Decomposition of Industrial Carbon Emission Drivers and Exploration of Peak Pathways: Empirical Evidence from China

Abstract

Against the backdrop of increasing extreme weather events associated with global climate change, regulating carbon dioxide emissions, a primary contributor to atmospheric warming, has emerged as a pressing global challenge. Focusing on China as a representative case study of major developing economies, this research examines industrial carbon emission patterns during 2001–2022. Methodologically, it introduces an innovative analytical framework that integrates the Generalized Divisia Index Method (GDIM) with the Low Emissions Analysis Platform (LEAP) to both decompose industrial emission drivers and project future trajectories through 2040. Key findings reveal that:the following: (1) Carbon intensity in China’s industrial sector has been substantially decreasing under green technological advancements and policy interventions. (2) Industrial restructuring demonstrates constraining effects on carbon output, while productivity gains show untapped potential for emission abatement. Notably, the dual mechanisms of enhanced energy efficiency and cleaner energy transitions emerge as pivotal mitigation levers. (3) Scenario analyses indicate that coordinated policies addressing energy mix optimization, efficiency gains, and economic restructuring could facilitate achieving industrial carbon peaking before 2030. These results offer substantive insights for designing phased decarbonization roadmaps, while contributing empirical evidence to international climate policy discourse. The integrated methodology also presents a transferable analytical paradigm for emission studies in other industrializing economies.

1. Introduction

Since the 1990s, nations all over the world have started to consider the unsustainable development model that puts the environment at risk, and international organizations have adopted a series of emission reduction initiatives, providing guidance for action on green development in all countries. Since then, it has gradually become the consensus of all mankind to protect our home planet, on which mankind depends for its survival. The shift toward low-carbon development and climate change mitigation is driven by national policies (Raza and Shakeel, 2025) [1]. Since China’s reform and opening up began four decades ago, its industrial ecosystem has undergone comprehensive maturation, delivering extraordinary developmental outcomes that now serve as a primary engine for global economic expansion. However, the long-standing, resource-dependent industrial development model has caused great damage to the ecosystem and impaired the quality of economic development (Chen and Chen, 2018) [2]. During the 75th UN General Assembly (2020), China formally pledged to reach a carbon emission peak before 2030 and attain carbon neutrality by 2060, a strategic vision subsequently termed the “dual-carbon” objectives. These targets demonstrate China’s courage and determination to contribute to global emissions reduction. In 2022, China’s energy consumption per 10,000 CNY of GDP was 0.1% lower than in 2021, and China’s carbon dioxide emissions per 10,000 CNY of GDP were 0.8% lower than in 2021, reflecting the success of China’s energy conservation and emission reduction initiatives. The exploration of carbon emission drivers and reduction pathways has become increasingly in-depth, providing sufficient theoretical support for better realizing the “dual-carbon” goal.

In recent years, however, the pace of green transformation has slowed. Affected by the unfavorable economic situation at home and abroad, countries around the world have focused their goals on economic recovery and development, while the pursuit of green development goals has been diluted. In fact, continuous efforts to conserve energy and reduce emissions are essential for sustainable global economic development. The periodic slackening of carbon emission reduction targets is not conducive to the overall development of the world economy and does great harm to the well-being and long-term interests of people around the world. As the world’s largest developing country, China’s phased targets and pathways for emission reduction will have a profound impact on other countries. Therefore, it is necessary to start from the characteristics of China’s industrial energy consumption and carbon emissions and explore a Chinese-characteristic emission reduction path based on the Chinese scenario in order to provide beneficial policy insights for countries around the world.

The primary motivation behind this paper is to systematically analyze the current state of industrial energy consumption and carbon emissions while enhancing the comprehension of key drivers behind China’s industrial carbon footprint. By doing so, this study aims to provide valuable theoretical and policy implications for shaping future emission reduction strategies. This study’s key components, their interconnections, and potential academic advancements are outlined in three key dimensions:

First, as the primary source of CO₂ emissions, China’s industrial sector plays a pivotal role in achieving the nation’s dual-carbon objectives. In recent years, we have made great efforts in energy conservation and emission reduction in the industrial field, but we also face the inherent tension between industrial development and green transformation. Against this backdrop, this study examines the latest patterns of industrial energy use and trends of carbon emissions, providing critical insights into the current state of China’s industrial green development and offering valuable benchmarks for targeted, implementable decarbonization strategies.

Second, our analysis reveals distinct carbon reduction progress and divergent decarbonization trajectories between high-carbon and low-carbon industries, highlighting the need for differentiated emission reduction strategies. To identify policy priorities, we integrated five absolute indicators and ten relative indicators into the GDIM framework, investigating the drivers of industrial carbon emissions through three dimensions: scale, structure, and efficiency, which not only advances the empirical understanding of sectoral decarbonization dynamics but also provides actionable insights for targeted policy design.

Third, amid escalating global industrial competition, achieving a sustainable equilibrium between economic growth and green transition is imperative. While existing studies often prioritize emission reduction targets without adequately considering their economic feasibility, this paper introduces a more nuanced approach. We construct three primary policy scenarios—baseline, low-carbon, and comprehensive scenarios—each further divided into high-growth and low-growth sub-scenarios. This dual-layered framework allows for dynamic policy calibration, ensuring that emission reduction measures remain adaptable to varying economic conditions while preserving flexibility for strategic adjustments.

2. Literature Review

To elucidate global carbon mitigation pathways, contemporary studies employ pluralistic analytical perspectives to disentangle complex causation networks. The existing literature comprehensively examines the driving mechanisms and decarbonization trajectories, with a predominant focus on four key dimensions: economic growth patterns, industrial structure evolution, energy efficiency gains, and technological innovation capacities. Empirical evidence demonstrates the pivotal role of economic expansion in emission escalations. Xu et al. (2006) quantitatively analyzed the influencing factors of China’s per capita carbon emissions from 1995 to 2004 and found that economic development exponentially pushed the growth of China’s per capita carbon emissions and that the inhibiting effect of energy efficiency improvement and energy structure optimization on the growth of China’s carbon emissions could not offset the positive driving effect of economic development [3]. Shahbaz et al. (2013) found a robust positive role of economic growth and energy consumption on carbon emissions [4]. Subsequent analyses by Wang et al. (2020) across eleven developing economies revealed that both financial sector development and GDP growth consistently cause higher emission outputs under Paris Agreement (COP21) compliance scenarios [5]. More recently, Abam et al. (2024) identified capital intensity as an emerging accelerator of industrial carbon emissions [6]. Jiang et al. (2024) pointed out that the industrial structure endogenously determines the energy demand, the energy structure determines the mode of energy supply, and the industrial structure and energy structure together determine the level of carbon emissions [7]. In terms of industry, manufacturing is one of the main sources of carbon emissions. Shao et al. (2017) used the GDIM (Generalized Divisia Index Method) and found that the investment scale is the primary contributor to carbon emissions in the manufacturing sector, while investment carbon intensity and output carbon intensity are the key factors leading to carbon emission reduction [8].

