Emissions embodied in bilateral trade between Canada and China

(1)

Emissions embodied in bilateral trade between

Canada and China

(2)

2

Abstract

Carbon emissions embodied in Canada-China trade are analysed during 1990-2014. First, input-output analysis is used to calculate the emissions embodied in trade, second, structural decomposition analysis is applied to identify the driving forces of embodied emissions. The results show that China is a net exporter of emissions in the China-Canada trade and the sector inducing the most emissions is electrical and machinery in both countries. The study also reveals that carbon emissions coefficients is the main driving force of the decrease of embodied emissions while trade scale is the main driver of the increase of embodied emissions.

(3)

3

1. Introduction

“Humans are influencing the climate and the earth's temperature by burning fossil fuels, cutting down rainforests and farming livestock”. These activities increase the amounts of greenhouse gases (GHG) naturally being in the atmosphere, which causes the greenhouse effect and climate change. Carbon dioxide (CO2) is the most produced GHG by human activities, and it is

responsible for 64% of man-made climate change. Moreover, its concentration now in the atmosphere is 40% higher than it was at the start of industrialisation (European Commission, 2019).

Due to the consequences of global climate change carbon reduction has become an urgent necessity in recent years (Stern, 2007). Considering the large amount of carbon emissions induced by demand abroad, quantification and analysis of carbon emissions embodied in trade has become an important part of academic research to address climate change (Sato, 2014).

Over the last twenty years, nearly 25% of all carbon dioxide emissions were related to the production activities of international trade (Peters et al., 2011). As the geographic distance between consumption and production extended, most developed economies became net carbon importers while emerging and resource-abundant countries became net carbon exporters of carbon emissions (Kanemoto et al., 2014).

One of the theories discussing this issue is the Pollution Haven Hypothesis (PHH). Dietzenbacher and Mukhopadhyay (2007) explained PHH as a result of the Heckscher-Ohlin (HO) theory. HO has been argued to include natural resources as a third factor (Leamer, 1980; Bowen et al., 1987). When pollution is restricted, then ‘‘emission permits’’ might play a role as a third factor. Developing countries are usually relatively abundant in emission permits and thus, based on HO theory, they have a comparative advantage in relatively emission intensive goods. As a result, trade will worsen existing environmental problems in the developing countries with relatively low regulations. In other words, developed countries have stricter pollution regulations or restrictions on greenhouse gas emissions than developing countries, thus developing countries will export ‘‘dirty’’ products and import ‘‘clean’’ products.

Carbon dioxide (CO2) emissions increased globally from 22.2 billion tons in 1990 to 36.1

(5)

5

According to a report of the Energy Information Administration, China overtook the USA as the biggest CO2 emitter in the world in 2009 and contributes 23% of total global carbon

emissions (Liu et al., 2017).

The value of CO2 emissions for China in 2014 has reached 10.3 billion tons (World Bank,

2019). From the international trade perspective, China has also become the world's largest importer and exporter, with a 47% dependence on foreign trade in 2012 (Liu et al., 2017). In recent years, tremendous number of researches have been carried out on embodied emissions. The leading research methodology is clearly the input-output analysis which was developed by Leontief (1936, 1941). Initially this method was applied to study the relationship between inputs and outputs in the field of economics but in the 1960s some researches also started to use this model in the environmental and energy fields (Leontief, 1970).

Since then, input-output analysis has been used in many different researches to calculate emissions embodied in trade, especially the biggest carbon emitter, China (Liu, 2017; Su, 2011). To further investigate CO2 emissions embodied in trade, several researches conducted

decomposition analysis (Dong, 2010; Su, 2012) or hypothetical scenario analysis (Yu and Chen, 2017).

Wu et al. (2016) also introduced the concept of dependence on traded CO2 which can reveal on

what extent an economy’s output is dependent on CO2 emissions abroad through importing

intermediate products but can also take into consideration both CO2 emission and economy.

(6)

6

Figure 1: Value of bilateral exports between Canada and China.

Source: UN Comtrade.

In 2005 the Chinese export to Canada is 11.65 billion US$ while in 2014 is about 30.00 billion US$. Canadian export to China is 5.96 US$ in 2005 and 17.45 US$ in 2014 which also means that Canada has a trade deficit with China in more recent years.

The development of bilateral trade between them also lead to the accelerated transport of productions and consumption of products which resulted in the redistribution of carbon emissions which makes a study on the relationship of the two countries necessary.

Furthermore, Canada’s carbon emission per capita is one of the highest in the world. Figure 2 shows the per capita carbon emissions in six OECD countries where Canada has the highest values together with the United States and Australia. Moreover, Canada’s emissions per capita have been rising in the last two years while the other countries’ emissions are decreasing. Moreover, since 2005, Canada's is a net importer of carbon emissions that can be attributed to an increase in imports from developing countries (China being the largest), that produce goods using a mix of energy sources that are more emissions-intensive than Canada's (Environment and Climate Change Canada, 2017). This thesis also aims to analyse if Canada has been a net importer of emissions against China since 2005 and further investigate what are the driving forces of embodied emissions.

0 5 10 15 20 25 30 35 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 b ill ion US$

(7)

7

Figure 2: Emissions per capita in 6 OECD countries between 1991 and 2014.

Source: World Bank.

Therefore, the research question of the paper is how much carbon emissions are embodied in the bilateral trade between Canada and China. This paper aims to fill a research gap by observing not only the quantity but also the structure changes of CO2 emissions embodied in

bilateral between China and Canada using Emissions Embodied in Bilateral Trade (EEBT) approach. In order to uncover the driving forces inducing changes in emissions, structural decomposition analysis is used to study the impacts of four effects, namely, carbon emissions coefficients (e), technology, trade structure and trade scale on embodied carbon emissions (Yu and Chen, 2017).

The remainder of the paper is organized as follows. Section 2 gives an overview about the China-Canada trade and how their relationship effects the environment. The literature review about carbon emissions embodied in trade is presented in Section 3. Section 4 describes the methodology used to calculate and analyse the embodied emissions in the trade of Canada and China. Section 5 introduces the data sources. Section 6 presents the further analysis and discussion on the results which includes input-output analysis and structural decomposition analysis. Finally, conclusions are given in Section 7.

(8)

8

2. China-Canada relations

The following chapter, first, gives a more detailed overview about the bilateral trade between China and Canada by discussing the main export products between the countries. Second, further opportunities about environmental and energy possibilities are described.

