Embodied Carbon Emissions in Export of Yangtze River Delta: Calculation and Decomposition of Driving Factors

Abstract

The total export of the Yangtze River Delta exceeds 1/3 of China. This region is the backbone of China’s emission reduction and green transformation exports. This paper adopts input–output analysis to calculate direct carbon emission coefficients and embodied carbon in exports. Using the LMDI method, the total embodied carbon emissions and driving factors are quantitatively analyzed by region in terms of scale effect, structural effect and technology effect. Based on the results of the study, the government should actively promote the achievement of carbon neutrality goals by setting carbon emission standards or introducing regulatory policies and incentive mechanisms. Meanwhile, the export structure needs to be continuously optimized. It is necessary to accelerate technological innovation while actively promoting and encouraging the adoption of new energy sources in order to take a low-carbon, clean and sustainable green development path.

1. Introduction

Carbon emission surges and global warming pose serious challenges to human society, and the resulting extreme climate change has become a worldwide issue. As an important factor affecting the economic development of a country or a region, the “carbon” problem generated by export has become a global issue. The research and discussion on green trade concepts such as carbon emission reduction and the improvement of environmental quality and living standards have become hot international issues. China is the world’s largest exporter. While improving its international competitiveness and influence, the expansion of its export scale has led to a large number of greenhouse gases remaining in the country’s atmosphere, causing a surge in total carbon dioxide emissions and further intensifying the conflict between the environment and trade. The 14th Five-Year Plan is the key period and window for China to reach the carbon peak. Under the influence of the green trade trend in the new era, China has elevated green development to a national strategy, strived to coordinate the relationship between export and carbon emissions and proactively assumed responsibility for international emission reduction.

The Yangtze River Delta region (Figure 1) is an important window for the development of China’s export because the total exports of the region have increased from 29.19% in 2000 to 37.6% in 2021, with Jiangsu and Zhejiang holding the second and third position in national exports, accounting for 28.8% of national exports. Meanwhile, the scale of carbon emissions in the Yangtze River Delta region is also expanding, increasing from 324,429,300 tons in 2000 to 803,195,100 tons in 2019. The main reason is that the Yangtze River Delta region relies heavily on high energy consumption and high pollution energy sources, with a low utilization rate of clean energy. The expansion of exports in the region promotes economic growth but leaves a large amount of carbon dioxide in the country, which aggravates domestic environmental pollution.

The innovation and contribution of this research are as follows:

• This paper takes the Yangtze River Delta as the research object to conduct detailed measurement and in-depth analysis, which not only helps to alleviate the prominent problem of embodied carbon emissions in this region but also provides abundant data from the region. As a matter of fact, analysis of the embodied carbon emissions from China’s exports is useful for international consideration of carbon emissions from China.

2. Literature Review

2.1. Review

Embodied carbon refers to all carbon dioxide produced in the complete production chain that a given product (containing individual components) undergoes, i.e., carbon emissions directly or indirectly generated throughout its life cycle from raw materials, processing and transportation to consumption [1]. Machado et al. studied embodied carbon emissions from export for Brazil [2], Chung and Rhee for South Korea [3], Lenzen for Australia [4], Mukhopadhyay for India [5] and Peters and Hertwich for Norway [6]. Shui and Harriss [7], Zhu [8], Qi et al. [9], Yan and Zhao [10], Yin [11] and Huang [12] measured China’s export embodied carbon emissions of China’s export. The embodied carbon measurement mainly uses the life-cycle approach and the input–output approach [13,14]. Liu et al. used the whole life cycle evaluation method to calculate that the carbon emissions of China’s export products accounted for approximately 14.4% of national carbon emissions [15]; Ackerman used the multi-regional input–output method to measure the scale of embodied carbon generated between the U.S. and Japan during trade, and concluded that Japan took up part of the carbon emissions from the U.S. [16]; Qi used the input–output method to estimate that the net exports of embodied carbon accounted for 29.4% of the total carbon emissions by 2006, and 29.28% of the total carbon emissions [9]. Zhang and Sun found that China exported nearly four times more produced carbon to developed countries than the value of imports during 1990−2015 based on cross-country input–output data, and in particular, China took up a large amount of produced carbon for the United States [17].

