.91 −0.32
1.80 −0.84 0.95 −0.43 −0.89 −1.32 −2.24 1.48 −0.76 −0.25 0.10 −0.15
0.55 19.01 19.56 −0.34 −0.20 −0.97 5.30 −6.49 −1.19 0.07 0.75 0.82
−23.40 −50.20 −73.60 0.02 −3.87 −3.85 −19.16 −13.04 −32.20 1.17 −1.73 −0.56
−2.16 29.62 27.47 1.43 6.59 8.02 −4.83 −9.28 −14.11 −0.24 0.02 −0.22
India
USA
Japan
Source: Author estimations.
4.2. International Comparison of Factors Affecting Air Pollutant Emissions from a Supply-Side Perspective Table 4 presents the SDA results for changes in air pollutant emissions for two developing countries (China and India) and two developed ones (the United States and Japan) for 1995–2009. In 1995–2009, economic activity and emissions intensity had the highest influence on air pollutant emissions across all four countries, whereas economic and allocation structure had varying impacts. Overall, economic and allocation structures decreased air pollutant emissions in India and the United States, but increased emission levels in China. In Japan, while economic structure decreased air pollutant emissions, allocation structure increased them. This suggests that there is significant potential to reduce supply-side air pollutant emissions in China. In particular, economic structure decreased air pollutant emissions in India, the United States, and Japan, respectively, by 1.32 Mt, 0.76 Mt, and 0.15 M, accounting for −16.46%, −5.39%, and −68.18%. In China, however, economic structure increased air pollutant emissions by 0.95 Mt (3.46%). Allocation structure decreased air pollutant emissions by 0.97 Mt (−12.09%) and 1.19 Mt (−8.43%) in India and the United States and significantly so in China (19.56 Mt; 71.20%) and Japan (0.82 Mt; 372.73%). During 1995–2001, economic structure decreased air pollutant emissions in India, the United States, and Japan, respectively, by 0.43 Mt, 2.24 Mt, and 0.25 Mt, accounting for −30.07%, −46.38%, and −104.17%, but increased them in China by 1.80 Mt (83.33%). Allocation structure decreased air pollutant emissions in India by 0.34 Mt (−23.78%), but increased them in China, the United States, and Japan, respectively, by 0.55 Mt, 5.30 Mt, and 0.07 Mt, thus accounting for 25.46%, 109.73%, and 29.17% of total emissions. In 2001–2009, economic structure caused air pollutant emissions in China and India to decrease by 0.84 Mt (2.84%) and 0.89 Mt (13.51%), but significantly increased it in the United States and Japan by 1.48 Mt (15.95%) and 0.10 Mt (500.00%). allocation structure led to an increase of 19.01 Mt and 0.75 Mt in China and Japan, accounting for 64.18% and 3750.00%, but a decrease in India and the United States of 0.2 Mt and 6.49 Mt, accounting for −3.03% and −69.94%. A comparison of 1995–2001 and 2001–2009 reveals that China’s economic structure is transforming to one that promotes the reduction of air pollutant emissions, while the opposite holds true for the United States and Japan. As for India, its economic structure has always been conducive to the reduction of air pollutant emissions. Allocation structure, however, seems unfavorable to the reduction of air pollutant emissions in China and Japan and a critical factor in increasing air pollutant emissions in the United States. However, it has always been conducive to reducing air pollutant emissions in India. In sum, economic and allocation structures depict opposing trends and have been conducive to air pollutant emission reduc-
tion in China. Nevertheless, to improve the distribution structure such that it promotes the reduction of air pollutant emission, highly pollutant intermediate inputs should be replaced by less polluting ones. 5. Conclusions and Policy Suggestions Since 2013, China has been experiencing frequent haze formation in its urban economic belts, particularly in the BeijingTianjin-Hebei region, Pearl River Delta, and Yangtze River Delta. While several studies have been conducted on factors contributing to environmental pollutants using the Leontief input-output model, they have largely done so from a demand perspective, which may be subject to limitations. Ghosh (1958) input-output model, which uses data from WIOD and those on SOx and NOx, decomposes factors influencing air pollutant emissions into economic activity, economic structure, allocation structure, and emissions intensity. Using his model, we identify the main factors but from a supplyside perspective. In addition, under the unified framework of SDA, we compare these factors between China and the United States, Japan, and India. The main results can be summarized as follows. First, the overall air pollutant emissions trend in China showed a slight decline with a substantial increase between 1995 and 2009. Economic and allocation structure, two supply-side factors, had varying impacts in 1995–2001 and 2001–2009. During 1995–2009, economic structure caused air pollutant emissions in