The Impact of Carbon Emissions Trading Policy on the ESG Performance of Heavy-Polluting Enterprises: The Mediating Role of Green Technological Innovation and Financing Constraints

Abstract

The carbon emissions trading policy is a key policy driving China’s low-carbon economic transition. Based on the policy environment of China’s carbon emissions trading pilot program from 2010 to 2021, this research selects representative heavy-polluting listed enterprises from this period as research subjects. Using the DID model, it investigates the impact of the carbon emissions trading policy on corporate ESG performance. The results of the research show that: (1) The carbon emissions trading policy significantly improves the ESG performance of heavy-polluting enterprises. (2) The carbon emissions trading policy enhances corporate ESG performance primarily by alleviating external financing constraints and stimulating green technological innovation within enterprises. (3) Due to differences in corporate characteristics and decision-making structures, the impact of carbon emissions trading policies on the ESG performance of heavily polluting enterprises exhibits heterogeneity. The findings of research are of great significance for a comprehensive understanding of the carbon emissions trading policy, facilitating fundamental changes in enterprises, promoting the construction of the carbon emissions trading market, and ensuring the timely achievement of the “dual carbon” goals.

1. Introduction

In the context of the current green development strategy and the “dual carbon” goals, traditional command-and-control environmental regulatory measures are facing limitations. The establishment of the environmental rights trading market, however, overcomes the bottleneck of directive regulation by assigning value to environmental rights and utilizing market-based incentives to strengthen enterprises’ responsibility for emission reduction, pollution control, and carbon emission reduction. This approach has effectively broken through the constraints of command-based regulation. Globally, the innovative practice of carbon emissions trading policy began in the European Union, where its practical operation has not only demonstrated its effectiveness and feasibility at the policy level but also highlighted its substantial impact on environmental protection and its potential economic benefits [1]. Compared to developed countries, although China was somewhat late in initiating carbon emissions trading policy, it has shown significant momentum to catch up and accelerate its development, adopting stronger policies and measures.

Since 2013, the National Development and Reform Commission selected eight regions—Shenzhen, Shanghai, Beijing, Guangdong, Tianjin, Hubei, Chongqing, and Fujian—as the first pilot zones to explore and implement the carbon emissions trading policy [2]. After years of development, in 2021, China launched a nationwide unified carbon financial market for trading. Today, this market has grown into one of the largest carbon emission trading systems in the world, with a wide geographical coverage and numerous participating enterprises. It has strongly supported China’s deep transition to a green, low-carbon development model and established a larger, more inclusive platform for green development transformation.

Current research on the effects of carbon emission trading policy mainly focuses on their economic and social impacts. Scholars have conducted extensive studies on the environmental, economic, and social effects of carbon emission trading policy on enterprises. From an economic perspective, carbon emission trading policy helps form economies of scale in regions with well-developed infrastructure, significantly increasing urban GDP [3]. It also promotes technological innovation at both the enterprise and urban levels [4], facilitates industrial structure upgrading, and enhances corporate productivity [5]. From a social perspective, a carbon emission trading policy reduces carbon emissions and energy consumption intensity, controls urban smog pollution, and improves air quality, effectively contributing to air pollution control [6]. However, some studies point out that with the deepening of policy issues such as inadequate policy implementation, imperfect evaluation systems and insufficient financial support have emerged, affecting the effectiveness of the policy.

Whether the carbon emissions trading policy has truly played a decisive role in driving enterprises to change their economic development model, and whether it has effectively expanded from mere emission control to profound adjustments in production, business operations, and internal governance systems, remains a critical question. The answers to these questions are central to advancing the national carbon emissions trading market and ensuring the timely achievement of the “dual carbon” goals. However, research on the effectiveness of this policy has mainly been limited to macro-level studies on regional economic growth [7], industrial structural transformation [8], and technological advancement [9]. At the micro level, studies have primarily focused on the role of carbon emissions trading policy in promoting green technological innovation in enterprises [10], with few comprehensive evaluations of its impacts on environmental protection, social responsibility, and internal governance within companies.

At the same time, China’s carbon emissions exhibit significant industry characteristics, with emissions from heavily polluting enterprises accounting for a scale comparable to that covered by the national carbon market at present. Heavy-pollution industries remain the main contributors to carbon emissions. According to the “Carbon Emission Ranking of Chinese Listed Companies (2021)”, nearly half of the nation’s carbon dioxide emissions are generated by the top 100 listed companies by carbon emissions, totaling 4.424 billion tons [11]. The extensive development model adopted by heavily polluting enterprises has not only exacerbated environmental pollution but also restricted the sustainable development of the enterprises themselves [12,13]. Therefore, in the process of implementing carbon reduction actions nationwide, heavily polluting enterprises undoubtedly bear significant responsibility and a crucial mission, playing an indispensable role in the carbon reduction efforts [14]. It can be said that the actions and effectiveness of heavily polluting enterprises in the carbon reduction endeavor will directly impact the successful implementation of the national low-carbon transformation strategy and the achievement of global climate governance goals.

However, research on the environmental, social, and governance (ESG) performance of heavily polluting enterprises in China is still in its early stages, especially concerning the relationship between carbon emission trading and ESG performance, which remains underexplored. Additionally, most existing studies on carbon trading focus on the macroeconomic level, with few delving into its specific impacts on the micro level of enterprises. At the same time, effectively controlling greenhouse gas emissions in key areas is crucial for advancing sustainable development goals, yet current research seldom focuses on heavily polluting enterprises. Therefore, exploring the impact of carbon emission trading policies on the ESG performance of heavily polluting enterprises holds both practical and innovative significance.

