In the 2000s, the rapid growth of CO₂ emitted in the production of exports from developing to developed countries, in which China accounted for the dominant share, led to concerns that climate polices had been undermined by international trade. Arguments on “carbon leakage” and “competitiveness”—which led to the refusal of the U.S. to ratify the Kyoto Protocol—put pressure on developing countries, especially China, to limit their emissions with Border Carbon Adjustments used as one threat. After strong growth in the early 2000s, emissions exported from developing to developed countries plateaued and could have even decreased since 2007. These changes were mainly due to China: In 2002–2007, China's exported emissions grew by 827 MtCO₂, amounting to almost all the 892 MtCO₂ total increase in emissions exported from developing to developed countries, while in 2007–2012, emissions exported from China decreased by 229 MtCO₂, contributing to the total decrease of 172 MtCO₂ exported from developing to developed countries. We apply Structural Decomposition Analysis to find that, in addition to the diminishing effects of the global financial crisis, the slowdown and eventual plateau was largely explained by several potentially permanent changes in China: Decline in export volume growth, improvements in CO₂ intensity, and changes in production structure and the mix of exported products. We argue that growth in China's exported emissions will not return to the high levels during the 2000s, therefore the arguments for climate polices focused on embodied emissions such as Border Carbon Adjustments are now weakened.

Plain Language Summary

In the 2000s, CO₂ emissions from production in developed countries flattened while emissions from their consumption grew. The rapid growth of exported emissions from developing countries to developed countries, with the largest contribution from China, played a significant role. This led to concerns that climate polices had been undermined by rapid growth in international trade. Since around 2007, growth in these exported emissions has plateaued, predominantly due to changes in China. Our study investigates China's changes, demonstrating that in addition to the effects of global financial crisis, China implemented potentially permanent structural changes. We argue that China's exported emissions are unlikely to return to the growth levels from the 2000s, and therefore trade-related climate polices will be much less relevant.

1 Introduction

In the 2000s, CO₂ emissions from the production in developed countries flattened, while emissions from their consumption grew (Figure 1), with a key contribution from the rapid growth of CO₂ emitted in the production of goods and services exported from developing countries to developed countries (Figure 2). These changes led to concerns that rapid growth in international trade and increased consumption in developed countries had offset the emission reductions in developed countries and thereby undermined climate policies [Peters et al., 2011; Kanemoto et al., 2014], and led to discussions on how to deal with these “emission transfers,” that is the difference between emissions embodied in exports and emissions embodied in imports, or equivalently, between production- and consumption-based emissions [Davis et al., 2011; Jakob and Marschinski, 2012; Grasso and Roberts, 2014; Springmann, 2014]. In response, some have suggested reallocating mitigation responsibilities from production- to consumption-based emission inventories [Munksgaard and Pedersen, 2001; Pan et al., 2008; Davis and Caldeira, 2010; Peters et al., 2011; Qi et al., 2013]. Some developed countries argued for Border Carbon Adjustments (BCAs) aiming at “carbon leakage” and “competitiveness concerns,” and encouraged developing countries to take on mitigation commitments [Pauwelyn, 2007; Böhringer et al., 2012; Cosbey et al., 2012]. These discussions put pressure on developing countries to limit their emissions, especially China, which was a key contributor to the increasing emission transfers [Davis and Caldeira, 2010; Peters et al., 2011].

EFT2-250-FIG-0001-c — **Figure 1**
Open in figure viewer PowerPoint

Production, exports, imports, and consumption CO₂ emissions of the U.S. (a), EU28 (b), China (c), and Annex B countries (d). The data are from the Global Carbon Budget [*Le Quéré et al*., 2015]. The consumption emissions can be estimated from the production emissions by subtracting the exports and adding the imports. The emission transfers (Net) is the difference between exports and imports, or equivalently, production and consumption. For the U.S. and EU28, and developed (Annex B) countries as a whole, the emission transfers are negative (net importers). The net import grew in the 2000s, but has plateaued in the last 10 years. China is a net exporter of emissions, and has shown strong growth in the 2000s, but is quite stable outside of that period. Annex B countries are those countries with emission limitations in the Kyoto Protocol, whether ratified or not.

EFT2-250-FIG-0002-c — **Figure 2**
Open in figure viewer PowerPoint

Exported CO₂ emissions from non-Annex B countries to Annex B countries. The data are from the Global Carbon Budget [*Le Quéré et al*., 2015]. The stacked bars show the total exported CO₂ emissions from non-Annex B countries to Annex B countries. Only the five countries with the largest changes over the period are shown. The changes in each period are decomposed into contributions of individual countries/regions, shown as floating bars (between 2002–2007 and 2007–2012) that are in the same color and order as the stacked bars. China has the largest share of the emission transfers, and is the main reason for changes in both periods. The years 2002, 2007, and 2012 were selected since these are the years with the necessary Chinese input–output data for later analysis.

After reaching the maximum in 2007, the difference between emissions from the production and consumption in developed countries has since narrowed (Figure 1d), mainly caused by the plateaued and even decreased emissions imported from developing countries (Figure 2). The plateauing of emissions imported from developing countries was largely due to the decrease in emissions exported from China to developed countries (Figure 2). In the previous period (2002–2007), China's emissions exported to developed countries increased by 827 Mt. CO₂, 93% of the total change in emissions exported from developing countries to developed countries (892 Mt. CO₂). Without China, there would probably have been only a marginal increase of emission transfers over the period. Consistent with the trend in emission transfers in developed countries, China's exported CO₂ emissions hit their highest level in 2007 (Figure 1c). From 2007 to 2012, the change of emissions exported from China to developed countries was −229 Mt. CO₂, and the change in total exported emissions from all developing countries to developed countries was −172 Mt. CO₂. As for the previous period, developing countries other than China made only a small contribution. Consequently, many arguments in favor of consumption-based climate policies, BCAs, and indeed, the U.S.' justification for not ratifying the Kyoto Protocol, would be less relevant.

Studies have shown that in the early and mid-2000s, CO₂ emissions from the production of Chinese exports accounted for a large amount of the fast-growing Chinese emissions. In fact, more than one fifth of China's CO₂ emissions during this period were from the production of exported products [Chen and Zhang, 2010; Davis and Caldeira, 2010]. The high value of the CO₂ intensity, combined with the volume of exports, was the dominant reason for the high level of China's exported CO₂ emissions [Guan et al., 2009; Liu et al., 2015a]. In terms of the growth in China's CO₂ emissions, more than one-half was due to the increase in China's exports between 2002 and 2005 [Guan et al., 2009; Minx et al., 2011]. The emission intensity had improved during this period. However, this efficiency gains were offset by the additional emissions caused by the increased final consumption, especially the increase in exports [Guan et al., 2009; Xu et al., 2011]. From 2002 to 2007, changes in production structure emerged to be a major driver in explaining the fast growth in China's exported emissions [Minx et al., 2011; Xu et al., 2011]. Studies at the regional level find that the increase in exported emissions was much more significant in China's eastern-coastal region in this period [Feng et al., 2012].However, the plateau in Chinese exported CO₂ emissions after 2007 has not been highlighted nor investigated. The global financial crisis in 2008/2009 could be a reason, but the inconsistent trends in China's exported emissions and its exports after the crisis suggest that there are potentially other reasons (Figure S6 of Supporting Information S1, China's exports declined in the financial crisis, but increased after the crisis, although less fast. Data source: the United Nations [Department of Economic and Social Affairs, 2016]). If the change in the trend in China's exported emissions were not simply indicative of a slow rebound after the global financial crisis, it would potentially have significant implications for climate policy discussions, particularly on whether emission transfers are a political justification for weaker climate policies in developed countries [Peters, 2014].We analyze the factors that drove changes in CO₂ emissions from fossil fuels and cement in the production of Chinese exports between 1997 and 2012, and investigate the policy implications. We construct constant-price input–output tables (IOTs) for China (see Section 2) to perform a Structural Decomposition Analysis (SDA; see Section 2), quantifying the contributions of the main changes at the sector level. We focus our analysis on 1997, 2002, 2007, and 2012 as these are the years with the necessary input–output data, and these years, particularly 2007, happen to coincide with relevant structural changes in China (see Section 2).