To systematically identify key determinants of carbon emissions, researchers have employed diverse analytical frameworks. Jiang et al. (2025) utilized the STIRPAT model to examine urban emission drivers, including urbanization rates, affluence levels, energy intensity, and energy structure [9]. Simultaneously, decomposition analysis has emerged as a predominant methodological approach, with Ang et al. (2005)’s Logarithmic Mean Divisia Index (LMDI) gaining widespread adoption due to its residual-free computation and mathematical tractability in CO₂-related studies [10]. Subsequent applications include the following: Zhou et al. (2016) expanded LMDI, incorporating carbon density, energy intensity, economic development, industrial structure, and economic strength into the analysis framework [11]. Chen et al. (2020) identified sector-specific drivers across China’s four carbon-intensive pillars by applying LMDI [12]. Solaymani et al. (2019) used it to conduct a comparative analysis of carbon emission trends and characteristics of the transportation sector in seven major economies, including the United States, China, India, Russia, Japan, Brazil, and Canada [13].

While LMDI offers computational simplicity, its limitations lie in the fact that the decomposition result is affected by factors’ dependence, and the impact of absolute indicators and relative indicators cannot be included simultaneously. This prompted the development of the Generalized Divisia Index Method (GDIM) (Vaninsky, 2014; Chen et al., 2023), making complex interaction analysis among fundamental drivers and the decomposition of emission pathways possible [14,15]. Empirical applications demonstrate GDIM’s analytical superiority in various industrial sectors, including the mining sector (Shao et al., 2016) [16], thermal power generation sector (Yan et al., 2019) [17], and so on.

The global transition toward sustainable development has catalyzed extensive research into sector-specific decarbonization pathways. Recent studies have systematically examined emission reduction strategies across key industries, including the transportation sector (Wang and Wu, 2024) [18], aviation (Fageda and Teixido, 2022) [19], the steel industry (Song et al., 2025) [20], and the construction industry (Li et al., 2020) [21]. Concurrently, rigorous policy evaluations have revealed distinct mechanisms for emission mitigation. Liu et al. (2024) found that green finance policies can reduce corporate carbon emissions by easing financing constraints and promoting green innovation, particularly in emission-intensive industries [22]. The EU Emissions Trading System (ETS) implementation in aviation achieved sectoral emission reductions through market-based incentives (Fageda and Teixidó, 2022) [19]. Wang et al. (2023) believed that carbon trading policies primarily reduce regional carbon emissions through technological innovation and the optimization of energy structures [23].

Regarding the exploration of emission reduction pathways, Wang et al. (2020) argued that a single policy intervention cannot solve the dilemma of economic growth and environmental quality and that a combination of policies is significantly better than a single policy [24]. Mo et al. (2018) found that the paths to achieving carbon peak targets differ significantly depending on policy choices and that a hybrid policy of carbon pricing and non-fossil energy subsidies could achieve peak targets at lower costs [25]. There are also Chinese scholars who have made reasonable predictions of the evolutionary trend of future GHG emissions based on the current economic and environmental conditions and future policy directions in order to judge the realization of China’s environmental protection goals at various time points. Liu and Zhao (2011) systematically identified the key factors hindering the development of China’s low-carbon economy and used the LEAP (Low Emissions Analysis Platform) to quantitatively simulate the level of China’s low-carbon economic development in 2050; the prediction results showed that future energy demands and carbon emissions are very likely to fluctuate within a large range and that there are large differences in emission reduction effects under different scenarios [26]. Wang and Kuai (2022) used Monte Carlo simulation to study the dynamic trend of carbon emissions from the transportation sector and its peak potential under different scenarios as well as policy pathways [27].

The literature has revealed a relatively comprehensive picture of the factors influencing industrial carbon emissions and has made many useful explorations of the paths for China to better realize the “dual-carbon” goal. However, most of the literature focuses on the impact of single factors such as economic development, structural adjustment, technological innovation, efficiency improvement, and environmental regulation on energy conservation and emission reduction; few studies build a comprehensive model with multiple factors such as scale, structure, and efficiency to explore the trajectory of carbon emissions under the mutual constraints of multiple factors in the process of economic and social development. Meanwhile, studies on future energy-saving, emission reduction target-setting, and carbon emission trajectory prediction mainly focus on the industrial sector as a whole or a single industrial sub-sector, lacking in-depth explorations of differentiated energy-saving and emission-reduction-target paths for different types of industrial sectors.

This paper contributes to the literature through three primary innovations: (1) In terms of research methodology, the GDIM was combined with the LEAP modeling framework. The GDIM was used to comprehensively and systematically identify the driving factors of China’s industrial carbon emissions. Building on this decomposition, we developed a sector-specific LEAP model for China’s industrial system, calibrating scenario parameters based on historical industrial trends and expert projections. This hybrid approach enabled a quantitative simulation of emission trajectories under baseline, low-carbon, and integrated scenarios. (2) With regard to the content of this study, anchoring our analysis to China’s 2030 carbon peak commitment, we propose differentiated decarbonization pathways for high- and low-carbon industries under varying economic growth assumptions. This sector-stratified approach ensures targeted policy alignment with industrial realities. (3) In terms of empirical rigor, our dataset reflects China’s unique industrial energy consumption patterns, with key parameters scientifically calibrated to enhance model validity. This empirically grounded framework improves the applicability of energy-emission policy assessments in the Chinese context.

The remainder of this paper proceeds as follows: Section 3 presents a sectoral analysis of energy consumption patterns and carbon emissions across China’s industrial landscape. Section 4 details the research methodology, including the integrated GDIM-LEAP analytical framework and its empirical implementation and specific application. Section 5 discusses the key findings from both the decomposition analysis and scenario projections. The final chapter synthesizes the main conclusion and policy implications, providing targeted recommendations for China’s industrial decarbonization pathway.

