- Email: [email protected]
Changes of CO2 emissions embodied in China–Japan trade: drivers and implications
Accepted Manuscript Changes of CO2 emissions embodied in China-Japan trade: Drivers and implications Rui Wu, Yong Geng, Professor, Huijuan Dong, Dr., Tsuyoshi Fujita, Xu Tian PII:
S0959-6526(15)00897-5
DOI:
10.1016/j.jclepro.2015.07.017
Reference:
JCLP 5823
To appear in:
Journal of Cleaner Production
Received Date: 19 April 2015 Revised Date:
12 June 2015
Accepted Date: 3 July 2015
Please cite this article as: Wu R, Geng Y, Dong H, Fujita T, Tian X, Changes of CO2 emissions embodied in China-Japan trade: Drivers and implications, Journal of Cleaner Production (2015), doi: 10.1016/j.jclepro.2015.07.017. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Changes of CO2 emissions embodied in China-Japan trade: drivers and implications
TE D
M AN U
SC
RI PT
Rui Wua,d, Yong Gengb*, Huijuan Dongc*, Tsuyoshi Fujitac, Xu Tiana,d a Key Lab on Pollution Ecology and Environmental Engineering, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang 110016, China b School of Environmental Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China c National Institute for Environmental Studies, Tsukuba 305-8506, Japan d University of Chinese Academy of Sciences, Beijing 100049, China *Corresponding author: Professor Yong Geng School of Environmental Science and Engineering Shanghai Jiao Tong University No. 800 Dongchuan Road, Minhang, Shanghai China 200240 E-mail: [email protected] Telephone: +86-21-54748019 Fax: +86-21-54740825. Dr. Huijuan Dong Email: [email protected] (H. Dong) Telephone: +81-29-850-2184, Fax: +81-29-850-2960;
AC C
EP
Abstract CO2 emission embodied in trade is an important aspect to respond international carbon mitigation. Half of China’s emission increase was due to production of exports. In order to analyze the reason of such a rapid emission increase, embodied CO2 emission flows between China and Japan for the period of 2000-2009 were estimated in this study by using emission embodied in bilateral trade (EEBT) approach in order to raise policy implications for both countries from trade perspective. Decomposition analysis was further conducted in order to identify driving forces underlying changes during the study period. Moreover, the concept of dependence on traded CO2 was proposed for analyzing mutual dependence of China and Japan’s carbon emission and economy. The results show that China was a net exporter of embodied CO2 emissions between China-Japan trade, but both China’s exported emissions and net emission transfer to Japan began to decrease after 2007. More emissions were embodied in more advanced and less carbon intensive products, especially for China’s exports. Driving force analysis shows that trade volume was the main driver for the increase of embodied emissions and technology effect contributed mainly to the decrease. The absolute value of technology effect was even larger than activity effect in some years. This study also reveals that Japan was relatively more dependent on China’s CO2 emissions and showed an increasing trend over the last decade, while China’s economic development was more dependent on imports from Japan and such a situation reversed after 2006. This study suggests that China should further reduce its emission intensity for narrowing 1
ACCEPTED MANUSCRIPT the gap of embodied emissions between the bilateral trade flows and Japanese energy-saving technologies should be transferred to China. Keywords: Embodied CO2 emissions; China-Japan trade; Input-output analysis; Driving forces.
AC C
EP
TE D
M AN U
SC
RI PT
1. Introduction United Nations Framework Convention on Climate Change (UNFCCC) requires countries to provide information on their national inventory of anthropogenic emissions by sources of greenhouse gases within national territories (Dong et al., 2014). However, more greenhouse gas emissions occur outside a country in order to meet all the consumption occurred in that country (‘weak carbon leakage’) in last 20 years (Peters, 2010; Springmann, 2014). The net emissions transferred from non-Annex B countries (countries that do not belong to Annex B and do not have emission limitations) to Annex B countries (39 developed countries or transition countries with quantified emission limitations listed in Annex B of Kyoto Protocol) increased from 0.4 Gt CO2 in 1990 to 1.6 Gt CO2 in 2008, more than the Kyoto Protocol emission reductions (Peters et al., 2011). Few evidences have been found that carbon-intensive industries are located in developing countries in direct response to climate policy (‘strong carbon leakage’) (Davis and Caldeira, 2010). The so-called ‘weak carbon leakage’ caused by supply chain extension made it very difficult to appropriately allocate emission responsibilities. China became the world top CO2 emitter in the last decade and half of the emission increase was due to production of exports (Guan et al., 2009). In 2005, around one third of China’s emissions were caused by production of exports (Weber et al., 2008). Japan ranks the sixth CO2 emitter in the world and emission embodied in China-Japan trade is one of the largest inter-regional fluxes (Davis et al., 2011). As the second and third largest economies in the world, China and Japan have a close connection through enormous bilateral trade. However, some changes occurred over the last decade. Japan had been ranked as China’s largest trade partner for 11 straight years until 2003, then was surpassed by EU, USA, ASEAN (Association of Southeast Asian Nations) and HK in the last decade. China-Japan trade grew rapidly during 2000-2011 except 2009, and China’s trade changed from a surplus to a deficit since 2007 (Fig.1). The share of imports from Japan in China’s total imports significantly declined, while the share of imports from China in Japan’s total imports increased steadily except 2008 in the last decade. Moreover, export structure to Japan changed, with declined proportion of food and textiles and increased proportion of electrical & optical equipment. Also, electrical & optical equipment and machinery accounted for half of the imports from Japan to China and such a structure kept stable during the same period. Such a reality results in that CO2 emission patterns embodied in China-Japan trade experienced similar changes with the bilateral trade. Academically, several studies have been done for investigating embodied CO2 emissions in China-Japan trade. For example, Liu et al. (2010) and Dong et al. (2010) found that emissions embodied in imports from Japan to China kept increasing from 1990 to 2000, whilst emissions embodied in exports from China to Japan greatly increased in the first half of 1990s owing to the growth of trade volume and then decreased in the second half of 1990s owing to dramatic decline of China’s emission intensity. Duan et al. (2012) found that China-Japan trade reduced 2
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
China’s CO2 emission while increased Japan’s emission compared to the ‘no China-Japan trade’ scenario. However, few studies focused on embodied CO2 emissions changes in China-Japan trade for recent years and examined the key driving forces on such changes, leading to less valuable policy insights to policy makers. Consequently, this paper aims to fill such a research gap by conducting a temporal study on observing both quantity and structure changes of CO2 emissions embodied in China-Japan trade using time-series Multi-Regional Input-Output (MRIO) tables. In order to uncover key driving forces inducing such changes, the conventional 3-factor index decomposition analysis (IDA) is selected to identity the impacts of three effects, namely activity, structure and technology effects (Xu and Ang, 2013). In summary, different from existing studies, our study will contribute more in several aspects. First, time serial analysis on CO2 emission embodied in China-Japan trade has been extended to more recent years (from 2000 to 2009), and decomposition analysis has been further conducted in order to identify corresponding driving forces so that solid policy recommendations can be provided from international trade perspective. Second, the concept of dependence on traded CO2 is proposed and discussed, which can not only quantitatively reveal to what extent one economy’s output relies on CO2 emissions abroad through importing intermediate products, but also can take into consideration of both CO2 emission and economy simultaneously. The remainder of the paper is organized as follows: after this introduction section, Section 2 describes the data sources and methodology. Section 3 presents detailed results of CO2 emission changes and corresponding driving forces. Then policy implications for both China and Japan are discussd in section 4. Finally conclusions are drawn in section 5. 2. Methods 2.1. Data Collection and Treatment In order to conduct such a complicated study, a number of measures on data collection were adopted, including direct surveys, governmental document reviews, key-informant interviews, and informal meetings. The national input-output tables used in this study were obtained from the World Input-Output Database (WIOD) (Timmer et al., 2012). The SUT-RAS method developed by Temurshoev and Timmer (2011) was employed for building a time series national IO tables. The national input-output tables consist of 35 sectors. Therefore, the raw international trade data, taken from UN COMTRADE and WIOD, were reclassified in accordance with these 35 sectors. CO2 emissions data at sectoral level were from environmental accounts of WIOD. The CO2 emissions include energy related emissions and non-energy related emissions, covering the period of 1995-2009. In order to avoid the influence of inflation, all years’ prices were adjusted into 2005 constant prices by using GDP deflator from World Bank. In addition, key-informant interviews with the relevant stakeholders can provide a valuable source of information and significant information on specific hard-to-find subject data and therefore has been used as one way to get additional information. Therefore, the authors contacted several key officials working in China custom and Japan custom so that more valuable information can be obtained. Finally, in order to further validate collected data and information, informal meetings with these stakeholders have been held so that valuable comments can be gathered and verified. 2.2. Emission embodied in bilateral trade (EEBT) Input-output analysis was originated by Leontief (1941). It was extended to interregional trade analysis by Isard (1951), Chenery (1953) and Moses (1955). There are two approaches to measure 3
ACCEPTED MANUSCRIPT
SC
RI PT
embodied emissions in trade at national level, namely EEBT approach and MRIO approach (Peters, 2008; Peters et al., 2011). The EEBT approach considers the total exports (as in bilateral trade data) from a country including intermediate and final products, while the MRIO approach is a more complex approach that uses a full multi-regional input-output table to determine the total global emissions to produce the products exported from a country. The EEBT approach was employed in this study because the main target of this paper is to focus on bilateral trade relationships and the EEBT approach is arguably more suitable and transparent (Peters and Hertwich, 2008). Part of the 40-regions’ MRIO table is extracted to constitute a two regions input-output table. The basic form of two-region input-output table is shown in table 1. RoW represents the rest of the world. The matrix Zrr represents domestic intermediate inputs in order to meet region r’s intermediate demands, while Zsr represents intermediate demands of region r from region s. The matrix Frr represents domestic final demands, and Fsr represents final demands of region r from region s. The column vector X represents output, while row vector V represents added value. The emission embodied in export from region r to region s (frs) can be expressed in Eq. (1).