China to substantially increase, whereas, in 2001–2009, it helped reduce them. This indicates that, overall, economic structure positively affected China’s air pollutant emissions. On the other hand, China’s allocation structure marginally increased air pollutant emissions from 1995 to 2009, but substantially so between 1995 and 2009. This indicates that China’s allocation structure of intermediate products was based on pollution-intensive industries and was deteriorating over time. Second, during 1995–2009, electricity, gas, and water supply; other non-metallic mineral and basic metals; and fabricated metal were the three main sectors contributing to increased total air pollutant emissions in China. The SDA results for these major polluting industries reveal that in 1995–2001, economic structure increased air pollutant emissions in electricity, gas, and water supply and decreased those in other non-metallic mineral and basic metals and fabricated metal. In 2001–2009, while economic structure decreased air pollutant emissions in electricity, gas, and water supply, it increased them in basic and fabricated metal. As for allocation structure, it transformed from a weak impact on air pollutant emissions in the former period to substantially increasing them in the latter through the electricity, gas, and water supply and basic metals and fabricated metal industries. Third, the international comparison revealed that during 1995–2009, economic structure decreased air pollutants in the
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005
G Model STRECO-712; No. of Pages 7
ARTICLE IN PRESS R. Xie et al. / Structural Change and Economic Dynamics xxx (2018) xxx–xxx
United States, Japan, and India but increased them in China. A comparison between 1995–2001 and 2001–2009 showed that China’s economic structure has transformed to one that promotes the reduction of air pollutant emissions, while those of the United States and Japan seem to contribute towards increasing them. India, by contrast, has always had an economic structure that is conducive to reducing air pollutant emissions. In 1995–2009, allocation structure reduced air pollutant emissions in the United States and India but increased them in China and Japan. In other words, allocation structure is transforming from being unfavorable to reduce air pollutant emissions in China and Japan but is becoming a major contributing factor to increased levels in the United States. India’s allocation structure has always been conducive to the reduction of air pollutant emissions in India. The main policy recommendations can be formulated as follows. First, although allocation structure promoted increased air pollutant emissions in China, it has been beneficial to reducing them in the United States and India. Therefore, China should adopt various measures to reduce the promoting impact of its distribution structure on air pollutant emissions by upgrading its high-polluting industries—i.e. electricity, gas, and water supply; other non-metallic mineral; and basic metals and fabricated metal—and decreasing or optimizing the distribution proportion of intermediate products to these high-polluting industries. Second, economic structure has been critical in reducing China’s air pollutant emissions. Thus, China should aim to further optimize its industrial structure, with its growth-stimulating industries switching from high-polluting industries to less-polluting ones, such as the information industry and financial intermediation, thus decoupling economic growth from air pollution. Finally, it is imperative that China undertakes to strengthen measures to reduce air pollution from both the demand and supply side. For the demand side, the production of pollution-intensive products should be reduced by encouraging consumers to choose greener products (Zhang et al., 2015). Compared to the end-of-pipe measures on the demand side, the supply side should curb primary sources of air pollution that aggravate the formation of haze (Zhang, 2010). These changes can improve the production, distribution, circulation, and consumption of products and lead to reduced air pollutant emissions. Acknowledgments Our heartfelt thanks should be given to the Editor, Dr Coffman, as well as anonymous reviewers, and Zhifu Mi for their helpful comments, which have allowed us to improve the manuscript. We thank the National Natural Science Foundation of China (NSFC) (71673083,71303076, 71203062 and 71303174), and Hunan Science and Technology Key Program of China (Grant no. 2015zk2002) for funding support. References Ang, B.W., 1995. Decomposition methodology in industrial energy demand analysis. Energy 20 (11), 1081–1095. Ang, B.W., Liu, F.L., 2001. ‘A new energy decomposition method: perfect in decomposition and consistent in aggregation’. Energy 26 (No. 6), 537–548.