3. Facts Characterizing Industrial Energy Consumption and Carbon Emissions in China

Carbon emissions primarily refer to the release of greenhouse gases, with carbon dioxide from fossil fuel combustion being the dominant component. Given that CO₂ emissions are intrinsically linked to energy consumption, this study collected data on coal, petroleum, and natural gas consumption, referring to the practice of Chen (2009) [28], adopting a measurement approach consistent with established methodologies in the field [29]. This was the basis for the data and methods used in this study. While prior research (Wang et al., 2017) [30] distinguishes between direct emissions (from fossil fuel combustion) and indirect emissions (from electricity consumption), this study focused exclusively on direct emissions to prevent potential double counting in the assessment. Specifically, data were compiled from authoritative sources, including the China Industry Statistical Yearbook and China Energy Statistical Yearbook, to serve as the foundation for quantifying industrial carbon emissions and analyzing their temporal trends from 2001 to 2022. Furthermore, data inconsistencies, including missing values and outliers in the original yearbook records, were addressed through appropriate imputation and exclusion procedures. In particular, some industries with serious missing data were excluded, and inconsistent industry classifications were merged or split according to the National Economic Industry Classification Standard issued by the Chinese government. The formal calculation was expressed as follows:

$$Z={\displaystyle \sum _{i=1}^{3}C{O}_2}={\displaystyle \sum _{i=1}^3{E}_i}\times NC{V}_i\times CE{F}_i\times CO{F}_i\times (\mathit{44}/\mathit{12})$$

(1)

$Z$ represents the total carbon emissions in tons from each energy type $i$, including coal, crude oil, and natural gas consumption. $E_i$ indicates the energy consumption. $NC{V}_i$ (Net Calorific Value) denotes the average lower heating value in kJ/kg, which refers to the heat released by a unit mass of fuel when it is completely burned. $CE{F}_i$ (Carbon Emission Factor) denotes the carbon content per unit calorific value in tons/trillion joules, which means the mass of the carbon element corresponding to each unit of calorific value when the fuel is burned. $CO{F}_i$ (Carbon Oxidation Factor) represents carbon oxidation rate, which refers to the proportion of carbon that is completely oxidized to carbon dioxide. The value 44/12 represents the stoichiometric ratio between the molecular weight of carbon dioxide (CO₂) and the atomic weight of carbon (C), serving as a critical conversion factor for quantifying CO₂ emissions based on carbon content.

As shown in Figure 1, overall, since 2001, with the advancement of China’s economic development process, industrial carbon dioxide emissions have experienced a thirteen-year-long rapid growth. In 2014, China’s industrial carbon dioxide emissions fell by 1.39%, which was the first decline since 2001. After 2014, China’s industrial carbon dioxide emissions showed a small downward trend for three consecutive years before showing a small rebound in 2016, but the overall growth rate was significantly lower than that of the previous decade. This shows that since 2012, China’s policies related to eco-environmental protection have had a significant effect, the pace of industrial green transformation has accelerated significantly, and the work of reducing pollution and carbon emissions has made stage-by-stage progress.

Industrial carbon emissions exhibit significant inter-sectoral disparities. Empirical analysis reveals that the top five carbon-intensive industries collectively contribute over 85% of total industrial CO₂ emissions, while the top ten account for more than 95%. This heavy skewness underscores the pivotal role of industrial structure in shaping aggregate emission trends. Consequently, targeted sectoral adjustments, particularly within these high-emission industries, will critically influence the trajectory of industrial decarbonization and the attainment of China’s “dual-carbon” objective.

Sectoral analysis reveals that a select group of energy-intensive industries, including electric power generation, petrochemicals, iron and steel production, nonferrous metals, building materials, and paper manufacturing, collectively dominate industrial carbon emissions, representing the largest contributors to CO₂ output. Among them, the electricity and gas supply sector ranks as the largest carbon emitter among all industrial sectors (

Table 1. Carbon emission share of China’s industry by sector (2010 and 2022).

Rank	2010	Percentage Share	2022	Percentage Share
1	Electricity, heat generation, and supply	41.22%	Electricity, heat generation, and supply	44.68%
2	Petroleum, coal, and other fuel processing industries	24.65%	Petroleum, coal, and other fuel processing industries	28.95%
3	Ferrous metal smelting and rolling industry	7.65%	Manufacture of chemical materials and products	6.37%
4	Nonmetallic mineral products industry	6.46%	Ferrous metal smelting and rolling industry	5.46%
5	Coal mining and washing industry	6.24%	Nonmetallic mineral products industry	4.91%
	Top 5 industries cumulative	86.21%	Top 5 industries cumulative	90.36%
6	Manufacture of chemical materials and products	5.84%	Non-ferrous metals smelting and rolling	2.80%
7	Non-ferrous metals smelting and rolling	1.57%	Coal mining and washing industry	2.72%
8	Paper and paper products industry	1.16%	Paper and paper products industry	0.61%
9	Oil and gas extraction industry	0.99%	Oil and gas extraction industry	0.59%
10	Textile industry	0.71%	Gas production and supply industry	0.49%
	Top 10 industries cumulative	96.48%	Top 10 industries cumulative	97.57%

Data sources: The data are from the China Statistical Yearbook, China Industrial Statistical Yearbook, and China Energy Statistical Yearbook. The specific values were calculated by the author.

). This phenomenon can be systematically attributed to the following factors: First, fossil fuels continue to dominate the energy mix for thermal power generation. Despite accelerated transitions in the sector’s energy structure in recent years, the inherent intermittency of renewable energy sources mandates ongoing reliance on thermal generation for ancillary services and system regulation. Furthermore, winter heating demands sustain elevated carbon emissions with distinct seasonal fluctuations. To address these structural and persistent decarbonization challenges, a dual-focused strategy targeting both existing infrastructure and new capacity is essential. On the one hand, energy efficiency must be retrofitted and operational flexibility enhancements must be implemented for the current infrastructure. Infrastructure modernization is required to improve renewable energy absorption capacity. On the other hand, overcoming the critical barrier of techno-economic feasibility is paramount. This demands intensified R&D investments in breakthrough technologies such as green hydrogen and advanced energy storage systems.

Carbon dioxide mainly comes from three major fossil energy sources, namely, coal, crude oil, and natural gas. From the carbon emission contribution of the three major fossil energy sources in China’s industry in Figure 2, coal has the largest carbon emission contribution, followed by crude oil, and natural gas is the lowest. Among them, the contribution of coal carbon emissions is more than 70% on average, which reached a peak of 81.76% in 2013 and then gradually declined to 75.52% in 2022. This indicates that among the three traditional fossil energy sources, coal combustion is undoubtedly the largest source of carbon emissions from industrial production.