M AN U
frs =Fr ( I - Arr ) −1 ers
(1)
NET = f rs − f sr
TE D
Where Fr represents the vector of direct CO2 emission intensity of region r; Arr represents the direct requirement matrix of domestic inputs of region r; ers represents export from region r to region s, I represents the identity matrix. The net emission transfer (NET) for region r equals to emissions embodied in exports from region r to region s minus emissions embodied in exports from region s to region r and can be calculated by Eq. (2). (2)
Where frs represents emission embodied in exports from region r to region s and fsr represents emission embodied in exports from region s to region r.
AC C
EP
2.3. Dependence on traded CO2 In this study, dependence on traded CO2 measures the extent that development of one country relies on CO2 emissions of another country, e.g. the more emissions of China driven by unit output of Japan, the more dependent of Japan on CO2 emissions embodied in imports from China. This concept also highlights the effect of domestic industrial structure and inter-regional input structure and deemphasizes the scale of economy by standardizing gross output. Economic linkage as well as CO2 dependence between economies is analyzed by using inter-regional and regional direct requirements matrices in the MRIO table. The economic dependence of region s (DEs), namely gross output of region r (∆Er) driven by unit output of region s, can be calculated by using Eq. (3).
DEs = ∆Er =( I - Arr )−1 Ars X% s ∆x
(3)
Where Ars represents inter-regional direct requirements matrix, in which all the intermediate inputs from region r to region s are listed. ܺ෨௦ represents output structure of region s and ∆x represents unit output. ܣ௦ ܺ෨௦ ∆ ݔrepresents the intermediate input from region r to region s driven by output ∆x of region s. And then ( ܫ− ܣ )ିଵ ܣ௦ ܺ෨௦ ∆ ݔrepresents the gross output of 4
ACCEPTED MANUSCRIPT region r through direct and indirect transactions between sectors within region r, when final demands ܣ௦ ܺ෨௦ ∆ ݔin region r. Likewise, dependence on traded CO2 of region s (DFs), namely CO2 emission of region r (∆fr) driven by unit output of region s, can be calculated by using Eq. (4).
DFs = ∆ fr =Fr ( I - Arr ) −1 Ars X% s ∆x
(4)
RI PT
2.4. Index decomposition analysis (IDA) Structure decomposition analysis (SDA) and index decomposition analysis (IDA) are two widely used methods for decomposing changes in indicators such as energy use, CO2 emissions, labor demand and value added at sector level. SDA uses the input-output framework while IDA uses only aggregate sector information. IDA is chosen due to its lower data requirement and easier interpretation of results. Changes in CO2 emissions embodied in trade are decomposed into three
SC
driving forces: namely activity effect (∆Cact), structure effect (∆Cstr) and technology effect (∆Ctec). The total change of CO2 emission embodied in unilateral trade (∆C) can be calculated by using Eq. (5).
M AN U
∆C = C T − C 0 = ∆Cact + ∆Cstr + ∆Ctec
(5)
Where superscript 0 and T stand for base year and terminal year, respectively. Activity effect represents the contribution of export volume change; structure effect represents the contribution of export structure change; technology effect represents the contribution of domestic emission intensity change as well as input-output coefficient change. According to Kaya identity, total CO2 emissions for all sectors embodied in trade can be calculated by using Eq. (6).
i
Qi Ci = ∑ Q •Si • Ti Q Qi i
TE D
C = ∑Q
(6)
AC C
EP
Where Q represents overall export volume and refers to activity effect; Qi represents export volume of sector I; Ci represents CO2 emission of sector I; Si represents activity share of sector i and refers to structure effect; Ti represents technology level of sector i and refers to technology effect. Additive form of Logarithmic Mean Divisia Index (LMDI) approach (Ang et al., 1998) is used to decompose the driving forces of changes in embodied CO2 emissions. LMDI method gives no residual and can be shown to converge when the zero values in the data set are replaced by a small positive number (Ang, 2004). Due to these two advantages, LMDI is selected as the decomposition analysis method. The equations for the three driving forces have a similar structure as follows:
QT ∆Cact = ∑ wi ln 0 i Q
(7)
ST ∆Cstr = ∑ wi ln i0 i Si
(8)
TiT = ∑ wi ln 0 i Ti
(9)
∆Ctec
5
ACCEPTED MANUSCRIPT CiT − Ci0 wi = ln CiT − ln Ci0
(10)
AC C
EP
TE D
M AN U
SC
RI PT
3. Results 3.1. Changes in embodied CO2 emissions Fig.2 shows the emissions embodied in China-Japan trade and the net emission transfer (NET) from 2000 to 2009. The results demonstrate that CO2 emissions embodied in bilateral trade first increased in the early 2000s and then decreased in the late 2000s for both sides. Embodied CO2 exported from China to Japan increased very quickly and peaked at 162 Mt in 2006, while embodied CO2 imported from Japan to China peaked at 37 Mt in 2007. Compared with 2000, emissions embodied in exports from China to Japan and imports from Japan to China increased by 61% and 144% in 2009, respectively. In addition, the emissions exported from China to Japan fluctuated in the last 20 years. For instance, it decreased during 1995-2000 (Dong et al., 2010). It is clear that emissions embodied in export from China to Japan is much bigger than that embodied in import from Japan to China, indicating that China is a net exporter of embodied CO2 emissions between China-Japan trade. Moreover, emissions embodied in imports from Japan to China grew faster than that in exports from China to Japan in the early 2000s, leading to the slight decrease of NET from 73 Mt to 69 Mt. However, it increased dramatically from 2003 to 2005, with an increase of 188% as a result of the rapid growth of emissions embodied in China’s exports. The NET peaked in year 2005, with a value of 130 Mt CO2. Since then this value began to fall, with a decrease of 19% from 2005 to 2009. Fig.3 shows a breakdown of emissions embodied in China-Japan trade by industrial sectors. Fig.3 a) demonstrates that substantial fraction of emissions of exports from China to Japan was mainly embodied in three sectors: electrical and optical equipment, textiles and textile products, basic metals and fabricated metal, with values of 15%, 20% and 16% for the year of 2000, respectively. However, the proportions of emissions embodied in these sectors changed significantly during the period of 2000-2009, with a fact that emissions from electrical & optical equipment sector rose to 25%, while emissions from textiles and textile products sector fell to 11% and emissions from basic metals and fabricated metal sector fell to 10%. This means that China’s exported emissions were gradually embodied in more advanced products for the study period, indicating an export structure change from primary products to more sophisticated products. For imports from Japan to China, electrical & optical equipment, basic metals and fabricated metal and chemicals & chemical products were the three main sectors inducing embodied CO2 emissions, with values of 20%, 32% and 17% for the year of 2000, respectively (see Fig.3 b). Similar to the trend of exported emissions, the structure of imported emissions changed significantly during the period of 2000-2009, with a fact that emission from electrical & optical equipment sector rose to 27% and emission from basic metals & fabricated metal sector decreased to 22%. The bilateral trade between China and Japan show that more exported CO2 emissions tend to be embodied in more advanced and less carbon intensive industrial products such as electrical and optical equipment rather than primary industrial products such as textiles and textile products, 6
ACCEPTED MANUSCRIPT basic metals and fabricated metal and non-metallic mineral. Such shifts imply that per exported embodied emission received more added values. However, the proportion of CO2 emission from another carbon intensive sector, chemicals & chemical products sector, didn’t decline, which indicates that such products are still needed by Japan.