Therefore, this study employs a multi-period difference-in-differences (DID) model, focusing on heavily polluting enterprises rather than all listed companies. It examines the impact of carbon emission trading policies on heavily polluting enterprises from the integrated perspectives of ESG. This study not only provides valuable insights for the further improvement of carbon trading markets in China and other low-carbon policy countries, utilizing both government and market forces to optimize resource allocation and achieve long-term macroeconomic sustainability and cost reduction and productivity enhancement at the microenterprise level but also further contributes to the research on carbon emission trading policies in the key area of heavily polluting enterprises.

The study makes significant contributions in the following areas: First, we conduct a detailed and comprehensive exploration of the specific impact of carbon emissions trading policy on the ESG performance of heavy-polluting enterprises from the perspective of micro-level enterprises. It focuses on revealing how, under policy intervention, enterprises adjust their business strategies and management approaches by participating in the carbon emissions trading market in order to cope with increasingly stringent carbon emission restrictions and growing demands for sustainable development. Second, in the process of green and low-carbon transformation, industries across various sectors need to change their development concepts, optimize industrial structures, improve technological processes, and phase out outdated capacities. Within this systemic transformation, exploring the mechanisms and effects of carbon emissions trading policies on corporate ESG performance is of practical significance for promoting the high-quality development of enterprises and the broader socio-economic system. Third, the carbon emissions trading policy in China and other developing countries is still in its early stages and faces many challenges and uncertainties. Therefore, the paper focuses on the differentiated characteristics at multiple levels, aiming to provide more precise and adaptable guidance for the further improvement and development of the policy.

The remainder of the paper consists of the following sections: Section 1 is the introduction. Section 2 presents a review of the literature on carbon emissions trading policy and ESG, and the development of research hypotheses. Section 3 discusses the research methodology. Section 4 provides the results of the empirical analysis. Section 5 offers a discussion of the findings. Section 6 describes the conclusions of the research and makes policy implications.

2. Literature Review

2.1. Carbon Emissions Trading Policy

As a system of environmental rights trading, carbon emissions trading plays a pivotal role in achieving the “dual carbon” goals. In the academic research field, scholars have conducted extensive and diversified studies on the effectiveness of carbon emissions trading pilot projects. These research findings can be summarized from two perspectives: macro and micro.

From a macro perspective, the carbon emissions trading pilot policy has successfully curbed the growth of carbon dioxide emissions within the pilot regions. For emission allowance sellers, the policy has significantly facilitated the implementation of emission reduction measures through various mechanisms, including market-based revenue incentives, innovation-driven approaches, and policy support. On the buyers’ side, enterprises have effectively reduced carbon emissions under the combined influence of cost pressure, the internal demand for production process improvement, and market-driven trends [15]. This policy has contributed to regional high-quality economic development by enhancing economic growth through technological innovation, industrial structure optimization, and the synergistic effects of environmental policies [16]. Moreover, the emission reduction effects of the pilot policy increase over time [17], and its spillover effects also demonstrate a suppressing influence on carbon emissions in neighboring regions [3]. Additionally, the policy stimulates increased investments in innovation and the growth of green patents, thereby vigorously promoting the development of a green economy [18].

At the micro level, the carbon emissions trading policy has not only reshaped corporate behavior but also improved investment efficiency in areas such as financing decisions and technological innovation. This has provided robust support for enterprises in adopting green development strategies and has driven the transition of the economy and society toward a sustainable development model [19]. The system effectively stimulates technological innovation within enterprises, with the level of carbon market liquidity positively correlating with its impact on technological advancement. However, it is worth noting that corporate cost-transfer strategies may, to some extent, diminish the positive incentive effects of such environmental policy measures. Particularly when enterprises operate in less competitive market environments and possess significant bargaining power in negotiations with customers and suppliers, the impact of the carbon emissions trading system on driving technological innovation may weaken. The policy also positively enhances corporate production efficiency, as enterprises participating in carbon trading exhibit significantly improved production quality compared to those in regions where such policies are not implemented [20]. It is important to recognize, however, that while the carbon emissions trading mechanism theoretically holds the potential to drive corporate transitions toward a low-carbon economy, its impact on long-term value creation for all enterprises may not immediately show significant positive effects. This is particularly true in the early implementation stages when lenient allocation strategies and relatively low carbon market prices may limit its influence, and the outcomes might vary across different enterprises.

2.2. Corporate ESG Performance

Since the concept of ESG was first introduced by the United Nations in 2004, it has gained increasing popularity and is now regarded as one of the most innovative and widely used sustainability indicators globally [21]. ESG can be defined across three dimensions [22,23]. The environmental dimension focuses on a company’s performance in relation to the natural environment, including its impact on air quality, biodiversity, carbon footprint, greenhouse gas emissions, waste, and water quality. The social dimension emphasizes the management of a company’s relationships with employees, suppliers, customers, and the broader community. Furthermore, human rights, privacy policies, and efforts to assist impoverished communities are key aspects of this pillar. The governance dimension primarily concerns corporate leadership style, internal controls, auditing, management diversity, executive compensation, and policies. Today, a company’s performance is often assessed based on these dimensions, and policies, regulations, investors, and stakeholders are increasingly demanding companies disclose their ESG performance, influencing corporate focus on sustainable development initiatives [24].

