What drives CO2 emissions from the transport sector? A linkage analysis

Energy 175 (2019) 195e204

Contents lists available at ScienceDirect

Energy journal homepage: www.elsevier.com/locate/energy

What drives CO2 emissions from the transport sector? A linkage analysis Huibin Du a, b, Zhenni Chen a, Binbin Peng a, *, Frank Southworth c, Shoufeng Ma a, Yuan Wang d a

College of Management and Economics, Tianjin University, Tianjin, 300072, China Center for Energy & Environmental Policy Research, Institute of Science and Development, Chinese Academy of Sciences, Beijing, 100190, China School of Civil and Environmental Engineering, Georgia Institute of Technology, 790 Atlantic Drive, SEB Building, Atlanta, GA, 30332, USA d School of Environmental Science and Engineering, Tianjin University, Tianjin, 300072, China b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 21 November 2018 Received in revised form 27 February 2019 Accepted 9 March 2019 Available online 12 March 2019

The objective of this study is an analysis of the relationship between the transportation sector and the rest of the Chinese economy as it impacts the generation of carbon dioxide (CO2) emissions. Using data from input-output tables of China for 2002, 2007, and 2012, the hypothetical extraction method (HEM) was used to quantify these inter-sectoral linkages and decompose the CO2 emissions associated with each transport modal sub-sector (i.e. the rail, road, water, and air sub-sectors). The transport services required by other sectors of the economy are found to induce more carbon emissions than the emissions associated with transportation's own final demands. As the principal CO2 emitter, the road sub-sector's export of CO2 emissions increased substantially from 11.31 to 55.22 million tons (Mt) between 2002 and 2012. In contrast, technological advances within the rail sub-sector resulted in a net decrease in exported CO2 emissions, from 9.99 Mt to 5.16 Mt. Taken as a whole, the transportation sector transferred large CO2 emissions through the economy's various supply chains, notably exporting to the service (63.78 Mt in 2012) and construction (83.88 Mt in 2012) sectors while importing a large amount of CO2 emissions from the nation's energy producing industry (223.52 Mt in 2012). © 2019 Elsevier Ltd. All rights reserved.

Keywords: CO2 emissions Transportation sector Linkage analysis Hypothetical extraction method

1. Introduction Climate change poses a grave threat to the sustainable socioeconomic development of all countries. The United Nations Climate Change Conference (COP 21), held in Paris in December 2015, reached a consensus agreement on climate among all participating nations to try to limit the global average temperature rise to below 2
C above pre-industrial levels [1]. To achieve this objective means controlling CO2, which accounts for about 80% of total GHG emissions [2,3]. As the largest developing country, China has the largest energy consumption and CO2 emissions among all countries in the world. According to the International Energy Agency (IEA), the CO2 emissions from fuel combustion in 2015 achieved 9084 Mt, which accounts for 28% of global CO2 emissions [4]. In China, the

* Corresponding author. E-mail addresses: [email protected] (H. Du), [email protected] (Z. Chen), [email protected] (B. Peng), [email protected] (F. Southworth), [email protected] (S. Ma), [email protected] (Y. Wang). https://doi.org/10.1016/j.energy.2019.03.052 0360-5442/© 2019 Elsevier Ltd. All rights reserved.

transportation sector is the third largest source of CO2 emissions, following electricity and heat production, manufacturing industries and construction. The transportation sector in China has developed rapidly with the nation's progress toward greater industrialization and urbanization. Nationally, passenger-kilometers increased by 132% over the decade 2002e2017, while freight ton-kilometers grew by 289% over this same period.1 This rapid growth brought with it an 8% per annum increase in CO2 emissions over the period 2002e2016, largely the result of its high dependence on petroleum products. According to the China Statistical Yearbook, between 2002 and 2016 the average growth rate of gasoline consumption was more than 9.5% and that of diesel more than 8.4%. The latest IEA report on CO2 emissions from fuel combustion (2017 edition) points out that in 2015 the CO2 emissions resulting directly from transportation

1 Passenger-kilometers (Freight ton-kilometers) refers to the sum of the product of the volume of transported passengers (cargo) multiplied by the transport distance. The data comes from the National Bureau of Statistics of China.

196

H. Du et al. / Energy 175 (2019) 195e204

services in China totaled some 836.6 Mt, accounting for 9% of China's total CO2 emissions: while in large metropolitan areas such as Beijing, transportation's share of total CO2 emissions reached as high as 22% in 2012 [5]. The acceleration of urbanization and industrialization in China seems likely to further increase this demand for transportation services, suggesting continued growth in CO2 emissions over the near term [6,7], and hence the interest in carbon mitigation policies targeting transportation services. To date the literature on CO2 emissions from the transportation sector in China has focused on the spatial and temporal variation in these emissions and on the driving factors influencing its continued growth [8e14]. For example, Zhou et al. [15], and Wang and He [7] evaluated the carbon performance of the transportation sector within 30 administrative regions in China during different periods and found the CO2 emissions efficiency of the coastal area of south China to be better than that of the other regions, while Zhang et al. [12]; and Cui and Li [16] found that energy intensity and changes in technology, as well as management factors, have had a great influence on carbon emissions from transportation. These past studies, however, focused most of their attention on the “direct” CO2 emissions within transportation, i.e. those emissions induced by energy consumption activities that are owned or controlled within the transportation sector. Much less attention has been given to the transportation sector's CO2 emissions by considering its inter-sectoral linkages, neglecting the “indirect” CO2 emissions2 resulting from both intermediate purchases and sales between the transportation sector and the rest of the economy, including some highly carbon-intensive sectors [17e19]). In this paper we consider these inter-sectoral linkages with transportation in order to identify its net real influence on the carbon emissions produced within the full national economy. There are three reasons for focusing on these upstream and also downstream linkages between transportation and the other sectors in the economy. First, besides its direct energy consumption which produces large CO2 emissions, the transportation sector also needs intermediate inputs (e.g. vehicle parts, fuels) from the rest of the economy, whose own production processes also consume large amounts of energy and produce large CO2 emissions [17,20e22]. Thus the above indirect, upstream CO2 emission embodied in these intermediate inputs between the transportation sector and the rest of the economy cannot be neglected. Second, consumption is the driving force behind this production. Previous transportation sector studies have mainly examined CO2 emissions from the production side. It is also necessary to analyze CO2 emissions from the consumption, or downstream, side in order to explore the fundamental causes triggering CO2 emissions from a service sector such as transportation [23]. A more complete upstream and downstream linkage analysis of CO2 emissions can determine the amount of embodied CO2 that flows from production sector to consumption sector and thereby help to clarify the role of transportation as either a CO2 ‘net exporter’ or ‘net importer’ within the national economy. Ignorance of such inter-sectoral CO2 emissions linkages underestimates the real influence of carbon emissions from the transportation sector and its spillover effects3 on other economic sectors, which goes against the realization of targets to reduce CO2 emissions within the entire economy. From a methodological perspective, a number of scholars have