There are two main methods for factor decomposition of embodied carbon emissions: One is the Index Decomposition Analysis (IDA), of which the Logarithmic Mean Divisia Index (LMDI) is widely used. The other is Structural Decomposition Analysis (SDA). Greening used the LMDI method to decompose the embodied carbon emissions of trade among the countries of the Organization for Economic Cooperation and Development and pointed out that technological upgrading can force carbon emissions to scale down [18]. Tong and Gao analyzed the factors influencing the embodied carbon emissions of export in Jiangsu from three aspects of scale effect, structural effect and technology effect using the LMDI method and concluded that the technology effect has the greatest inhibitory effect [19]. Yin analyzed the carbon emissions of export in China using the SDA method and showed that the technology structural effect and export effect are the main factors for the change of the embodied carbon emissions intensity of export factors [11]. Zhao and Zheng used the SDA method to analyze the effect of total embodied carbon emissions in the Yangtze River Delta region to show that technological progress is the main driver of carbon emission reduction [20].

These existing studies reveal that the assessment method of embodied carbon emissions and the analysis of the driving factors of trade embodied carbon are mature, which provides a solid theoretical foundation and argumentative method for scholars’ subsequent studies. Given the economic status and export influence of the Yangtze River Delta, we take the Yangtze River Delta as the research object and use the input–output method and LMDI method for detailed measurement and analysis.

2.2. Research Gap

We arranged the survey of related work in

Table 1. Survey on Related Work.

Refs.	Method	Driving Factors	Case Study
Wyckoff and Roop [13]	I-O	-	6 OECD countries
Machado et al. [2]	I-O	-	Brazil
Chung and Rhee [3]	I-O	-	Korea-Japan trade
Lenzen M [4]	I-O	-	Australia
Zhang and Sun [17]	I-O	-	China
Greening [18]	LMDI	Technology effect	10 OECD countries
Mukhopadhyay Kakali [5]	I-O	-	Indian foreign trade
Peters Glen and Hertwich E.G [6]	LCA	-	Norway
Shui Bin and Harriss Robert C [7]	EIO-LCA	-	US-China Trade
Liu et al. [15]	LCA	-	China Export
Ackerman et al. [16]	I-O	-	Japan-US trade
Qi et al. [9]	I-O	-	Chinese foreign trade
Zhu, Q [8]	I-O	-	Chinese Export
Yan and Zhao [10]	I-O	-	Chinese Export
Fu [21]	I-O	-	Chinese Export
Yin [11]	I-O SDA	Direct carbon emission coefficient intermediate input structure export effect	Chinese Export
Guo and Luo [14]	I-O	-	Yangtze River Economic Belt of China
Huang et al. [12]	I-O LMDI	Scale effect structural effect intensity effect	Yangtze River Economic Belt of China
Tong and Gao [19]	I-O LMDI	Scale effect structural effect technology effect	Jiangsu, China
Zhao and Zheng [20]	I-O SDA	Intensity effect energy structural effect industrial structural effect demand scale effect	Yangtze River Delta of China
This research	I-O LMDI	Scale effect structural effect technology effect	Yangtze River Delta of China

.

As can be seen, most studies have taken countries as the subject, while few studies have taken provinces and cities as the subject. The literature on measuring embodied carbon in export by region and sector is even less. Therefore, our study of embodied carbon emissions from exports in the Yangtze River Delta fills a research gap.

Similar to Fu [21], we use input–output analysis to calculate the embodied carbon emissions. In addition, the driving factors were quantitatively analyzed by region in terms of scale effect, structural effect and technology effect using the LMDI method.

3. Methodology

3.1. Problem Statement

In view of the economic and export status of the Yangtze River Delta region, we choose the Yangtze River Delta as the research object and raise some important problems related to it: First, with the continuous expansion of the export and carbon emission scale, what will be the trend of the embodied carbon emission from export in the Yangtze River Delta? Second, what are the driving factors that bring about this trend? To measure the embodied carbon emissions of export in the Yangtze River Delta, we adopt the input–output model and further decompose the embodied carbon emissions from exports at different stages through the LMDI method to quantify the effects of three driving factors: scale effect, structural effect and technology effect.

3.2. Notation List

The variables we used are listed in

Table 2. Variables and Description.