2.1 China-Canada trade

In 2016, Canada and China announced that they were initiating a cooperative process to explore the possibility of negotiating a free trade agreement (FTA) between the two countries (Blanchfield, 2016).

After the United States, China is Canada’s second most important trade partner. In 2017, China accounted for 4% and 13% of Canada’s product exports and imports, respectively. In comparison, it only accounted for 1% and 2% of the country’s product exports and imports in 1995, respectively (UN Comtrade, 2019).

Canada’s trade with China has experienced rapid growth since the beginning of the 2000s, exports to China tripled as seen in Figure 1, reaching $21 billion in 2016. During the same period, Canadian product imports from China more than doubled, reaching $64 billion, thereby slightly surpassing Canadian imports from the European Union.

We can see the top export products from Canada to China in 2017 in Figure 3. The majority of products are coming from two sectors: agriculture and natural resources. From the top-10 categories of Canadian export products to China, only the following three were not based on natural resources: nuclear reactors, boilers, machinery; organic chemicals; vehicles other than railway, tramway.

(9)

9

Figure 3: Top 10 export products from Canada to China in 2017.

Figure 4: Top 10 export products from China to Canada in 2017.

This represents the activity of electronic and electrical equipment manufacturers to move their final product assembly to China over the last two decades in order to take advantage of the China’s cheaper labour, even if a large part of the end products’ added value is exported to China in the form of higher-value intermediate inputs (Van Assche, 2012). Aside from nuclear reactors, boilers, machinery; vehicles other than railway, tramway and articles of iron or steel China’s top product exports to Canada are traditional in nature, including furniture, clothing, apparel and footwear as well as toys, games and sports products.

0 0,5 1 1,5 2 2,5 3 3,5

Oil seed, oleagic fruits, grain, seed, fruit, etc Pulp of wood, fibrous cellulosic material, waste etc Wood and articles of wood, wood charcoal Vehicles other than railway, tramway Ores, slag and ash Mineral fuels, oils, distillation products, etc Fish, crustaceans, molluscs, aquatic invertebrates Organic chemicals Animal,vegetable fats and oils, cleavage products Nuclear reactors, boilers, machinery, etc

billion US$

0 1 2 3 4 5 6

Electrical, electronic equipment Nuclear reactors, boilers, machinery, etc Furniture, lighting, signs, prefabricated buildings Toys, games, sports requisites Articles of apparel, accessories, not knit or crochet Plastics and articles thereof Articles of iron or steel Articles of apparel, accessories, knit or crochet Vehicles other than railway, tramway Footwear, gaiters and the like, parts thereof

(10)

10

2.2 Energy and environmental opportunities

China’s demand for energy and natural resources will continue to increase significantly in the coming years, even if the rate of growth will be slower than before as China’s economy grows at rates that average 5-6% rather than 8-10%. For instance, McKinsey & Company (2016) estimates that “China will account for over 40 percent of the global growth in energy demand by 2030.” This would explain why Chinese officials and experts often mention agriculture and natural resources as “objects of desire” (Mazereeuw, 2016).

(11)

11

3. Literature Review

This section presents a review of recent literature on analysing the emissions embodied in international trade. Several methods are reported in the literature to address this issue so I will focus on the findings of recent papers that used the same methodology, namely input-output analysis and decomposition analysis, by highlighting the results and emphasizing the gaps. Over time, an extensive literature has developed on environmental input-output analysis and China being the largest carbon emitter in the world has drawn the attention of many researchers in the past decade. Emissions embodied in the trade of China has been discussed by a great number of authors in literature, especially with OECD countries, for example, the USA, South-Korea, Australia or Japan (see for example: Dong et al.,2010; Pan et al., 2008; Weber et al., 2008; Xu et al., 2011; Su et al., 2013; Su and Ang, 2013; Su and Ang, 2014; Jiang et al., 2015; Zhang and Chen, 2010; Zhang et al., 2014; Zhang et al., 2015).

For instance, research has provided evidence for China being a net exporter of carbon emissions in the China-Japan trade during 1990-2000. Liu et al. (2010) calculated the CO2 emissions

embodied in bilateral trade between the two countries by using a two-step input-output analysis. First, they conducted vectors of embodied carbon emissions coefficients at sector levels for both countries. Second, they multiplied these carbon emission coefficients by the corresponding volume of exports and summed them up to get the total carbon emissions embodied in China-Japan trade. They found that carbon emissions embodied exported from China-Japan to China increased during 1990-2000 due to the increase of export volume. Exports from China to Japan also increased gradually during the same period, but the carbon emissions embodied in export only increased in the first half of the period, and then decreased in the second half of the period. Dong et al. (2010) also confirmed that the carbon emissions embodied in the exports from China to Japan were much larger than the “reverse flow of emissions” in this period. Therefore, this leads to the first two hypotheses:

Hypothesis 1: In the bilateral trade between Canada and China, China is net exporter of carbon emissions.

(12)

12

A recent study by Liu et al. (2017), however, concluded that there are huge differences in the estimations of embodied carbon emissions in the trade of China. They identified three aspects that can lead to these differences in the calculations and databases used in the literature. First, most studies have been based on competitive input–output tables instead of non-competitive tables (e.g. Pan et al., 2008). Second, many studies apply the simple assumption of emissions avoided by imports. This means that these studies assume that China's imports and exports use the same technology, by identically adopting the country's carbon intensity index (e.g. Yan et al., 2012; Chang, 2013). Finally, majority of papers have neglected the impact of processing trade on the estimation of embodied emissions, assuming that all firms use the same technology to produce goods and services.

Based on different assumption, output analysis has different types: single-regional input-output (SRIO) model and multi-regional input-input-output (MRIO) model. Moreover, Peters’ (2008) suggested treating MRIO as two approaches, namely, the emissions embodied in bilateral trade (EEBT) approach and MRIO approach. (Further details of the approach are discussed in Section 4.)

Yu and Chen (2017) found using EEBT approach that China is a net exporter of emissions in the trade between South-Korea and China and that trade surplus was not the reason for the embodied carbon emission surplus. Wu et al. (2016) using the same approach also confirmed China being a net exporter of carbon emissions.

Su and Ang (2014) applied a hybrid method of EEBT and MRIO to study the emissions embodied in Chinese trade. They used EEBT at the national level and MRIO at the regional level in order to increase transparency at the national level and the interregional feedback effect at the regional level. They also found evidence that the developed countries are net importers while developing economies are net exporters of embodied emissions from interregional and international bilateral trade.