2 Materials and Methods

Full details of materials and methods are given in the Text S3 of Supporting Information S1, including a more detailed discussion on potential limitations of our analysis.

2.1 Input–Output Tables

We obtain the necessary IOTs of China in current price from the Chinese National Bureau of Statistics (NBS) for 1997 with 124 sectors, 2002 with 122 sectors, 2007 with 135 sectors, and 2012 with 139 sectors [NBS of China, 1999, 2006a, 2009, 2015a]. These years, particularly 2007, also happen to coincide with the relevant changes in China: The absolute amounts of emission transfers of both China and developed countries reached their highest level in 2007 (Figure 1), and the share of exports in China's gross domestic product (GDP) reached a maximum in 2007 (Figure S6 of Supporting Information S1), suggesting that 2007 is the key year to study the changes in Chinese exported emissions.

While multiregional input–output tables (MRIOTs) have been widely used to study trade related issues [e.g., Davis and Caldeira, 2010; Peters et al., 2011], we use China's single-region IOTs (SRIOTs) instead of global MRIOTs for two reasons. First, our focus is on Chinese emissions embodied in Chinese exports, for which the difference between using MRIOTs and SRIOTs is that the former can track the emissions embodied in the products which are exported from China but arrive back in China through the global supply chains. However, the amount of these emissions is small, and would not alter our conclusions. Second, when constructing an MRIOT, the original IOT of China must be adjusted due to harmonization of the data from different resources. Thus, some of the information in the original IOT will be lost, resulting in potential bias in data. Therefore, we choose not to trace the small effects at the risk of introducing more uncertainties.

2.1.1 Constant Price IOTs

To perform the SDA, we construct IOTs in constant price of 2002 using the method of double deflation [United Nations, 1999]. The sector-level chained price indices [Eurostat, 2013] are used, which are taken from the China Price Statistical Yearbook [NBS of China, 2013], the China Statistical Yearbooks [NBS of China, 1998b], as well as the National Data [NBS of China, 2016] (see details in Table S2 of Supporting Information S1). The producer price index (PPI) is used for the nonservice sectors (except for the construction sector, for which the Price Index for Investment in Fixed Assets is used). However, since the PPI is not available for the service sectors, we use the consumer price index (CPI) for the service sectors. Importantly, the retail price index is used for the retail sector, and the value-added index for tertiary industry is used for the service sectors not associated with products from the CPI. The deflators (covering 53 sectors) are then mapped to the IOT sectors (more than 100 sectors).

2.1.2 Splitting Domestic and Imported Products

Since our study focuses on the CO₂ emissions released in the territory of China for Chinese exports, we split the imported from the intermediate use and final demand in the IOTs based on two assumptions: (1) There are no reexports; (2) For each sector and final demand category, imports account for the same proportion as the overall share of imports in total domestic use.

2.1.3 Sector Harmonization

Given that the classifications of the IOTs were revised throughout the years, all four constant price IOTs are harmonized into 99 sectors (Table S3 of Supporting Information S1).

2.1.4 Adjustment for the Electricity System Reform

The IOTs of 2007 and 2012 are adjusted because of the large amount of self-purchase of the electricity sector following Chinese electricity system reform in 2002 [State Council of the People's Republic of China, 2002] (see the details in Text S1 of Supporting Information S1). We also perform comparisons between the results and uncertainties of the Structural Decomposition Analyses based on the adjusted and the original IOTs, showing that there are notable differences between the results (Figure S3 and Text S2 of Supporting Information S1), but not sufficient to alter our conclusions. The differences in the results also suggest that “self-purchases” (e.g., chemical industry purchasing from the chemical industry) could be problematic, but additional analysis suggests our findings remain robust (see Text S1 and Table S6 of Supporting Information S1).

2.2 CO₂ Emissions

We update our previous methods [Peters et al., 2006] to estimate China's sectoral CO₂ emissions. The emissions are related to 27 types of fossil fuels of 48 sectors (17 types before 2010 and 46 sectors before 2012) and the production of cement.

2.2.1 CO₂ Emissions from Fossil-Fuel Consumption

Fossil-fuel consumption data are obtained from China Energy Statistical Yearbooks (CESY). Since the historical energy consumption data are revised after each National Economic Census (NEC) [Korsbakken et al., 2016], we adopt the energy data for 2002, 2007, and 2012 in the CESY 2014 [NBS of China, 2015b] that is based on the third NEC, and 1997 in the CESY 2009 [NBS of China, 2010] that is based on the second NEC. The oxidation rates and emission factors are taken from China National Greenhouse Gas Inventory 2005 (CNGHG 2005) [China National Development and Reform Commission, 2014].

For the CO₂ emissions from the final energy consumption: (1) Certain proportions of non-energy use are subtracted from the corresponding sectors according to CNGHG 1994 [Office of National Climate Change Coordination Committee and National Development and Reform Commission, 2007]. (2) Losses of petroleum and natural gas are added to corresponding sectors [NBS of China, 2006b]. (3) CO₂ emissions are estimated using the IPCC tier-2 [IPCC, 2006] method. Particularly, an IPCC tier-3 [IPCC, 2006] calculation for coal is introduced to obtain the average emission factors and oxidation rates for each sector and coal type. For the fossil fuel consumed in the transformation processes, we estimate the CO₂ emissions by the carbon “lost,” which refers to the net carbon inputs in the processes, then allocate the emissions to associated sectors.

2.2.2 CO₂ Emissions from Cement Production

We adopt the IPCC tier-2 [IPCC, 2006] method to calculate the CO₂ emissions from cement production. The clinker production data are taken from China Cement Almanac [China Cement Association, 2015] (for 2002, 2007, and 2012) and China Industry Economy Statistical Yearbook [NBS of China, 1998a] (for 1997). The emission factor for clinker and the correction factor for cement kiln dust are taken from CNGHG 2005.