Variable	Description
$i$	Index of sectors $i ϵ I=\left\{1,2,\dots ,i\right\}$
$j$	Index of sectors $j ϵ J=\left\{1,2,\dots ,j\right\}$
$k$	Index of energy source $k ϵ K=\left\{1,2,\dots ,k\right\}$
$a_{ij}$	Direct consumption coefficients: the value of sector $i$ that needs to be consumed directly by sector $j$ to produce a unit value of product
$A$	Matrix of $a_{ij}$ (n × n square)
$X$	Total output matrix of each sector
$Y$	End-use matrix of each sector
$E$	Unit matrix
${(E-A)}^{-1}$	Leontief inverse matrix
$X_{ij}$	Intermediate use of $i$ by the $j$ sector
$X_j$	The total output of the $j$ sector
$b_{ij}$	Complete consumption coefficients: total value of direct and indirect consumption of the $i$ sector required to produce a unit value of product in the $j$ sector
$B$	Matrix of $b_{ij}$
$d{c}_i$	Direct carbon emission factor in sector $i$
$C_i$	Total direct carbon emissions generated by sector $i$
$DC$	Matrix of direct carbon emission factors $d{c}_i$
$TC$	Complete carbon emission factor matrix
$A^d$	Actual direct carbon emission factor matrix
$EC$	Embodied carbon emissions from exports
$EX$	Scale factor: total value of export
$M_i$	Structural factor: share of exports in sector $i$
$U_i$	Technology factor: complete carbon emission factor in sector $i$
$E{C}^0$	Total embodied carbon from export in the base period
$E{C}^t$	Total embodied carbon from export in $t$ period
$\mathsf{\Delta }E{C}^{0-t}$	Combined effect from base period to period t
$\mathsf{\Delta }E{C}_{ex}$	Scale effect
$\mathsf{\Delta }E{C}_m$	Structural effect
$\mathsf{\Delta }E{C}_u$	Technology effect
${\lambda }_k$	CO2 emission factor from the $k$ energy source
$NC{V}_k$	Average low-level heat generation from the $k$ energy source
$C{C}_k$	Carbon emission factor from the $k$ energy source

.

3.3. Mathematical Model

3.3.1. Input–Output Analysis

The method of measuring the embodied carbon emissions from exports of the Yangtze River Delta in this paper is based on the input–output method proposed by Leontief (1936), which can be expressed as Equation (1):

$$AX+Y=X.$$

(1)

Equation (1) can be further organized to obtain Equation (2):

$$X={(E-A)}^{-1}Y.$$

(2)

The direct consumption coefficient $a_{ij}$ reflects the degree of production linkage between a sector and others. It can be expressed by Equation (3)

$$a_{ij}=\frac{X_{ij}}{X_j}.$$

(3)

Complete consumption is the total value of all products consumed by a given product throughout the chain, including direct and indirect consumption. The matrix of complete consumption coefficients B can be expressed by Equation (4):

$$B={\left(E-A\right)}^{-1}-E.$$

(4)

When applying input–output analysis to measure embodied carbon emissions, a direct carbon emission factor $d{c}_i$ should be constructed first, which refers to the energy use efficiency of the products produced by the $i$ sector.

$$d{c}_i=\frac{C_i}{X_i}.$$

(5)

According to the Leontief matrix, the complete carbon emission factor matrix $TC$ can be expressed by Equation (6):

$$TC=\left[{\left(E-A^d\right)}^{-1}-E\right]·DC,$$

(6)

where only domestically produced inputs are considered, and foreign imported inputs are excluded.

Embodied carbon emissions from exports can be expressed by Equation (7):

$$EC=TC·EX=\left[{\left(E-A^d\right)}^{-1}-E\right]·DC·EX.$$

(7)

The energy sources covered in the Energy Balance Sheet and the Energy Statistical Yearbook are mainly raw coal, washed coal, other washed coal, coal, coke, coke oven gas, crude oil, gasoline, kerosene, diesel, fuel oil, liquefied petroleum gas, refinery dry gas and natural gas, a total of 14 types. Based on the relevant parameters in the IPCC Guidelines for National Greenhouse Gas Inventories, the CO₂ emission factor of the $k$ energy source can be calculated as ${\lambda }_k$, with Equation (8):

$${\lambda }_k=NC{V}_k·C{C}_k· \frac{44}{12} ,$$

(8)

where 44/12 is the ratio of the relative molecular mass of CO₂ to C.

3.3.2. LMDI Method

The LMDI method (Logarithmic Mean Divisia Index) is based on the IDA method, which is widely used in the energy field initially and later in the carbon emission field. Compared with the Laspeyres index method and other methods, the LMDI model can be decomposed by constructing indicators for multi-factor decomposition. There are no residual terms after decomposition, and the conclusion is more general.