As illustrated in Figure 3, the structural composition of fossil energy consumption in China exhibited dynamic shifts between 2000 and 2022. Coal, the dominant energy source, initially increased its share from 70.90% (2000) to a peak of 76.72% (2008), followed by a general decline (despite intermittent fluctuations) to 69.72% by 2022. Concurrently, crude oil’s proportion gradually decreased from 26.72% to 22.33%, while natural gas demonstrated sustained growth, rising from 2.38% to 7.95% over the same period. These trends reflect a gradual transition toward a cleaner fossil energy mix, though coal remains the primary component of China’s energy consumption. Therefore, to align economic growth with emission reduction goals, near-term strategies must prioritize enhancing energy utilization efficiency and accelerating the energy structure’s low-carbon transition.

Building upon the industrial reporting and verification framework established by China’s Ministry of Ecology and Environment for carbon emissions and utilizing the sectoral CO₂ emission data quantified in preceding analyses, this study classified 36 industrial subsectors into two distinct groups: 13 high-carbon and 23 low-carbon industries, ranked by emission magnitude (see Appendix A). Figure 4 reveals divergent trends in carbon emission intensity across China’s industrial sectors from 2001 to 2022. While low-carbon industries maintained a steady decline in emission intensity, high-carbon industries exhibited slowing decarbonization progress, with intensity levels actually rising in recent years. This divergence has contributed to an overall deceleration in industrial emission intensity reduction, even showing signs of potential rebound.

The data demonstrate that China’s industrial development has gradually reduced its dependence on energy consumption and CO₂ emissions. This trend reflects structural improvements in industrial organization, more rational energy consumption patterns, and enhanced production efficiency through technological upgrades, collectively indicating the effectiveness of China’s energy conservation and emission reduction policies. Notably, high-carbon industries continue to dominate total industrial emissions. Achieving sustainable industrial development will therefore require the strategic optimization of industrial structure through reduced reliance on high-carbon sectors and the targeted acceleration of low-carbon transition in these emission-intensive industries.

4. Research Design

4.1. Methodology for Decomposition of Carbon Emission Drivers

The Generalized Divisia Index Method (GDIM) represents an advanced decomposition approach derived from the fundamental Kaya identity. As an extension of traditional Logarithmic Mean Divisia Index (LMDI) decomposition, GDIM offers superior analytical flexibility by simultaneously incorporating both absolute and relative factors in driver decomposition analysis. This methodological advancement enables a more comprehensive examination of the complex determinants underlying changes in target variables. Building upon established decomposition frameworks, this study applied GDIM to systematically analyze industrial CO₂ emission drivers in China. Utilizing sector-specific energy consumption data and emission measurements, we constructed a GDIM model following Vaninsky’s approach [14]. The model decomposes emission variations into multiple interacting factors, which provides robust empirical evidence to inform China’s emission reduction strategy development and climate policy optimization.

The main idea of the GDIM is to change the driving factor, capturing the partial change of a target variable with the change in factor variables. Thus, according to the basic principles of the GDIM, it is first assumed that the target variable $Z$ is the product of the factor $X_1,X_2\dots X_n$, as shown in Formula (2). The factor indicators are interconnected in the following way to form a system of factor interconnections, shown in Equation (3), which is rewritten in matrix form to obtain (4).

$$Z=f(X)=f(X_1,X_2\cdot \cdot \cdot X_n)$$

(2)

$${\varphi }_j(X_1,X_2\dots X_n)=0,j=1,2\cdot \cdot \cdot k$$

(3)

$$\varphi (X)=0$$

(4)

$$\Delta Z={\displaystyle {\int }_{L}d}Z={\displaystyle {\int }_L{f}_1^{\prime }}d{X}_1+{\displaystyle {\int }_L{f}_2^{\prime }}d{X}_2+\cdots +{\displaystyle {\int }_L{f}_n^{\prime }}d{X}_n={\displaystyle {\int }_L\nabla }{Z}^{T}dX$$

(5)

The method assumes the continuous change of factorial indicators over time; thus, there is no need to order the factorial indicators. In Formula (5), $L$ is the time span and $\nabla Z=\langle f_1^{\prime },f_2^{\prime }\cdot \cdot \cdot f_n^{\prime }\rangle$ is the column gradient vector of the function $Z=f(X)=f(X_1,X_2\cdots X_n)$. Map changes in the target variable in vector form on the surface are defined by the set of factorial interconnected equations:

$$\Delta Z[X\mid \varphi ]={\displaystyle {\int }_L\nabla }{Z}^T(I-{\varphi }_X{\varphi }_X^{+})dX$$

(6)

where $T$ denotes the transpose, $I$ denotes the unit matrix, ${\varphi }_X$ denotes the Jacobi matrix formed by the partial differentiation of the set of factorial interconnected equations, and $+$ denotes the generalized inverse matrix. If the Jacobi matrix ${\varphi }_X$ is a column that is linearly uncorrelated, then there will be ${\varphi }_X^{+}={({\varphi }_X^T{\varphi }_X)}^{-1}{\varphi }_X^T$.

Specifically, this study broke down the drivers of industrial sub-sector CO₂ emissions as follows.

$$C{O}_2={\displaystyle \frac{C{O}_2}{R}}\times R={\displaystyle \frac{C{O}_2}{E}}\times E={\displaystyle \frac{C{O}_2}{P}}\times P={\displaystyle \frac{C{O}_2}{I}}\times I={\displaystyle \frac{C{O}_2}{T}}\times T$$

(7)

We then simplified the above equation to Equation (8):

$$Z=X_2\times X_1=X_4\times X_3=X_6\times X_5=X_8\times X_7=X_{10}\times X_9$$

(8)

In summary, the drivers of industrial carbon emissions were decomposed into 5 absolute and 10 relative indicators, which are summarized in

Table 2. Summary of decomposition indicators for carbon emission drivers.

Target Variables	CO2	Z	Carbon Dioxide Emissions (tons)
Absolute indicators	R	x1	Revenue from main operations (million CNY)
	E	x3	Total energy consumption (tons of standard coal)
	P	x5	Average number of workers employed by industrial enterprises above designated size (10,000 persons)
	I	x7	Investment in fixed assets (million CNY)
	T	x9	R&D expenditure (million CNY)
Relative indicators	CO2/R	x2	Output carbon intensity
	CO2/E	x4	Carbon intensity of energy consumption
	CO2/P	x6	Employment carbon intensity
	CO2/I	x8	Investment carbon intensity
	CO2/T	x10	R&D carbon intensity
	R/P	x11	Output per employee (labor productivity)
	E/R	x12	Energy use per unit of output (Energy intensity)
	T/I	x13	R&D expenditure share in fixed asset investment (R&D expenditure intensity)
	R/I	x14	Fixed asset investment productivity (investment efficiency)
	R/T	x15	Economic output per unit of R&D investment (R&D efficiency)

below. Five absolute indicators, $R,E,P,I,T$ were mainly used to reflect the size of the industry, and the impact of the scale of factor inputs, including energy, labor, fixed asset investment, and R&D funding, on CO₂ emissions was part of the scale effect. Relative indicators, including $X_2=C{O}_2/R$, $X_4=C{O}_2/E$, $X_6=C{O}_2/P$, $X_8=C{O}_2/I$, and $X_{10}=C{O}_2/T$, mainly reflected the change in the degree of the low-carbon development of the industry driven by factor inputs, which was part of the intensity effect.