(13)

13

Over time, an extensive literature has developed on the driving factors of carbon emissions.

Structural decomposition analysis (SDA), which is based on the environmental input–output model, is traditionally used to study the observed changes in the level and mix of output (Rose and Casler, 1996), but has also been used to identify the driving factors of carbon emissions embodied in trade (Zhao et al, 2016).

Historically, there has been a great deal of confusion in the literature regarding the driving forces of emissions embodied. For example, Feng et al. (2012) studied the changes in population, technology, economic structure, urbanization and household consumption patterns on regional carbon emissions in China. They found that urbanization and associated income and lifestyle changes were the most important driving factors for the increase of carbon emissions in most regions.

Su and Ang (2015) found that the emissions intensity effect is estimated to contribute the most to the decrease in the aggregate emission intensity. On the other hand, Liu and Liang (2017) suggested that the main driver of decreasing of emissions was technical effects.

Zhao et al. (2016) also used SDA to investigate the driving forces of carbon emissions embodied in China–US trade during 1995–2009. They found that two factors “trade structure of intermediate products at home” and “export market shares of final products at home” presented the largest positive impacts to the increase in carbon emissions embodied in Chinese exports to the US. While the decrease was generated by changes in “energy intensities at home.” The increase in carbon emissions embodied in US exports to China was mostly contributed by “total demands abroad”. Impacts of other driving factors were much smaller.

(14)

14

Hypothesis 3: Trade scale and trade structure contribute to the increase of emissions embodied in bilateral trade between China and Canada.

Hypothesis 4: Technology and emission intensity contribute to the decrease of emissions embodied in bilateral trade between China and Canada.

From the short literature review above, some key findings emerge: China is a net exporter of carbon emissions to OECD countries, although the results differ greatly based on the methodology used. The driving forces of carbon emissions embodied are studied by applying decomposition analysis. Some papers focused on the drives of CO2 embodied in bilateral trade

of China and found that trade structure, technology, final demand and energy intensity all contribute to the changes to embodied emissions exported from one country to another.

(15)

15

4. Methodology

In this section I describe the methodology used in the analysis in order to have a better understanding of the CO2 emissions embodied in Canada-China trade between 1990 and 2014.

My model is based on two studies: Yu and Chen (2017) investigated the emissions embodied in the bilateral trade between China and South Korea during 2000-2010 by using input-output analysis, hypothetical scenario analysis and structural decomposition analysis while Wu et al. (2016) estimated the CO2 emission embodied in China-Japan trade between 2000 and 2010 by

using the emissions embodied in bilateral trade approach, the dependence on traded CO2

concept and index decomposition analysis.

In my thesis, I combine the methodology of the two papers mentioned above. Therefore, first, I use input-output analysis to calculate the total emissions embodied in bilateral trade as well as the emissions embodied by sectors to find which country is the net exporter of emissions and which sector is the most emissions inducing sector. Then, I apply structural decomposition analysis to calculate the contribution of carbon emissions coefficients, intermediate technology, trade scale and trade structure to the emissions embodied in trade.

4.1 Input-output analysis

Input–output analysis, a useful analytical framework developed by Wassily Leontief (1936), has also been used to describe and analyse economy–environment relationships.

Table 2: World Input-Output Table

Intermediate demands Final demands Total

outputs

Region R Region S RoW Region R Region S RoW

Inter-mediate inputs Region R ZRR ZRS … FDRR FDRS … TR Region S ZSR ZSS … FDSR FDSS … TS RoW … … … … Value added VR VS … … … … … Total inputs (TR)T (TS)T … … … … …

(16)

16

The n x n matrix Z gives the intermediate deliveries. The matrix ZRR gives domestic intermediate

input in order to meet region R’s intermediate demands, while ZSR represents the intermediate

input of region S in order to meet the intermediate demand of region R.

The matrix n x k FD gives the deliveries to the final demand categories. Matrix FDRR represent

domestic final demand and FDRS represent final demands of region S from region R.

The column vector T represent total output while row vector V represent value added.

Based on different assumptions, we can differentiate two types of input-output analysis: single-regional input-output (SRIO) model (Dietzenbacher and Mukhopadhyah, 2007) and multi-regional input output (MRIO) model (Moran et al, 2013).

However, Peters et al. (2008) also differentiated two different approaches to model emissions embodied in trade at national level. The first approach determines the domestic carbon emissions in each country to produce the bilateral trade with another country. This method is the more transparent but does not assess the imports required to produce the bilateral trade. This is known as the emissions embodied in bilateral trade (EEBT) approach. The EEBT approach only considers total direct trade between two regions including intermediate and final products (Yu and Chen, 2017).

The second, more complex approach is also the MRIO model which determines the global emissions for an exogenous final consumption with global trade determined endogenously. Both methodologies give the same global emissions, but the national emissions differ in the method of allocating intermediate consumption (Peters et al., 2008).

In this paper I employ the simplified version as I focus on two countries and the bilateral trade between two countries: China and Canada so it is more suitable and transparent than MRIO for quantitative analysis.

(17)

17

In the input–output model, the total output of an economy, 𝑋, can be expressed as the sum of intermediate consumption, 𝐴𝑋, and final demand, 𝐹𝐷 (Leontief, 1970):

𝑋 = 𝐴𝑋 + 𝐹𝐷

where 𝑋 is the total output vector, 𝐹𝐷 is the final demand vector, and 𝐴 is the direct input coefficients matrix, describing the relationship between all sectors of the economy. 𝐴𝑋 denotes the intermediate input vector which can be obtained by multiplying the direct input coefficient matrix by the total output vector. And we can get

𝑀 = (𝐼 − 𝐴)−1

where 𝑀 is known as the Leontief inverse matrix and shows the requirement of total production for per unit of final consumption and 𝐼 is identity matrix. And then total output vector can be rewritten as:

𝑋 = (𝐼 − 𝐴)−1_{∗ 𝐹𝐷 = 𝑀 ∗ 𝐹𝐷}

In addition, the vector of emissions coefficients of sectors are defined by 𝐸𝐶 = (𝑒𝑐₁, 𝑒𝑐₂, … , 𝑒𝑐₂₆)

𝐸𝐶 = 𝐸𝑀/𝑋

where 𝐸𝐶 is the vector of emission coefficient by sectors and 𝐸𝑀 is the emissions by sectors. Lastly, the carbon emissions embodied in trade using EEBT approach are calculated. The emissions embodied in the export from Canada to China can be expressed as:

𝐸𝐸_𝐶𝐴 = 𝐸𝐶_𝐶𝐴∗ (𝐼 − 𝐴)−1∗ 𝐸𝑋_𝐶𝐴

where 𝐸𝐸𝐶𝐴 is the embodied emission in Canadian export to China, 𝐸𝐶𝐶𝐴is the column vector

of carbon emissions of Canadian sectors, (𝐼 − 𝐴)−1 is the Leontief inverse matrix based on Canadian input-output table and 𝐸𝑋_𝐶𝐴 represents the trade volume that Canada exports to China.