2 Indirect CO2 emissions are a consequence of the activities of the transportation sector, but occur at sources owned or controlled by other sectors in the rest of the economy. 3 According to Piaggio et al. [17], spillover effect refers to the CO2 emissions produced by the rest of the economy to meet the transportation sector's final demand.

used a variety of techniques to investigate both upstream and downstream inter-sectoral linkages when considering carbon emissions, with a focus on measuring sectoral linkages caused by the per unit output of different economic sectors. For example, Liu et al. [24] used an input-output framework to ‘map’ the most important inter-sectoral relationships associated with CO2 emissions. Chen et al. [25] used the classical multiplier method (CMM), a well-established linkage indicator of backward and forward linkages, to quantify the dependencies of CO2 emissions between regions and sectors in Australia. Liu et al. [26] used integrated backward and forward linkage analysis, structural path analysis, and Index decomposition analysis to investigate CO2 emissions linkages associated with the iron and steel industry and construction materials industry in the context of the entire economic system. Zhang et al. [27] calculated a carbon emission multiplier effect, a spillover effect, and a feedback effect in their study of the linkage of CO2 emissions from regional and sectoral perspectives. Of particular interest to this present study, with its focus on a single sector (i.e. transportation) is the hypothetical extraction method (HEM), which has been widely applied to analyze sectoral linkages and reveal the role and position of individual regions and sectors within an entire economic system. HEM is a method used to analyze sectoral linkages by first of all extracting hypothetically a single sector from the economy. That is, the extracted sector cannot sell products to other sectors and also cannot purchase products from these other sectors [18]. Inter-sectoral linkages are then quantified by computing the output loss to the rest of the economy after the extraction. An advantage of HEM here is that it considers not only the relative magnitude of final demand in each sector but also the relative effect of this demand on overall output [19]. In recent studies HEM has been used not only to identify the economic role of an individual sector, such as in the case of the construction sector [28], but also to evaluate resource usage and environment effects linkages, such as water use [29], as well as the production of CO2 emissions [30]. Wang et al. [18], Ali [30] and Zhao et al. [31] all adopted HEM to investigate the in-depth characteristics of CO2 emissions linkages among sectors in China, Italy, and South Africa, respectively. Wang et al. [18] found that the energy industry, basic industry and transportation sectors have the highest direct CO2 emissions, which mainly flow to service, construction and technology industries. Meanwhile, the transportation sector had the largest share of carbon emission imports. Fang et al. [32] conducted a linkage analysis by using HEM to explore the water-carbon nexus among provinces. Bai et al. [33] explored fuel-related CO2 flows among 30 Chinese industrial sectors in 2012 using HEM, and revealed the specificities of these flows on total CO2 emissions abatement for the whole economy. Liao et al. [34] utilized a modified HEM to analyze CO2 linkages among sectors in Beijing. HEM has also been extended to analyze the interdependent nexus between sectors and households in the context of CO2 emissions [35,36]. Despite the growing concern for sectoral linkages in terms of CO2 emissions, the existing literature mainly focuses on the CO2 emissions linkages of all sectors in the entire economic system. In this study then we focus our attention on the inter-sectoral linkage analysis of a specific sector: the transportation sector's CO2 emissions in China. We accomplish this by applying HEM to China's national input-output tables for the years 2002, 2007 and 2012, using the HEM approach to not only consider the CO2 emissions from the direct energy consumption of transportation from the perspective of its production, but also consider the indirect CO2 emissions resulting from the production of intermediate inputs associated with the transportation sector's final consumption. Furthermore, because of the rapidly evolving role played by the transportation sector in the entire national economy over time, we

H. Du et al. / Energy 175 (2019) 195e204

conduct an inter-temporal comparison to capture the changes in these CO2 emissions linkages over time. We take four different transport modes (i.e. rail, road, water, and air transportation) as target sub-sectors and calculate their CO2 emissions linkages with the rest of the economy. This subdivision helps us to better understand the heterogeneity of CO2 emissions linkages among the various transport modes, which in turn helps us to propose better targeted policy recommendations. In doing so, we use the transportation sector as an example to show in detail how to investigate the CO2 emission impacts of a specific sector on the whole economic system from a sectoral linkage perspective, noting that the conceptual and mathematical framework we describe could also be applied to other sectors within a national or regional economic system. The rest of this study is organized as follows. Section 2 introduces the methodology and data resources used. Section 3 then presents the work's most significant empirical results. Finally, section 4 offers a brief summary and discussion of the broad policy implications these results suggest. 2. Methodology and data

197

Ad ¼ A  Am

(5)

where Am is the direct input coefficients matrix of imported intermediate input and Ad is the direct input coefficients matrix of domestic intermediate input. M is an N  1 vector of imports, E is an N  1 vector of exports. Thus, the output of the whole economic system should be:

1  X ¼ I  Ad Yd

(6)

and where Yd is the final demand satisfied by the domestic products. A sector not only directly discharges CO2 by consuming energy during the process of production, but also indirectly discharges CO2 by consuming energy through intermediate inputs from other sectors. Total CO2 emissions intensity represents the sum of direct and indirect CO2 emissions caused by the production of per-unit final demand, which reflects CO2 emissions intensity in the whole supply chain. It is calculated by multiplying the direct CO2 emissions intensity row vector by the Leontief inverse matrix:

2.1. Direct and total CO2 emissions intensity

1  B ¼ C I  Ad

CO2 emissions intensity is defined here as the volume of CO2 emissions per unit of economic output. As such, it is taken to be a measure of the relationship between CO2 emissions and economic growth. The direct CO2 emissions intensity is calculated by dividing CO2 emissions by the total output [18]:

where B is the total CO2 emissions intensity row vector and C ¼ ðci Þ is the direct CO2 emissions intensity vector. Total CO2 emissions can be obtained by the follow equation:

P8 ci ¼

m¼1 pm fim

Xi

(2)

where X is an N  1 vector with elements of total output Xi (i ¼ 1, …, N); A ¼ (aij) is the direct input coefficients matrix reflecting the input-output relationships among all sectors, in which aij ¼ xij/Xj (i, j ¼ 1, 2, …, n) is the amount of direct input from sector i used in the per unit output of sector j; and Y is an N  1 vector of final demands Yi. Then Eq. (2) can be transformed as follows:

X ¼ ðI  AÞ

1

Y

(3)

where I is the identity matrix; ðI  AÞ1 is the Leontief inverse matrix, in which elements indicate the product of a sector i required directly and indirectly to satisfy per unit final demand of sector j. In an open economic system, the direct input includes domestic and imported products, that is A ¼ Am þ Ad .