With the growing emphasis on sustainable development and responsible business practices, corporate ESG performance is gaining widespread attention from policymakers and market investors [25]. For instance, the Hong Kong Stock Exchange began requiring listed companies to disclose ESG reports in stages starting in 2015 and implemented this requirement comprehensively in July 2020. In June 2021, the China Securities Regulatory Commission revised the format and guidelines for listed companies’ periodic reports, adding two chapters on ESG and clarifying disclosure requirements, urging companies to take primary responsibility for addressing climate change and promoting green, low-carbon development. Corporate ESG practices are shaped by external factors as part of a company’s development strategy. Existing literature on corporate ESG largely treats it as an antecedent variable, focusing on its impact on internal factors, such as the effect of ESG performance on corporate performance [26], the influence of ESG performance on capital structure, and the impact of ESG performance on corporate financing costs [27].

2.3. The Relationship Between Carbon Emissions Trading Policy and Corporate ESG Performance

The carbon emissions trading mechanism essentially represents a market-based approach to regulating environmental impact. Its core objective is to internalize the external costs of pollutant emissions, incorporating them into enterprises’ internal cost structures as critical factors for decision-making [18]. Studies on low-carbon city pilot policies indicate that these policies can enhance corporate performance across ESG dimensions. Moreover, the adoption of carbon emissions trading mechanisms significantly improves the transparency of corporate carbon emissions disclosures. This, in turn, encourages companies to present more accurate information regarding their carbon footprints and carbon efficiency indicators in social responsibility reports [28].

Under the general guidance of carbon emissions trading policy, the State Council has introduced a series of supplementary documents, including the green credit policy, a core component of the green finance policy framework. This policy provides financial support to enterprises for adopting environmentally friendly raw materials, deploying clean energy technologies, and transitioning to energy-efficient production methods [29].

On the one hand, Traditional economic theories often suggest that implementing environmental policies might suppress economic growth [30]. This view stems from the notion that such policies increase production costs for enterprises, such as investments in pollution control facilities, adoption of costlier environmentally friendly raw materials, or modifications to existing production processes to reduce emissions. These adjustments inevitably raise operational costs and may weaken the competitive advantage of firms, especially in the context of intense global market competition. If one region’s environmental standards exceed those of others, local firms might face a price disadvantage, leading to reduced market share and export capacity [31].

On the other hand, Porter’s Theory [32] argues that environmental regulations can incentivize enterprises to accelerate technological innovation, enhance pollution control technologies, and adopt more efficient resource utilization and advanced production processes. These improvements not only reduce environmental impacts but also increase the technological value of products, fostering new market advantages. For instance, companies can attract environmentally conscious consumer groups by offering sustainable and eco-friendly products.

Furthermore, environmental regulation-driven innovation and managerial changes can enhance corporate profitability [33]. By improving energy efficiency, reducing waste, and minimizing resource consumption, firms can lower operational costs. Additionally, the development of green products and services enables businesses to explore new markets, strengthen brand value, and offset the initial costs of complying with environmental policies. This ultimately promotes long-term corporate sustainability. Thus, strict environmental regulations are not necessarily obstacles to economic growth; under certain conditions, they can drive innovation and efficiency, fostering a greener and more sustainable economic transition. Based on this analysis, the following hypothesis is proposed:

H1. 

Carbon emissions trading policies significantly enhance the ESG performance of highly polluting enterprises.

2.4. The Mediating Role of Green Technological Innovation

Hicks’ induced innovation theory suggests that stringent environmental regulations lead to an increase in both factor input prices and compliance costs, thereby compelling enterprises to pursue technological innovation [34]. The carbon emission trading pilot policy, introduced as part of China’s climate change strategy, serves as a key environmental regulation targeting sectors such as industry, construction, transportation, energy supply, agriculture, forestry, and waste management [35,36]. The policy aims to control greenhouse gas emissions in critical industries and promote low-carbon development, which in turn stimulates the advancement of green innovation [37].

The central government’s allocation of pilot opportunities to local regions reflects both honor and responsibility. To enhance local environmental performance, government authorities actively guide enterprises toward green innovation and industrial restructuring, thereby fostering corporate green technological innovation [38]. In its notification on launching low-carbon city pilot programs, the National Development and Reform Commission (NDRC) explicitly required regions to strengthen low-carbon development capacity and talent pool building. The low-carbon city pilot policy enhances green technological innovation by increasing enterprise investment in scientific personnel and funding [39].

Current studies on the impact of green innovation on corporate ESG performance highlight that environmental rights trading markets, as a regulatory tool, provide benefit incentives for technologically advanced, high-pollution enterprises. Under the macro context of the “dual carbon” goals, green technological innovation enables high-pollution industries to mitigate environmental risks and reduce pollution costs effectively [40]. Some heavily polluting enterprises, after participating in carbon trading, face high carbon pricing costs and competitive pressures, prompting them to allocate significant funds to technological innovation, which drives performance growth through productivity improvements [41,42]. However, Chiou et al. [43] argue that while green process innovation, green product innovation, and green management innovation can reduce production costs, enhance product quality, and help enterprises gain competitive advantages and improve environmental performance, the relationship between green management innovation and environmental performance is not significant.