Moreover, the carbon emissions embodied in the export from China to Canada can be formulated as follows:

𝐸𝐸_𝐶𝐻 = 𝐸𝐶_𝐶𝐻∗ (𝐼 − 𝐴)−1_{∗ 𝐸𝑋} 𝐶𝐻

where 𝐸𝐸𝐶𝐻 is the embodied emission in Chinese export to Canada, 𝐸𝐶𝐶𝐻 is the column vector

(18)

18

4.2 Structural decomposition analysis

After calculating the emissions embodied, I use decomposition analysis to have a better understanding of the driving factors that influence carbon emissions embodied in China-Canada trade during 1990-2014.

In recent decades, decomposition analysis has been extensively used to investigate the driving factors of changes of an aggregate indicator over time. There are two widespread decomposition techniques: the index decomposition analysis (IDA) and the structural decomposition analysis (SDA).

SDA and IDA have been developed independently so they are used differently in studies. IDA is often applied in energy papers in order to have a better understanding of the driving forces of energy use and energy-related emissions in a specific energy consumption sector, such as transportation while SDA is used mostly by academics who are familiar with input–output analysis and extend it to investigate changes in energy consumption or emissions in the economy (Su et al., 2012)

The main advantage of IDA over SDA is a lower data requirement as it uses only aggregate sector information. However, this also means a disadvantage because IDA is not capable of detailed decompositions of the economic structure. For more complex analysis, it is advised to use SDA because SDA can distinguish between a range of technological effects and final demand effects which are not possible in the IDA method (Hoekstra, 2003).

Moreover, another advantage of SDA is that the input–output model includes indirect demand effects. Indirect effects occur when a direct demand increase in one sector leads to increases in the demand for inputs from other sectors. It is a spill-over effect of demand that is captured by the Leontief inverse of the input–output model (Miller and Blair, 1985).

There are different determinant effects can be distinguished in SDA decompositions. For example, the production effect which measures the effect of total output change on the indicator, or the Leontief effect which indicates the effect of the changes in the Leontief inverse coefficients (can also be interpreted as a technological effect of changes in the intermediate input structure) (Hoekstra, 2003).

(19)

19

The equation I used to calculate the emission embodied in Canada-China trade above is as follows:

𝐸𝐸 = 𝐸𝐶 ∗ (𝐼 − 𝐴)−1∗ 𝐸𝑋

where 𝐸𝑋 means the column vector of the total export volume of sectors.

Changes in CO2 emissions embodied in export are decomposed into the four following driving

factors, namely, carbon emissions coefficients effect (∆𝐸𝐶), technology effect (∆(𝐼 − 𝐴)−1), trade structure effect (∆𝑝) and trade scale effect (∆𝑣). In order to examine the effect of trade structure effect and trade scale on embodied carbon emissions a deformed equation is needed. The new equation can be expressed as:

𝐸𝐸 = 𝐸𝐶 ∗ (𝐼 − 𝐴)−1∗ 𝑝𝑣

where trade structure and trade scale are expressed as: 𝑝_𝑖/𝑣 where 𝑝_𝑖 represents the proportion of export volume in sector i (trade structure) and 𝑣 represents the total export volume (trade scale).

I assume that 0 means the base period and 1 means the value of report period so the decomposition formula can be expressed as follows:

∆𝐸𝐸 = ∆𝐸𝐸(∆𝐸𝐶) + ∆𝐸𝐸[∆(𝐼 − 𝐴)−1] + ∆𝐸𝐸(∆𝑝) + ∆𝐸𝐸(∆𝑣) where

the carbon emissions coefficients effect is expressed as: ∆𝐸𝐸(∆𝐸𝐶) = 1 2⁄ [∆𝐸𝐶(𝐼 − 𝐴)−1

0𝑝0𝑣0+ ∆𝐸𝐶(𝐼 − 𝐴)−11𝑝1𝑣1]

the technology effect is expressed as:

∆𝐸𝐸[∆(𝐼 − 𝐴)−1_{] = 1 2}⁄ [𝐸𝐶1∆(𝐼 − 𝐴)−1𝑝0𝑣0+ 𝐸𝐶0∆(𝐼 − 𝐴)−1𝑝1𝑣1]

the trade structure effect is expressed as:

∆𝐸𝐸(∆𝑝) = 1 2⁄ [𝐸𝐶1(𝐼 − 𝐴)1∆𝑝𝑣0 + 𝐸𝐶0(𝐼 − 𝐴)−1₀∆𝑝𝑣1]

and the trade scale effect is expressed as:

∆𝐸𝐸(∆𝑣) = 1 2⁄ [𝐸𝐶₁(𝐼 − 𝐴)−1_𝑝

(20)

20

5. Source Data

This section describes the two data sources, the EORA database and the UN Comtrade (United Nations International Trade Statistics) database which were used for the analysis. I use the EORA database in order to calculate the carbon emissions embodied in bilateral trade between Canada and China and compare the results with the raw data from the UN Comtrade database in order to compare the emissions with the trade volume.

The EORA database has been developed by Lenzen et al. (2012) in order to create a new series of environmentally extended multi-region input-output (MRIO) tables for 1990-2015 with applications in carbon, water and ecological footprint and Life-Cycle Assessment. The EORA global supply chain database consists of a MRIO model that provides a time series of IO tables with matching environmental and social satellite accounts for 190 countries.

In order to construct the EORA MRIO tables, the following data sources were used: national statistical offices, Eurostat, IDE-JETRO, OECD, the UN National Accounts Main Aggregates Database, the UN National Accounts Official Data, the UN Comtrade international trade database, and the UN Servicetrade international trade database.

The database is constructed in current US$, so that that countries can be compared against each other. The conversion of national currencies into current US$ is based on the exchange rates of the International Monetary Fund (IMF) Official Exchange Rates. It is provided in basic prices and in purchasers prices, but I use basic prices as it recommended for use for environmentally extended IO analysis.