EP
TE D
M AN U
SC
RI PT
3.2. Driving forces for changes of embodied CO2 emissions Fig. 4 a) shows the driving forces of CO2 emissions embodied in exports from China to Japan. It is clear that CO2 emissions embodied in exports from China to Japan peaked in 2006. Activity effect was the biggest driver for such an increase, and always exhibited positive effect except 2008-2009. Technology effect usually contributed to CO2 reduction during the study period, and played an important role in reducing CO2 emissions embodied in exports from China to Japan, particularly from 2005 to 2008. However, technology effect showed a positive effect from 2002 to 2004, because of the rise of China’s carbon emission intensity. Moreover, direct requirement matrix in the IO table can determine the total carbon emission coefficient. Therefore, during 2008-2009, technology effect was also positive even though carbon emission intensity decreased. Structure effect was usually not significant and its effect varied across different years. Only in 2008-2009 structure change of China’s exports played an important role on reducing carbon emissions because the export structure was adjusted due to the global financial crisis. In the case of CO2 emissions embodied in imports from Japan to China, the value is much smaller than CO2 emissions embodied in exports from China to Japan. However, driving forces for CO2 emissions embodied in imports from Japan to China have similar features as those for CO2 emissions embodied in exports from China to Japan. In this regard, activity effect was also the major driver for the 7 straight years’ increases of embodied CO2 emissions in imports from Japan to China before 2007 (Fig.4 b). The exception occurred during 2008-2009 because the import trade volume from Japan to China decreased, leading to obvious decline of embodied CO2 emissions. Technology effect was the second major driving force. However, its contribution was sometimes positive and sometimes negative. For instance, technology effect was the key factor leading to decline of embodied CO2 emissions during 2007-2008. Structure effect was negligible.
AC C
3.3. Driving forces for changes of net emission transfer (NET) The NET changes are related to 6 driving forces including activity, structure and technology effects from both sides. Since structure effect is very marginal and can be neglected, only activity effect (trade volume balance) and intensity effect (CO2 emission per GDP) are demonstrated in Fig.5. The NET experienced three different stages during the study period, namely, slight decline (2000-2002), rapid increase (2002-2005) and obvious decline (2005-2009). Variation of difference of CO2 emission intensity was more consistent with the trend of the NET than net trade volume transfer. As trade volume balance decreased, the increase of the NET during 2002-2004 was driven by the rise of CO2 emission intensity. Trade volume balance increased sharply during 2008-2009, but the NET still decreased during the same period, implying that the decline of CO2 emission intensity drove the decline of the NET. Although CO2 emission intensity in China is much higher than that of Japan, its effective reduction on intensity has narrowed the gap of embodied emissions between the bilateral trade flows. 7
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
RI PT
3.4. Dependence on traded CO2 In terms of economic analysis, greenhouse gas emission, which causes global climate change, represents an environmental externality under production-based emission accounting at present. Dependence on traded CO2 evaluates the negative externality of overseas CO2 emission caused by domestic demand of an economy. Overseas emission driven by unit output of an economy measures the degree of dependence of its industrial structure on traded CO2. Such an indicator reveals inherent linkages between two economies in terms of CO2 emission and contributes to strengthening technological and financial cooperation of emission mitigation. Fig. 6 shows CO2 emission dependence as well as economic dependence between China and Japan. It can be found that unit output of Japan always drove more CO2 emission of China, i.e. Japan was more dependent on CO2 emission of its trading partner than China. Moreover, CO2 emission dependence of China on Japan decreased gradually (blue solid line in Fig. 6), while dependence of Japan on China increased (red solid line) before 2007 and then decreased after 2007. Though emission intensity gap between them reduced during the study period, dependence difference between them increased, which means emission dependence could be largely influenced by economic linkage. In terms of economic dependence (dotted lines), unit output of China drove more outputs of Japan before 2008, i.e. China’s economic development was more dependent on imports from Japan. However, such a circumstance reversed in 2009, and since then Japan’s economic development has been more dependent on imports from China. IN summary, Japan increasingly relied on China’s CO2 emissions over the last decade while China became less dependent on Japan’s which is in line with its economic change.
AC C
EP
4. Policy Implications Carbon emission embodied in trade has been widely discussed to reveal the “carbon leakage” between Annex B and non-Annex B countries, since large quantities of embodied CO2 emissions are translated into environmental damages (Su and Ang, 2014). However, on the other hand, these countries that exported embodied CO2 can gain economic benefits through international trade. Therefore, it is important to study not only CO2 emission dependence but also economic dependence on international trade so that valuable policy implications can be obtained. CO2 emission intensity is a link between economic dependence and carbon dependence according to Eq. (4). Therefore, emission intensity is an important factor to affect both embodied emissions and carbon dependence. The emission intensities per trade volume and per GDP for both China and Japan are compared. Fig.7 lists the related research outcomes. It is obvious that China’s CO2 emission intensity was much higher, with values of about 7 times higher than those of Japan. This amplified the difference of dependence on traded CO2. Moreover, both of China’s CO2 emissions per export trade volume and per GDP significantly declined, while Japan’s CO2 emission intensity decreased slightly. Policy implications for China and Japan are discussed separately in the following paragraphs by considering carbon intensity. For China, CO2 emissions embodied in exports to Japan began to decrease in the late 2000s, which was not in line with the global increasing trend of CO2 emissions transferred from non-Annex B countries to Annex B countries (Peters et al., 2011). It was mainly driven by the 8
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
reduction of domestic emission intensity and industrial restructure. China set a goal that energy consumption per unit GDP would decrease by 20% in its 11th Five-Year Plan (2006-2010). This contributed to the reduction of emission intensity by implementing clean technologies and improving energy efficiency. A technology spillover effect of Japanese international companies in China also helped China reduce emission intensity. Since Japan has a long history of implementing energy conservation after oil crisis in 1970s, to learn Japan’s world leading energy-saving technologies should be an effective way to reduce the NET. These technologies include efficient combined heat and power generation (CHP) and heat pump in energy supply sector; energy management and energy-saving equipment for manufacturing enterprises; solar energy and wind power technologies; circular economy characterized by 3R (reduce, reuse and recycle) of wastes. Clean development mechanism (CDM) projects may be a good way to help improve China’s energy efficiency (Shimazaki et al., 2000). Decline of China’s emission intensity was the main factor for narrowing its NET gap. Due to the large emission intensity difference, China still has a potential to further reduce its NET. Effective measures and efforts should be taken to accelerate further technical progress in particular western China, the lowest energy efficiency region of China (Wang et al., 2014). Industrial restructure can significantly reduce total emission intensity. The experience of the structural change of Japanese manufacturing industry and adjustment of energy structure indicates that new strategies are needed to promote energy-saving structural change for Chinese manufacturing industry and accelerate renewable energy utilization (Wang et al., 2015; Zhao et al., 2014). Changes in structure of export commodities also contributed to the decline of CO2 emissions embodied in exports to Japan. Restriction on exporting ‘highly energy-consuming and emission intensive’ products and slowing down the development of related sectors should continue. With the development of Chinese economy, China-Japan trade became less vertically complementary but more horizontally competitive. Trade structure between two countries became similar, therefore, structure effect has little potential for reducing the NET. In the long term, emission intensity is a critical factor for reducing the net CO2 flows. For Japan, its manufacturing industry is regarded as one of the most energy-efficient in the world. Japan’s CO2 emission intensity was quite low and stable (see Fig.7). It was closely related with the low-carbon society policy currently being implemented in Japan. During the periods of 2000-2002, 2005-2006 and 2008-2009, technology effect was positive. Such a fact implies the Japanese government’s pressure on keeping low CO2 emission intensity was still high, especially after the closure of Fukushima nuclear power plant. Under such a circumstance, Japanese government decided to further support low carbon technologies. For instance, the famous eco-town project in Japan led to significant CO2 emission reduction from their industrial parks through the application of urban/industrial symbiosis (Geng et al., 2010). In addition, China’s export structure became similar to Japan’s since China has made a great effort to upgrade its industrial structure by encouraging industrial innovations. Under such a circumstance, more intra-industrial trade will occur in bilateral trade, which means that Japanese products should be more competitive in order to enter the Chinese market. This requires that they should continue to support their low carbon technologies so that their products can be more energy efficient. Finally, technology transfer should be supported so that advanced energy efficient technologies and products can be shared by their Chinese counterparts. Exporting energy efficient equipment to China contributes to decreasing the dependence on CO2 emissions of China. 9
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
5. Conclusions China-Japan trade is one major part of global trade due to its large volume, hence, CO2 emissions embodied in China-Japan trade need to be clarified for responding carbon mitigation problems from international trade perspectives. In this paper, time-series analysis of CO2 emissions embodied in China-Japan trade during 2000-2009 was estimated based on the EEBT approach. Then index decomposition method was further adopted to uncover the driving forces of the embodied emission amount changes. Not only carbon dependence, but also economic dependence between the two countries was measured. Policy implications of emission reduction for both countries from trade perspective were raised. The main research findings include: (1) China was always a net CO2 emission exporter of embodied CO2 emissions between China-Japan trade, whereas the NET began to decrease after 2005. Moreover, for both sides, CO2 emissions embodied in exports showed an increase trend in the early 2000s and then a decrease trend in the late 2000s. (2) Decrease of emission intensity was one key factor for the decline of emissions embodied in China-Japan trade. Increase of trade volume was one major driving force for the growth of embodied emissions. (3) Dependence on traded CO2 analysis revealed that Japan was more dependent on China’s CO2 emissions and showed an increasing trend over the last decade, while China’s economic development was more dependent on imports from Japan but such a situation reversed after 2006. (4) Low carbon technology transfer from Japan to China has a greater potential to help narrow the NET than trade volume and structure adjustment. Furthermore, total domestic CO2 emissions reduction can benefit from it. In general, this paper can provide valuable policy insights to international trade policy makers and CO2 emission reduction policy makers so that they can work together in order to raise appropriate mitigation policies. It highlights that closer collaboration between international trade partners can help mitigate the overall CO2 emission, while at the same time maintaining healthy economic development, so that a win-win situation can be achieved. Similar study can be further applied to embodied CO2 emission between China and other countries so that more specific findings can be found. One limitation of this paper is that driving forces of NET change were demonstrated using trade volume balance and difference of CO2 emission intensity. However, it is correlated with 6 factors: activity effect, structure effect and technology effect of both countries. Future studies could search an innovative method to decompose NET for an in-depth understanding of NET change. Acknowledgements This study is funded by the Natural Science Foundation of China (NSFC. 71461137008, 71325006) and the NSFC-JSPS Joint Research Grant (71311140172). References Ang, B., 2004. Decomposition analysis for policymaking in energy: which is the preferred method? Energy policy. 10
ACCEPTED MANUSCRIPT Ang, B.W., Zhang, F.Q., Choi, K.H., 1998. Factorizing changes in energy and environmental indicators through decomposition. Energy 23, 489-495. Chenery, H.B., 1953. The structure and growth of the Italian economy. United States of America, Mutual Security Agency. Davis, S.J., Caldeira, K., 2010. Consumption-based accounting of CO2 emissions. Proceedings of the National Academy of Sciences of the United States of America 107, 5687-5692. Davis, S.J., Peters, G.P., Caldeira, K., 2011. The supply chain of CO2 emissions. Proceedings of the
RI PT
National Academy of Sciences of the United States of America 108, 18554-18559.