The formula for the embodied carbon driving factors of export in the Yangtze River Delta can be expressed as Equation (9):

$$EC={{\displaystyle \sum }}_{i=1}^{n}E{C}_i={{\displaystyle \sum }}_{i=1}^{n}EX•\frac{E{X}_i}{EX}•\frac{E{C}_i}{E{X}_i}$$

(9)

Equation (8) is decomposed into three driving factors, as shown in Equation (10):

$$EC={{\displaystyle \sum }}_{i=1}^{n}EX·M_i·U_i.$$

(10)

The three driving factors can be expressed by Equation (11):

$$\{\begin{array}{l}\mathsf{\Delta }E{C}_{ex}={\displaystyle \sum }_{i=1}^n\frac{E{C}_i^t-E{C}_i^0}{lnE{C}_i^t-lnE{C}_i^0}·ln(\frac{E{X}^t}{E{X}^0})\\ \mathsf{\Delta }E{C}_m={\displaystyle \sum }_{i=1}^n\frac{E{C}_i^t-E{C}_i^0}{lnE{C}_i^t-lnE{C}_i^0}·ln(\frac{M_i^t}{M_i^0})\\ \mathsf{\Delta }E{C}_u={\displaystyle \sum }_{i=1}^n\frac{E{C}_i^t-E{C}_i^0}{lnE{C}_i^t-lnE{C}_i^0}·ln(\frac{U_i^t}{U_i^0})\end{array}$$

(11)

Then, the total effect of embodied carbon emissions from export can be expressed by Equation (12):

$$\mathsf{\Delta }EC=\mathsf{\Delta }E{C}_{ex}+\mathsf{\Delta }E{C}_m+\mathsf{\Delta }E{C}_u.$$

(12)

4. Results and Discussions

4.1. Comparison of Embodied Carbon Emission among Different Regions in the Yangtze River Delta

The CO₂ emission factors for each energy source calculated according to Equation (8) are shown in

Table 3. CO₂ emission factors for each energy source.

Energy	Average Low-Level Heat Generation NCV (kJ/kg)	Carbon Emission Factor CC (kg/106 kJ)	Carbon Dioxide Emission Factor (kg-CO2/kg)
Raw coal	20,908	26.32	2.02
washed refined coal	26,344	26.32	2.54
Other coal washing	12,545	26.32	1.21
Coal	18,025	26.32	1.74
Coke	28,435	31.38	3.27
Coke oven gas	17,981	21.49	1.42
Crude oil	41,816	20.08	3.08
Gasoline	43,070	18.90	2.98
Kerosene	43,070	19.60	3.10
Diesel	42,652	20.20	3.16
Fuel oil	41,816	21.10	3.24
Liquefied petroleum gas	50,179	20.00	3.68
Refinery dry gas	45,998	20.20	3.41
Natural gas	38,931	15.32	2.19

.

Based on the energy consumption data in the Energy Balance Sheet and the CO₂ emission factors of each energy source in Table 1, the CO₂ emissions of the Yangtze River Delta are measured. The total embodied carbon emissions of export in the Yangtze River Delta from 2000−2017 can be calculated based on Equation (7). The calculation results are shown in Figure 2.

As can be seen from Figure 2, carbon emissions in the Yangtze River Delta region grew from 324,429,300 tons in 2000 to 73,308,700 tons in 2017, an overall increase of 1.3 times. From 2000 to 2004, the embodied carbon share of export rose rapidly. This is due to the unique geographical advantages and comparative environmental advantages of the coastal cities in the Yangtze River Delta, which have attracted a large amount of international capital and taken over the polluting industries of developed countries. In 2005, there was a short slip. Then, in 2006, it started to rebound with a vengeance. During 2008–2013, it remained stable at around 20%. From 2014 to 2017, the share of exported embodied carbon emissions dropped significantly and gradually stabilized under the influence of the sustainable development concept. Overall, the embodied carbon from exports accounts for about one-fifth of the total carbon emissions in the region, so the embodied carbon from export in the Yangtze River Delta should be taken seriously.