2.2.3 Uncertainty in Estimating Chinese Emissions

There is a great deal of uncertainty and several inconsistent estimates of Chinese emissions [Guan et al., 2012; Liu et al., 2015b]. Issues include both internal inconsistencies and significant revisions of official activity data [in particular energy data, see Korsbakken et al., 2016 and references therein for an overview] and inconsistent data and assumptions for emission factors [e.g., Liu et al., 2015b]. However, these issues are unlikely to affect our main conclusions appreciably, as the uncertainties mainly apply to absolute emission levels, and not growth rates [Korsbakken et al., 2016]. Furthermore, the CO₂ emission estimates in this article use the most up to date official energy consumption data from the NBS of China [2010, 2015b], combined with emission factors and methodology from the most recent national greenhouse gas inventory[China National Development and Reform Commission, 2014]. The energy data include the most recent official revisions, which greatly reduced internal inconsistencies found in earlier data. This is the only truly comprehensive, independent data set of Chinese energy consumption, as all other data sets and CO₂ emission estimates are based directly or indirectly on it. For emission factors, Liu et al. [2015b] provide an ostensibly independent alternative to the data of the official greenhouse gas inventory, but their data and the interpretation of it have been challenged [Teng, 2015; Teng and Zhu, 2015]. We therefore take the conservative approach of only using official data.

2.3 Environmental Input–Output Analysis

In this study, we adopt environmental input–output analysis, which determines the economy-wide environmental repercussions stemming from economic activity [Peters et al., 2007]. Since the sector classification of the emission data (less than 50 sectors) is more aggregated than the IOTs' (more than 100 sectors, but varying by year), and there are overlaps between these two classifications, a bidirectional concordance between emission sectors and IOT sectors is required. We assume that the IOT sectors mapped to the same aggregate sector have the same emission intensity.

2.4 Structural Decomposition Analysis

The SDA is based on environmental input–output analysis [Miller and Blair, 2009]. In terms of the specific decomposition techniques, the method proposed by Dietzenbacher and Los [1998] (D&L) and the logarithmic mean Divisia index methods (LMDI) [Ang and Choi, 1997; Ang and Liu, 2001; Ang et al., 1998, 2003] are the most used. Here we adopted D&L for the justifications of the consistency between the results from D&L and LMDI [Su and Ang, 2012], and the general adoption of the D&L method in similar previous studies [Peters et al., 2007; Guan et al., 2009; Minx et al., 2011] (see details in Text S3 of Supporting Information S1). The change in emissions from the production of exports can be decomposed as

$urn:x-wiley:23284277:media:eft2250:eft2250-math-0001$ (1)

where q represents exported emissions; f is the vector of emission intensity. L is the Leontief inverse; y_s and y_v represent export structure and export volume, respectively. $urn:x-wiley:23284277:media:eft2250:eft2250-math-0002$ is the contribution of changes in emission intensity; $urn:x-wiley:23284277:media:eft2250:eft2250-math-0003$ is the contribution of changes in production structure; $urn:x-wiley:23284277:media:eft2250:eft2250-math-0004$ is the contribution of changes in export structure; and $urn:x-wiley:23284277:media:eft2250:eft2250-math-0005$ is the contribution of changes in export volume.

A major problem of SDA is that it has nonunique and equally acceptable decomposition forms [Dietzenbacher and Los, 1998]. The D&L method we adopt takes the full range of first-order index combinations for the measuring factors, assuming that both the start-point-index and the end-point-index weight for each factor [Hoekstra and van den Bergh, 2003]. Following the suggestion of Dietzenbacher and Los [1998], we take the mean of the full set of the decompositions, which are further sorted by factor to improve the computing efficiency [Rørmose and Olsen, 2005]. We also show the results of all the decomposition forms (Figure S5 of Supporting Information S1), and discuss the uncertainties of the decomposition involved by the nonunique decomposition forms (Text S2 of Supporting Information S1).

When performing an SDA, an assumption is that the determinants are independent. However, this assumption may involve problems if two or more determinants are technically dependent [Dietzenbacher and Los, 2000]. Several decomposition forms are provided in [Dietzenbacher and Los, 2000] trying to fix the dependency problems. However, due to a lack of clear theoretical justification of the interrelations, the assumptions involved to fix the dependency problems will possibly cause new problems [Dietzenbacher et al., 2004]. Therefore, as in other studies, we choose not to correct the potential dependency problem [Dietzenbacher et al., 2004; Minx et al., 2011; Owen et al., 2014], and stress that this is a potential problem of all traditional SDA on energy and emission issues.

2.5 Limitations on Data and Methods

In this study, we perform a standard SDA based on China's national IOTs. However, due to data limitations, we do not conduct an analysis at the subnational level that is suggested by Su and Ang [2010]. Furthermore, distinction between normal exports and processing exports would also have potential influences to the results [Dietzenbacher et al., 2012; Su et al., 2013]. At a more detailed level, firm heterogeneity is also an issue worth addressing when data are available [Meng et al., 2013].

3 Results

3.1 CO₂ Emissions in Chinese Exports

CO₂ emissions from the production of Chinese exports tripled between 1997 and 2012, with the largest growth in 2002–2007 (over 150%, Table 1). Meanwhile, the share of exported emissions in China's total CO₂ emissions went up from 19% in 2002 to a peak of 29% in 2007, surpassing households as the second largest final demand causing emissions (Figure S1 of Supporting Information S1). After 2007, the exported emissions started to decrease and dropped dramatically in 2009 coinciding with the global financial crisis. With global trade recovering slower than expected after the financial crisis, China's exported emissions have not returned to pre-2009 growth levels and only grew slowly back to 1951 Mt. CO₂ in 2012, which is still 8% lower than the amount of 2007. Correspondingly, in the year 2012, the share of exported emissions in the total CO₂ emissions of China decreased to a similar level as 2002 (21%). In 2007–2012, the largest increase in China's CO₂ emissions came from capital investments, which is due to massive government investments in response to the global financial crisis. In 2012, the non-exported emissions were associated with capital investments (50% of 2012's emissions; Figure S1 of Supporting Information S1), household consumption (24%), and government consumption (5%). There is a great deal of uncertainty and several inconsistent estimates of Chinese emissions [Guan et al., 2012; Liu et al., 2015b], but these issues are unlikely to have a major impact on the conclusions (see Section 2 for detailed discussions).

Table 1. China's Exported CO₂ Emissions From 1997 to 2012, Showing the Absolute Levels, Share of Total CO₂ Emissions, and the Percentage Growth Between Periods

	1997	2002	2007	2012
Exported CO₂ (Mt)	603	805	2112	1951
Share in total (%)	19	21	29	21
Growth (%)	–	33	162	−8

3.2 Explaining the Factors of Changes in China's Exported Emissions

To help understand the changes in China's exported CO₂ emissions from 1997 to 2012, we perform a SDA, which decomposes the changes in exported emissions into four key components at the sector level: CO₂ intensity (emissions per unit of total output in economic value), export structure (type of products exported), production structure (relationship between sectors), and export volume. The results of the SDA show that CO₂ intensity, production, and export structures played significant roles in the decrease of China's exported emissions from 2007 to 2012 (Figure 3). Over the entire period of our analysis (1997–2012), the rapid growth in export volume was the major driver of the changes in Chinese exported CO₂ emissions.