Moreover, based on social responsibility theory, enterprises may adopt two approaches to fulfill social responsibilities depending on their motivations: one is philanthropic activities aimed at achieving short-term reputational effects, and the other is sustainable development as a long-term strategic choice [44,45]. Enterprises are more likely to adopt the former to improve their ESG performance [46]. The implementation of low-carbon city pilot policies can effectively enhance corporate energy efficiency, reduce environmental costs, and improve financial performance, thereby alleviating pressures related to social, environmental, and corporate governance responsibilities, and promoting sustainable development [47]. To compensate enterprises that adopt low-carbon production technologies, governments typically provide financial subsidies and preferential interest rates to offset market imperfections [48].

Furthermore, green technological innovation, through the adoption of clean energy production technologies such as environmentally friendly materials, alternative energy sources, and resource recycling, alleviates the constraints of non-renewable energy, improves resource and energy utilization efficiency, and reduces pollution at its source. This effectively fulfills the energy conservation and emission reduction requirements of government environmental regulations, thus contributing to environmental improvement [49]. Based on the above analysis, this paper proposes the following hypotheses:

H2. 

Carbon emissions trading policies enhance corporate ESG performance by stimulating green technological innovation.

2.5. The Mediating Role of Financing Constraints

Capital flows are critical to corporate survival and development. In recent years, the government has placed great emphasis on ESG-related financing policies, encouraging enterprises to improve their ESG performance and disclose relevant information transparently [50].

Myers et al. [51] argue that information asymmetry is the primary cause of corporate financing constraints. With the gradual development and improvement of the carbon trading market, heavily polluting enterprises face increasingly stringent dynamic information verification requirements. According to signaling theory and stakeholder theory, after participating in carbon reduction, enterprises voluntarily disclose more compliant and transparent carbon emission information. This enables internal and external stakeholders to quickly receive signals about the enterprise’s high-quality development transition, thereby alleviating concerns about the uncertainty of the enterprise’s future development. Issues such as “moral hazard”, “principal-agent problems”, and “adverse selection” can be effectively mitigated. Consequently, corporate reputation and moral capital improve, while creditors’ risk premium expectations decrease, effectively easing the debt financing constraints faced by enterprises [52].

Additionally, resource dependence theory suggests that sustainable corporate development requires alignment with societal and environmental needs. By actively fulfilling social and environmental responsibilities and building a favorable external image, enterprises can secure more favorable debt financing conditions [53]. Solid corporate performance is a critical guarantee for repaying debt. After being included in the carbon emission trading system, heavily polluting enterprises are obligated to meet stricter information disclosure requirements [54]. For enterprises in their growth and maturity stages within heavily polluting industries, such obligations significantly alleviate their financing constraints [55,56]. By actively participating in carbon emissions trading, these enterprises can demonstrate to external investors their commitment to low-carbon development, their determination to combat climate change, their sustainable development capabilities, and the rational allocation of funds to support future transformation [57]. Financial institutions such as banks can also use the disclosed information to more accurately assess the credit risk of enterprises in heavily polluting industries, enabling reasonable pricing and ultimately reducing their debt financing costs [58]. Based on this, the following hypothesis is proposed:

H3. 

Carbon emissions trading policies improve corporate ESG performance by alleviating financing constraints.

3. Methodology

3.1. Sample and Data

In this study, the classification of heavily polluting enterprises is based on the “Guidelines for Environmental Information Disclosure of Listed Companies” published by the Ministry of Environmental Protection. Enterprises in 16 major industrial sectors, including thermal power generation, steel smelting, cement manufacturing, electrolytic aluminum, coal mining and processing, metallurgy, chemical industry, petrochemicals, building materials manufacturing, paper industry, brewing, pharmaceuticals, fermentation, textiles, leather processing, and mining, are categorized as heavily polluting enterprises.

The research sample consists of all A-share listed companies classified as heavily polluting during the period from 2010 to 2021. The data processing steps are as follows:

a. 

Exclude samples of companies marked as ST (Special Treatment) or *ST.

b. 

Remove samples with missing observations for any variables.

c. 

Exclude all non-highly polluting enterprises.

d. 

To mitigate the influence of outliers on the research results, all continuous variables were Winsorized at the 1% and 99% levels.

After these steps, a total of 9593 observations were retained for the analysis. The data utilized in the article are sourced from the CSMAR and Wind databases.

3.2. Definition of Variables

ESG Performance: The dependent variable in the article is the ESG performance of enterprises. Following Lin et al. [59] and Yan and Wang [60], we adopt the Huazheng ESG Ratings as a proxy for ESG performance. The Huazheng ESG rating system consists of nine levels, ranked from lowest to highest as C, CC, CCC, B, BB, BBB, A, AA, and AAA. These ratings are assigned values from 1 to 9 in ascending order, with higher scores indicating better ESG performance. The annual ESG score of each enterprise is thus determined based on this rating system.

Carbon Emission Trading Policy: The explanatory variable is the interaction term between the carbon emission trading pilot policy and a dummy variable for pilot regions. Given the different implementation timelines of the policy across pilot regions, the study carefully distinguishes the policy’s effective dates to ensure the accuracy of data processing and the validity of analytical results. Specifically, the year is divided into two periods:

a. 

January to June (first half of the year): Policies initiated in this period are considered effective in the same year.

b. 

July to December (second half of the year): Policies initiated in this period are treated as effective in the following year.

Control Variables: To minimize errors caused by omitted variables, this study incorporates control variables after accounting for the characteristics of heavily polluting enterprises. The main control variables include: Firm size, Total number of employees, Firm age, Debt-to-asset ratio, Return on assets, and Fixed asset investment [20,61]. These control variables aim to capture enterprise-specific characteristics that could influence ESG performance, ensuring robust results.