Dong, H., Geng, Y., Fujita, T., Jacques, D.A., 2014. Three accounts for regional carbon emissions from both fossil energy consumption and industrial process. Energy 67, 276-283.
Dong, Y., Ishikawa, M., Liu, X., Wang, C., 2010. An analysis of the driving forces of CO2 emissions embodied in Japan–China trade. Energy Policy 38, 6784-6792.
Duan, Y., Yang, C., Zhu, K., 2012. Embodied Energy and CO2 Emission in Sino-Japan Trade. 20th
SC
International Input - Output Conference. Bratislava, Slovakia. Conference proceedings (24 - 29 June, 2012).
Geng, Y., Tsuyoshi, F., Chen, X., 2010. Evaluation of innovative municipal solid waste management
M AN U
through urban symbiosis: a case study of Kawasaki. Journal of Cleaner Production 18, 993-1000. Guan, D., Peters, G.P., Weber, C.L., Hubacek, K., 2009. Journey to world top emitter: An analysis of the driving forces of China's recent CO2 emissions surge. Geophysical Research Letters 36. Isard, W., 1951. Interregional and regional input-output analysis: a model of a space-economy. The review of Economics and Statistics, 318-328.
Leontief, W.W., 1941. Structure of American economy, 1919-1929.
Liu, X., Ishikawa, M., Wang, C., Dong, Y., Liu, W., 2010. Analyses of CO2 emissions embodied in Japan–
TE D
China trade. Energy Policy 38, 1510-1518.
Moses, L.N., 1955. The stability of interregional trading patterns and input-output analysis. The American Economic Review, 803-826.
Peters, G.P., 2008. From production-based to consumption-based national emission inventories. Ecological Economics 65, 13-23.
EP
Peters, G.P., 2010. Policy Update: Managing carbon leakage. Carbon Management 1, 35-37. Peters, G.P., Hertwich, E.G., 2008. CO2 embodied in international trade with implications for global climate policy. Environmental Science & Technology 42, 1401-1407.
AC C
Peters, G.P., Minx, J.C., Weber, C.L., Edenhofer, O., 2011. Growth in emission transfers via international trade from 1990 to 2008. Proceedings of the National Academy of Sciences of the United States of America 108, 8903-8908.
Shimazaki, Y., Akisawa, A., Kashiwagi, T., 2000. A model analysis of clean development mechanisms to reduce both CO2 and SO2 emissions between Japan and China. Applied Energy 66, 311-324. Springmann, M., 2014. Integrating emissions transfers into policy-making. Nature Climate Change 4, 177-181. Su, B., Ang, B.W., 2014. Input–output analysis of CO2 emissions embodied in trade: A multi-region model for China. Applied Energy 114, 377-384. Temurshoev, U., Timmer, M.P., 2011. Joint estimation of supply and use tables. Papers in Regional Science 90, 863-882. Timmer, M., Erumban, A., Francois, J., Genty, A., Gouma, R., Los, B., Neuwahl, F., Pindyuk, O., Poeschl, J., Rueda-Cantuche, J., 2012. The World Input-Output Database (WIOD): Contents, Sources and 11
ACCEPTED MANUSCRIPT Methods. WIOD Background document available at www. wiod. org. Wang, Z., Feng, C., Zhang, B., 2014. An empirical analysis of China's energy efficiency from both static and dynamic perspectives. Energy 74, 322-330. Wang, Z., Wang, C., Yin, J., 2015. Strategies for addressing climate change on the industrial level: affecting factors to CO2 emissions of energy-intensive industries in China. Nat Hazards 75, 303-317. Weber, C.L., Peters, G.P., Guan, D., Hubacek, K., 2008. The contribution of Chinese exports to climate change. Energy Policy 36, 3572-3577.
RI PT
Xu, X.Y., Ang, B.W., 2013. Index decomposition analysis applied to CO2 emission studies. Ecological Economics 93, 313-329.
Zhao, Y., Ke, J., Ni, C.C., McNeil, M., Khanna, N.Z., Zhou, N., Fridley, D., Li, Q., 2014. A comparative
study of energy consumption and efficiency of Japanese and Chinese manufacturing industry. Energy
AC C
EP
TE D
M AN U
SC
Policy 70, 45-56.
12
ACCEPTED MANUSCRIPT Table. 1. Two regions input-output table Intermediate demands Region r Region s RoW Zrs Zrr Intermed- Region r iate Zsr Zss Region s inputs RoW Output
Input
Fss
T
(Xs)
EP
TE D
M AN U
SC
(Xr)
AC C
Xs
Vs T
Total inputs
Fsr
Total outputs Xr
RI PT
Vr
Value added
Final demands Region r Region s RoW Frs Frr
ACCEPTED MANUSCRIPT 25%
200 180 160
20%
US$ Billion
140 120
15%
Import from Japan Export to Japan
100 80
10%
Share of China's total imports
RI PT
60 40
5%
20 0
0%
AC C
EP
TE D
M AN U
SC
Fig. 1. Trade volume and shares of total imports.
Share of Japan's total imports
ACCEPTED MANUSCRIPT 180 160 Import from Japan
140
100
Export to Japan
80
RI PT
Mt CO2
120
60
Net emission transfer
40 20 0
SC
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
AC C
EP
TE D
M AN U
Fig. 2. CO2 emissions embodied in China-Japan trade for the period of 2000-2009.
ACCEPTED MANUSCRIPT 100% 90%
Others
80%
Electrical & Optical Equipment
70% Machinery
60%
Basic Metals & Fabricated Metal
50%
RI PT
Other Non-Metallic Mineral
40%
Rubber & Plastics
30%
Chemicals & Chemical Products
20% 10%
Textiles & Textile Products
0%
Food, Beverages & Tobacco
SC
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
a) CO2 emissions embodied in China’s exports to Japan by industrial sectors.
M AN U
100% 90% 80%
Others
70%
Transport Equipment
60%
Electrical & Optical Equipment
50%
Machinery
40%
Basic Metals & Fabricated Metal
20% 10% 0%
TE D
30%
Other Non-Metallic Mineral Chemicals & Chemical Products
EP
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
AC C
b) CO2 emissions embodied in China’s imports from Japan by industrial sectors. Fig. 3. Structure of emissions embodied in China-Japan trade.
ACCEPTED MANUSCRIPT 30 25 20 15 technology
RI PT
Mt CO2
10
srtucture
5
activity
0
total
2000-1 2001-2 2002-3 2003-4 2004-5 2005-6 2006-7 2007-8 2008-9 -5
SC
-10 -15
M AN U
-20
a) Drivers of changes in CO2 emissions embodied in export from China to Japan. 8 6
0
2000-1 2001-2 2002-3 2003-4 2004-5 2005-6 2006-7 2007-8 2008-9 -2 -4
technology srtucture activity total
EP
Mt CO2
2
TE D
4
AC C
-6 -8
b) Drivers of changes in CO2 emissions embodied in import from Japan to China. Fig. 4. Decomposition analysis results of changes in embodied CO2 emissions
ACCEPTED MANUSCRIPT 200%
100%
0%
-100%
-200% NET -300% CO2 emission intensity difference
-400%
M AN U
Trade volume balance
SC
RI PT
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
-500%
AC C
EP
TE D
Fig. 5. Changes of NET, difference of CO2 emission intensity and trade volume balance between China and Japan.