Am ¼

M A XþME

1  C ¼ CX ¼ C I  Ad Yd

(8)

(1)

where ci is the direct CO2 emissions intensity of the sector i; Xi is the total output of sector i. fim is the consumption of energy m in sector i. In this study, there are eight energy types: coal, coke, crude oil, gasoline, kerosene, diesel oil, fuel oil, and natural gas (m ¼ 1, 2, …,8); and pm is the CO2 emissions factor associated with energy type m. According to the input-output table, total output is equal to the combined inputs of products in each economic sector. The equation expressing input-output equilibrium for an economy with n sectors (n ¼ 34 in this study) is expressed as follows (see Ref. [31]; for example):

X ¼ AX þ Y

(7)

(4)

2.2. Calculation of CO2 emissions linkages by the hypothetical extraction method The principle behind the hypothetical extraction method (HEM) for measuring CO2 emissions linkages is to compare the CO2 emissions of the whole economy with that of a hypothetical economy in which one or more target sectors are extracted. For simplicity, the whole economy is divided into two blocks defined as Qs and Q-s. Qs is the group of target sectors with similar characteristics and in this study Qs is the four specific transportation subsectors, namely rail, road, water and air sub-sector respectively. Q-s is the group of remaining economic sectors. The relationship between Qs and Q-s can be described as follows [19]:

 Q¼

Qs; s Qs; s

Qs;s Qs;s

 (9)

The total CO2 emissions of the whole economy can be written as:



Cs Cs



 ¼

Cs 0 

0 C s



Ds ;s Ds;s Ds; s Ds;s



Ys Ys

 (10)

  Ys is the vector of total CO2 emissions; Y ¼ Ys   D D s; s s;s is is the vector of final demands and ðI  AÞ1 ¼ where C ¼

Cs Cs



Ds; s Ds;s

the Leontief inverse matrix. In the hypothetical economy, the block Qs is extracted, that is, the supply and demand relationship between Qs and Q-s is removed. The total CO2 emissions of this resulting hypothetical economy can be expressed as:

198



H. Du et al. / Energy 175 (2019) 195e204

C *s C *s 

¼  ¼



 ¼

Cs 0

Cs 0

0 C s

Cs 0

0 C s

0 C s





X *s X *s

As; s 0



0 As;s

1 "  I  As; s 0



X *s X *s



 þ

0

Ys Ys

 1 I  As;s



#

Ys Ys

 (11)

The change in CO2 emissions that reflects the impact of the block Qs on the whole economy's carbon emissions is given by Ref. [30]:

C  C* ¼





¼

Cs 0 0 C s   Ys  Ys

¼

  

"

 Ds; s  I  As; s 1

Ds; s

  1  C s Ds; s  I  As; s

"

¼



Cs  C *s Cs  C *s

Ys Ys



C s Ds; s

Us; s Us;s Us; s Us;s

Ds;s  Ds;s  I  As;s 1

#

C s Ds;s   1  C s Ds;s  I  As;s

from Qs to Q-s to satisfy the final demand of Q-s. In the above four indicators of CO2 emissions linkage, IE, ME and NFLE reflect the CO2 emissions generated in the production process of block Qs to produce goods, which are not only used by block Qs but also input to Q-s. On the other hand, IE, ME and NBLE reflect the CO2 emissions generated in block Qs or imported from Q-s to satisfy the final demand of Qs. They are the total CO2 emissions caused by the consumption of block Qs. So in order to analyze the resulting CO2 emissions linkages from the perspective of both total production and consumption, we define output emissions (OE) and demand emissions (DE) as follows:

OE ¼ IE þ ME þ NFLE

(17)

DE ¼ IE þ ME þ NBLE

(18)

And the net transfer emissions (NTE) of block Qs can be written as [19]:

NTEs ¼ OEs  DEs ¼ NFLEs  NBLEs

#

(19)

where NTEs >0 indicates that block Qs exports CO2 emissions to the economic system; NTEs <0 indicates that block Qs imports CO2 emissions from the economic system; and NTEs ¼ 0 indicates that exports and imports of CO2 emissions between Qs and the rest of the economy offset each other. 2.4. Decomposition of net transfer emissions from a sectoral perspective



Ys Ys

 (12)

2.3. Decomposition of CO2 emissions linkages by modified HEM

In order to further analyze the CO2 emissions linkages among different sectors, the NFE and NBE are divided [18]. Assume that sector t is a sector in block Q-s, the total net CO2 emissions transferred from block Qs to Q-s can be decomposed by CO2 emissions transfers to each sector in Q-s. The net forward linkage emissions (NFLE) can be described as:

X X 0 NFLEs/ t ¼ us C s Ds; t Yt ; ðt2  sÞ

According to the method described by Duarte et al. [37], the CO2 emissions related to the block Qs can be decomposed into four separate components, namely, internal emissions (IE), mixed emissions (ME), net backward linkage emissions (NBLE), and net forward linkage emissions (NFLE), each of which are defined as follows [37]:

NFLEs ¼

 1 0 IEs ¼ us C s I  As;s Ys

(13)

The difference between NFLEs/ t and NBLEt/ s is the net CO2 emissions linkage between sector t and s, namely NTs/ t .

h  1 i 0 Ys MEs ¼ us C s Ds;s  I  As;s

(14)

NTEs/ t ¼ NFLEs/t  NBLEt/s ¼ us C s Ds;t Yt  ut C t Dt;s Ys

0

NBLEs ¼ us C s Ds;s Ys 0

NFLEs ¼ us C s Ds;s Ys

Similarly, the net CO2 emissions transferred from block Q-s to Qs can be described as:

NBLEs ¼

X X 0 NBLEt / s ¼ ut C t Dt;s Ys ; ðt2  sÞ

0

(15) (16)