Table 1. Variable definitions.

Variable Name	Variable Symbol	Variable Definition
Dependent variable
ESG Performance	ESG	The Huazheng ESG rating index ranges from 1 to 9.
Independent variable
Carbon Emissions Trading Policy	CETP	If a firm is included in the carbon emissions trading system in a given year, the variable is assigned a value of 1; otherwise, it is assigned a value of 0.
Intermediary variable
Green technology innovation	GTI	Number of green patent applications/Number of total patent applications.
Financing Constraints	WW	The financing constraints faced by a firm are measured using the WW index.
Control variables
Firm size	Size	The natural logarithm of the firm’s total assets.
Total number of employees	Emp	The natural logarithm of the total number of employees.
Firm age	Age	The number of years since the firm’s establishment up to the current period.
Debt-to-asset ratio	Lev	Total liabilities divided by total assets.
Return on assets	ROA	Net profit/Total assets.
Fixed asset investment	FAI	The natural logarithm of the firm’s fixed asset investment.

outlines the definitions of the variables.

3.3. Modeling

To investigate how the carbon emission trading policy affects corporate ESG performance, the article constructs a DID model to evaluate the net effect of policy implementation. This method measures the differences between the experimental group and the control group before and after the implementation of CETP. The baseline model is constructed as follows:

$$ES{G}_{irt}=\alpha +\theta CET{P}_{tr}+X_{irt}+{\delta }_i+{\mu }_t+{\epsilon }_{irt}$$

(1)

In this model, the dependent variable $ES{G}_{irt}$ reflects the overall performance of a specific firm $i$ in region $r$ and time $t$ regarding environmental, social, and governance dimensions. Specifically, $i$, $r$, and $t$ represent the firm, region, and time dimensions, respectively. $CET{P}_{tr}$ = Treat × Pilot, Treat indicates whether region $r$ implemented the carbon emission trading pilot in year $t$. If region $r$ initiated the pilot program in the current or any subsequent year, the variable takes a value of 1; otherwise, it is 0. Pilot is a dummy variable for pilot regions, which takes a value of 1 if a region is part of the policy pilot program and 0 otherwise. $X$ includes a set of firm-level control variables, $\delta$ represents firm-fixed effects, $\mu$ denotes time-fixed effects, and $\epsilon$ is the random error term.

4. Empirical Analysis

4.1. Descriptive Statistics

Table 2. Descriptive statistics.

Variable	N	Mean	Std. Dev.	Min	Max
ESG	9593	4.160	1.078	1	7
CETP	9593	0.228	0.420	0	1
GTI	9593	0.788	1.097	0	4.234
WW	9593	−1.017	0.095	−3.111	−0.409
Size	9593	22.140	1.302	19.046	28.636
Emp	9593	7.688	1.233	2.639	13.223
Age	9593	2.089	0.927	0	3.401
Lev	9593	0.404	0.219	0.008	3.262
ROA	9593	0.040	0.056	−0.178	0.199
FAI	9593	18.868	1.829	0	26.525

presents the descriptive statistics of the main variables. The maximum value of corporate ESG performance is 7, the minimum value is 1, and the mean is 4.160, with a standard deviation of 1.078, indicating significant variation in ESG performance across firms. Additionally, the average value of CETP is 0.228, suggesting that relatively few firms in the sample participate in the carbon emissions trading pilot. Finally, the sample description shows no outliers, and the other control variables are generally consistent with existing studies.

4.2. Baseline Regression Analysis

Table 3. Base regression results.

	(1) ESG	(2) ESG	(3) ESG
CETP	0.523 *** (14.271)	0.106 *** (2.850)	0.108 *** (2.701)
Size		0.362 *** (15.114)	0.356 *** (15.130)
Emp		0.104 *** (6.015)	0.161 *** (6.923)
Age		0.013 (1.271)	0.017 * (1.655)
Lev		−0.198 ** (−2.139)	−0.117 (−1.288)
ROA		0.191 (0.919)	0.156 (0.764)
FAI		1.132 (1.635)	1.480 ** (2.160)
_cons	0.656 *** (54.495)	−7.898 *** (−15.503)	−7.723 *** (−15.437)
Year	No	Yes	Yes
Individual	No	No	Yes
N	9593	9593	9593
adj. R2			0.182

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively, with T-values reported in parentheses.

shows the impact of carbon emissions trading policy on corporate ESG performance. Column (1) presents the regression coefficient for policy on ESG performance without including control variables, which is 0.280 and statistically significant at the 1% level. Column (2) adds control variables, and the regression coefficient for CETP on ESG performance remains positive and significant. Column (3) introduces time-fixed effects and individual fixed effects, showing that the regression coefficient for CETP on ESG performance is 0.108 and statistically significant at the 1% level. This indicates that, when controlling for other factors, the implementation of carbon emissions trading policy has a significant positive effect on the ESG performance of high-pollution enterprises. Therefore, H1 can be confirmed.

4.3. Robustness Tests

4.3.1. Parallel Trend Test

The parallel trend test is a core step in the DID methodology for policy effect evaluation. The main goal is to verify whether, in the absence of policy intervention, the development trends of the experimental and control groups would have been parallel or similar before and after the policy implementation. To ensure the accuracy of the policy effect evaluation, the study defines the implementation year of China’s carbon emissions trading pilot policy as the time point and subdivides the policy implementation phases for different regions.