EFT2-250-FIG-0003-c — **Figure 3**
Open in figure viewer PowerPoint

Explaining the factors of changes in CO₂ emissions from production of Chinese exports. The bars of the factors show the mean of all the decomposition forms, while the error bars show the ranges of the decompositions (see Figure S5 for more details). The exported emissions have been largely driven by export volume growth (light blue), while offset by CO₂ intensity improvements (dark blue). In 2007–2012, the exported emissions actually reduced due to a smaller growth in export volume that was largely offset by stronger improvements in CO₂ intensity and structures of production and export. The data here are calculated from China's IOTs, so the actual numbers differ, but consistent with Figures 1 and 2. The years 1997, 2002, 2007, and 2012 were selected since these are the years with the necessary Chinese input–output data.

We now focus our analysis on the subperiods, 2002–2007 and 2007–2012. During 2002–07, China's exported emissions increased by 1307 Mt. CO₂, largely driven by the growth of export volume with a small offset from improvement in CO₂ intensity. In this period, the contributions of production structure and CO₂ intensity were small for exported emissions (the same for China's total emissions, see Figure S7 of Supporting Information S1)—this contrasts with some previous studies [Guan et al., 2009; Minx et al., 2011] because we make necessary adjustments to the IOTs for the Chinese electricity system reform in 2002 (see Text S1 of Supporting Information S1). In the following period, 2007–2012, exported emissions actually declined by 161 Mt. CO₂ due to much smaller growth in export volume, with the largest counterbalance (64%) from improvement in CO₂ intensity, but also accelerated improvements in structures of production and export. These effects are significant even though there are uncertainties involved by the non-unique decomposition forms (the error bars in Figure 3). We discuss the uncertainties of the decomposition in detail in the Text S2 of Supporting Information S1.

The smaller growth in Chinese export volume had a direct effect on China's exported emissions. The global financial crisis in 2008–2009 is an important consideration to understand the slowdown in Chinese exports. Since 2008, the collapse of global demand, especially the demand of advanced economies, led to a slowdown in Chinese exports [Ahuja et al., 2012]. In 2009, the value of China's exports dropped by 5% in real terms (Figure S6 of Supporting Information S1). In the following years, although global trade volumes have since recovered, their growth has been much slower than before the slump. Hence, the emissions transferred from China to other regions via exports also grew much more slowly.

A sector-level analysis shows a clearer picture of what happened in China during 2002–2012. Efficiency improvements in power generation played an important role in slowing the growth of China's exported emissions (Figure 4a). The electricity sector accounted for 46% of total exported emission savings from CO₂ intensity improvement in 2002–2007 and 26% in 2007–2012. In physical terms, the CO₂ intensity of Chinese electricity production decreased by 23% between 2002 and 2012 (Figure S4 of Supporting Information S1), driven both by improved standards and by a reduction in the share of thermal power from 82% to 79% [Chinadialogue, 2016]. Further significant contributions to emission savings from decreased CO₂ intensity in 2007–2012 came from the sectors producing nonmetal products, raw chemical products, and steel.

EFT2-250-FIG-0004-c — **Figure 4**
Open in figure viewer PowerPoint

Contributions of the 99 individual sectors to changes in exported CO₂ emissions by factor. (a) Emission changes due to the CO₂ intensity changes of each sector. (b) Emission changes due to the changes in export structure. For individual sectors, export structure refers to the shares of each sector of the total exports. (c) Emission changes caused by the changes of electricity consumptions of sectors. Full sector list is shown in the Table S3, including the split between Primary, Secondary, and Tertiary sectors (gray bar at the top).

When looking at the effects of production structure at the sector level, electricity inputs in the economy acted as the main factor causing changes, with 95% of the change in exported emissions caused by changes in production structure between 2002 and 2007, and 74% in the following period (Figure 4c). From 2002 to 2007, the manufacturing sectors, except for the electronic products sectors, used more electricity in production, which led to an increase in the exported emissions. This situation reversed in the period 2007–2012, with a reduced share of electricity input in China's production helping save 187 Mt. exported CO₂ emissions, indicating that the Chinese economy was using electricity more efficiently.

At the sector level, the effects of export structure in the changes in China's exported emissions have been increasing over time (Figure 4b). CO₂ emissions from the production of steel exports played a significant role in the changes in China's exported emissions related to export structure: In 2002–2007, the increased share of steel in exports led to 115 Mt. additional CO₂ emissions, though the increase was largely offset by the decrease in the wearing apparel sector. In 2007–2012, this situation reversed, as the share of steel in China's exports decreased, 89 Mt. CO₂ emission was saved, accounting for 56% of the total emission reduction gained from export structure change. The share of another significant export from China, wearing apparel, decreased in both 2002–2007 and 2007–2012, contributing 62% of the emissions related to export structure changes over these two periods. In contrast, the shares of the electronic products sectors increased in both periods, such that they exerted the largest upward pressure on the total emission change from export structure change. In general, changes in China's export structure over time have tended to lower emissions, with declining shares of carbon-intensive products and increasing shares of technology-intensive products. A decomposition of Chinese total CO₂ emissions shows similar structural changes to the exported structure (Figure S8 of Supporting Information S1).

A closer analysis of key changes in China's export structure using export data in both physical and monetary terms supports our analysis with input–output data. The most significant change was seen in steel exports. In 2009, the volume of China's steel exports declined by 58% in physical terms (Figure S9 of Supporting Information S1), with its share in Chinese merchandize exports (monetary) dropping from 3.2% to 1.3% (Figure S10 of Supporting Information S1) [NBS of China, 2016]. With the gradual recovery of global demand, the share of steel in China's exports climbed back to 2.3% in 2014 (Figure S10 of Supporting Information S1), which is still much lower than precrisis levels. Another noteworthy change happened to the wearing apparel sector, which was also hit in the financial crisis. In 2007–2012, the share of wearing apparel products in Chinese merchandize exports decreased from around 6% by nearly 2 percentage points and has not since recovered (Figure S10 of Supporting Information S1), even though China raised the export value-added tax (VAT) rebate of textiles and wearing apparel products during the crisis to encourage their production and exports (Figure S2 of Supporting Information S1, [State Administration of Taxation, 2017]).

When considering the changes in China's exported emissions gained from export structure changes, it is worth mentioning the role of the adjustment to export VAT rebates. Outside the global financial crisis, export VAT rebate was also used as a tool to control the growth of energy- and emission-intensive industries [Gourdon et al., 2016; Eisenbarth, 2017], especially steel which accounts for a significant share of Chinese exports and grew rapidly in the early 2000s. Over the 11th FYP period (2006–2010), the VAT rebates of steel products were reduced or removed [Figure S2 of Supporting Information S1, State Administration of Taxation, 2017], combining with the impacts of the financial crisis, emissions exported via steel declined over the period of 2007–2012. In contrast, the exports of electronic products, which are low in emission intensities, were encouraged in the 11th FYP period by high and even further raised VAT rebates [Figure S2 of Supporting Information S1, State Administration of Taxation, 2017]. The VAT rebates of electronic products were kept at a high level (76% of the VAT is rebated). For some electronic products, the VAT were even fully rebated. The high VAT rebates would have contributed to the increase in the export shares of electronic products.