Here IE is the CO2 emissions related to the block Qs itself. In that process, the block Qs does not trade with block Q-s and only consumes the intermediate goods of block Qs. That is, the CO2 emissions come from the intermediate goods produced, sold and purchased inside the block Qs. ME is the CO2 emissions in block Qs which are generated by the goods that sold from Qs to Q-s to form part of goods in Q-s and then repurchased by Qs from Q-s as intermediate inputs of Qs. Therefore, ME combines the results of backward linkage (emissions from Q-s to Qs) and forward linkage (emissions from Qs to Q-s). NBLE is the net CO2 emissions released from goods imported from Q-s to satisfy the final demand of Qs, while NFLE is the net CO2 emissions released from goods exported

(20)

(21)

0

(22)

A positive value of NTEs/ t indicates net output of CO2 emissions from sector s to t, while a negative value of NTEs/ t reflects a net input of CO2 emissions from sector t to s. 2.5. Data source and processing In this study, we analyze the CO2 emissions linkages of the transportation sector in 2002, 2007 and 2012 based on the inputoutput tables from the National Bureau of Statistics of China. The energy consumption data also comes from the website of the National Bureau of Statistics of China. Energy-related CO2 emissions are estimated based on the Guidelines for National Greenhouse Gas Inventories [38]. In order to discover the difference in CO2 emissions linkages associated with the transportation sector in different years, we modified the input-output table of China using the consumer price index of 2002 as a benchmark to obtain comparable price input-output tables for 2002, 2007 and 2012. Through the

above procedure, we eliminate the influence of price change on CO2 emissions. According to the classification method used in the national Input-Output Table and the energy statistics for China, we allocated more than 100 sectors4 in the economy to 34 more aggregate industrial sectors. The four transportation sub-sectors, namely the rail, road, water and air transport sector, are treated as the target sectors in this study. 3. Results and discussion

carbon intensity (t/104 RMB)

H. Du et al. / Energy 175 (2019) 195e204

199

8.00 7.00 6.00 5.00 4.00 3.00 2.00 1.00 0.00 2002 2007 2012 Rail

2002 2007 2012 Road

direct carbon intensity

2002 2007 2012 Water

2002 2007 2012 Air

indirect carbon intensity

3.1. CO2 emissions and intensity We estimate that the CO2 emissions of the whole transportation sector increased from 239.9 Mt to 600.3 Mt during the decade 2002 to 2012, based on its energy consumption and energy-to-carbon emissions factors reported in Ref. [38]. Of note, Guo et al. [39] used the same accounting method to calculate the total CO2 emissions from transportation within the eastern, central and western regions in China, with their estimated aggregate CO2 emissions from transportation increasing from some 200 Mt to 600 Mt during the decade 2002e2012 [39], a result consistent with our own calculation. Fig. 1 shows both the direct and indirect carbon intensities we obtained. In the four transportation sub-sectors (mode), the indirect carbon intensities all exceeded the direct carbon intensities in each year, representing more than 60% of total (direct þ indirect) emissions intensities. In the transportation sector, the indirect CO2 emissions embodied in intermediate inputs have a significant effect on the total CO2 emissions produced to satisfy the final demand of transportation. Thus the sectoral linkages between transportation sector and other economic sectors cannot be ignored when seeking to promote low carbon development of the whole economy. The direct CO2 emissions intensities of rail, road and water transportation decreased due to the development of technology between 2002 and 2012. For example in the rail sub-sector, the rapid development of railway electrification and use of high-speed trains in China has led to a significant adjustment in the mode's energy structure. According to data from Bulletin of Statistics for the Chinese Railway, between 2002 and 2012 the energy consumption of the rail sector decreased from 1992 Mt of standard coal to 1746 Mt of standard coal, while the economic output of the rail sector increased by 86% over this same decade. Thus the direct CO2 emissions intensity of the rail sector between 2002 and 2012 decreased significantly. In contrast, the air transportation subsector had a higher direct CO2 emissions intensity in 2012 than in 2002, perhaps reflecting a longer aircraft technology replacement cycle over this same period. 3.2. Decomposition of carbon linkage emissions from the transportation sector Fig. 2 shows the carbon emissions linkages of the transportation sector. The influence of transportation on the whole economy's CO2 emissions kept increasing over the 2002e2012 period. Taking the four transportation sub-sectors as a whole, between 2002 and 2012 the output emissions (OE) of the transportation sector increased from 233 Mt to 470 Mt while demand emissions (DE) increased from 191 Mt to 423 Mt. Thus the transportation sector had an increasingly important impact on CO2 emissions from both production and consumption perspectives. From a single sub-sector

4 There are 123 sectors in the input-output table of China in 2002, 135 sectors in 2007 and 139 sectors in 2012.

Fig. 1. CO2 emissions intensity of the four transportation sub-sectors in three different years.

perspective, OE and DE both increased in the road and air subsectors over this same decade, as the nation's rapid economic growth, accelerated urbanization process, and improved living standards supported significant growth in the demand for both road and air travel. This was especially true for the road sub-sector. Its OE increased from 76 Mt to 291 Mt between 2002 and 2012, and its DE increased from 65 Mt to 235 Mt during the same period, representing an average five-year growth rate of OE and DE of more than 90%. In China, more than 90% of passenger trips and about 75% of freight traffic volume occurred on the nation's roads during 2002e2012, so that by 2012 the road sub-sector had the largest OE and DE among the four modes of transport, with an increase in net exported CO2 emissions (i.e. OE being larger than DE). In 2012, the air sector also became a net CO2 emissions exporter, i.e. its OE was more than its DE, as the air sector produced more CO2 emissions to meet a growing downstream demand. It is clear from Fig. 2 that NFLE and NBLE account for a larger share than IE and ME in the transportation sector, particularly in the rail and road sub-sectors, where the proportion of NFLE in OE is more than 70%, indicating that the transportation sector emits a large part of CO2 emissions to satisfy the final demands of the rest of the economy. That is, the transportation sector has a large push effect on the whole economy's CO2 emissions by supplying transport services to downstream sectors. Meanwhile, for all transportation sub-sectors the proportion of NBLE in DE is more than 60%, showing that the provision of transport services, as expected, requires significant purchased inputs from its upstream sectors and thus the final demand from the transportation sector also has a large pull effect on the whole economy's CO2 emissions, a conclusion consistent with the results from, for example, Wang et al. [18]. Technological change has also influenced significantly the CO2 emissions linkages of transportation during the current century. For example, it is worth noting that compared with 2002, the rail sub-sector's OE and DE decreased between 2002 and 2007 as well as between 2002 and 2012. Over the decade 2002e2012 diesel locomotives replaced steam locomotives, with a subsequent move to electric locomotive development: supported by a shift from coalfired dominated to diesel-fired dominated and then electricity dominated energy feedstocks. With these technological advances in the rail sub-sector the proportion of NBLE in DE increased from 68.7% in 2002 to 74.7% in 2012, indicating a greater dependence on such industries as the smelting and pressing of ferrous metals, and electric power generation. As a result, the amount of net CO2 emissions transferred from the rail sub-sector to the whole economy decreased from 9.99 Mt to 5.16 Mt between 2002 and 2012. Within the water sub-sector, OE and DE decreased between 2007 and 2012: due in large part to the 2008 global financial crisis, which significantly affected international shipping. In trade between China and overseas countries water transportation plays a