Figure 1 shows that the trends of change between the experimental group and the control group were approximately the same during the three years prior to the policy implementation. However, in the year following the policy implementation, the coefficients start to show significant changes and pass the 5% significance test. This suggests that before the implementation of the carbon emission trading policy, the experimental and control groups maintained similar parallel trends in ESG performance, but after the policy was implemented, the ESG performance differences between the two groups became significant. From the perspective of dynamic effects, the carbon emission trading policy has a certain lag and lasting impact on improving corporate ESG performance. The parallel trend assumption is thus satisfied, confirming the validity of the DID model constructed in this study.

4.3.2. PSM-DID Test

As this research primarily focuses on examining the impact of the carbon emission trading policy on the ESG performance of heavily polluting enterprises, the research sample is non-random. Additionally, the randomness of whether a province is included in the carbon trading policy cannot be guaranteed. To address these issues, the PSM-DID method is employed to reduce the potential sample bias affecting the baseline regression results. First, all control variables are selected as matching covariates, and 1:1 nearest-neighbor matching is conducted, excluding unmatched samples. The balance test confirms that covariates show no significant differences between groups, indicating good matching quality. Subsequently, the main regression model is applied for analysis. The regression results after PSM, as shown in

Table 4. PSM-DID test.

	(1) ESG
CETP	0.119 ** (2.034)
Size	0.341 *** (9.247)
Emp	0.014 (0.935)
Age	0.008 (0.958)
Lev	−0.126 (−0.882)
ROA	0.078 (0.242)
FAI	1.454 (1.443)
_cons	−7.081 *** (−9.092)
Year	YES
Individual	YES
N	6150
adj. R2	1.454

Note: *** and ** indicate statistical significance at the 1% and 5% levels, respectively, with T-values reported in parentheses.

, indicate that the impact coefficient of CETP on corporate ESG performance is 0.119, which is significant at the 5% level. This finding suggests that after eliminating sample selection bias, carbon emission trading policy has a significant positive effect on the ESG performance of heavily polluting enterprises, and the results remain robust.

4.3.3. Placebo Test

To further exclude the influence of non-policy factors on the ESG performance of heavily polluting enterprises, this study also conducted a placebo test. Based on the original control variables, simulated policy variables were constructed by randomly selecting pilot cities and implementation times. Then, under control for time effects and industry effects, 1000 regression analyses were conducted. The estimated coefficient p-values are shown in Figure 2, the estimated kernel density distribution closely aligns with a normal distribution, with the estimated values centered around 0, and most of the estimated coefficients not reaching statistical significance. Based on the test results, the placebo test is successful, indicating that the improvement in the ESG performance of heavily polluting enterprises in the pilot regions is indeed attributable to the implementation of the carbon emission trading policy, rather than random events.

4.3.4. Substituting Pilot Regions

Given that the initial batch of pilot regions includes many provinces with relatively high economic levels, we re-examine Hypothesis 1 by selecting enterprises from eight non-pilot provinces, assuming they were also affected by the indirect or diffusion effects of carbon emissions trading policy. The results are presented in

Table 5. Substituting pilot regions.

	(1) ESG
CETP	0.105 *** (2.580)
Size	0.365 *** (15.253)
Emp	0.014 (1.375)
Age	0.011 ** (2.102)
Lev	−0.124 (−1.367)
ROA	0.173 (0.895)
FAI	1.327 * (1.932)
_cons	−8.024 *** (−13.987)
Year	YES YES
Individual
N	9593
adj. R2	0.187

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively, with T-values reported in parentheses.

, the signs of the coefficients for the explanatory variables remain unchanged and are significant at the 1% level, with results showing little difference from the main regression findings.

4.4. Mediating Effect of Green Technological Innovation

Under the guidance of the nationally advocated “Dual Carbon” strategic goals, enterprises face a pressing need to enhance their innovation capabilities and drive technological advancements, particularly in the optimization process of Environmental, Social, and Governance performance. Corporate research and development investment plays a critical driving role in this process. At the policy level, the government has consistently emphasized that enterprises should actively respond to the “Dual Carbon” goals by increasing investment in the research, development, and application of green and low-carbon technologies. This innovation-driven approach aims to promote industrial upgrading, achieving both economic and social benefits.

Based on existing literature and national policy directions, the carbon emission trading policy, as a market-based policy instrument, is regarded as an important environmental regulatory measure for the government to internalize enterprises’ pollution costs. This policy not only enhances enterprises’ awareness of ecological protection but also actively guides them toward adopting green and low-carbon transformation strategies. By emphasizing technological innovation, carbon emission trading policy contributes to improving enterprises’ environmental sustainability, strengthening governance capabilities, and fostering social responsibility, thereby positively influencing ESG ratings.

To verify the mediating effect of corporate green technological innovation, this study measures it using the number of green patent applications [62]. As shown in

Table 6. Mediating effect of green technological innovation.

	(1) GTI	(2) ESG
CETP	0.690 ** (2.191)	0.105 *** (2.623)
GTI		0.005 *** (3.150)
Size	1.304 *** (7.057)	0.350 *** (14.836)
Emp	−0.031 (−0.389)	0.017 * (1.670)
Age	−0.024 (−0.600)	0.012 ** (2.318)
Lev	1.260 * (1.763)	−0.123 (−1.352)
ROA	−0.940 (−0.587)	0.163 (0.787)
FAI	−0.914 *** (−2.766)	−0.392 * (−1.708)
_cons	−14.586 ** (−6.989)	−7.633 *** (−15.241)
N	9593	9593
R2	0.020	0.182
Year	Yes	Yes
Individual	Yes	Yes

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively, with T-values reported in parentheses.