30
10
25
8
20
6
15
4
10
2
Million US $
12
RI PT
Kt CO2
ACCEPTED MANUSCRIPT
5
0
0
2002
2003
2004
2005
2006
2007
2008
2009
SC
2001
Emission dependence of China
Emission dependence of Japan
Economic dependence of China
Economic dependence of Japan
M AN U
2000
AC C
EP
TE D
Fig. 6. CO2 emission dependence and economic dependence of China and Japan Note: The values show CO2 emission and economic output of its trading partner driven by output of 1 billion US $
ACCEPTED MANUSCRIPT 2.5
1.5
China per GDP China per export
1.0
RI PT
Kg CO2 / 2005 US $
2.0
Japan per GDP
Japan per export
0.5
0.0
SC
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
AC C
EP
TE D
M AN U
Fig. 7. CO2 emissions per trade volume and per GDP
ACCEPTED MANUSCRIPT
Highlights
Embodied CO2 emissions flows between China and Japan were estimated; Emission embodied in bilateral trade (EEBT) approach was employed
RI PT
for our analysis;
China was a net exporter of embodied CO2 emissions between
Trade volume was the main driver for the increase of embodied
M AN U
emissions;
Technology effect contributed most to the decrease of embodied
EP
TE D
emissions.
AC C
SC
China-Japan trade;
S0959-6526(15)00897-5
DOI:
10.1016/j.jclepro.2015.07.017
Reference:
JCLP 5823
To appear in:
Journal of Cleaner Production
Received Date: 19 April 2015 Revised Date:
12 June 2015
Accepted Date: 3 July 2015
Please cite this article as: Wu R, Geng Y, Dong H, Fujita T, Tian X, Changes of CO2 emissions embodied in China-Japan trade: Drivers and implications, Journal of Cleaner Production (2015), doi: 10.1016/j.jclepro.2015.07.017. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Changes of CO2 emissions embodied in China-Japan trade: drivers and implications
TE D
M AN U
SC
RI PT
Rui Wua,d, Yong Gengb*, Huijuan Dongc*, Tsuyoshi Fujitac, Xu Tiana,d a Key Lab on Pollution Ecology and Environmental Engineering, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang 110016, China b School of Environmental Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China c National Institute for Environmental Studies, Tsukuba 305-8506, Japan d University of Chinese Academy of Sciences, Beijing 100049, China *Corresponding author: Professor Yong Geng School of Environmental Science and Engineering Shanghai Jiao Tong University No. 800 Dongchuan Road, Minhang, Shanghai China 200240 E-mail: [email protected] Telephone: +86-21-54748019 Fax: +86-21-54740825. Dr. Huijuan Dong Email: [email protected] (H. Dong) Telephone: +81-29-850-2184, Fax: +81-29-850-2960;
AC C
EP
Abstract CO2 emission embodied in trade is an important aspect to respond international carbon mitigation. Half of China’s emission increase was due to production of exports. In order to analyze the reason of such a rapid emission increase, embodied CO2 emission flows between China and Japan for the period of 2000-2009 were estimated in this study by using emission embodied in bilateral trade (EEBT) approach in order to raise policy implications for both countries from trade perspective. Decomposition analysis was further conducted in order to identify driving forces underlying changes during the study period. Moreover, the concept of dependence on traded CO2 was proposed for analyzing mutual dependence of China and Japan’s carbon emission and economy. The results show that China was a net exporter of embodied CO2 emissions between China-Japan trade, but both China’s exported emissions and net emission transfer to Japan began to decrease after 2007. More emissions were embodied in more advanced and less carbon intensive products, especially for China’s exports. Driving force analysis shows that trade volume was the main driver for the increase of embodied emissions and technology effect contributed mainly to the decrease. The absolute value of technology effect was even larger than activity effect in some years. This study also reveals that Japan was relatively more dependent on China’s CO2 emissions and showed an increasing trend over the last decade, while China’s economic development was more dependent on imports from Japan and such a situation reversed after 2006. This study suggests that China should further reduce its emission intensity for narrowing 1
ACCEPTED MANUSCRIPT the gap of embodied emissions between the bilateral trade flows and Japanese energy-saving technologies should be transferred to China. Keywords: Embodied CO2 emissions; China-Japan trade; Input-output analysis; Driving forces.
AC C
EP
TE D
M AN U
SC
RI PT
1. Introduction United Nations Framework Convention on Climate Change (UNFCCC) requires countries to provide information on their national inventory of anthropogenic emissions by sources of greenhouse gases within national territories (Dong et al., 2014). However, more greenhouse gas emissions occur outside a country in order to meet all the consumption occurred in that country (‘weak carbon leakage’) in last 20 years (Peters, 2010; Springmann, 2014). The net emissions transferred from non-Annex B countries (countries that do not belong to Annex B and do not have emission limitations) to Annex B countries (39 developed countries or transition countries with quantified emission limitations listed in Annex B of Kyoto Protocol) increased from 0.4 Gt CO2 in 1990 to 1.6 Gt CO2 in 2008, more than the Kyoto Protocol emission reductions (Peters et al., 2011). Few evidences have been found that carbon-intensive industries are located in developing countries in direct response to climate policy (‘strong carbon leakage’) (Davis and Caldeira, 2010). The so-called ‘weak carbon leakage’ caused by supply chain extension made it very difficult to appropriately allocate emission responsibilities. China became the world top CO2 emitter in the last decade and half of the emission increase was due to production of exports (Guan et al., 2009). In 2005, around one third of China’s emissions were caused by production of exports (Weber et al., 2008). Japan ranks the sixth CO2 emitter in the world and emission embodied in China-Japan trade is one of the largest inter-regional fluxes (Davis et al., 2011). As the second and third largest economies in the world, China and Japan have a close connection through enormous bilateral trade. However, some changes occurred over the last decade. Japan had been ranked as China’s largest trade partner for 11 straight years until 2003, then was surpassed by EU, USA, ASEAN (Association of Southeast Asian Nations) and HK in the last decade. China-Japan trade grew rapidly during 2000-2011 except 2009, and China’s trade changed from a surplus to a deficit since 2007 (Fig.1). The share of imports from Japan in China’s total imports significantly declined, while the share of imports from China in Japan’s total imports increased steadily except 2008 in the last decade. Moreover, export structure to Japan changed, with declined proportion of food and textiles and increased proportion of electrical & optical equipment. Also, electrical & optical equipment and machinery accounted for half of the imports from Japan to China and such a structure kept stable during the same period. Such a reality results in that CO2 emission patterns embodied in China-Japan trade experienced similar changes with the bilateral trade.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
China’s CO2 emission while increased Japan’s emission compared to the ‘no China-Japan trade’ scenario. However, few studies focused on embodied CO2 emissions changes in China-Japan trade for recent years and examined the key driving forces on such changes, leading to less valuable policy insights to policy makers. Consequently, this paper aims to fill such a research gap by conducting a temporal study on observing both quantity and structure changes of CO2 emissions embodied in China-Japan trade using time-series Multi-Regional Input-Output (MRIO) tables. In order to uncover key driving forces inducing such changes, the conventional 3-factor index decomposition analysis (IDA) is selected to identity the impacts of three effects, namely activity, structure and technology effects (Xu and Ang, 2013). In summary, different from existing studies, our study will contribute more in several aspects. First, time serial analysis on CO2 emission embodied in China-Japan trade has been extended to more recent years (from 2000 to 2009), and decomposition analysis has been further conducted in order to identify corresponding driving forces so that solid policy recommendations can be provided from international trade perspective. Second, the concept of dependence on traded CO2 is proposed and discussed, which can not only quantitatively reveal to what extent one economy’s output relies on CO2 emissions abroad through importing intermediate products, but also can take into consideration of both CO2 emission and economy simultaneously. The remainder of the paper is organized as follows: after this introduction section, Section 2 describes the data sources and methodology. Section 3 presents detailed results of CO2 emission changes and corresponding driving forces. Then policy implications for both China and Japan are discussd in section 4. Finally conclusions are drawn in section 5. 2. Methods 2.1. Data Collection and Treatment In order to conduct such a complicated study, a number of measures on data collection were adopted, including direct surveys, governmental document reviews, key-informant interviews, and informal meetings. The national input-output tables used in this study were obtained from the World Input-Output Database (WIOD) (Timmer et al., 2012). The SUT-RAS method developed by Temurshoev and Timmer (2011) was employed for building a time series national IO tables. The national input-output tables consist of 35 sectors. Therefore, the raw international trade data, taken from UN COMTRADE and WIOD, were reclassified in accordance with these 35 sectors. CO2 emissions data at sectoral level were from environmental accounts of WIOD. The CO2 emissions include energy related emissions and non-energy related emissions, covering the period of 1995-2009. In order to avoid the influence of inflation, all years’ prices were adjusted into 2005 constant prices by using GDP deflator from World Bank. In addition, key-informant interviews with the relevant stakeholders can provide a valuable source of information and significant information on specific hard-to-find subject data and therefore has been used as one way to get additional information. Therefore, the authors contacted several key officials working in China custom and Japan custom so that more valuable information can be obtained. Finally, in order to further validate collected data and information, informal meetings with these stakeholders have been held so that valuable comments can be gathered and verified. 2.2. Emission embodied in bilateral trade (EEBT) Input-output analysis was originated by Leontief (1941). It was extended to interregional trade analysis by Isard (1951), Chenery (1953) and Moses (1955). There are two approaches to measure 3
ACCEPTED MANUSCRIPT
SC
RI PT
embodied emissions in trade at national level, namely EEBT approach and MRIO approach (Peters, 2008; Peters et al., 2011). The EEBT approach considers the total exports (as in bilateral trade data) from a country including intermediate and final products, while the MRIO approach is a more complex approach that uses a full multi-regional input-output table to determine the total global emissions to produce the products exported from a country. The EEBT approach was employed in this study because the main target of this paper is to focus on bilateral trade relationships and the EEBT approach is arguably more suitable and transparent (Peters and Hertwich, 2008). Part of the 40-regions’ MRIO table is extracted to constitute a two regions input-output table. The basic form of two-region input-output table is shown in table 1.