200

H. Du et al. / Energy 175 (2019) 195e204

Fig. 2. Decomposition of CO2 linkage emissions to transportation in 2002, 2007 and 2012.

significant role because of its relatively low shipping costs. From the production perspective, the NFLE that reflects the CO2 emissions exported from the water sub-sector to the rest of the Chinese economy decreased from 66.16 Mt in 2007 to 36.74 Mt in 2012. In terms of economic output from the water sub-sector in the inputoutput tables of China, between 2007 and 2012 the use of water transportation services as intermediate inputs to other sectors of the economy decreased by 32.93%, as the 2008 global financial crisis impacted the demand for ocean transportation from these downstream industries. In addition, substitution effects among different transport modes also influences the CO2 emissions linkages of each subsector. For example, the road transportation sub-sector plays an increasingly important role in the national economy, while the role of the water transportation sub-sector plays in the national economy tends to be declining. According to data from input-output tables of China, we found that in 2002 the road sub-sector accounted for 40% and the water sub-sector accounted for 32% of the total economic output of the whole transportation sector. By 2012 the road sub-sector proportion had increased to 67%, while that of the water sub-sector had decreased to 12%. Thus the gap of OE and DE between the road and water sub-sectors expanded as road transportation prospered greatly and became a partial substitute for water transportation. Of note here, China launched a four trillion Yuan ($586 billion) plan to address the 2008 global financial crisis. One part of its central stimulus package was to invest 1.5 trillion Yuan in large-scale infrastructure projects including railways, roads, airports and its national electricity grid [40]. With the rapid development of other transport modes, water transport showed a decrease in its overall demand from other economic sectors. As a result, between 2002 and 2012 the role of the water sub-sector in the CO2 emissions of the whole economy changed, from a CO2 exporter in 2002 to a CO2 importer after 2007. In contrast, the role of the air sub-sector changed from a CO2 importer before 2007 to a CO2 exporter by 2012, and the influence of CO2 emissions from the air sub-sector had become greater than that

from the water sub-sector. In 2012, the air sub-sector's NFLE was 11 Mt more than that of water sub-sector, indicating a growing volume of CO2 emissions exported to other economic sectors and an increasing CO2 emission influence on the whole economy. 3.3. Net transfer emissions from a sectoral perspective In order to have a clear understanding of the transportation's CO2 emission linkages we integrated the remaining 3 non-target transportation sub-sectors and 30 non-transportation sectors of the economy into 7 groups, termed respectively agriculture, energy industry, basic industry, light industry, technology industry, service, and construction (as shown in Appendix B Table B1). According to Eq. (22) the net effect of CO2 emissions between each transportation sub-sector and the remaining sectors in the economy was obtained as shown in Fig. 3. From the perspective of emission imports, as shown in Fig. 3, the imported CO2 emissions to the transportation sector are mainly from the energy industry, with imported CO2 emissions from the energy industry to the whole transportation sector increasing from 100.91 Mt in 2002 to 223.52 Mt in 2012. The energy producing industries such as the oil processing industry and electric power industry are also energy intensive sectors. In China, more than 57% of oil products (i.e. gasoline oil and diesel oil) are used in the transportation sector. For the road sub-sector, the imported CO2 emissions from the energy industry increased from 34.09 Mt to 125.72 Mt between 2002 and 2012, doubling every five years during the 10-year period, and the imported CO2 emissions from the oil processing industry accounted for more than 60% of the total emissions produced by the energy industry. The development of the rail sub-sector also had a pull effect on the CO2 emissions in the electric power sector. In 2012, the imported CO2 emissions from the electric power and heat power subsectors totaled an estimated 7.25 Mt, which accounted for 54% of the total imported CO2 emissions of the rail sub-sector. Based on the input-output relationships it is found that the proportion of

H. Du et al. / Energy 175 (2019) 195e204

201

Fig. 3. Decomposition of net transfer of each transportation sub-sector (Mt). Note: In the middle are the four targeted transportation sub-sectors, each of them imports CO2 emissions from the sectors on the left and exports CO2 emissions to the sectors on the right. The specific amount of net transfer emissions between each transportation sub-sector and other sectors are shown in Table C1 in Appendix C.

intermediate input from electric power and heat power increased from 3.9% to 30.3% between 2002 and 2012. In the wake of development in high-speed railways, the numbers of electric locomotives in China grew from 3876 in 2002 to 10,047 in 2012, making extensive use of electricity products which came mainly from coalfired thermal power generation. That is, increased electricity demand from the rail sub-sector induced a significant spillover effect, increasing CO2 emissions within the electric power industry. This spillover effect results from rail technology innovation and its associated energy sourcing (feedstock) adjustments, supported by environmental policy [41e43]. To mitigate this CO2 spillover effect, adoption of clean electricity can help the rail sub-sector substantially reduce its imported CO2 emissions. From the perspective of emission exports from the transportation sector as a whole, the output of CO2 is mainly transferred to the service and construction sectors. Between 2002 and 2012, the exported CO2 emissions from the whole transportation sector to service sector increased from 36.44 Mt to 63.78 Mt and the exported CO2 emissions from the whole transportation sector to construction increased from 45.43 Mt to 83.88 Mt. The service sector accounted for the largest proportion of these emissions in the rail and air sub-sectors: especially in the air sub-sector, where the proportion of exported CO2 emissions attributable to service sector was more than 50%, and the amount of CO2 emissions transferred from the air transportation sub-sector to the service sector increased from 3.75 Mt in 2002 to 22.21 Mt in 2012. The tourist industry falls under this service sector, and with a growing demand for leisure and entertainment, tourism has been a fastgrowing sector in the Chinese economy. The development of tourism also appears to have contributed to a rapid increase in the demand for, and development of the air sub-sector. From Fig. 3 it is also clearly seen that the road sub-sector outputs significant CO2 emissions to the rapid urbanization-driven construction sector, with a growth rate of 349% between 2002 and 2012. The road sub-sector also exported significant CO2 emissions to both technology and light industries. The emissions linked to technology industry increased from 9 Mt to 48.9 Mt between 2002 and 2012, while those to light industry increased from 7.92 Mt to 29.96 Mt over the same decade. In contrast, the CO2 emissions