4 Discussion and Conclusions

Our results show that between 2002 and 2007, China experienced rapid growth in exported CO₂ emissions primarily due to rapid growth in the volume of exports. After 2007, Chinese exported emissions unexpectedly plateaued. The reasons for this plateauing were the much slower export volume growth, and the accelerated improvements in CO₂ intensity and structures of production and export. Our findings indicate that the external demand for Chinese products is the key factor in the plateauing of Chinese export-related CO₂ emissions, while internal factors (CO₂ intensity, production structure and export structure) are the main factors offsetting emissions caused by continued export growth.

Improvements in CO₂ intensity and production structure were the major internal drivers in the plateau in China's exported emissions from 2007 to 2012, which were primarily associated with changes in the electricity sector. During this period, per unit production of electricity emitted less CO₂. In addition, with less relative electricity input in the production, Chinese economy was using electricity in a more efficient way. Several actions of China focusing on efficiency improvements in power generation as well as energy saving were important to these changes. During the 11th FYP period, around 77 GW of small thermal power units with lower efficiency were shut down [Du, 2011]. Meanwhile, the power generated by high-efficiency thermal power units was given priority access to grids, while the access given to power from small units was limited [State Council of the People's Republic of China, 2007, 2011]. Renewable energy has also been strongly encouraged: the share of non-fossil electricity generation increased by 10 percentage points from 2007 to 2015 [Chinadialogue, 2016], further decreasing the CO₂ intensity of electricity generation. The downward pressure on China's exported emissions improvements in the electricity sector highlight the significant role of electricity in Chinese exported emissions via the domestic supply chains. The improvements in the production and use of electricity also indicate that the Chinese economy is shifting onto a lower carbon pathway.

Changes in export structure also played an important role in reducing China's exported emissions in 2007–2012, with decreased shares of steel, wearing apparel, and an increase in the share of electronics. Other than the effect of the global financial crisis, changes in the export structure are associated with a number of Chinese policies in this period, such as the adjustment to export VAT, which has proved to be effective in China [Song et al., 2015]. In addition, similar structural changes are also observed in China's total final demand, except that there was an increase in the emissions related to the share of construction sector. The consistency in the structural changes of both exports and total final demand strongly suggests that China's economy as a whole has been undergoing these same structural changes. This is consistent with the recent phenomenon that the low-value-added industries are moving from China to the less developed countries, due to the increasing costs of labor, environment protection, and similar [Qi et al., 2016].

Based on these findings, and considering current economic conditions and climate policies, we see little reason to expect China's exported CO₂ emissions to return to the growth rates seen before 2007. There are four arguments for this. First, after the global financial crisis in 2008–2009, the world economy recovered, but has grown at a slower rate than expected [United Nations Conference on Trade and Development, 2016]. Correspondingly, global trade has also slowed [International Monetary Fund, 2016b], and though its growth revived in 2016, global trade is still not as robust as that in the precrisis decades [Gurría and Mann, 2017]. The International Monetary Fund expects that the slow global trade will most likely persist due to a limited pickup in global activity and weak investment growth, and claims that even as the global trade grows faster eventually, it is unlikely to return to the precrisis rates [International Monetary Fund, 2016a]. Second, although China's share in global trade is growing [World Trade Organization, 2017], its export structure is becoming less emission-intensive. The increasing cost of productions is weakening China's competitiveness in the global market of resource- and labor-intensive products. As a result, China is moving to production with higher technology, efficiency and value-added. A policy released early 2016 focusing on processing trade also emphasized promoting China to a higher level in the global value chain [State Council of the People's Republic of China, 2016]. Third, China's dramatic economic growth is now decelerating and in the stage of transforming and rebalancing. Recently released policies have shown that China is taking this chance to transform onto a more efficient and ecologically balanced path [State Council of the People's Republic of China, 2015; Green and Stern, 2016]. For CO₂ emissions specifically, having over fulfilled its goal in the 12th FYP period (2011–2015), China set a goal of reducing its CO₂ emissions per unit of GDP by 18% in the following 13th FYP (2016–2020) [National Development and Reform Commission, 2016], further strengthening China's efforts to mitigate its CO₂ emissions. Fourth, the pressure of controlling local air pollution and the signing of the Paris Agreement [The United Nations, 2015] (and a series of joint statements on climate change with the U.S. and the European Union), as well as higher expectations of China's efforts on climate change due to the announced withdrawal of the U.S. from the Paris Agreement, are all reinforcing China's efforts to develop efficiently and sustainably.

Given the probable persistent slow growth in global trade and China's policies to clean its production, we argue that Chinese exported CO₂ emissions will grow slowly or stabilize. As an economy that accounts for a large share of global emission transfers between countries, China's emission transfers (net exported emissions) have remained flat since 2007. This indicates that the threat of climate policies such as BCAs, aimed at reducing “carbon leakage” from developing countries, especially China, will have significantly less relevance in the years ahead. This is further reinforced by the Paris Agreement where all countries contribute, and as ambition is increased across countries, the tendency for carbon leakage will be reduced.

There are concerns about potential increases in emissions exported from other developing countries such as India, whose exported emissions increased between 2007 and 2012. However, considering that these countries account for much smaller proportions in exported emissions from developing to developed countries than China, and the signing of the Paris Agreement as well as the rapid development of new energy, we see no reason to expect that emissions exported from other developing countries will cause a significant increase in the tendency of carbon leakage. India, for example, having proposed construction of coal-fired power in conflict with its climate commitments [Shearer et al., 2017], has since canceled 13.7 GW of planned coal power projects as solar energy is becoming cheaper [Kalra, 2017]. Another concern is the withdrawal of the U.S. from the Paris Agreement. However, since the Paris Agreement is a bottom-up process with pledges put forward by countries themselves, it is possible the withdrawal will have little negative effect on the contributions of other countries. Instead, the withdrawal could stimulate further efforts from regions and institutions within the U.S. as well as efforts from other countries, and inspire positive opportunities on climate change [Kemp, 2017]. Hence, we do not expect increases in the tendency for carbon leakage caused by the withdrawal of the U.S.. Going forward, we suggest less policy focus on emission transfers, and rather more focus on policies aiming to reduce emissions efficiently inside China.

Acknowledgments

C.P. was funded by the State Scholarship Fund of China, the Funding of Jiangsu Innovation Program for Graduate Education (no. KYZZ15_0096). C.P. (in part), S.L. (in part), G.P.P., R.M.A., and J.I.K. were funded by the Research Council of Norway (no. 235523), including a Visiting Researcher Grant to C.P. C.P. (in part), D.Z. (in part), P.Z. were funded by the National Natural Science Foundation of China (no. 71573121, 71,625,005,71,573,119). C.P. (in part), D.Z. (in part) were funded by the Fundamental Research Funds for the Central Universities (no. NP2017107). The data used in the analysis is publicly available, and can be found following the references. The data to generate the figures is provided in the Supporting Information, excluding those that are referenced.