In addition, $X_{11}=X_6/X_2=R/P$, $X_{12}=X_2/X_4=E/R$, $X_{13}=X_8/X_{10}=T/I$, $X_{14}=X_8/X_2=R/I$, and $X_{15}=X_{10}/X_2=R/T$ were constructed to reflect the impact of energy dependence on production, R&D intensity, and factor input–output efficiency on CO₂ emissions. The main data were primarily sourced from the official website of the National Bureau of Statistics, China Statistical Yearbook, China Industrial Statistical Yearbook, and CEIC database.

Further transforming the selected decomposition metrics into a system of factor-interconnected equations:

$$\begin{array}{c}Z=X_1{X}_2\hfill \\ {\phi }_1:X_1{X}_2-X_3{X}_4=0\hfill \\ {\phi }_2:X_1{X}_2-X_5{X}_6=0\hfill \\ {\varphi }_3:X_1{X}_2-X_7{X}_8=0\hfill \\ {\phi }_4:X_1{X}_2-X_9{X}_{10}=0\hfill \\ {\phi }_5:X_1-X_5{X}_{11}=0\hfill \\ {\phi }_6:X_3-X_1{X}_{12}=0\hfill \\ {\phi }_7:X_9-X_7{X}_{13}=0\hfill \\ {\phi }_8:X_1-X_7{X}_{14}=0\hfill \\ {\varphi }_9:X_1-X_9{X}_{15}=0\hfill \end{array}$$

(9)

We obtained the column gradient vector $\nabla Z$ and Jacobi matrix ${\varphi }_X$, using the R and completed a decomposition of the drivers of carbon emissions.

$$\nabla Z={\left(\begin{array}{c}{X}_2,X_1,0,0,0,0,0,0,0,0,0,0,0,0\end{array}\right)}^T$$

(10)

$${\varphi }_2={\left(\begin{array}{ccccccccccccccc}{X}_2& X_1& -X_4& -X_3& 0& 0& 0& 0& 0& 0& 0& 0& 0& 0& 0\\ X_2& X_1& 0& 0& -X_6& -X_5& 0& 0& 0& 0& 0& 0& 0& 0& 0\\ X_2& X_1& 0& 0& 0& 0& -X_8& -X_7& 0& 0& 0& 0& 0& 0& 0\\ X_2& X_1& 0& 0& 0& 0& 0& 0& -X_{10}& -X_9& 0& 0& 0& 0& 0\\ 1& 0& 0& 0& -X_{11}& 0& 0& 0& 0& 0& -X_5& 0& 0& 0& 0\\ -X_{12}& 0& 1& 0& 0& 0& 0& 0& 0& 0& 0& -X_1& 0& 0& 0\\ 0& 0& 0& 0& 0& 0& -X_{13}& 0& 0& 0& 0& 0& -X_7& 0& 0\\ 1& 0& 0& 0& 0& 0& -X_{14}& 0& 0& 0& 0& 0& 0& -X_7& 0\\ 1& 0& 0& 0& 0& 0& 0& 0& -X_{15}& 0& 0& 0& 0& 0& -X_9\end{array}\right)}^T$$

(11)

4.2. Scenario Analysis and Carbon Emission Forecasts

The Low Emissions Analysis Platform (LEAP) is an energy–environmental accounting modeling tool that can be applied to detail technologies and activities for energy demand and environmental impact. This study constructed a LEAP analysis framework under the scenario analysis method for the Chinese industry. Parameters were set according to the policy orientation, industry carbon emission indicators, industry energy structure, energy technology development level, and experts’ and scholars’ outlook on the future socio-economic development trend of China. Combined with the results of the decomposition of carbon emission drivers by the GDIM method in the previous section, the impact of the combined effect of multiple factors, such as scale, structure, and efficiency, on environmental development was considered comprehensively. Finally, the future industrial carbon dioxide emissions and the size of the peak carbon value under different scenarios were derived to scientifically predict the effect of energy-saving and emission reduction policies.

This study established LEAP modeling with the following structure. First, sectoral classification: the energy demand sector was categorized into industrial and non-industrial components. To enhance modeling precision, the industrial sector was further disaggregated into high-carbon and low-carbon subsectors based on their distinct energy consumption patterns, intensity levels, and emission characteristics. Next, temporal parameters: the model adopted 2022 as the base year, with a projection horizon spanning 2022–2040 at annual intervals. The framework estimated future energy demand and associated CO₂ emissions through the following formulations:

$$ENDEMAN{D}_{t,k}={\displaystyle \sum _{i}ACTILE{V}_{t,i}*ENINTENSIT{Y}_{t,i,k}}$$

(12)

$$CARBONEM{I}_t={\displaystyle \sum _{k}E}NDEMAN{D}_{t,k}*EMIFACTO{R}_k$$

(13)

where $t$ denotes the forecast period, $i$ shows the industry type, and $k$ indicates the energy type. To avoid double-counting problems, only three primary energy sources, coal, crude oil, and natural gas, were considered for energy consumption types. $ENDEMAN{D}_{t,k}$ represents the total consumption of energy $k$ by industry in period $t$. $ACTILE{V}_{t,i}$ denotes the level of activity in the industrial sub-sector in period $t$. $ENINTENSIT{Y}_{t,i,k}$ shows the energy consumption per unit of economic activity $k$ in the industrial sub-sector in period $t$. $CARBONEM{I}_t$ indicates the total industrial CO₂ emissions in period $t$. $EMIFACTO{R}_k$ denotes the carbon emission factor for energy source $k$. Instead of measuring carbon emissions with the help of LEAP’s default technical environmental database, carbon emissions were estimated based on the carbon emission factors of different energy sources measured by us in the previous section. Industrial carbon emissions depended on sub-sector activity levels, energy use per unit activity, and energy source emission factors, as shown in Equations (12) and (13).

The framework of the research is shown below(Figure 5).

5. Results and Analyses

5.1. Decomposition of Carbon Emission Driving Factors

Given data limitations and economic volatility considerations, this study adopted differentiated time frames for decomposition analysis: 2001–2020 for aggregate industrial emissions and 2006–2015 for subsector-level analysis. This approach ensured methodological robustness while accommodating data availability constraints. The corresponding decomposition results are presented in

Table 3. Decomposition of industrial carbon emission driving factors.