exported from the water transport sub-sector to the construction sector decreased from 24.5 Mt to 8.23 Mt. 4. Conclusions With the rapid development of industrialization and urbanization the transportation sector plays an increasingly important role in the national economy of China, and in its people's daily life. Given its importance, the environmental performance of the sector deserves serious consideration. To achieve a 17% reduction in the CO2 emissions intensity of the whole economy, the Chinese government should not ignore the carbon emissions in the sector's supply and demand chain. To address this issue, we used the national inputoutput tables and a HEM model to trace the CO2 emission linkages between this transportation sector and the rest of economy, and carried out an inter-temporal and mode specific comparison analysis to identify any trends within the transportation sector's CO2 emissions linkages. A number of conclusions can be drawn from our analysis: First, the transportation sector has an increasingly important impact on the whole economy's CO2 emissions, from both production and consumption perspectives. In particular, and as expected, the transportation sector transfers significant CO2 emissions “downstream” to the service and construction sectors, through the national economy's supply chains, while itself importing large CO2 emissions from the energy producing sector. Specifically, we find that the output emissions (OE) of the transportation sector increased from 233 Mt to 470 Mt while its demand emissions (DE) increased from 191 Mt to 423 Mt between 2002 and 2012. A large proportion of these OE emissions are attributable to net forward linkage emissions (NFLE), indicating that the transportation sector has a large push effect on the economy's carbon emissions by providing transport services to other economic sectors. A similarly large proportion (more than 50%) of transportation's DE emissions are attributable to net forward linkages (NBLE) indicating a transportation sector that also has a large pull effect on the economy's carbon emissions by importing intermediate goods from different sectors within the economy. For CO2 emissions reduction purposes therefore these forward and

202

H. Du et al. / Energy 175 (2019) 195e204

backward inter-sectoral CO2 emissions linkages should be considered carefully when formulating carbon reduction policies. Second, while as a whole the transportation sector is a CO2 emissions exporter, with a high dependence on other economic sectors, both technological and economic developments have affected these carbon emissions linkages in different ways across the four transportation modes studied. According to the net transfer emissions (NTE) of each transportation sub-sector, the rail and road sub-sectors are CO2 emission exporters, but with the road sub-sector increasing its net exported carbon emissions from 11.31 Mt to 55.22 Mt, while the rail sub-sector decreased its net exported carbon emissions from 9.99 Mt to 5.16 Mt between 2002 and 2012. In contrast, during this same period the water transport sub-sector changed from a CO2 emissions exporter to importer, as it saw a reduction in its final demand for services: while the air transport sub-sector changed from a CO2 emissions importer to exporter, with a noticeable push effect on CO2 emissions within non-transportation sectors of economy. To achieve significant CO2 emissions reductions for the whole economy, it is therefore necessary to put forward suitable policies according to the different roles played by each transportation subsector. By far the largest net exporter of CO2 emissions comes from the road transportation sub-sector, requiring a significant reduction in the current dependence on fossil fuels (gasoline and diesel), as well as improvements in vehicle operating efficiencies. Where feasible, a shift to other modes of travel, notably to rail, should also be considered, possibly through the use of preferential use policy instruments (e.g. subsidies). Based on our calculations, the rail transportation sub-sector has the lowest carbon intensity as well as the lowest carbon linkage emissions from the year 2007. With the development of high-speed, electrified railways, reduced CO2 emissions from rail transportation make it a good choice for policy makers to promote further: including increased high-speed rail investment and the expansion of rail infrastructure construction to promote the use of high-speed rail transport services. This said, the rail transportation sub-sector imports a growing volume of CO2 emissions from the electric power and heat power sub-sectors, due to an increase in the number of electric locomotives, and as long as

the electricity industry is the main source of coal consumption in China this limits the mode's current response to further CO2 reductions: so that adoption of cleaner sources of electricity can help the rail sub-sector to substantially reduce its embodied CO2 emissions. A growing air transportation sub-sector also warrants careful study as the demand for its services grows, and as further economic growth produces a more time-sensitive and more affluent traveling public. As a CO2 emissions exporter, the air transportation subsector has an increasingly large influence on the whole economy's CO2 emissions, suggesting the investigation and promotion of cleaner energy sources in air transportation also, as an important step towards achieving a goal of economy-wide carbon reduction. The water transportation sub-sector was a CO2 importer by 2012, importing large CO2 emissions from the nation's energy industry, particularly its oil processing industry, so that moving towards cleaner energy sources and more efficient vessels and port operations is also an important topic for policy development. Finally, the considerable volume of exported CO2 emissions from the transportation sector to the service and construction sectors within the Chinese economy increased steadily between 2002 and 2012. Regulating demands from these downstream industries (notably the service and construction sectors) is also an effective way to achieve CO2 emissions abatement of the transportation sector and even the whole economic system. Thus it is also necessary to find ways to encourage those sectors to use low carbon transportation. Acknowledgements This study was supported by the National Key R&D Program of China (No. 2018YFC0213600), National Natural Sciences Foundation of China (Nos. 71834004, 71431005, 71673198, 71273185 and 41571522). Appendix A

Table A1 Nomenclature. Symbols ci fij pj Xi xij X Y A aij I Am Ad B C C C*

D u’ IE ME Mt NBLE NFLE OE DE NTE

direct CO2 emissions intensity of the sector i consumption of energy j in sector i CO2 emissions factor of energy j total output of sector i direct input from sector i for the output of sector j total output vector final demand vector direct input coefficients matrix amount of direct input from sector i used in the per unit output of sector j Identity matrix direct input coefficients matrix of imported intermediate input direct input coefficients matrix of domestic intermediate input total CO2 emissions intensity row vector direct CO2 emissions intensity vector total CO2 emissions of the whole economy total CO2 emissions of the hypothetical productive relationships elements of Leontief inverse matrix unit vector internal emissions mixed emissions million tons net backward linkage emissions net forward linkage emissions output emissions demand emissions net transfer