Supporting Information

References

Ahuja, A., N. A. Chalk, M. Nabar, P. N'Diaye, and N. Porter (2012), An end to China's imbalances?, Working paper from International Monetary Fund.
Google Scholar
Ang, B. W., and K.-H. Choi (1997), Decomposition of aggregate energy and gas emission intensities for industry: A refined Divisia Index Method, Energy J., 18(3), 59– 73.
CrossrefWeb of Science®Google Scholar
Ang, B. W., and F. L. Liu (2001), A new energy decomposition method: Perfect in decomposition and consistent in aggregation, Energy, 26(6), 537– 548, https://doi.10.1016/S0360-5442(01)00022-6.
CrossrefCASWeb of Science®Google Scholar
Ang, B. W., F. Q. Zhang, and K.-H. Choi (1998), Factorizing changes in energy and environmental indicators through decomposition, Energy, 23(6), 489– 495, https://doi.10.1016/S0360-5442(98)00016-4.
CrossrefWeb of Science®Google Scholar
Ang, B. W., F. L. Liu, and E. P. Chew (2003), Perfect decomposition techniques in energy and environmental analysis, Energy Policy, 31(14), 1561– 1566, https://doi.10.1016/S0301-4215(02)00206-9.
CrossrefWeb of Science®Google Scholar
Böhringer, C., E. J. Balistreri, and T. F. Rutherford (2012), The role of border carbon adjustment in unilateral climate policy: Overview of an energy modeling forum study (EMF 29), Energy Econ., 34, S97– S110, https://doi.10.1016/j.eneco.2012.10.003.
CrossrefWeb of Science®Google Scholar
Chen, G. Q., and B. Zhang (2010), Greenhouse gas emissions in China 2007: Inventory and input–output analysis, Energy Policy, 38(10), 6180– 6193, https://doi.10.1016/j.enpol.2010.06.004.
CrossrefCASWeb of Science®Google Scholar
China Cement Association (2015), China Cement Almanac (2014), China Cement Assoc., Beijing.
Google Scholar
China National Development and Reform Commission (2014), The People's Republic of China National Greenhouse Gas Inventory 2005, China Environ. Sci. Press, Beijing.
Google Scholar
Chinadialogue (2016), Climate, energy and China's 13th Five-Year Plan in Graphics, Chinadialogue, London.
Google Scholar
Cosbey, A., S. Droege, C. Fischer, J. Reinaud, J. Stephenson, L. Weischer, and P. Wooders (2012), A guide for the concerned: Guidance on the elaboration and implementation of border carbon adjustment, SSRN Electron. J., https://doi.10.2139/ssrn.2178312.
CrossrefWeb of Science®Google Scholar
Davis, S. J., and K. Caldeira (2010), Consumption-based accounting of CO₂ emissions, Proc. Natl. Acad. Sci. U. S. A., 107(12), 5687– 5692, https://doi.10.1073/pnas.0906974107.
CrossrefCASADSPubMedWeb of Science®Google Scholar
Davis, S. J., G. P. Peters, and K. Caldeira (2011), The supply chain of CO₂ emissions, Proc. Natl. Acad. Sci. U. S. A., 108(45), 18554– 18559, https://doi.10.1073/pnas.1107409108.
CrossrefCASADSPubMedWeb of Science®Google Scholar
Department of Economic and Social Affairs (2016), United Nations Statistics Division, Department of Economic and Social Affairs, New York, N. Y.
Google Scholar
Dietzenbacher, E., and A. R. Hoen (1999), Double deflation and aggregation, Environ. Plan. A, 31(9), 1695– 1704, https://doi.10.1068/a311695.
CrossrefWeb of Science®Google Scholar
Dietzenbacher, E., and B. Los (1998), Structural decomposition techniques: Sense and sensitivity, Econ. Syst. Res., 10(4), 307– 324, https://doi.10.1080/09535319800000023.
CrossrefGoogle Scholar
Dietzenbacher, E., and B. Los (2000), Structural decomposition analyses with dependent determinants, Econ. Syst. Res., 12(4), 497– 514, https://doi.10.1080/09535310020003793.
CrossrefGoogle Scholar
Dietzenbacher, E., M. L. Lahr, and B. Los (2004), The decline in labor compensation's share of GDP: A structural decomposition analysis for the United States, 1982 to 1997, Wassily Leontief Input-Output Econ., 138-185, https://doi.10.1017/CBO9780511493522.013.
CrossrefGoogle Scholar
Dietzenbacher, E., J. Pei, and C. Yang (2012), Trade, production fragmentation, and China's carbon dioxide emissions, J. Environ. Econ. Manage., 64(1), 88– 101, https://doi.10.1016/j.jeem.2011.12.003.
CrossrefWeb of Science®Google Scholar
Du, Y. (2011), Qian Zhimin: China will Actively Build Low-Carbon Demonstration Towns, and Green and Low-Carbon Industrial System (), People – Energy ().
Google Scholar
Eisenbarth, S. (2017), Is Chinese trade policy motivated by environmental concerns? J. Environ. Econ. Manage., 82, 74– 103, https://doi.10.1016/j.jeem.2016.10.001.
CrossrefWeb of Science®Google Scholar
Eurostat (2013), Glossary: Chain index, [Available at: http://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:Chain_index].
Google Scholar
Feng, K., Y. L. Siu, D. Guan, and K. Hubacek (2012), Analyzing drivers of regional carbon dioxide emissions for China, J. Ind. Ecol., 16(4), 600– 611, https://doi.10.1111/j.1530-9290.2012.00494.x.
Wiley Online LibraryCASWeb of Science®Google Scholar
Gourdon, J., S. Monjon, and S. Poncet (2016), Trade policy and industrial policy in China: What motivates public authorities to apply restrictions on exports? China Econ. Rev., 40, 105– 120, https://doi.10.1016/j.chieco.2016.06.004.
CrossrefWeb of Science®Google Scholar
Grasso, M., and T. Roberts (2014), A compromise to break the climate impasse, Nat. Clim. Change., 4(7), 543– 549, https://doi.10.1038/nclimate2259.
CrossrefCASADSWeb of Science®Google Scholar
Green, F., and N. Stern (2016), China's Changing economy: Implications for its carbon dioxide emissions, Clim. Policy, 17(4), 423– 442, https://doi.10.1080/14693062.2016.1156515.
CrossrefWeb of Science®Google Scholar
Guan, D., G. P. Peters, C. L. Weber, and K. Hubacek (2009), Journey to world top emitter: An analysis of the driving forces of China's recent CO₂emissions surge, Geophys. Res. Lett., 36(4), https://doi.10.1029/2008gl036540.
Wiley Online LibraryWeb of Science®Google Scholar
Guan, D., Z. Liu, Y. Geng, S. Lindner, and K. Hubacek (2012), The gigatonne gap in China's carbon dioxide inventories, Nat. Clim. Change, 2(9), 672– 675, https://doi.10.1038/nclimate1560.
CrossrefCASADSWeb of Science®Google Scholar
Gurría, A., and C. L. Mann (2017), OECD Economic Outlook, June 2017: Better, But Not Good Enough, Organization for Economic Co-operation and Development, Paris.
Google Scholar
Hoekstra, R., and J. C. J. M. van denBergh (2003), Comparing structural decomposition analysis and index, Energy Econ., 25(1), 39– 64, https://doi.10.1016/S0140-9883(02)00059-2.
CrossrefWeb of Science®Google Scholar
International Monetary Fund (2016a), World Economic Outlook, October 2016 : Subdued Demand: Symptoms and Remedies, International Monetary Fund, Washington, D. C.
Google Scholar
International Monetary Fund (2016b), World Economic Outlook, April 2016 : Too Slow for Too Long, International Monetary Fund, Washington, D. C.
CrossrefGoogle Scholar
IPCC (2006), 2006 IPCC Guidelines for National Greenhouse Gas Inventories, IGES, Kanagawa, Jpn.
Google Scholar
Jakob, M., and R. Marschinski (2012), Interpreting trade-related CO₂ emission transfers, Nat. Clim. Change, 3(1), 19– 23, https://doi.10.1038/nclimate1630.
CrossrefADSWeb of Science®Google Scholar
Kalra, S. (2017), India cancels mega plans to build coal power stations due to falling solar energy prices, Indiatimes. [Available at: http://www.indiatimes.com/news/india/india-cancels-mega-plans-to-build-coal-power-stations-due-to-falling-solar-energy-prices-322553.html].
Google Scholar
Kanemoto, K., D. Moran, M. Lenzen, and A. Geschke (2014), International trade undermines national emission reduction targets: New evidence from air pollution, Global Environ. Change., 24, 52– 59.
CrossrefWeb of Science®Google Scholar
Kemp, L. (2017), Better out than in, Nat. Clim. Change, 7, 458– 460, https://doi.10.1038/nclimate3309.
CrossrefADSWeb of Science®Google Scholar
Korsbakken, J. I., G. P. Peters, and R. M. Andrew (2016), Uncertainties around reductions in China's coal use and CO₂ emissions, Nat. Clim. Change, 6(7), 687– 690, https://doi.10.1038/nclimate2963.
CrossrefCASADSWeb of Science®Google Scholar
Le Quéré, C., et al. (2015), Global carbon budget 2015, Earth Syst. Sci. Data, 7(2), 349– 396, https://doi.10.5194/essd-7-349-2015.
CrossrefADSWeb of Science®Google Scholar
Liu, Z., S. J. Davis, K. Feng, K. Hubacek, S. Liang, L. D. Anadon, B. Chen, J. Liu, J. Yan, and D. Guan (2015a), Targeted opportunities to address the climate–trade dilemma in China, Nat. Clim. Change, 6(2016), 201– 206, https://doi.10.1038/nclimate2800.
CrossrefADSPubMedWeb of Science®Google Scholar
Liu, Z., et al. (2015b), Reduced carbon emission estimates from fossil fuel combustion and cement production in China, Nature, 524(7565), 335– 338, https://doi.10.1038/nature14677.
CrossrefCASADSPubMedWeb of Science®Google Scholar
Meng, B., Z. Wang, and R. Koopman (2013), How are global value chains fragmented and extended in China's domestic production networks? IDE Discuss. Paper 424.