Shanghai, Jiangsu, Zhejiang and Anhui in the Yangtze River Delta have large differences in export value and embodied carbon emissions from exports. The percentage of export value and the percentage of embodied carbon emissions from exports of each of the three provinces and cities from 2000−2017 were measured, as shown in Figure 3. The percentage of export value in Shanghai has been decreasing in the Yangtze River Delta, and the percentage of embodied carbon emission is smaller than the percentage of export value except in 2002. In 2011, the total export value of Zhejiang surpassed Shanghai and was located second in the Yangtze River Delta. Jiangsu’s exports are the second largest in China, accounting for approximately 40% of the total exports in the Yangtze River Delta. However, the embodied carbon emission ratio of exports is higher than that of exports except for 2002 and 2003. Anhui developed later than other cities. As a result, the proportion of exports has been low, and the share of embodied carbon emissions from exports has been larger than its share of export value. Therefore, Jiangsu and Anhui should pay more attention to export value while not forgetting to strengthen environmental protection.

4.2. Comparison of the Driving Factors among Different Regions in the Yangtze River Delta

The LMDI decomposition results of the total embodied carbon emissions from export in the Yangtze River Delta according to Equation (11) and the related data are shown in

Table 4. LMDI decomposition of total embodied carbon in export by region (unit: million tons).

		Shanghai	Jiangsu	Zhejiang	Anhui
Stage 1	$\mathsf{\Delta }E{C}^{2002-2007}$	1095.21	4180.56	2266.14	274.67
	$\mathsf{\Delta }E{C}_{ex}^{2002-2007}$	2992.72	4238.08	2694.29	478.65
	$\mathsf{\Delta }E{C}_m^{2002-2007}$	8.49	353.71	101.98	−40.28
	$\mathsf{\Delta }E{C}_u^{2002-2007}$	−1906.00	−411.23	−530.13	−163.70
Stage 2	$\mathsf{\Delta }E{C}^{2007-2012}$	520.88	370.72	20.44	234.09
	$\mathsf{\Delta }E{C}_{ex}^{2007-2012}$	3797.06	7104.50	3802.47	831.07
	$\mathsf{\Delta }E{C}_m^{2007-2012}$	142.84	291.68	−92.98	22.49
	$\mathsf{\Delta }E{C}_u^{2007-2012}$	−3419.03	−7025.46	−3689.05	−619.47
Stage 3	$\mathsf{\Delta }E{C}^{2012-2017}$	475.38	−2035.10	−646.65	71.60
	$\mathsf{\Delta }E{C}_{ex}^{2012-2017}$	952.36	362.13	937.17	461.78
	$\mathsf{\Delta }E{C}_m^{2012-2017}$	−262.62	210.23	−38.82	−94.95
	$\mathsf{\Delta }E{C}_u^{2012-2017}$	−214.36	−2607.47	−1545.00	−295.23

Data source: Calculated from Equation (10), Equation (11) and related data.

The total effect shows three stages of rapid increase, convergence increment and gradual decrease. Furthermore, the scale effect is in the shape of an inverted “U”. The line first grows steadily, then rises rapidly, and finally falls back slowly. The structural effects are significantly different, with no significant correlation. In addition, the decomposition value of the technology effect is always negative. It indicates that the technology effect suppresses carbon emissions, and the inhibitory effect is most prominent in the second stage.

The total amount of embodied carbon exported by Shanghai increases with the increase in export scale, but the growth rate tends to slow down gradually. In the first stage, the scale effect is dominant, and its contribution value reaches 29,927,200 tons. In the second stage, the improvement of technology plays a good role in emission reduction, and the decomposition value of the technology effect reaches −34,190,300 tons, which weakens the high growth caused by the expansion of scale. In the third stage, the decomposition value of the structural effect is −26,262,200 tons, which indicates that proper adjustment of export structure can also suppress the growth of embodied carbon emissions.

As for Jiangsu Province, the total embodied carbon emission of export has experienced three stages of high growth, slow growth and rapid decline. The decomposition value of the scale effect in the first stage is 42,380,800 tons; thus, the expansion of the export scale is the main reason for the high growth of total embodied carbon emissions. The technology effect in the second stage is −70,254,600 tons, so the improvement of the technology level inhibits the growth of embodied carbon emissions. In the third stage, the technology effect is −26,074,700 tons. The inhibition of the technology effect is much higher than that of the structural effect and scale effect, which is an important factor in promoting emission reduction.