M AN U
frs =Fr ( I - Arr ) −1 ers
(1)
NET = f rs − f sr
TE D
Where Fr represents the vector of direct CO2 emission intensity of region r; Arr represents the direct requirement matrix of domestic inputs of region r; ers represents export from region r to region s, I represents the identity matrix. The net emission transfer (NET) for region r equals to emissions embodied in exports from region r to region s minus emissions embodied in exports from region s to region r and can be calculated by Eq. (2). (2)
Where frs represents emission embodied in exports from region r to region s and fsr represents emission embodied in exports from region s to region r.
AC C
EP
2.3. Dependence on traded CO2 In this study, dependence on traded CO2 measures the extent that development of one country relies on CO2 emissions of another country, e.g. the more emissions of China driven by unit output of Japan, the more dependent of Japan on CO2 emissions embodied in imports from China. This concept also highlights the effect of domestic industrial structure and inter-regional input structure and deemphasizes the scale of economy by standardizing gross output. Economic linkage as well as CO2 dependence between economies is analyzed by using inter-regional and regional direct requirements matrices in the MRIO table. The economic dependence of region s (DEs), namely gross output of region r (∆Er) driven by unit output of region s, can be calculated by using Eq. (3).
DEs = ∆Er =( I - Arr )−1 Ars X% s ∆x
(3)
Where Ars represents inter-regional direct requirements matrix, in which all the intermediate inputs from region r to region s are listed. ܺ෨௦ represents output structure of region s and ∆x represents unit output. ܣ௦ ܺ෨௦ ∆ ݔrepresents the intermediate input from region r to region s driven by output ∆x of region s. And then ( ܫ− ܣ )ିଵ ܣ௦ ܺ෨௦ ∆ ݔrepresents the gross output of 4
ACCEPTED MANUSCRIPT region r through direct and indirect transactions between sectors within region r, when final demands ܣ௦ ܺ෨௦ ∆ ݔin region r. Likewise, dependence on traded CO2 of region s (DFs), namely CO2 emission of region r (∆fr) driven by unit output of region s, can be calculated by using Eq. (4).
DFs = ∆ fr =Fr ( I - Arr ) −1 Ars X% s ∆x
(4)
RI PT
2.4. Index decomposition analysis (IDA) Structure decomposition analysis (SDA) and index decomposition analysis (IDA) are two widely used methods for decomposing changes in indicators such as energy use, CO2 emissions, labor demand and value added at sector level. SDA uses the input-output framework while IDA uses only aggregate sector information. IDA is chosen due to its lower data requirement and easier interpretation of results. Changes in CO2 emissions embodied in trade are decomposed into three
SC
driving forces: namely activity effect (∆Cact), structure effect (∆Cstr) and technology effect (∆Ctec). The total change of CO2 emission embodied in unilateral trade (∆C) can be calculated by using Eq. (5).
M AN U
∆C = C T − C 0 = ∆Cact + ∆Cstr + ∆Ctec
(5)
Where superscript 0 and T stand for base year and terminal year, respectively. Activity effect represents the contribution of export volume change; structure effect represents the contribution of export structure change; technology effect represents the contribution of domestic emission intensity change as well as input-output coefficient change. According to Kaya identity, total CO2 emissions for all sectors embodied in trade can be calculated by using Eq. (6).
i
Qi Ci = ∑ Q •Si • Ti Q Qi i
TE D
C = ∑Q
(6)
AC C
EP
Where Q represents overall export volume and refers to activity effect; Qi represents export volume of sector I; Ci represents CO2 emission of sector I; Si represents activity share of sector i and refers to structure effect; Ti represents technology level of sector i and refers to technology effect. Additive form of Logarithmic Mean Divisia Index (LMDI) approach (Ang et al., 1998) is used to decompose the driving forces of changes in embodied CO2 emissions. LMDI method gives no residual and can be shown to converge when the zero values in the data set are replaced by a small positive number (Ang, 2004). Due to these two advantages, LMDI is selected as the decomposition analysis method. The equations for the three driving forces have a similar structure as follows:
QT ∆Cact = ∑ wi ln 0 i Q
(7)
ST ∆Cstr = ∑ wi ln i0 i Si
(8)
TiT = ∑ wi ln 0 i Ti
(9)
∆Ctec
5
ACCEPTED MANUSCRIPT CiT − Ci0 wi = ln CiT − ln Ci0
(10)
AC C
EP
TE D
M AN U
SC
RI PT
3. Results 3.1. Changes in embodied CO2 emissions Fig.2 shows the emissions embodied in China-Japan trade and the net emission transfer (NET) from 2000 to 2009. The results demonstrate that CO2 emissions embodied in bilateral trade first increased in the early 2000s and then decreased in the late 2000s for both sides. Embodied CO2 exported from China to Japan increased very quickly and peaked at 162 Mt in 2006, while embodied CO2 imported from Japan to China peaked at 37 Mt in 2007. Compared with 2000, emissions embodied in exports from China to Japan and imports from Japan to China increased by 61% and 144% in 2009, respectively. In addition, the emissions exported from China to Japan fluctuated in the last 20 years. For instance, it decreased during 1995-2000 (Dong et al., 2010). It is clear that emissions embodied in export from China to Japan is much bigger than that embodied in import from Japan to China, indicating that China is a net exporter of embodied CO2 emissions between China-Japan trade. Moreover, emissions embodied in imports from Japan to China grew faster than that in exports from China to Japan in the early 2000s, leading to the slight decrease of NET from 73 Mt to 69 Mt. However, it increased dramatically from 2003 to 2005, with an increase of 188% as a result of the rapid growth of emissions embodied in China’s exports. The NET peaked in year 2005, with a value of 130 Mt CO2. Since then this value began to fall, with a decrease of 19% from 2005 to 2009.
ACCEPTED MANUSCRIPT basic metals and fabricated metal and non-metallic mineral. Such shifts imply that per exported embodied emission received more added values. However, the proportion of CO2 emission from another carbon intensive sector, chemicals & chemical products sector, didn’t decline, which indicates that such products are still needed by Japan.
EP
TE D
M AN U
SC
RI PT
3.2. Driving forces for changes of embodied CO2 emissions Fig. 4 a) shows the driving forces of CO2 emissions embodied in exports from China to Japan. It is clear that CO2 emissions embodied in exports from China to Japan peaked in 2006. Activity effect was the biggest driver for such an increase, and always exhibited positive effect except 2008-2009. Technology effect usually contributed to CO2 reduction during the study period, and played an important role in reducing CO2 emissions embodied in exports from China to Japan, particularly from 2005 to 2008. However, technology effect showed a positive effect from 2002 to 2004, because of the rise of China’s carbon emission intensity. Moreover, direct requirement matrix in the IO table can determine the total carbon emission coefficient. Therefore, during 2008-2009, technology effect was also positive even though carbon emission intensity decreased. Structure effect was usually not significant and its effect varied across different years. Only in 2008-2009 structure change of China’s exports played an important role on reducing carbon emissions because the export structure was adjusted due to the global financial crisis.