Google Scholar
Miller, R. E., and P. D. Blair (2009), Input–Output Analysis: Foundations and Extensions, Cambridge, U. K, Cambridge University Press.
CrossrefGoogle Scholar
Minx, J. C., G. Baiocchi, G. P. Peters, C. L. Weber, D. Guan, and K. Hubacek (2011), A "carbonizing dragon": China's Fast growing CO₂ emissions revisited, Environ. Sci. Technol., 45(21), 9144– 9153, https://doi.10.1021/es201497m.
CrossrefCASADSPubMedWeb of Science®Google Scholar
Munksgaard, J., and K. A. Pedersen (2001), CO₂ accounts for open economies: Producer or consumer responsibility? Energy Policy, 29(4), 327– 334, https://doi.10.1016/s0301-4215(00)00120-8.
CrossrefWeb of Science®Google Scholar
National Bureau of Statistics of China (1998a), China Industry Economy Statistical Yearbook (1998), China Stat. Press, Beijing.
Google Scholar
National Bureau of Statistics of China (1998b), China Statistical Yearbook (1998∼2013), China Stat. Press, Beijing.
Google Scholar
National Bureau of Statistics of China (1999), Input–Output Tables of China (1997), China Stat. Press, Beijing.
Google Scholar
National Bureau of Statistics of China (2006a), Input–Output Tables of China (2002), China Statistics Press, Beijing.
Google Scholar
National Bureau of Statistics of China (2006b), Handbook of Energy Statistics Knowledge, China Stat. Press, Beijing.
Google Scholar
National Bureau of Statistics of China (2009), Input–Output Tables of China (2007), China Stat. Press, Beijing.
Google Scholar
National Bureau of Statistics of China (2010), China Energy Statistical Yearbook (2009), China Stat. Press, Beijing.
Google Scholar
National Bureau of Statistics of China (2013), China Price Statistical Yearbook (2013), China Stat. Press, Beijing.
Google Scholar
National Bureau of Statistics of China (2015a), Input–Output Tables of China (2012), China Stat. Press, Beijing.
Google Scholar
National Bureau of Statistics of China (2015b), China Energy Statistical Yearbook (2014), China Stat. Press, Beijing.
Google Scholar
National Bureau of Statistics of China (2016), National data. [Available at: http://data.stats.gov.cn/english/easyquery.htm?cn=C01.]
Google Scholar
National Development and Reform Commission (2016), The 13th five-year plan for economic and social development of the people's republic of China ("), National Development and Reform Commission, Beijing.
Google Scholar
Office of National Climate Change Coordination Committee, and National Development and Reform Commission (2007), The People's Republic of China National Greenhouse Gas Inventory, China Environ. Sci. Press, Beijing.
Google Scholar
Owen, A., K. Steen-Olsen, J. Barrett, T. Wiedmann, and M. Lenzen (2014), A structural decomposition approach to Comparing MRIO databases, Econ. Syst. Res., 26(3), 262– 283, https://doi.10.1080/09535314.2014.935299.
CrossrefWeb of Science®Google Scholar
Pan, J., J. Phillips, and Y. Chen (2008), China's balance of emissions embodied in trade: Approaches to measurement and allocating international responsibility, Oxf. Rev. Econ. Policy, 24(2), 354– 376, https://doi.10.1093/oxrep/grn016.
CrossrefCASWeb of Science®Google Scholar
Pauwelyn, J. (2007), U.S. Federal climate policy and competitiveness concerns: The limits and options of International Trade Law, Nicholas Inst. for Environ. Policy Sol. WP 07-02, https://doi.10.2139/ssrn.1669336
Google Scholar
Peters, G. P. (2014), Policy update: Managing carbon leakage, Carbon Manage., 1(1), 35– 37, https://doi.10.4155/cmt.10.1.
CrossrefWeb of Science®Google Scholar
Peters, G. P., C. L. Weber, and J. Liu (2006), Construction of Chinese Energy and Emissions Inventory. Rep. and Working Papers from NTNU, 4/2006.
Google Scholar
Peters, G. P., C. L. Weber, D. Guan, and K. Hubacek (2007), China's Growing CO₂ emissions a race between increasing consumption and efficiency gains, Environ. Sci. Technol., 41(17), 5939– 5944.
CrossrefCASADSPubMedWeb of Science®Google Scholar
Peters, G. P., J. C. Minx, C. L. Weber, and O. Edenhofer (2011), Growth in emission transfers via international trade from 1990 to 2008, Proc. Natl. Acad. Sci. U. S. A, 108(21), 8903– 8908, https://doi.10.1073/pnas.1006388108.
CrossrefCASADSPubMedWeb of Science®Google Scholar
Qi, Y., H. Li, and T. Wu (2013), Interpreting China's carbon flows, Proc. Natl. Acad. Sci. U. S. A, 110(28), 11221– 11222, https://doi.10.1073/pnas.1309470110.
CrossrefCASADSPubMedWeb of Science®Google Scholar
Qi, Y., N. Stern, T. Wu, J. Lu, and F. Green (2016), China's Post-coal growth, Nat. Geosci., 9(8), 564– 566, https://doi.10.1038/ngeo2777.
CrossrefCASADSWeb of Science®Google Scholar
Rørmose, P., and T. Olsen (2005), Structural decomposition analysis of air emissions in Denmark 1980–2002. paper presented at 15th International Conference on Input-Output Techniques, Beijing, China.
Google Scholar
Shearer, C., R. Fofrich, and S. J. Davis (2017), Future CO₂ emissions and electricity generation from proposed coal-fired power plants in India, Earth's Future, 5(4), 408– 416, https://doi.10.1002/2017EF000542.
Wiley Online LibraryCASADSWeb of Science®Google Scholar
Song, P., X. Mao, and G. Corsetti (2015), Adjusting export tax rebates to reduce the environmental impacts of trade: Lessons from China, J. Environ. Manage., 161, 408– 416, https://doi.10.1016/j.jenvman.2015.07.029.
CrossrefPubMedWeb of Science®Google Scholar
Springmann, M. (2014), Integrating emissions transfers into policy-making, Nat. Clim. Change, 4(3), 177– 181, https://doi.10.1038/nclimate2102.
CrossrefADSWeb of Science®Google Scholar
State Administration of Taxation (2017), Taxation Regulations. [Available at: http://www.chinatax.gov.cn/].
Google Scholar
State Council of the People's Republic of China (2002), Reform Programme of Electricity System (), State Council of the People's Republic of China, Beijing.
Google Scholar
State Council of the People's Republic of China (2007), Comprehensive Working Plan for Energy Saving and Emission Reduction (), State Council of the People's Republic of China, Beijing.
Google Scholar
State Council of the People's Republic of China (2011), Comprehensive Working Plan for Energy Saving and Emission Reduction of the 12th Five-Year (``), State Council of the People's Republic of China, Beijing.
Google Scholar
State Council of the People's Republic of China (2015), Opinions about Accelerating the Construction of Ecological Civilization (), State Council of the People's Republic of China, Beijing.
Google Scholar
State Council of the People's Republic of China (2016), Several Opinions about Promoting the Innovation and Development of Processing trade (), State Council of the People's Republic of China, Beijing.
Google Scholar
Su, B., and B. W. Ang (2010), Input–output analysis of CO₂ emissions embodied in trade: The effects of spatial aggregation, Ecol. Econ., 70(1), 10– 18, https://doi.10.1016/j.ecolecon.2010.08.016.
CrossrefWeb of Science®Google Scholar
Su, B., and B. W. Ang (2012), Structural decomposition analysis applied to energy and emissions: Some methodological developments, Energy Econ., 34(1), 177– 188, https://doi.10.1016/j.eneco.2011.10.009.
CrossrefWeb of Science®Google Scholar
Su, B., B. W. Ang, and M. Low (2013), Input–output analysis of CO₂ emissions embodied in trade and the driving forces: Processing and normal exports, Ecol. Econ., 88, 119– 125, https://doi.10.1016/j.ecolecon.2013.01.017.
CrossrefWeb of Science®Google Scholar
Teng, F. (2015), Carbon: Resolve ambiguities in China's emissions, Nature, 525(7570), 455– 455, https://doi.10.1038/525455c.
CrossrefCASADSPubMedWeb of Science®Google Scholar
Teng, F., and S. Zhu (2015), Which estimation is more accurate? A technical comments on nature paper by Liu et al on overestimation of China's emission ( Nature CO₂), Science & Technology Review (), 22, 112– 116.
Google Scholar
The United Nations (2015), The Paris Agreement, The United Nations, New York, N. Y.
Google Scholar
United Nations (1999). Handbook of input-output table compilation and analysis. Studies in Methods Series F, No 74. Handbook of National Accounting, United Nations.
Google Scholar
United Nations Conference on Trade and Development (2016), Structural Transformation for Inclusive and Sustained Growth, United Nations Conference on Trade and Development, Geneva, Switz.
Google Scholar
World Trade Organization (2017), World Trade Organization Statistics Database, World Trade Organization, Geneva, Switz.
Google Scholar
Xu, M., R. Li, J. C. Crittenden, and Y. Chen (2011), CO₂ emissions embodied in China's exports from 2002 to 2008: A structural decomposition analysis, Energy Policy, 39(11), 7381– 7388, https://doi.10.1016/j.enpol.2011.08.068.
CrossrefCASWeb of Science®Google Scholar