Timescale			2001–2020	2001–2020	2006–2015	2006–2015
	Industry Type		Industry (Revenue from Main Operations)	Industry (Value Added)	High-Carbon Industries	Low-Carbon Industries
Indicator Type
Absolute indicators	R	x1	2.16	1.50	0.43	0.55
	E	x3	−0.20	−0.18	−0.11	−0.14
	P	x5	0.63	0.64	0.15	0.16
	I	x7	−0.01	−0.01	0.00	0.06
	T	x9	0.21	0.29	0.04	0.08
Relative indicators	CO2/R	x2	0.58	0.53	0.15	0.14
	CO2/E	x4	0.45	0.45	0.29	0.33
	CO2/P	x6	−0.21	−0.21	−0.14	−0.17
	CO2/I	x8	0.46	0.46	0.21	0.29
	CO2/T	x10	−0.21	−0.21	−0.09	−0.14
	R/P	x11	−1.30	−0.68	−0.21	−0.25
	E/R	x12	−0.02	−0.01	−0.01	−0.03
	T/I	x13	0.00	0.00	−0.01	−0.01
	R/I	x14	−0.01	−0.02	−0.01	−0.02
	R/T	x15	−0.01	−0.02	0.00	0.00

.

In terms of industry as a whole or industry sectors, both size and main business income were the most important CO₂ emission-promoting factors among the 15 decomposition factors examined. Compared to high-carbon industries, low-carbon industries had a more pronounced effect on contributing to CO₂ emissions from their main business income. This means that the development of low-carbon sectors, which currently account for a higher share of main business income in China’s industry, requires energy support. In the short term, strong emission reduction requirements for low-carbon industries that bring greater economic benefits will, to some extent, come at the expense of economic growth. However, due to the larger base of carbon dioxide emissions from high-carbon industries, the expansion of the scale of high-carbon industries contributes more to the absolute value of carbon dioxide emissions. This reflects that the internal structure of industries has a significant impact on the total amount of industrial CO₂ emissions in China.

Enhanced labor productivity and R&D efficiency demonstrated significant negative correlations with CO₂ emissions across industries, confirming efficiency improvement as a critical decarbonization pathway. While low-carbon industries exhibited stronger proportional emission reductions from efficiency gains, high-carbon sectors contributed larger absolute reductions due to their greater emission baselines. Notably diverging from conventional findings, this study reveals that both total energy consumption and energy intensity (consumption per unit output) exhibited negative correlations with CO₂ emissions, a counterintuitive result that reflects the countervailing potential where cleaner energy structures may offset or outweigh scale effects and the mediating role of structural decarbonization in China’s energy system. The observed inverse relationship between energy consumption and emissions growth reflects China’s ongoing transition toward greener energy systems, suggesting that structural transformation is gradually decoupling energy demand growth from emission trajectories.

Analyzing the data in Table 3, carbon intensity of output, energy consumption, and investment were positive contributors of carbon dioxide emissions, while carbon intensity of employment and R&D had a certain decreasing effect. Employment and R&D investment growth (conventional indicators of industrial expansion) exhibited strong positive correlations with CO₂ emissions; however, their underlying quality dimensions’ improvement, including human capital enhancement and technological advancement, did not demonstrate significant mitigation potential.

To summarize, industrial scale expansion and profitability pursuit through factor input intensification were the primary drivers of emission growth, while the underutilized decarbonization effects from human capital enhancement (via labor quality improvement) and technological advancement (via R&D efficacy advancement) were not fully manifested. The growth of energy consumption and its intensity under the improvement of the energy consumption structure was instead an important factor contributing to the reduction of emissions. In addition, sectoral composition dynamics within industrial systems were was an important aspect affecting carbon dioxide emissions. These findings collectively underscore the critical interplay between production scale, input quality, and structural transformation in shaping emission trajectories. Future mitigation strategies should prioritize factor input quality upgrading over quantity expansion, the strategic alignment between energy consumption growth and clean energy penetration, and industrial restructuring toward low-carbon subsectors.

5.2. Scenario Analysis and Peak Carbon Forecasts

5.2.1. Parameter Adjustment Carving

(1)

Size of the Economy

The trajectory of carbon emissions is intrinsically linked to a nation’s economic development phase and growth patterns. China’s transition from rapid expansion to high-quality development implies a balanced growth structure and a stable growth rate (Zhao et al., 2019) [31]. Projections from authoritative sources converge on a gradual economic slowdown: China Energy Research Institute (2030 Outlook) predicts 5.5% avg. GDP growth (2021–2025), declining to 4.5% (2026–2030) [32]. State Council Development Center predicts 4–6% medium-low growth (2020–2030), slowing to 2–4% (2031–2050). CASS Economic Research Institute predicts 5.42% (2021–2025) and 4.92% (2026–2030) potential growth (Tang et al., 2020) [33]. Building on this consensus, we established two distinct development pathways: (1) High-growth scenario: from 5.5% (2024–2030) to 5% (2031–2040). (2) Low-growth scenario: from 5% (2024–2030) to 4.5% (2031–2040). This captured the probable range of China’s economic evolution while maintaining consistency with national policy orientations and research benchmarks.

(2)

Industrial Structure

China’s ongoing industrial restructuring reflects a marked shift toward advanced rationalization, with projections indicating sustained growth in the tertiary sector’s dominance. According to the China Energy Outlook 2030, the service sector is expected to account for 65% of GDP by 2030, while the secondary industry’s share will account for 30.5%. After excluding construction activities, the manufacturing sector’s proportion is projected to decline to approximately 26%.

Given the significant disparity in emission intensity between high-carbon and low-carbon industries, strategic resource allocation should prioritize structural optimization to rebalance industrial composition toward cleaner, higher-value-added subsectors and energy transition to ensure that energy supply systems align with end-user demand while minimizing carbon intensity. This dual approach leverages structural change as a key decarbonization mechanism, complementing technological efficiency improvements.

(3)

Energy Intensity

Projections from China Energy Outlook 2030 indicate a deceleration in China’s energy consumption growth, with annual averages of 1.3% (2021–2025) and 0.7% (2026–2030), signaling a gradual decoupling of economic growth from energy demand. However, sectoral analyses reveal divergent trajectories, namely, high-carbon industries are expected to maintain substantial fossil fuel dependence in the near-to-medium term, while low-carbon industries demonstrate greater potential for clean energy substitution. Building on these trends, this study established energy intensity parameters through historical benchmarking (2001–2022 sectoral energy consumption data), taking short-term technological constraints and medium-to-long-term decarbonization potential into account. The resulting framework ensures staged, achievable targets for energy efficiency improvements across industrial subsectors and energy types.