In the database the environmentally extended MRIO tables have satellite accounts in physical units in order to complement the monetary table with nonmonetary inputs to production. Therefore, it includes the conventional economic inputs (steel, machinery, labour and capital) as well as resources (land, energy and water) and environmental impacts (emissions and biodiversity loss). The advantage of this format is that both the monetary MRIO and the satellite accounts follow the same sector classification.

I choose to examine only the CO2 emissions as during the combustion process, most of carbon

emissions are immediately emitted as CO2. Moreover, some carbon emitted as non-CO2 species

like carbon monoxide (CO) or methane (CH4) eventually oxidises to CO2 as well in the

(21)

21

EORA provides CO2 emissions inventory satellite accounts rows from three different data

sources: EDGAR, CDIAC, and the PIK PRIMAPHIST model. The sectoral allocation of emissions also follows the original pattern described above. That means that users can take the territorial emissions inventory from these three data providers as the starting point for their analysis. In my thesis, I use the data provided by EDGAR as it has information over CO2

emissions in all the 26 sectors in both countries while in the other two, there are missing columns.

Eora is available in three formats, namely, individual country IO tables, EORA26 which is a complete global MRIO table, plus environmental satellite account, in a harmonized 26-sector classification and Full EORA which is the complete EORA MRIO table.

In this paper, the EORA26 is used as the other tables have mixed structures and making the consumption categories of countries comparable, which is out of the scope of this paper. In this simplified version all countries have been aggregated to a common 26-sector classification and contains only symmetric product-by-product and industry-by-industry IO tables. The sector classification is shown in Table 2. The only disadvantage of the simplified model is that is known to be slightly less accurate.

Table 2: EORA sectors

Source: EORA database.

The MRIO tables are divided into different matrices, namely, the transactions matrix (T), primary inputs or value added (VA), final demand block (FD), and satellite accounts or environmental extensions (Q). In this analysis, only the transactions matrix, the final demand and the environmental extensions are used.

1 Agriculture 14 Construction

2 Fishing 15 Maintenance and Repair

3 Mining and Quarrying 16 Wholesale Trade

4 Food & Beverages 17 Retail Trade

5 Textiles and Wearing Apparel 18 Hotels and Restaurants

6 Wood and Paper 19 Transport

7 Petroleum, Chemical and Non-Metallic Mineral Products 20 Post and Telecommunications

8 Metal Products 21 Financial Intermediation and Business Activities 9 Electrical and Machinery 22 Public Administration

10 Transport Equipment 23 Education, Health and Other Services

11 Other Manufacturing 24 Private Households

12 Recycling 25 Others

(22)

22

6. Results

In this section, I present the results that have been calculated using the methodology described in Section 4. First, the results of the input-output analysis, namely the total embodied emissions, as well as the breakdown of emissions by industrial sectors in the bilateral trade between China and Canada are presented.

Second, the results of the structural decomposition analysis are discussed where the contributions of carbon emissions coefficients, intermediate technology, trade scale and trade structure to the changes in embodied carbon emissions are explained. All the figures and tables are made in Excel based on the results exported from MATLAB.

6.1 Input-output analysis

Figure 1 represents the carbon emissions embodied in bilateral trade between Canada and China and the net emission transfer (NET) between 1990 and 2014.

Figure 1: CO2 emissions embodied in Canada-China trade for the period of 1990-2014.

The results demonstrate that CO2 emissions embodied in bilateral between China and Canada

increased for both sides from 1990 to 2014. However, while embodied emissions in export from Canada to China increased by 3579% (from 22.38 kilotons to 801.15 kilotons), embodied emission exported from China to Canada only increased by 511% (from 668.26 kilotons to 3415.36 kilotons). 0 500 1000 1500 2000 2500 3000 3500 4000 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 CO 2 kilo to n n es

(23)

23

During the investigated period, the former, the embodied emissions exported from Canada to China had a steady increase with almost no fluctuations, peaked in the last year, 2014 with a value of 801.15 kilotons and had the lowest point in 1991 with a value of 22.27 kilotons. The average annual growth rate of the emission embodied in export from Canada to China is 1.17%. Meanwhile, the latter, the carbon emissions embodied in export from China to Canada fluctuated more, had local peaks, where embodied emissions started decreasing for a few years and then started increasing again. This happened, for example, in 1998.

In 1998, emissions embodied in export started decreasing, except for a small increase in 2000 and continued decreasing until 2002. After 2002, emissions embodied in export from China to Canada started increasing rapidly, reaching its peak in 2006, with a value of 3557.13 kilotons. After 2006, it started decreasing again reaching a local low point in 2009, with a value of 2601.9 kilotons.

After 2009, embodied carbon emissions exported from China to Canada started increasing again, but only until reaching the 2008 value again. Since 2011 the CO2 emissions embodied in

export remained almost the same. The embodied emissions in export from China to Canada was also the lowest in 1991, with a value of 514.54 kilotons and had an average annual growth rate of 1.09%.

It is clear from Figure 1 that embodied emissions in export from China to Canada are larger than from Canada to China so we can state that China is a net exporter of embodied CO2

emissions in Canada-China trade. The line of the net emission transfer (NET) follows the embodied emissions in export from China to Canada. The NET peaked in year 2006 with a value of 3297.92 kilotons (3.3 Mt) and was the lowest in 1991 with a value of 492.27 kilotons. This result also confirms the papers mentioned in Section 2 (Liu et al., 2010 and Dong et al.,2010) that China is a net exporter of embodied emissions to major OECD countries including Japan or the USA.

(24)

24

In order to find out whether trade surplus can explain the position of China as the net exporter of emissions in the bilateral trade with Canada I compared the trade values of export from China to Canada and from Canada to China with the emissions embodied in these exports for the years of 1995, 2000, 2005, 2010 and 2014. Table 3 shows the export values and the emissions embodied in the bilateral trade between Canada and China.

Table 3: Export values and emissions embodied in export.

Value of export from China to Canada (billion US$) Emissions embodied in export from China

to Canada (CO2 kilotons) Value of export from Canada to China (billion US$) Emissions embodied in export from Canada to China (CO2 kilotons) 1995 1.53 1263.81 2.53 101.20 2000 3.16 1665.42 2.44 143.11 2005 11.65 2822.96 5.96 239.98 2010 22.22 2878.37 12.85 555.74 2014 30.00 3415.36 17.45 801.15

The export values both from Canada to China and from China to Canada have increased significantly between 1995 and 2014. The value of Chinese export to Canada was 1.53 billion US$ in 1995, and 30.00 billion US$ in 2014. The value of Canadian export to China was 2.53 billion US$ in 1995, and 17.45 billion US$ in 2014. The embodied carbon emissions were also increasing but in 1995 China had a trade deficit with Canada while the embodied carbon emissions were higher in that year too. Therefore, we can conclude that trade surplus was not the reason for China being a net exporter of emissions embodied in Canada-China trade. To further investigate if trade surplus a reason of for China being a net exporter of emissions, we can calculate the emissions embodied in the case when both countries produce products with the same value, 1 billion $.