7
Ang, B.W., Liu, F.L., Chew, E.P., 2003. Perfect decomposition techniques in energy and environmental analysis. Energy Policy 31 (No. 14), 1561–1566. Dietzenbacher, E., Los, B., 1998. Structural decomposition techniques: sense and sensitivity. Econ. Syst. Res. 10 (4), 307–324. Dietzenbacher, E., Los, B., 2000. Structural decomposition analyses with dependent determinants. Econ. Syst. Res. 12 (No. 4), 497–514. Ghosh, A., 1958. Input-output approach in an allocation system. Economica 25 (No. 97), 58–64. Guan, D.B., Hubacek, K., Weber, C.L., Reiner, D.M., 2008. The driver of Chinese CO2 emissions from 1980 to 2030. Global Environ. Change 18 (No. 4), 626–634. Guan, D.B., Peters, G.P., Weber, C.L., 2009. Journey to world top emitter: an analysis of the driving forces of China’s recent CO2 emissions surge. Geophys. Res. Lett. 36 (4), 1–5. He, H., Jiang, G.B., 2014. Scientific and reason explain haze problem of China. Qiushi (No. 6), 48–50, In Chinese. He, H., Wang, X.M., Wang, Y.S., Wang, Z.F., Liu, J.G., Chen, Y.F., 2013. The tracking and control of atmospheric haze. Bull. Chin. Acad. Sci. (No. 7), 344–352, In Chinese. Hoekstra, R., Van den Bergh, J.C.J.M., 2003. Comparing structural and index decomposition analysis. Energy Econ. (No. 1), pp. 39– 64. Li, F.Y., Song, Z.Y., Liu, W.D., 2014. ‘China’s energy consumption under the global economic crisis: decomposition and sectoral analysis’. Energy Policy 64, 193–202. Li, H.N., Wei, Y.M., 2015. Is it possible for China to reduce its total CO2 emissions? Energy 83, 438–446. Lyu, W., Li, Y., Guan, D., Zhao, H., Zhang, Q., Liu, Z., 2016. Driving forces of Chinese primary air pollution emissions: an index decomposition analysis. J. Clean. Prod. 133, 136–144. Meng, L., Crijns-Graus, W.H., Worrell, E., Huang, B., 2018. Impacts of booming economic growth and urbanization on carbon dioxide emissions in Chinese megalopolises over 1985–2010: an index decomposition analysis. Energy Efficiency 11 (1), 203–223. Meng, B., Wang, J., Andrew, R., Xiao, H., Xue, J., Peters, G.P., 2017. Spatial spillover effects in determining China’s regional CO2 emissions growth: 2007–2010. Energy Econ. 63, 161–173. Mi, Z., Meng, J., Guan, D., Shan, Y., Liu, Z., Wang, Y., Wei, Y.M., 2017. Pattern changes in determinants of Chinese emissions. Environ. Res. Lett. 12 (7), 074003. Nie, H., Kemp, R., Vivanco, D.F., Vasseur, V., 2016. Structural decomposition analysis of energy-related CO2 emissions in China from 1997 to 2010. Energy Efficiency 9 (6), 1351–1367. Rormose, P., Olsen, T., 2005. Structural decomposition analysis of air emissions in Denmark 1980-2002 Beijing, China. 15th International Input-Output Conference. Rose, A., Casler, S., 1996. Input-output structural decomposition analysis: a critical appraisal. Econ. Syst. Res. 8 (No. 1), 33–62. Su, B., Ang, B.W., 2012. Structural decomposition analysis applied to energy and emissions: some methodological developments. Energy Econ. 34 (1), 177–188. Su, B., Ang, B.W., 2016. Multi-region comparisons of emission performance: the structural decomposition analysis approach’. Ecol. Indic. 67, 78–87. Su, B., Ang, B.W., 2017. Multiplicative structural decomposition analysis of aggregate embodied energy and emission intensities’. Energy Econ. (No. 65), 137–147. Su, B., Ang, B.W., Li, Y., 2017. Input-output and structural decomposition analysis of Singapore’s carbon emissions. Energy Policy 105, 484–492. Wang, H., Ang, B.W., Su, B., 2017. A multi-region structural decomposition analysis of global CO2 emission intensity. Ecol. Econ. 142, 163–176. Weir, M., 1998. Sources of changes in emissions from energy: a structural decomposition analysis. Econ. Syst. Res. 10, 99–112. Wood, R., 2009. ‘Structural decomposition analysis of Australia’s greenhouse gas emissions’. Energy Policy 37 (No. 11), 4943–4948. Xie, S.C., 2014. ‘The driving forces of China’s energy use from 1992 to 2010: an empirical study of input-output and structural decomposition analysis’. Energy Policy 73, 401–415. Yang, C.J., Zhou, Y., Jackson, R.B., 2014. ‘China’s fuel gas sector: history, current status, and future prospects’. Utilities Policy 28, 12–21. Zhang, Y.G., 2009. Structural decomposition analysis of sources of decarbonizing economic development in China: 1992–2006. Ecol. Econ. 68 (No. 8), 2399–2405. Zhang, Y.G., 2010. Supply-side structural effect on carbon emissions in China. Energy Econ. 32 (No. 1), 186–193.
Please cite this article in press as: Xie, R., et al., Supply-side structural effects of air pollutant emissions in China: A comparative analysis. Struct. Change Econ. Dyn. (2018), https://doi.org/10.1016/j.strueco.2018.04.005