5.2.2. Scenario Setting Description

The scenario-setting module set up three main scenarios: the baseline scenario, the low-carbon scenario, and the composite scenario. Each scenario contained two sub-scenarios, one for high growth and one for low growth. This study established three primary analytical scenarios, including the baseline scenario, the low-carbon scenario, and the integrated scenario. Each of them incorporated high-growth and low-growth variants, serving as the reference case for policy evaluation. (1) The baseline scenario assumed (a) the continuation of current economic growth patterns, (b) no additional emission reduction policies, and c) fixed industrial structures, energy mixes, and technological levels. (2) The low-carbon scenario incorporated moderate structural adjustment policy targeting, including industrial upgrading and sectoral transformation, while simultaneously maintaining baseline energy parameters and excluding major technological breakthroughs in energy efficiency. (3) The integrated scenario represented the implementation of comprehensive mitigation measures, including industrial restructuring, energy mix optimization, technological innovation in energy systems, and sector-specific energy intensity adjustments. This scenario rationalized the energy intensity parameters by industry category and energy type to ensure that the goal of peak carbon was basically achieved by 2030. The scenarios and parameter settings are shown in

Table 4. Summary of scenario settings and parameter settings.

Scenario		Baseline Scenario				Low-Carbon Scenario
Sub-Scenario		High Growth		Low Growth		High Growth		Low Growth
Stage		2025–2030	2031–2040	2025–2030	2031–2040	2025–2030	2031–2040	2025–2030	2031–2040
Economic Growth Rate		5.50	5.00	5.00	4.50	5.50	5.00	5.00	4.50
Industrial Structure (Percentage of Industry)		(2024, 30.05)				(2024, 30.05; 2030, 26.00; 2040, 26.00)
Industrial Ssector Structure (Proportion of High-Carbon Industries)		(2022, 41.34)				(2022, 41.34; 2030, 34.00; 2040, 28.00)
Average Annual Growth Rate of Energy Intensity		0.00				0.00
Scenario		Integrated Scenario
Sub-Scenario		High Growth				Low Growth
Stage		2025–2030		2031–2040		2025–2030		2031–2040
Economic Growth Rate		5.50		5.00		5.00		4.50
Industrial Structure (Percentage of Industry)		(2024, 30.05; 2030, 26.00; 2040, 26.00)
Industrial Sector Structure (Proportion of High-Carbon Industries)		(2022, 41.34; 2030, 36.00; 2040, 28.00)
Industry Type		High Carbon	Low Carbon	High Carbon	Low Carbon	High Carbon	Low Carbon	High Carbon	Low Carbon
Average Annual Growth Rate of Energy Intensity	coal	−0.02	−0.10	−0.05	−0.20	−0.01	−0.08	−0.04	−0.18
	crude oil	−0.01	0.00	−0.03	0.00	−0.01	0.00	−0.02	0.00
	gas	0.06	0.01	0.00	0.01	0.03	0.01	0.00	0.00

.

5.2.3. Carbon Emission Projections

(1)

Analysis of projected results

Industrial carbon dioxide emissions increased sharply each year in the baseline scenario and were projected to reach 17.0 billion tons by 2030 and 27.4 billion tons by 2040 under high-growth conditions, which is about three times as much as the total industrial carbon emissions in 2022. This value was predicted to decrease slightly under the low-growth scenario, but China’s 2030 carbon peaking target would be far from being achieved under the condition that economic structure and energy intensity remained unchanged.

Structural reforms in the low-carbon scenario yielded substantially different pathways. Emission growth rates decelerated markedly compared to those of the baseline scenario. China’s industrial CO₂ declined by 25.1% (2030) and 45.1% (2040) under the low-carbon scenario, making structural transformation the key to realizing China’s energy-saving and emission reduction goals. The low-growth scenario emitted 357.3 Mt CO₂ less than the high-growth scenario in 2030 and 1.1 billion tons less in 2040. Nevertheless, neither structural adjustments alone nor growth moderation sufficed to achieve the 2030 industrial emission peak, underscoring the necessity for more transformative interventions.

As can be seen from the forecast results in Figure 6, the integrated scenario could control industrial CO₂ emissions within 11.6 billion tons in 2030, and this value was controlled below 11.2 billion tons under the low-growth scenario. The overall industrial carbon emissions level was predicted to drop to below 9.1 billion tons in 2040, which is lower than the level in 2020. Compared to the baseline scenario, the combined scenario reduced emissions by 31.9% in 2030 and 67.2% in 2040.

As shown in Figure 7, under the high-growth integrated scenario, coal’s contribution to fossil fuel-related CO₂ emissions declined from 74.1% (2022) to 71.2% by 2030. Driven by the accumulation of technological advancements in energy systems and accelerated structural shifts, it was expected that coal’s contribution to the carbon emissions of the three types of primary fossil energy sources would be greatly reduced to 63.1% by 2040.

(2)

“Peak Carbon 2030” target-setting statement

The established decarbonization roadmap sectoral transition targets were set as follows: Overall industrial share reduced to 26% by 2030. High-carbon industries’ proportion decreasing to 36% (2030) and further to 28% (2040). For high-carbon industries, given China’s coal-dominant energy infrastructure and the long-cycle characteristics of energy technology change, it is difficult to realize the transformation of energy consumption structure and the improvement of energy utilization efficiency in the short term. Hence, an average annual coal energy intensity reduction rate of 2% was set for the period up to 2030. With the goal of crossing the historical inflection point of China’s industrial carbon emissions from increasing to decreasing and gradually bringing down the total industrial carbon emissions, it was necessary to realize an accelerated 5% annual reduction in coal energy intensity during 2030–2040, enabled by technological breakthroughs in energy systems.

Given that crude oil accounts for less than one-third of industrial fossil fuel consumption and contributes under 25% of CO₂ emissions relative to coal, we set 1% and 3% annual reductions in crude oil energy intensity during 2025–2030 and 2030–2040, respectively. This was done considering appropriately relaxing the requirements for the declining rate in the energy intensity of crude oil and the principle of ensuring the rationality of policy intensity. This showed the stepwise characteristic of policy strength from weak to strong.

Due to natural gas’s superior environmental profile among fossil fuels, strategic prioritization of its industrial utilization was warranted. Based on the average annual coal energy carbon intensity reduction rate of 12% from 2001 to 2022 in the low-carbon sector, we found that industrial sectors with low energy dependence were moving away from coal dependence and realizing clean energy substitution. Therefore, the average annual coal energy intensity reduction rates of 2025–2030 and 2030–2040 for low-carbon industries were set to be 10% and 20%, respectively.