, the results in Column (1) indicate that the coefficient of carbon emission trading policy on corporate green technological innovation is 0.690, which is significant at the 5% level, suggesting that the policy significantly promotes green technological innovation. The results in Column (2) show that the estimated coefficient of corporate green technological innovation on ESG performance is 0.105, significant at the 1% level, indicating that green technological innovation improves ESG performance. Combining the results of Columns (1) and (2), it can be concluded that under the influence of the carbon emission trading policy, heavily polluting enterprises improve their ESG performance by promoting green technological innovation among enterprises. Therefore, H2 can be confirmed.

4.5. Mediating Effect of Financing Constraints

Funding plays a critical role in the normal operation and long-term development of enterprises. The sustained and stable operation of a business must rely on a steady flow of funds. Only with sufficient capital can an enterprise achieve further development. Our research further explores how carbon emissions trading policy, as a market-oriented tool, affects corporate financing ability, arguing that through the establishment of carbon emissions trading policy, the government indirectly imposes financing constraints on enterprises through market mechanisms. Specifically, a company’s carbon emission status directly influences its ease of financing and the associated costs. The study uses the WW index, a representative indicator of corporate financing constraints, to analyze how carbon emissions trading policy impacts financing constraints and, in turn, affects ESG performance. The results in

Table 7. Mediating effect of financing constraints.

	(1) WW	(2) ESG
CETP	−0.108 *** (−2.616)	0.093 ** (2.269)
WW		−0.009 ** (−2.011)
Size	−0.324 *** (−14.130)	−0.356 *** (−15.132)
Emp	−0.017 * (−1.655)	−0.100 ** (−2.260)
Age	0.037 *** (0.479)	0.056 (0.266)
Lev	0.020 (0.280)	0.052 (0.212)
ROA	0.011 * (1.875)	0.150 (0.736)
FAI	−0.001 (−0.627)	0.005 (0.878)
_cons	−1.153 *** (−6.113)	−3.723 *** (−8.367)
N	9593	9593
R2	0.248	0.925
Year	Yes	Yes
Individual	Yes	Yes

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively, with T-values reported in parentheses.

reflect the mediation effect of financing constraints in the transmission mechanism between the main effects. The results in Column (1) show that the implementation of CETP has a significant negative impact on WW, indicating that the financing constraints faced by heavily polluting enterprises can be alleviated. Column (2) shows that, holding other variables constant, the coefficients of CETP and WW are 0.093 and −0.009, respectively, both significant at the 5% level. This suggests that financing constraints mediate the effect of carbon emission trading policy on the ESG performance of heavily polluting enterprises. In other words, as carbon emission trading policy is implemented, financing constraints are further alleviated, leading to improvements in ESG performance. Therefore, H3 can be confirmed.

In summary, the carbon emission trading policy significantly impacts the green technological innovation and financing constraints of heavily polluting enterprises. Green technological innovation and financing constraints play mediating roles in the process by which the carbon emission trading policy enhances the ESG performance of these enterprises. Through the implementation of this policy, the level of green technological innovation in heavily polluting enterprises can be improved, and their financing constraints can be alleviated, thereby further promoting their ESG performance.

4.6. Heterogeneity Analysis

To explore the generalizability of the study’s conclusions, this research conducts detailed classification and comparative analysis of the sample firms to investigate whether the conclusions hold across firms with different ownership structures and governance models. First, the sample is divided into two groups based on ownership type: state-owned enterprises (SOEs) and non-state-owned enterprises (NSOEs), to explore the differences in how firms under different ownership systems respond to carbon emissions trading policy. Furthermore, the sample was categorized into enterprises with a combined CEO-chair structure (dual-role enterprises) and those with a separate CEO-chair structure (split-role enterprises). The former refers to enterprises where decision-making and execution powers are concentrated in a single entity, while the latter refers to firms where these powers are separated, with an independent board and management team.

As shown in

Table 8. Heterogeneity analysis.

	(1) ESG	(2) ESG	(3) ESG	(4) ESG
CETP	0.084 (1.641)	0.155 ** (2.419)	0.013 (0.280)	0.282 *** (3.423)
Size	0.366 *** (10.949)	0.396 *** (11.063)	0.320 *** (12.140)	0.431 *** (8.957)
Emp	0.002 (0.158)	0.043 ** (2.418)	0.016 (1.376)	0.018 (0.895)
Age	0.025 ** (3.874)	−0.017 ** (−1.995)	−0.230 *** (−3.714)	−0.206 *** (−5.607)
Lev	−0.136 (−0.966)	−0.030 (−0.236)	−0.126 (−1.191)	−0.088 (−0.512)
ROA	0.257 (0.842)	−0.025 (−0.090)	−0.080 (−0.341)	0.669 * (1.706)
FAI	1.800 (1.570)	1.378 (1.563)	1.169 (1.294)	2.056 * (1.890)
_cons	−8.007 *** (−11.304)	−8.647 *** (−11.391)	−7.102 *** (−12.654)	−9.174 *** (−8.948)
N	3656	5796	2545	6925
R2	0.245	0.146	0.199	0.173
Year	Yes	Yes	Yes	Yes
Individual	Yes	Yes	Yes	Yes

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively, with T-values reported in parentheses.