Table 4: Carbon emissions in 1 billion $ of export.

(25)

25

By comparison, China emitted more carbon dioxide than Canada when they produce goods with the same value. For example, China exported 30 billion US$ products and emitted 3415 kilotons CO2 in 2014 which means that China emitted 114 thousand tons per billion US dollars on

average. In comparison, Canada only emitted 46 thousand tons per billion US dollars on average. This means that Chinese export sectors are more carbon-intensive than Canadian export sectors. Therefore, in the next step the export sectors will be analysed.

Figure 2 and Figure 3 show the breakdown of emissions embodied in Canada-China trade by industrial sectors.

Figure 2 shows the main sectors inducing embodied CO2 emissions in the export from Canada

to China. These sectors are electrical and machinery, food and beverages, transport equipment, transport, agriculture, and petroleum, chemical and non-metallic mineral products, with values of 23%, 22%, 14%, 13%, 11% and 8% for the year of 2014, respectively. The remainder sectors in the export structure from Canada to China have a share of 8%.

Figure 2: CO2 emissions embodied in Canada’s export to China by sectors.

The export structure has undergone some significant changes regarding the share of some sectors in the period of 1990-2014. Emissions embodied in export increased in the sectors of food and beverages from 20% to 22%, transport from 10% to 13%, and petroleum chemical and non-metallic mineral products from 7% to 8% while slightly decreased in the sector of transport equipment from 16% to 14% and remained the same in the sector of agriculture with the share of 11%. The biggest change happened in the sector of electrical and machinery, the share of the sector decreased from 28% to 23%.

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 Others

Petroleum, Chemical and Non-Metallic Mineral Products Agriculture

Transport

(26)

26

Some of these sectors are also represented in Section 2, Figure 3 where the top-10 export product categories from Canada to China are listed. This list mainly has products from two sources: agriculture and natural resources. These two categories are also present here above: in the agriculture sector and in the food and beverages sector. Moreover, sectors of petroleum, chemical and non-metallic mineral products and transport equipment are also presented in both. Therefore, we can state that large proportion of emissions embodied in the export of these sectors is due to large volume of exports from Canada to China in these segments.

Machinery can also be found on both lists but while based on its volume it should be the last one inducing carbon emissions in export, yet, based on the calculations, it is the first one inducing CO2 emissions in export from Canada to China. This indicates that the machinery

sector is a carbon intensive sector.

Figure 3 represents the sectors inducing emissions embodied in export from China to Canada. Substantial emissions embodied in export are present in the following seven sectors: electrical and machinery, textiles and wearing apparel, other manufacturing, petroleum, chemical and non-metallic mineral products, metal products, transport equipment, food and beverages, with values of 48%, 15%, 11%, 10%, 6%, 4% and 3% for the year of 2014, respectively. The remaining sectors only make up 3% of the total emissions embodied in export.

Figure 3: CO2 emissions embodied in China’s export to Canada by sectors.

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 Others

Food & Beverages Transport Equipment Metal Products

(27)

27

The structure of exported emissions has changed significantly in the investigated period. The emissions embodied in the export of electrical and machinery increased the most, in 1990 it only had a share of 17% which increased to 48% in 2014. Metal products and transport equipment also increased, from 5% to 6% and from 1% to 4%, respectively. Carbon emissions embodied in the export of textiles and wearing apparel, other manufacturing and food and beverages decreased substantially, from 24% to 15%, from 27% to 11% and 6% to 3%, respectively while petroleum, chemical and non-metallic mineral products only slightly decreased, from 11% to 10%. This indicates that export structure changed from primary products to more sophisticated products in the study period.

The results, here, also show similarities with Section 2, Figure 4 which represents the top-10 export product categories from China to Canada. Electrical and machinery is top export sector inducing emissions based on the results, and this sector makes up the first two export products categories which means that embodied emissions in these sectors are due to their large export volume. Other sectors can also be found in both, which means that trade scale can be a driving force for embodied emissions. These sectors are other manufacturing, textiles and wearing apparel, transport equipment and metal products.

(28)

28

6.2 Structural decomposition analysis

In this section, I use SDA to make further investigation about the driving factors, namely the effect of carbon emission coefficients, intermediate technology, trade structure and trade scale on embodied carbon emissions in the bilateral trade between China and Canada in the period of 1990-2014.

Figure 5: Structural decomposition analysis of exported emissions by Canada between 1990 and 2014.

The total embodied emissions in export from Canada to China increased almost for the whole investigated period, except for 3 years (1991, 1996, 2001) when it experienced a small decrease (Figure 1). Figure 5 demonstrates how the four driving factors mentioned above contributed to this increase of emissions embodied in the export from Canada to China in every year between 1990 and 2014.

Carbon emissions coefficients and trade scale gradually influenced the emissions embodied in export during the investigated period while intermediate technology and trade scale had no or very little effect on the embodied emissions.

Carbon emissions coefficients played an important role in reducing the CO2 emissions

embodied in export in almost every year during the study period. Meanwhile, the trade scale - the other dominant driving force – was the biggest driver for increasing the exported embodied emissions. -100 -50 0 50 100 150 200 250

(29)

29

There are only a few exceptions where these two effects contributed to the changes in embodied emissions differently. Between 1995 and 1996, trade scale also helped reducing the embodied emissions and trade structure became dominant factor inducing carbon emissions embodied. During 2008-2009, trade scale contributed to the decrease of embodied emissions and carbon emissions coefficients contributed to the increase. This is the result of the global financial crisis when the trade volume also decreased. The trade structure also changed which helped reducing the emissions embodied.

The trade scale effect peaked between 2009 and 2010, followed by 2010-2011 because after the crisis, the volume of export started increasing again, reaching its 2008-value again in 2011. After 2011 the trade scale effect went back to its normal state. The effect of carbon emissions coefficients only becomes dominant after 2002 and remains dominant until 2011. In this period, Canada adjusted energy consumption structure by the decline of carbon-intensive energy consumption which led to the decrease of carbon emissions coefficients and thus had a significant impact on reducing carbon emissions embodied in export.