It is worth noting that the integrated scenarios presented herein estimate only the minimum parameter thresholds required for achieving peak industrial carbon emissions by 2030. Accelerated realization of carbon peaking would necessitate expedited industrial restructuring beyond current projections, breakthrough innovations to enhance energy efficiency at unprecedented rates, and an accelerated shift from high-carbon to carbon-neutral energy infrastructure. These interventions would collectively enable anticipatory emission stabilization prior to the 2030 target horizon.

5.2.4. Research Summary

First of all, this study acknowledges the phased achievements in China’s energy conservation and emission reduction initiatives. However, as critically noted by Aziz et al. (2023) [34], the full potential of environmental policy instruments, particularly renewable energy adoption and environmental tax policies, remains underutilized, necessitating further governmental refinement of the policy framework to maximize ecological effectiveness.

Second, the literature consistently demonstrates a robust, positive association between economic expansion and carbon emissions (Chen et al., 2020; Solaymani, 2019; Shao et al., 2016; Wang and Yang, 2015) [12,13,16,35], indicating that economic growth typically induces higher emission outputs, a finding congruent with our driver analysis. However, sectoral exceptions exist: Wang and Wu (2024) revealed that industrial scaling in Silk Road Economic Belt transportation systems did not elevate sectoral emissions, attributable to substantial energy structure improvements [18]. Zhou et al. (2016) discovered through index decomposition analysis that the decoupling effect of industrial energy carbon emissions from economic growth in China is not significant, mainly attributed to the coal-dominated energy structure and suboptimal emission reduction technology levels [11]. These findings collectively underscore that China’s decarbonization pathway necessitates a dual focus on energy system restructuring and energy efficiency improvement, which is also consistent with the basic conclusion of this study.

Third, departing from conventional single-sector analyses (e.g., transportation, mining, aviation), this study adopted a holistic approach by examining China’s industrial sectors, given their substantial contribution to national emissions. We implemented a novel classification framework, segmenting industries into high- and low-carbon categories based on energy consumption intensity profiles. Through scenario modeling incorporating baseline, low-carbon, and integrated development pathways, we systematically evaluated peak carbon trajectories under varying economic growth regimes. While Chen et al. (2020) cautioned about potential delays in China’s emission peaking due to lagging decarbonization in key sectors [12], our analysis presents an optimistic yet evidence-based perspective: the strategic coordination of three critical levers, including economic restructuring, energy transition, and efficiency improvements.

6. Conclusions and Policy Implications

6.1. Conclusions

On the basis of revealing the facts about the characteristics of China’s industrial energy consumption and carbon emissions, this study applied the GDIM to systematically and comprehensively decompose the key drivers of sectoral emissions. Incorporating these determinants into China’s industrial LEAP analysis framework, setting key scenario parameters, we quantitatively predicted industrial CO₂ emission trajectories, assessing peak potential under dynamic interactions of economic growth, structural shifts, and efficiency gains. The conclusions are summarized below:

First, China’s industrial sector has demonstrated accelerated progress in green transition, achieving significant decoupling of economic growth, energy consumption, and CO₂ emissions. While coal remains the predominant contributor to industrial carbon emissions—accounting for 75% of the total—the adoption of cleaner alternatives such as natural gas remains limited, representing only 5% of emissions. Carbon emissions are unevenly distributed across industries within the industrial sector, and the slowdown in the green and low-carbon transformation of a few high-carbon industries has slowed down the overall green and low-carbon development process of China’s industry.

Secondly, the expansion of industrial scaling driven by factor inputs is the most significant contributor to the growth of industrial carbon emissions. In contrast, the improvement of human capital through labor inputs and the effects of technological progress driven by R&D inputs have failed to demonstrate significant carbon emission reduction effects. With the increase in energy consumption and its intensity, the improvement of the energy consumption structure is the key factor in reducing carbon emissions. Furthermore, intra-industrial structural reallocation exerts substantial influence over aggregate emission trajectories.

Thirdly, neither the baseline scenario nor the low-carbon scenario—confined to economic restructuring alone—can achieve China’s 2030 industrial emission peak target. Only through an integrated strategy combining moderate industrial restructuring with transformative energy technological advancements, including large-scale energy mix decarbonization and systematic efficiency gains, can the sector successfully reach the critical inflection point marking sustained emission reductions.

6.2. Research Implications

6.2.1. Balancing Industrial Growth with Low-Carbon Transition

We should insist on consolidating the dominant position of industry and promote the synergistic development of economic growth and carbon mitigation. Our findings demonstrate that under integrated scenarios, the rational adjustment of emission shares from coal, crude oil, and natural gas enables carbon reduction targets even during high-growth trajectories. Policymakers should adopt flexible emission controls aligned with macroeconomic cycles, implementing transitional measures during downturns to preserve industrial stability. A tiered target system should be established based on sectoral characteristics, granting 3–5 year transition periods for energy-intensive sectors (mining, power, steel, etc.) to facilitate technological upgrades while ensuring clean energy supply for low-carbon industries (electronics, equipment manufacturing, etc.). This differentiated approach will ensure systematic industrial emission decline.

6.2.2. Multidimensional Energy Conservation System for Structural Transformation

We should build a multi-dimensional energy conservation and emission reduction system to promote structural changes in industry. Benchmark and low-carbon scenarios prove insufficient for achieving industrial emission peaking by 2030, whereas integrated adjustments across industrial composition, energy efficiency, and consumption structure can trigger the inflection point. Three synergistic pathways emerge: First, industrial restructuring—accelerating the green retrofitting of traditional sectors while cultivating emerging low-carbon clusters to transition toward high-value-added, low-energy-intensive models. Second, energy efficiency enhancement—implementing dynamic benchmarking for key industries alongside digital monitoring platforms for real-time energy flow optimization. Third, systematic energy mix transition—prioritizing clean coal technologies in the short term while scaling renewable energy integration in the long term to establish diversified industrial energy systems.

6.2.3. Institutionalizing Sustainable Governance Mechanisms

Pilot demonstrations should spearhead sectoral decarbonization, particularly in power, steel, and mining, through “zero-carbon factory” initiatives and green industrial parks. Originating technology workshops and best practice repositories enable efficient low-carbon knowledge transfer. Integrated policy instruments should combine, first, positive incentives (VAT rebates, preferential green loans) for adopters, and, second, constraint mechanisms (stricter carbon quotas, differential electricity pricing) for laggards. There must also be complementary institutional reforms. We must establish industrial carbon accounting standards and enhance MRV (monitoring, reporting, verification) capacity building.