H. Du et al. / Energy 175 (2019) 195e204

203

Appendix B Table B1 Categorization of sectors. No

Sectors

Groups

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

Agriculture, hunting, forestry and Fishing Mining and washing of coal Extraction of petroleum and natural gas Coke, refined petroleum and nuclear fuel Electric power and heat power production and distribution Gas production and distribution Water production and distribution Food, beverages and tobacco Textiles Textile, leather, leather and footwear Wood and products of wood and cork Pulp, paper, printing and publishing Mining and processing of metal ores Mining and processing of non-metal ores Chemicals and chemical products Non-metallic mineral Smelting and pressing of metals Mental products General and special purpose machinery Transport equipment Electrical machinery and equipment Communication equipment, computers and other electronic equipment Artwork and other manufacturing Recycling and disposal of waste Construction Rail transport Road transport Water transport Air transport Pipeline transport Warehousing Postal service Wholesale and retail trade, hotel and restaurants Others

Agriculture Energy industry

Light industry

Basic industry

Technology industry

Construction Service

Appendix C

Table C1 Decomposition of net transfer of each transportation sub-sector (Mt). Year

sector

2002

Agriculture Industry

Sub-sector

Energy industry Light industry Basic industry Technology industry

2007

Service Construction Net transfer Agriculture Industry Energy industry Light industry Basic industry Technology industry

2012

Service Construction Net transfer Agriculture Industry Energy industry Light industry Basic industry Technology industry Service Construction Net transfer

rail

road

water

air

1.27 5.45 12.07 2.94 0.20 3.88 8.50 5.67 9.99 0.53 1.77 8.50 2.61 0.14 3.98 4.91 3.50 7.17 0.58 5.55 13.38 3.20 0.14 4.77 5.33 4.80 5.16

3.36 17.63 34.09 7.92 0.46 9.00 11.51 14.07 11.31 2.83 29.82 66.19 14.36 0.10 22.11 22.58 35.25 30.84 4.19 44.31 125.72 29.96 2.55 48.9 32.17 63.17 55.22

4.12 11.64 40.24 11.05 2.90 14.65 12.61 24.5 29.59 1.50 33.68 62.42 10.75 0.72 17.27 12.03 15.35 4.80 0.83 31.44 46.22 5.21 0.85 10.42 4.06 8.23 18.32

0.24 13.98 14.51 0.73 1.52 1.32 3.82 1.19 8.73 0.27 44.61 47.31 2.03 3.19 3.86 10.07 3.07 31.20 0.44 25.56 38.2 3.2 0.93 10.37 22.22 7.68 4.78

204

H. Du et al. / Energy 175 (2019) 195e204

References [1] Rhodes CJ. The 2015 Paris climate change conference: COP21. Sci Prog 2016;99:97e104. https://doi.org/10.3184/003685016X14528569315192. [2] Li H, Lu Y, Zhang J, Wang T. Trends in road freight transportation carbon dioxide emissions and policies in China. Energy Pol 2013a;57:99e106. https:// doi.org/10.1016/j.enpol.2012.12.070. [3] Li H, Lu Y, Zhang J, Wang T. Trends in road freight transportation carbon dioxide emissions and policies in China. Energy Pol 2013b;57:99e106. https:// doi.org/10.1016/j.enpol.2012.12.070. [4] IEA. CO2 emissions from fuel combustion. International Energy Agency; 2017. [5] Fan F, Lei Y. Decomposition analysis of energy-related carbon emissions from the transportation sector in Beijing. Transp Res Part D Transp Environ 2016;42:135e45. https://doi.org/10.1016/j.trd.2015.11.001. [6] Chai J, Lu Q-Y, Wang S-Y, Lai KK. Analysis of road transportation energy consumption demand in China. Transp Res Part D Transp Environ 2016;48: 112e24. https://doi.org/10.1016/j.trd.2016.08.009. [7] Wang Z, He W. CO2 emissions efficiency and marginal abatement costs of the regional transportation sectors in China. Transp Res Part D Transp Environ 2017;50:83e97. https://doi.org/10.1016/j.trd.2016.10.004. [8] Liu LQ, Liu CX, Sun ZY. A survey of China's low-carbon application practice opportunity goes with challenge. Renew Sustain Energy Rev 2011;15: 2895e903. https://doi.org/10.1016/j.rser.2011.02.034. [9] Mao X, Yang S, Liu Q, Tu J, Jaccard M. Achieving CO2 emission reduction and the co-benefits of local air pollution abatement in the transportation sector of China. Environ Sci Pol 2012;21:1e13. https://doi.org/10.1016/ j.envsci.2012.03.010. [10] Xu B, Lin B. Factors affecting carbon dioxide (CO2) emissions in China's transport sector: a dynamic nonparametric additive regression model. J Clean Prod 2015;101:311e2. https://doi.org/10.1016/j.jclepro.2015.03.088. [11] Xu B, Lin B. Differences in regional emissions in China's transport sector: determinants and reduction strategies. Energy 2016;95:459e70. https:// doi.org/10.1016/j.energy.2015.12.016. [12] Zhang M, Li H, Zhou M, Mu H. Decomposition analysis of energy consumption in Chinese transportation sector. Appl Energy 2011;88:2279e85. https:// doi.org/10.1016/j.apenergy.2010.12.077. [13] Liu WM, Luk MKR. Reform and opening up: way to the sustainable and harmonious development of air transport in China. Transport Pol 2009;16: 215e23. https://doi.org/10.1016/j.tranpol.2009.08.007. [14] Cheng YH, Chang YH, Lu IJ. Urban transportation energy and carbon dioxide emission reduction strategies. Appl Energy 2015;157:953e73. https://doi.org/ 10.1016/j.apenergy.2015.01.126. [15] Zhou G, Chung W, Zhang X. A study of carbon dioxide emissions performance of China's transport sector. Energy 2013;50:302e14. https://doi.org/10.1016/ j.energy.2012.11.045. [16] Cui Q, Li Y. An empirical study on the influencing factors of transportation carbon efficiency: evidences from fifteen countries. Appl Energy 2015;141: 209e17. https://doi.org/10.1016/j.apenergy.2014.12.040. ntara V, Padilla E. The materiality of the immaterial: service [17] Piaggio M, Alca sectors and CO2, emissions in Uruguay. Ecol Econ 2015;110:1e10. https:// doi.org/10.1016/j.ecolecon.2014.12.003. [18] Wang Y, Wang W, Mao G, Cai H, Zuo J, Wang L, Zhao P. Industrial CO2 emissions in China based on the hypothetical extraction method: linkage analysis. Energy Pol 2013;62:1238e44. https://doi.org/10.1016/ j.enpol.2013.06.045. [19] Zhao Y, Liu Y, Wang S, Zhang Z, Li J. Inter-regional linkage analysis of industrial CO2 emissions in China: an application of a hypothetical extraction method. Ecol Indicat 2016;61:428e37. https://doi.org/10.1016/ j.ecolind.2015.09.044. ntara V, Padilla E. Inputeoutput subsystems and pollution: an application [20] Alca to the service sector and CO2, emissions in Spain. Ecol Econ 2008;68:905e14. https://doi.org/10.1016/j.ecolecon.2008.07.010. [21] Daly HE, Scott K, Strachan N, Barrett J. Indirect CO2 emission implications of energy system pathways: linking IO and TIMES models for the UK. Environ Sci Technol 2015;49:10701e9. https://doi.org/10.1021/acs.est.5b01020. [22] Limmeechokchai B, Suksuntornsiri P. Embedded energy and total greenhouse