The overall trend of the embodied carbon of export in Zhejiang Province is arch-shaped. The value of the scale effect in the first stage is 26,946,900 tons, and the expansion of the export scale is the main reason for the high growth of total embodied carbon emission. In the second stage, the decomposition value of the technology effect is −36,890,500 tons, and its emission reduction effect rises significantly. The decomposition value of the technology effect in the third stage is −15,450,000 tons, far exceeding the value of the scale effect of 9,371,900 tons, indicating that improving technology can promote the harmonious development of the environment and trade.

Anhui Province owns a small export volume while the overall embodied carbon emission scale keeps a slow growth trend. The value of scale effect in the first stage is 4,786,500 tons, which has the largest share and is the main reason for its increase. In the second stage, the value of the scale effect increased 73.63% compared with the previous stage, and the decomposition value of the technology effect is −6,194,700 tons, while the value of the structural effect is positive. It means that Anhui Province may have undertaken a large number of polluting industries transferred from developed countries or regions. During the third stage, the total embodied carbon of export is in the slow-rising period. The scale effect is still in the main position, but the inhibition of the structural effect and technology effect is gradually obvious. In conclusion, the optimization of export structure and the improvement of production technology have effectively suppressed the growth of total embodied carbon.

5. Conclusions and Outlook

5.1. Conclusions

This paper uses the input–output analysis to measure the total embodied carbon emissions from export in the Yangtze River Delta from 2002–2017 and analyzes the driving factors in terms of scale effect, structural effect and technology effect by region with the LMDI method. The main findings are as follows:

• The overall export embodied carbon emissions are highly consistent with the export value. It indicates that as the scale of exports continues to expand, more carbon dioxide gas stays in the country, causing a broken environment in the Yangtze River Delta region. Since 2013, embodied carbon emissions have been decreasing, mainly due to carbon emission reduction measures. However, overall, the embodied carbon emissions from exports account for approximately 15% of the total carbon emissions in the region, which is still at a high level.
• Shanghai has been at the forefront of resource conservation, technological innovation and industrial structure optimization. The export scale of Zhejiang has been growing steadily. As e-commerce in Zhejiang is developing rapidly, its processing trade share is low, and the share of embodied carbon emissions in exports continues to decline. The export structure of Jiangsu and Anhui is on the sloppy side. Therefore, it is necessary to optimize and upgrade the export structure and advocate technological innovation.
• The scale effect is positively related to the embodied carbon in exports. It is also the main reason for the significant growth in emissions. However, as the export size of the Yangtze River Delta is in a period of steady growth, the contribution of this effect gradually decreases. In comparison, the structural effect is relatively small and differs greatly depending on the period. The technology effect is the most important driving force of carbon-emission reduction, and its effect is becoming more and more obvious with the technology advance.

5.2. Policy Implications

Based on these findings, the policy implications are as follows:

• We need to implement the concept of green trade. High-tech industries should be encouraged to export. Financial support should be given to enterprises that use new energy to replace highly polluting fossil energy sources. Instead, stricter carbon emission standards are required for high polluters. They are encouraged to produce their own or switch to import. Companies should be classified into different levels of energy consumption and pollution so that the government can impose different carbon tariffs according to the different levels.
• It is necessary to optimize the structure of export commodities. The only way is to optimize the export structure and change the crude export mode to a green trade mode. Meanwhile, we should upgrade the quality of export products through technological innovation, guide the shift of export commodities to high value-added and low energy consumption and promote the gradual transformation of export industries to technology-intensive and resource-green industries.
• We need to encourage companies to engage in technological innovation. There is an urgent need to actively encourage the development of new technologies and give full play to the curbing effect of technology effects on the embodied carbon emissions of export. Specific measures are to increase technological innovation and make full use of highly skilled personnel, improve regional R&D, focus on high-tech industries and modern service exports and further stimulate the technology effect of emission reduction.
• Promoting the use of clean energy is also a necessary measure. The authority can increase the financial investment of intensive green clean energy such as natural gas and electricity. At the same time, they may actively promote the adoption of new green energy, such as wind, solar and biomass, increase the proportion of green energy consumption and take a low-carbon, clean and sustainable path of green development.

5.3. Limitation and Future Research

There are still some limitations that can be improved in future research. I-O analysis, the method we have chosen, is not an advanced methodology and may have some drawbacks. However, our method fits our data, and the results are reliable. In future research, some new methods should be tried to analyze the more specific measurement of carbon emissions, such as a CGE model. In addition, only some of the major driving factors are selected for embodied carbon emission in this study. Therefore, all driving factors can be considered more comprehensively in future studies.