AC C
3.3. Driving forces for changes of net emission transfer (NET) The NET changes are related to 6 driving forces including activity, structure and technology effects from both sides. Since structure effect is very marginal and can be neglected, only activity effect (trade volume balance) and intensity effect (CO2 emission per GDP) are demonstrated in Fig.5. The NET experienced three different stages during the study period, namely, slight decline (2000-2002), rapid increase (2002-2005) and obvious decline (2005-2009). Variation of difference of CO2 emission intensity was more consistent with the trend of the NET than net trade volume transfer. As trade volume balance decreased, the increase of the NET during 2002-2004 was driven by the rise of CO2 emission intensity. Trade volume balance increased sharply during 2008-2009, but the NET still decreased during the same period, implying that the decline of CO2 emission intensity drove the decline of the NET. Although CO2 emission intensity in China is much higher than that of Japan, its effective reduction on intensity has narrowed the gap of embodied emissions between the bilateral trade flows.
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
RI PT
3.4. Dependence on traded CO2 In terms of economic analysis, greenhouse gas emission, which causes global climate change, represents an environmental externality under production-based emission accounting at present. Dependence on traded CO2 evaluates the negative externality of overseas CO2 emission caused by domestic demand of an economy. Overseas emission driven by unit output of an economy measures the degree of dependence of its industrial structure on traded CO2. Such an indicator reveals inherent linkages between two economies in terms of CO2 emission and contributes to strengthening technological and financial cooperation of emission mitigation. Fig. 6 shows CO2 emission dependence as well as economic dependence between China and Japan. It can be found that unit output of Japan always drove more CO2 emission of China, i.e. Japan was more dependent on CO2 emission of its trading partner than China. Moreover, CO2 emission dependence of China on Japan decreased gradually (blue solid line in Fig. 6), while dependence of Japan on China increased (red solid line) before 2007 and then decreased after 2007. Though emission intensity gap between them reduced during the study period, dependence difference between them increased, which means emission dependence could be largely influenced by economic linkage. In terms of economic dependence (dotted lines), unit output of China drove more outputs of Japan before 2008, i.e. China’s economic development was more dependent on imports from Japan. However, such a circumstance reversed in 2009, and since then Japan’s economic development has been more dependent on imports from China. IN summary, Japan increasingly relied on China’s CO2 emissions over the last decade while China became less dependent on Japan’s which is in line with its economic change.
AC C
EP
4. Policy Implications Carbon emission embodied in trade has been widely discussed to reveal the “carbon leakage” between Annex B and non-Annex B countries, since large quantities of embodied CO2 emissions are translated into environmental damages (Su and Ang, 2014). However, on the other hand, these countries that exported embodied CO2 can gain economic benefits through international trade. Therefore, it is important to study not only CO2 emission dependence but also economic dependence on international trade so that valuable policy implications can be obtained. CO2 emission intensity is a link between economic dependence and carbon dependence according to Eq. (4). Therefore, emission intensity is an important factor to affect both embodied emissions and carbon dependence. The emission intensities per trade volume and per GDP for both China and Japan are compared. Fig.7 lists the related research outcomes. It is obvious that China’s CO2 emission intensity was much higher, with values of about 7 times higher than those of Japan. This amplified the difference of dependence on traded CO2. Moreover, both of China’s CO2 emissions per export trade volume and per GDP significantly declined, while Japan’s CO2 emission intensity decreased slightly. Policy implications for China and Japan are discussed separately in the following paragraphs by considering carbon intensity. For China, CO2 emissions embodied in exports to Japan began to decrease in the late 2000s, which was not in line with the global increasing trend of CO2 emissions transferred from non-Annex B countries to Annex B countries (Peters et al., 2011). It was mainly driven by the 8
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
reduction of domestic emission intensity and industrial restructure. China set a goal that energy consumption per unit GDP would decrease by 20% in its 11th Five-Year Plan (2006-2010). This contributed to the reduction of emission intensity by implementing clean technologies and improving energy efficiency. A technology spillover effect of Japanese international companies in China also helped China reduce emission intensity. Since Japan has a long history of implementing energy conservation after oil crisis in 1970s, to learn Japan’s world leading energy-saving technologies should be an effective way to reduce the NET. These technologies include efficient combined heat and power generation (CHP) and heat pump in energy supply sector; energy management and energy-saving equipment for manufacturing enterprises; solar energy and wind power technologies; circular economy characterized by 3R (reduce, reuse and recycle) of wastes. Clean development mechanism (CDM) projects may be a good way to help improve China’s energy efficiency (Shimazaki et al., 2000). Decline of China’s emission intensity was the main factor for narrowing its NET gap. Due to the large emission intensity difference, China still has a potential to further reduce its NET. Effective measures and efforts should be taken to accelerate further technical progress in particular western China, the lowest energy efficiency region of China (Wang et al., 2014). Industrial restructure can significantly reduce total emission intensity. The experience of the structural change of Japanese manufacturing industry and adjustment of energy structure indicates that new strategies are needed to promote energy-saving structural change for Chinese manufacturing industry and accelerate renewable energy utilization (Wang et al., 2015; Zhao et al., 2014). Changes in structure of export commodities also contributed to the decline of CO2 emissions embodied in exports to Japan. Restriction on exporting ‘highly energy-consuming and emission intensive’ products and slowing down the development of related sectors should continue. With the development of Chinese economy, China-Japan trade became less vertically complementary but more horizontally competitive. Trade structure between two countries became similar, therefore, structure effect has little potential for reducing the NET. In the long term, emission intensity is a critical factor for reducing the net CO2 flows. For Japan, its manufacturing industry is regarded as one of the most energy-efficient in the world. Japan’s CO2 emission intensity was quite low and stable (see Fig.7). It was closely related with the low-carbon society policy currently being implemented in Japan. During the periods of 2000-2002, 2005-2006 and 2008-2009, technology effect was positive. Such a fact implies the Japanese government’s pressure on keeping low CO2 emission intensity was still high, especially after the closure of Fukushima nuclear power plant. Under such a circumstance, Japanese government decided to further support low carbon technologies. For instance, the famous eco-town project in Japan led to significant CO2 emission reduction from their industrial parks through the application of urban/industrial symbiosis (Geng et al., 2010). In addition, China’s export structure became similar to Japan’s since China has made a great effort to upgrade its industrial structure by encouraging industrial innovations. Under such a circumstance, more intra-industrial trade will occur in bilateral trade, which means that Japanese products should be more competitive in order to enter the Chinese market. This requires that they should continue to support their low carbon technologies so that their products can be more energy efficient. Finally, technology transfer should be supported so that advanced energy efficient technologies and products can be shared by their Chinese counterparts. Exporting energy efficient equipment to China contributes to decreasing the dependence on CO2 emissions of China. 9
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
5. Conclusions China-Japan trade is one major part of global trade due to its large volume, hence, CO2 emissions embodied in China-Japan trade need to be clarified for responding carbon mitigation problems from international trade perspectives. In this paper, time-series analysis of CO2 emissions embodied in China-Japan trade during 2000-2009 was estimated based on the EEBT approach. Then index decomposition method was further adopted to uncover the driving forces of the embodied emission amount changes. Not only carbon dependence, but also economic dependence between the two countries was measured. Policy implications of emission reduction for both countries from trade perspective were raised. The main research findings include: (1) China was always a net CO2 emission exporter of embodied CO2 emissions between China-Japan trade, whereas the NET began to decrease after 2005. Moreover, for both sides, CO2 emissions embodied in exports showed an increase trend in the early 2000s and then a decrease trend in the late 2000s. (2) Decrease of emission intensity was one key factor for the decline of emissions embodied in China-Japan trade. Increase of trade volume was one major driving force for the growth of embodied emissions. (3) Dependence on traded CO2 analysis revealed that Japan was more dependent on China’s CO2 emissions and showed an increasing trend over the last decade, while China’s economic development was more dependent on imports from Japan but such a situation reversed after 2006. (4) Low carbon technology transfer from Japan to China has a greater potential to help narrow the NET than trade volume and structure adjustment. Furthermore, total domestic CO2 emissions reduction can benefit from it. In general, this paper can provide valuable policy insights to international trade policy makers and CO2 emission reduction policy makers so that they can work together in order to raise appropriate mitigation policies. It highlights that closer collaboration between international trade partners can help mitigate the overall CO2 emission, while at the same time maintaining healthy economic development, so that a win-win situation can be achieved. Similar study can be further applied to embodied CO2 emission between China and other countries so that more specific findings can be found. One limitation of this paper is that driving forces of NET change were demonstrated using trade volume balance and difference of CO2 emission intensity. However, it is correlated with 6 factors: activity effect, structure effect and technology effect of both countries. Future studies could search an innovative method to decompose NET for an in-depth understanding of NET change. Acknowledgements This study is funded by the Natural Science Foundation of China (NSFC. 71461137008, 71325006) and the NSFC-JSPS Joint Research Grant (71311140172). References Ang, B., 2004. Decomposition analysis for policymaking in energy: which is the preferred method? Energy policy. 10
ACCEPTED MANUSCRIPT Ang, B.W., Zhang, F.Q., Choi, K.H., 1998. Factorizing changes in energy and environmental indicators through decomposition. Energy 23, 489-495. Chenery, H.B., 1953. The structure and growth of the Italian economy. United States of America, Mutual Security Agency. Davis, S.J., Caldeira, K., 2010. Consumption-based accounting of CO2 emissions. Proceedings of the National Academy of Sciences of the United States of America 107, 5687-5692. Davis, S.J., Peters, G.P., Caldeira, K., 2011. The supply chain of CO2 emissions. Proceedings of the
RI PT
National Academy of Sciences of the United States of America 108, 18554-18559.