[23]

[24] [25]

[26]

[27]

[28]

[29]

[30]

[31]

[32] [33]

[34]

[35]

[36]

[37] [38] [39]

[40]

[41]

[42]

[43]

gas emissions in final consumptions within Thailand. Renew Sustain Energy Rev 2007;11:259e81. https://doi.org/10.1016/j.rser.2005.01.005. Huo H, Zhang Q, Guan DB, Su X, Zhao HY, He KB. Examining air pollution in China using production- and consumption-based emissions accounting approaches. Environ Sci Technol 2014;48:14139e47. https://doi.org/10.1021/ es503959t. Liu L, Cao D, Wei Y. What drives intersectoral CO2 emissions in China? J Clean Prod 2016;133:1053e61. https://doi.org/10.1016/j.jclepro.2016.05.126. Chen G, Hadjikakou M, Wiedmann T. Urban carbon transformations: unravelling spatial and inter-sectoral linkages for key city industries based on multi-region input-output analysis. J Clean Prod 2017;163:224e40. https:// doi.org/10.1016/j.jclepro.2016.04.046. Liu S, Tian X, Cai W, et al. How the transitions in iron and steel and construction material industries impact China's CO2 emissions: comprehensive analysis from an inter-sector linked perspective. Appl Energy 2018;211: 64e75. https://doi.org/10.1016/j.apenergy.2017.11.040. Zhang J, Yu B, Wei Y, et al. Heterogeneous impacts of households on carbon dioxide emissions in Chinese provinces. Appl Energy 2018;229:236e52. https://doi.org/10.1016/j.apenergy.2018.07.110. Song Y, Liu C, Langston C. Linkage measures of the construction sector using the hypothetical extraction method. Constr Manag Econ 2006;24:579e89. https://doi.org/10.1080/01446190500435358. rez Blanco CD, Thaler T. An input-output assessment of water productivity Pe n Region (Spain). Water 2014;6:929e44. https://doi.org/ in the castile and Leo 10.3390/w6040929. Ali Y. Measuring CO2 emission linkages with the hypothetical extraction method (HEM). Ecol Indicat 2015;54:171e83. https://doi.org/10.1016/ j.ecolind.2015.02.021. Zhao Y, Zhang Z, Wang S, Zhang Y, Liu Y. Linkage analysis of sectoral CO2 emissions based on the hypothetical extraction method in South Africa. J Clean Prod 2015;103:916e24. https://doi.org/10.1016/j.jclepro.2014.10.061. Fang D, Chen B. Linkage analysis for water-carbon nexus in China. Appl Energy 2018;225:682e95. https://doi.org/10.1016/j.apenergy.2018.05.058. Bai H, Feng X, Huo H, et al. Mapping inter-industrial CO2 flows within China. Renew Sustain Energy Rev 2018;93:400e8. https://doi.org/10.1016/ j.rser.2018.05.054. Liao H, Andrade C, Lumbreras J, et al. CO2 emissions in Beijing: sectoral linkages and demand drivers. J Clean Prod 2017;166:395e407. https:// doi.org/10.1016/j.jclepro.2017.08.033. Tian J, Andrade C, Lumbreras J, et al. Integrating sustainability into city-level CO2 accounting: social consumption pattern and income distribution. Ecol Econ 2018;153:1e16. https://doi.org/10.1016/j.ecolecon.2018.06.019. Zhang Y, Zhang K. The linkage of CO2 emissions for China, EU, and USA: evidence from the regional and sectoral analyses. Environ Sci Pollut Control Ser 2018;25:20179e92. https://doi.org/10.1007/s11356-018-1965-7. Duarte R, Sa J, Bielsa J. Water use in the Spanish economy: an input-output approach. Ecol Econ 2002;43:71e85. IPCC. IPCC guidelines for national greenhouse gas inventories. Institute for Global Environmental Strategies; 2006. Guo B, Geng Y, Franke B, Hao H, Liu Y, Chiu A. Uncovering China's transport CO2 emission patterns at the regional level. Energy Pol 2014;74:134e46. https://doi.org/10.1016/j.enpol.2014.08.005. Yuan C, Liu S, Xie N. The impact on Chinese economic growth and energy consumption of the Global Financial Crisis: an input-output analysis. Energy 2010;35:1805e12. https://doi.org/10.1016/j.energy.2009.12.035. Ji Q, Xia T, Liu F, et al. The information spillover between carbon price and power sector returns: evidence from the major European electricity companies. J Clean Prod 2019;208:1178e87. https://doi.org/10.1016/ j.jclepro.2018.10.167. Ji Q, Zhang D, Geng J. Information linkage, dynamic spillovers in prices and volatility between the carbon and energy markets. J Clean Prod 2018;198: 972e8. https://doi.org/10.1016/j.jclepro.2018.07.126. Wang-Helmreich H, Kreibich N. The potential impacts of a domestic offset component in a carbon tax on mitigation of national emissions. Renew Sustain Energy Rev 2019;101:453e60. https://doi.org/10.1016/j.rser.2018.11.026.