Figure 6: Structural decomposition analysis of exported emissions by China between 1990 and 2014. -1500 -1000 -500 0 500 1000 1500

(30)

30

Embodied emissions in the export from China to Canada have periods of increasing and decreasing during the study period (Figure 1). Figure 6 demonstrates how the carbon emissions coefficients, intermediate technology, trade structure and trade scale influenced these periods of increases and decreases regarding the emissions embodied in the export from China to Canada.

The results presented in Figure 5 and Figure 6 are similar regarding the driving factors that contribute to the changes in embodied emissions in export. In this regard, carbon emissions coefficients also reduced significantly the emissions embodied in export while trade scale was also the dominant factor inducing the increase of exported embodied emissions in every year, except for 1990-1991 and 2008-2009.

In 2008-2009 both trade scale and carbon emissions coefficients contributed to the decrease of carbon emissions embodied in export. This exception also occurred due to the global financial crisis where export volume decreased, leading to the obvious decline of embodied carbon emissions.

Trade structure and intermediate technology, however, had almost no significant effect on the CO2 emissions embodied in export from China to Canada. The effects of trade structure can be

seen, for example, between 1997 and 1998.

(31)

31

6.3 Evaluation of hypotheses

Looking at the results described above, the four hypotheses stated in Section 3 can be evaluated. Based on the results, China is a net exporter of emissions in the bilateral trade between Canada and China, so the first hypothesis is accepted.

Hypothesis 1: In the bilateral trade between Canada and China, China is net exporter of carbon emissions.

China has a trade surplus with Canada, but not in every year. Until 1996, China has a trade deficit, so trade surplus cannot be the reason for China being the exporter of carbon emissions. Therefore, the second hypothesis is rejected.

Hypothesis 2: The trade surplus is the reason for China being the net exporter of carbon emissions in China-Canada trade.

Emission intensity (measured in the analysis as carbon emissions coefficients) contributed to the increase of carbon emissions embodied in bilateral trade, while trade scale contributed to the decrease of embodied emissions (for most of the years). Trade structure and technology (measured as intermediate technology) had no significant effect on the embodied CO2

emissions. Therefore, the third and the fourth are partially accepted.

Hypothesis 3: Trade scale and trade structure contribute to the increase of emissions embodied in bilateral trade between China and Canada.

(32)

32

7. Conclusion

In my thesis, I have calculated and analysed the emissions embodied in Canada-China trade between 1990 and 2014. Furthermore, I applied structural decomposition analysis in order to find what factors influenced the emissions embodied in their export. The main takeaways of my paper are the following.

First, emissions embodied in bilateral trade of China and Canada increased during 1990-2014. However, while embodied emissions in export from Canada to China increased by 3579%, embodied emission exported from China to Canada only increased by 511%. Moreover, embodied emissions in export from China to Canada are larger than from Canada to China which means that China is a net exporter of embodied CO2 emissions in Canada-China trade.

Second, the export structure has experienced some significant changes regarding the share of sectors in both countries. In the case of emissions embodied in export from Canada to China, the biggest change happened in the sector of electrical and machinery, the share of the sector decreased from 28% to 23%. While, in the case of embodied emissions exported from China to Canada, the sector of electrical and machinery increased the most, from 17% to 48%. Carbon emissions embodied in the export of textiles and wearing apparel, other manufacturing and food and beverages also decreased substantially, which means that export structure changed from primary products to more sophisticated products in the study period.

Third, trade scale was responsible for the increase of exported carbon emissions in both countries. Trade structure and intermediate technology had no significant effects on CO2

emissions embodied in trade. And carbon emissions coefficients effect was the dominant driving factor reducing the emissions embodied in bilateral trade between Canada and China. This thesis has also numerous limitations due to the limited scope of the research of a master thesis. First, the analysis only focuses on the trade of two countries, namely, China and Canada, so a comprehensive evaluation of the impacts of CO2 emissions embodied in global trade is not

possible based on the results of the analysis. Second, the sector classification of the EORA database is not exactly in line with reality, which may generate bias of CO2 emission

(33)

33

In summary, although China is biggest CO2 emitter in the world, large part of its CO2 emissions

is based on foreign consumption of Chinese products which means China cannot be blamed for the increase in its carbon emissions. Moreover, although economic development is important for China, which is driven by export, it must be in line with the international trends of reducing CO2 emissions. Therefore, the national policy should emphasis on the harmony of economic

development, CO2 reduction, energy security, and environmental protection by shifting the

(34)

34

References

2006 IPCC Guidelines for National Greenhouse Gas Inventories Retrieved from: /http://www.ipcc-nggip.iges.or.jp/public/2006gj/index.html.

Andrew R. M., Davis S. J., Peters G. P., 2013. Climate policy and dependence on traded carbon. Environmental Research Letters 8(3)

Arto, I., Dietzenbacher E. 2014. Drivers of the Growth in Global Greenhouse Gas Emissions. Environmental Science & Technology 48, 5388–5394

Baiocchi, G., Minx J. 2010. Understanding Changes in the UK’s CO2 Emissions: A Global

Perspective. Environmental Science & Technology 44, 1177–1184.

Blanchfield, M. 2016. Trudeau confirms Canada exploring free trade agreement with China.

The Toronto Star, September 22. Retrieved from:

www.thestar.com/news/canada/2016/09/22/chinese-premier-li-keqiang-welcomed-toparliament-hill-by-justin-trudeau.html.

Bowen, H. P., Leamer E. E., Sveikauskas L.,1987. Multicountry Multifactor Tests of the Factor Abundance Theory. American Economic Review 77, 791–809.

Chang, N., 2013. Sharing responsibility for carbon dioxide emissions: a perspective on border tax adjustments. Energy Policy 59 (8), 850-856.

Dawson, L., Ciuriak D., 2016. Chasing China: Why an economic agreement with China is necessary for Canada’s continued prosperity. January. Ottawa, ON: Dawson Strategic and Ciuriak Consulting. Retrieved from: ccbc.com/whitepapers/ChasingChina.pdf.

Dietzenbacher E., Mukhopadhyay K., 2007. An Empirical Examination of the Pollution Haven Hypothesis for India: Towards a Green Leontief Paradox? Environmental & Resource Economics 36 (2007), 427–449.