Dong, H., Geng, Y., Fujita, T., Jacques, D.A., 2014. Three accounts for regional carbon emissions from both fossil energy consumption and industrial process. Energy 67, 276-283.
Dong, Y., Ishikawa, M., Liu, X., Wang, C., 2010. An analysis of the driving forces of CO2 emissions embodied in Japan–China trade. Energy Policy 38, 6784-6792.
Duan, Y., Yang, C., Zhu, K., 2012. Embodied Energy and CO2 Emission in Sino-Japan Trade. 20th
SC
International Input - Output Conference. Bratislava, Slovakia. Conference proceedings (24 - 29 June, 2012).
Geng, Y., Tsuyoshi, F., Chen, X., 2010. Evaluation of innovative municipal solid waste management
M AN U
through urban symbiosis: a case study of Kawasaki. Journal of Cleaner Production 18, 993-1000. Guan, D., Peters, G.P., Weber, C.L., Hubacek, K., 2009. Journey to world top emitter: An analysis of the driving forces of China's recent CO2 emissions surge. Geophysical Research Letters 36. Isard, W., 1951. Interregional and regional input-output analysis: a model of a space-economy. The review of Economics and Statistics, 318-328.
Leontief, W.W., 1941. Structure of American economy, 1919-1929.
Liu, X., Ishikawa, M., Wang, C., Dong, Y., Liu, W., 2010. Analyses of CO2 emissions embodied in Japan–
TE D
China trade. Energy Policy 38, 1510-1518.
Moses, L.N., 1955. The stability of interregional trading patterns and input-output analysis. The American Economic Review, 803-826.
Peters, G.P., 2008. From production-based to consumption-based national emission inventories. Ecological Economics 65, 13-23.
EP
Peters, G.P., 2010. Policy Update: Managing carbon leakage. Carbon Management 1, 35-37. Peters, G.P., Hertwich, E.G., 2008. CO2 embodied in international trade with implications for global climate policy. Environmental Science & Technology 42, 1401-1407.
AC C
Peters, G.P., Minx, J.C., Weber, C.L., Edenhofer, O., 2011. Growth in emission transfers via international trade from 1990 to 2008. Proceedings of the National Academy of Sciences of the United States of America 108, 8903-8908.
Shimazaki, Y., Akisawa, A., Kashiwagi, T., 2000. A model analysis of clean development mechanisms to reduce both CO2 and SO2 emissions between Japan and China. Applied Energy 66, 311-324. Springmann, M., 2014. Integrating emissions transfers into policy-making. Nature Climate Change 4, 177-181. Su, B., Ang, B.W., 2014. Input–output analysis of CO2 emissions embodied in trade: A multi-region model for China. Applied Energy 114, 377-384. Temurshoev, U., Timmer, M.P., 2011. Joint estimation of supply and use tables. Papers in Regional Science 90, 863-882. Timmer, M., Erumban, A., Francois, J., Genty, A., Gouma, R., Los, B., Neuwahl, F., Pindyuk, O., Poeschl, J., Rueda-Cantuche, J., 2012. The World Input-Output Database (WIOD): Contents, Sources and 11
ACCEPTED MANUSCRIPT Methods. WIOD Background document available at www. wiod. org. Wang, Z., Feng, C., Zhang, B., 2014. An empirical analysis of China's energy efficiency from both static and dynamic perspectives. Energy 74, 322-330. Wang, Z., Wang, C., Yin, J., 2015. Strategies for addressing climate change on the industrial level: affecting factors to CO2 emissions of energy-intensive industries in China. Nat Hazards 75, 303-317. Weber, C.L., Peters, G.P., Guan, D., Hubacek, K., 2008. The contribution of Chinese exports to climate change. Energy Policy 36, 3572-3577.
RI PT
Xu, X.Y., Ang, B.W., 2013. Index decomposition analysis applied to CO2 emission studies. Ecological Economics 93, 313-329.
Zhao, Y., Ke, J., Ni, C.C., McNeil, M., Khanna, N.Z., Zhou, N., Fridley, D., Li, Q., 2014. A comparative
study of energy consumption and efficiency of Japanese and Chinese manufacturing industry. Energy
AC C
EP
TE D
M AN U
SC
Policy 70, 45-56.
12
ACCEPTED MANUSCRIPT Table. 1. Two regions input-output table Intermediate demands Region r Region s RoW Zrs Zrr Intermed- Region r iate Zsr Zss Region s inputs RoW Output
Input
Fss
T
(Xs)
EP
TE D
M AN U
SC
(Xr)
AC C
Xs
Vs T
Total inputs
Fsr
Total outputs Xr
RI PT
Vr
Value added
Final demands Region r Region s RoW Frs Frr
ACCEPTED MANUSCRIPT 25%
200 180 160
20%
US$ Billion
140 120
15%
Import from Japan Export to Japan
100 80
10%
Share of China's total imports
RI PT
60 40
5%
20 0
0%
AC C
EP
TE D
M AN U
SC
Fig. 1. Trade volume and shares of total imports.
Share of Japan's total imports
ACCEPTED MANUSCRIPT 180 160 Import from Japan
140
100
Export to Japan
80
RI PT
Mt CO2
120
60
Net emission transfer
40 20 0
SC
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
AC C
EP
TE D
M AN U
Fig. 2. CO2 emissions embodied in China-Japan trade for the period of 2000-2009.
ACCEPTED MANUSCRIPT 100% 90%
Others
80%
Electrical & Optical Equipment
70% Machinery
60%
Basic Metals & Fabricated Metal
50%
RI PT
Other Non-Metallic Mineral
40%
Rubber & Plastics
30%
Chemicals & Chemical Products
20% 10%
Textiles & Textile Products
0%
Food, Beverages & Tobacco
SC
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
a) CO2 emissions embodied in China’s exports to Japan by industrial sectors.
M AN U
100% 90% 80%
Others
70%
Transport Equipment
60%
Electrical & Optical Equipment
50%
Machinery
40%
Basic Metals & Fabricated Metal
20% 10% 0%
TE D
30%
Other Non-Metallic Mineral Chemicals & Chemical Products
EP
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
AC C
b) CO2 emissions embodied in China’s imports from Japan by industrial sectors. Fig. 3. Structure of emissions embodied in China-Japan trade.
ACCEPTED MANUSCRIPT 30 25 20 15 technology
RI PT
Mt CO2
10
srtucture
5
activity
0
total
2000-1 2001-2 2002-3 2003-4 2004-5 2005-6 2006-7 2007-8 2008-9 -5
SC
-10 -15
M AN U
-20
a) Drivers of changes in CO2 emissions embodied in export from China to Japan. 8 6
0
2000-1 2001-2 2002-3 2003-4 2004-5 2005-6 2006-7 2007-8 2008-9 -2 -4
technology srtucture activity total
EP
Mt CO2
2
TE D
4
AC C
-6 -8
b) Drivers of changes in CO2 emissions embodied in import from Japan to China. Fig. 4. Decomposition analysis results of changes in embodied CO2 emissions
ACCEPTED MANUSCRIPT 200%
100%
0%
-100%
-200% NET -300% CO2 emission intensity difference
-400%
M AN U
Trade volume balance
SC
RI PT
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
-500%
AC C
EP
TE D
Fig. 5. Changes of NET, difference of CO2 emission intensity and trade volume balance between China and Japan.
30
10
25
8
20
6
15
4
10
2
Million US $
12
RI PT
Kt CO2
ACCEPTED MANUSCRIPT
5
0
0
2002
2003
2004
2005
2006
2007
2008
2009
SC
2001
Emission dependence of China
Emission dependence of Japan
Economic dependence of China
Economic dependence of Japan
M AN U
2000
AC C
EP
TE D
Fig. 6. CO2 emission dependence and economic dependence of China and Japan Note: The values show CO2 emission and economic output of its trading partner driven by output of 1 billion US $
ACCEPTED MANUSCRIPT 2.5
1.5
China per GDP China per export
1.0
RI PT
Kg CO2 / 2005 US $
2.0
Japan per GDP
Japan per export
0.5
0.0
SC
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
AC C
EP
TE D
M AN U
Fig. 7. CO2 emissions per trade volume and per GDP
ACCEPTED MANUSCRIPT
Highlights
Embodied CO2 emissions flows between China and Japan were estimated; Emission embodied in bilateral trade (EEBT) approach was employed
RI PT
for our analysis;
China was a net exporter of embodied CO2 emissions between
Trade volume was the main driver for the increase of embodied
M AN U
emissions;
Technology effect contributed most to the decrease of embodied
EP
TE D
emissions.
AC C
SC
China-Japan trade;