Citing Literature

Volume5, Issue9

September 2017

Pages 934-946

Filename	Description
EFT2250-sup-0001-2017EF000625-SI.pdfPDF document, 1 MB	Supporting Information S1
EFT2250-sup-0002-2017EF000625-ts01.xlsExcel spreadsheet, 38 KB	Table S1
EFT2250-sup-0003-2017EF000625-ds01.xlsExcel spreadsheet, 28.5 KB	Data Set S1
EFT2250-sup-0004-2017EF000625-ds02.xlsExcel spreadsheet, 43.5 KB	Data Set S2
EFT2250-sup-0005-2017EF000625-ds03.xlsExcel spreadsheet, 26.5 KB	Data Set S3
EFT2250-sup-0006-2017EF000625-ds04.xlsExcel spreadsheet, 30.5 KB	Data Set S4
EFT2250-sup-0007-2017EF000625-ds05.xlsExcel spreadsheet, 69 KB	Data Set S5
EFT2250-sup-0008-2017EF000625-ds06.xlsExcel spreadsheet, 27 KB	Data Set S6
EFT2250-sup-0009-2017EF000625-ds07.xlsExcel spreadsheet, 28.5 KB	Data Set S7
EFT2250-sup-0010-2017EF000625-ds08.xlsExcel spreadsheet, 44 KB	Data Set S8
EFT2250-sup-0011-2017EF000625-ds09.xlsExcel spreadsheet, 26 KB	Data Set S9
EFT2250-sup-0012-2017EF000625-ds10.xlsExcel spreadsheet, 27 KB	Data Set S10

Emissions embodied in global trade have plateaued due to structural changes in China