Dobson, W., Evans P., 2015. The Future of Canada’s Relationship with China. IRPP Policy Horizons Essay, November 17. Montreal: Institute for Research on Public Policy. Retrieved from: irpp.org/wp-content/uploads/2015/11/policy-horizons-2015-11-17.pdf.

Dong Y., Ishikawa M., Liu X., Wang C., 2010. An analysis of the driving forces of CO2

(35)

35

Environment and Climate Change Canada, 2017. Canadian Environmental Sustainability Indicators: Carbon Dioxide Emissions from a Consumption Perspective. Retrieved from:

www.ec.gc.ca/indicateurs-indicators/default.asp?lang=en&n=7BA2C0DA-1

European Commission, 2019. Causes of climate change. Retrieved from:

https://ec.europa.eu/clima/change/causes_en

Feng K., Siu Y. L., Guan D., Klaus H., 2012. Analyzing Drivers of Regional Carbon Dioxide Emissions for China A Structural Decomposition Analysis. Journal of Industrial Technology 16(4), 600-611.

Hoekstra R., van der Bergh J. J. C. J. M. 2003. Comparing structural and index decomposition analysis. Energy Economics 25(2003), 39–64.

Jiang, X., Liu, Y., Zhang, J., Zu, L., Wang, S., Green, C., 2015. Evaluating the role of international trade in the growth of China's CO2 emission. J. Syst. Sci. Complex. 28(4),

907-924.

Kanemoto, K., Moran, D., Lenzen, M., Geschke, A., 2014. International trade undermines national emission reduction targets: new evidence from air pollution. Global Environmental Change 24, 52-59.

Leamer, E. E., 1980. The Leontief Paradox, Reconsidered, Journal of Political Economy 88, 495–503.

Leblond P., 2017. Toward a Free Trade agreement with China. Opportunities, Challenges and Building Blocks for Canada. Centre for International Governance Innovation.

Leontief, W. W., 1936. Quantitative Input and Output Relations in the Economic System of the United States, The Review of Economic and Statistics 18, 105-25.

Leontief, W. W., 1941. In: The Structure of American Economy, 1919–1929: An Empirical Application of Equilibrium Analysis. Harvard University Press, Cambridge.

Leontief, W. W., 1970. Environmental Repercussions and the Economic Structure: An Input-Output Approach, The Review of Economics and Statistics, August, 262-72.

(36)

36

Liu X., Ishikawa M., Wang C., Dong Y., Liu W., 2010. Analyses of CO2 emissions embodied

in Japan–China trade. Energy Policy, 38(2010), 1510–1518.

Liu Y., Chen S., Chen B., Yang W., 2017. Analysis of CO2 emissions embodied in China's

bilateral trade: a noncompetitive import input-output approach. Journal of Cleaner Production, 163(2017), 410-419.

Mazereeuw, Peter. 2016. In Leblond P., 2017. Toward a Free Trade agreement with China. Opportunities, Challenges and Building Blocks for Canada. Centre for International Governance Innovation.

Miller, R.E., Blair, P.D., 1985. Input–Output Analysis: Foundations and Extensions. Prentice-Hall, Englewood-Cliffs, New Jersey.

Pan, J., Phillips, J., Chen, Y., 2008. China's balance of emissions embodied in trade: approaches to measurement and allocating international responsibility. Oxford Review Economic Policy 24(2), 354-376.

Peters, G. P., Hertwich, E.G., 2008. CO2 embodied in international trade with implication for

global climate policy. Environmental Science Technology 42(5), 1401-1407.

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 U.S.A. 108, 8903-8908.

Rose A., Casler S., 1996. Input–output structural decomposition analysis: a critical appraisal. Economic Systems Research 8(1), 33-62.

Sato, M., 2014. Embodied carbon in trade: a survey of the empirical literature. Journal of Economi Surveys 28, 831-861.

Stern, N., 2007. The Economics of Climate Change: the Stern Review. Cambridge University Press.

Su B., Ang B. W., 2011. Multiregion input–output analysis of CO2 emissions embodied in trade:

The feedback effects. Ecological Economics 71(1), 42-53.

Su B., Ang B. W., 2012. An analysis of the driving forces of CO2 emissions embodied in Japan–

China trade. Energy Economics,34(1), 177-188.

Su B., Ang B. W., 2014. Input–output analysis of CO2 emissions embodied in trade: A

(37)

37

Su, B., Ang, B.W., 2012. Structural decomposition analysis applied to energy and emissions: Some methodological developments. Energy Economics 34(1), 177-188.

Su, B., Ang, B.W., 2013. Input-output analysis of CO2 emissions embodied in trade: competitive versus non-competitive imports. Energy Policy 56 (5), 83-87.

Su, B., Ang, B.W., 2015. Multiplicative decomposition of aggregate carbon intensity change using input-output analysis. Applied Energy 154, 13-20.

Su, B., Ang, B.W., Low, M., 2013. Input-output analysis of CO2 emissions embodied in trade

and the driving forces: processing and normal exports. Ecological Economics 88, 119-125. UN Comtrade database

Van Assche, A., 2012. Global Value Chains and Canada’s Trade Policy: Business as Usual or Paradigm Shift? IRPP Study No. 32. Montreal, QC: Institute for Research on Public Policy. Weber, C.L., Peters, G.P., Guan, D.B., Hubacek, K., 2008. The contribution of Chinese exports to climate change. Energy Policy 36(9), 3572-3577.

World Bank Database, 2019. CO2 emissions (kt).

Wu et al. (2016)

Xu, M., Li, R., Crittenden, J.C., Chen, Y., 2011. CO2 emissions embodied in China's exports

from 2002 to 2008: a structural decomposition analysis. Energy Policy 39(11), 7381-7388. Xu, Y., Dietzenbacher E. 2014. A Structural Decomposition Analysis of the Emissions Embodied in Trade. Ecological Economics 101, 10–20.

Yan, Y., Zhao, Z., 2012. CO2 emissions embodied in China's international trade: A perspective

of allocating international responsibilities. J. Int. Trade 1, 131-142. Yu and Chen, 2017

Zhang, B., Chen, G.Q., 2010. Methane emissions by Chinese economy: Inventory and embodiment analysis. Energy Policy 38 (8), 4304e4316.

Zhang, B., Chen, Z.M., Qiao, H., Chen, B., Hayat, T., Alsaedi, A., 2015. China's non-CO2

greenhouse gas emissions: inventory and input-output analysis. Ecol. Inf. 26, 101e110.

(38)

38

Emissions embodied in bilateral trade between Canada and China