, the results in Column (1) represent the SOEs, where the regression coefficient is 0.084. In Column (2), which represents the NSOEs, the regression coefficient is 0.155, significant at the 5% level. Since the sample is concentrated on heavily polluting enterprises, the promotion effect of carbon emission trading policy on ESG performance is more pronounced for NSOEs. First, in terms of policy transmission, compared to NSOEs, SOEs tend to respond more comfortably to policy regulations due to government shareholding and strong government-business relationships. For NSOEs, the external pressures from regulations are greater, which drives them to adopt transformation strategies more actively. Second, from a long-term strategic perspective, SOEs are managed by the State-Owned Assets Supervision and Administration Commission (SASAC), which exercises significant supervision over them annually. These enterprises tend to have more standardized management over the long term and already possess good technological reserves and forward-looking strategic frameworks, making their policy responses less pronounced than those of NSOEs.

Column (3) represents dual-role enterprises, with a regression coefficient of 0.013. Column (4) represents split-role enterprises, with a regression coefficient of 0.282, which is significant at the 1% level. This indicates that the positive effect of carbon emission trading policy on split-role enterprises is greater than that on dual-role enterprises. For dual-role enterprises, where decision-making and execution powers are highly concentrated, the impact of carbon emission trading policy on ESG performance is smaller. This reflects a tendency for such enterprises to prioritize short-term economic gains over long-term sustainable development when making decisions. In contrast, split-role enterprises exhibit a more positive ESG response to carbon emission trading policy. Their decision-making processes balance economic interests with social responsibility and environmental protection, resulting in greater attention to energy conservation, pollution reduction, and improved corporate governance while pursuing commercial success.

5. Discussion

The findings of this study confirm that the implementation of CETP has indeed yielded positive effects, effectively incentivizing and promoting heavily polluting enterprises to enhance their ESG performance. This drives technological upgrades and optimizes resource allocation, enabling more sustainable development across environmental, social, and corporate governance dimensions. These results align with the conclusions of Rubashkina [33]. Furthermore, the study demonstrates that additional energy consumption for environmental management does not lead to price fluctuations caused by supply-demand imbalances, nor does it disrupt optimal investment decisions under the original market mechanism, thereby avoiding reductions in production efficiency and competitiveness. This refutes the argument by Greenstone [63] that strong regulatory policies negatively impact corporate productivity. Robustness tests further validate the reliability of the empirical results. In addition, China’s carbon emission trading policy pilot was introduced prior to the establishment of the New Collective Quantified Goals (NCQGs). The implementation of this policy has provided valuable experience for China’s heavily polluting enterprises in supporting and advancing the realization of the NCQGs.

According to endogenous growth theory, technological innovation can drive enterprises to improve output efficiency while innovating their methods of profit acquisition, thereby facilitating their transition to a high-quality development stage. Under the promotion of carbon emission trading pilot policies, companies can engage in green technological innovation, adopt or improve production processes, reduce material consumption in manufacturing, and thus promote the efficient use of resources, optimize development quality, and enhance their performance in environmental, social, and corporate governance aspects. This view aligns with the perspectives of Wang et al. [40] and Seman et al. [64] who argue that environmental policies motivate heavy-polluting companies to actively focus on green technological innovation and upgrading to enhance technological capabilities, reduce pollution emissions, and achieve comprehensive improvement in ESG performance.

Regarding the relationship between corporate financing constraints and ESG performance, the implementation of carbon emission trading pilot policies can alleviate the financing constraints of heavily polluting enterprises, thereby improving their ESG performance [65]. In this process, forward-looking and responsible companies will enhance their ESG performance through green technological innovation and investment in low-carbon projects. This not only helps improve the company’s public image and social reputation but also boosts investor confidence, attracting more social capital and green finance support [53,56]. Additionally, under policy guidance, banks and other financial institutions will pay more attention to a company’s environmental performance and carbon emission management when making loan disbursements and investment decisions. This provides more favorable financing conditions and additional financing channels for enterprises committed to emission reduction and good environmental performance [58]. As a result, the difficulty and cost of financing for these companies are reduced, thereby alleviating their financing constraints to some extent.

While the article strives for clarity in structure and logical coherence in examining the impact of carbon emissions trading policy evolution on the ESG performance of heavy-polluting enterprises, certain limitations still exist. First, this paper only examines the impact of carbon emission trading policy on heavily polluting enterprises. In future research, we plan to further explore the effect of carbon emission trading policy on the ESG performance of non-heavily polluting enterprises and conduct a comparative analysis alongside this study to form a comprehensive research topic. Second, in the selection of mediating variables, this study focused on representative factors such as internal green technology innovation and external financing constraints. However, other potential mediating factors, including market competition level, intellectual property protection strength, customer satisfaction, and resource allocation efficiency, may also play a role in the impact of the carbon emission trading policy on enterprises’ ESG performance. Future research should collect and analyze relevant data on these factors to enhance the rigor and comprehensiveness of the study. Third, given that carbon finance is still rapidly developing and the carbon emissions trading policy has expanded from pilot regions to national implementation over the past decade, this study can only offer a preliminary exploration of the policy’s short-term effects on corporate ESG performance based on current pilot cases. The long-term impact on the sustained ESG performance of heavily polluting enterprises requires further investigation with more data in the future.

6. Conclusions and Policy Implications

6.1. Conclusions

This study selects heavily polluting enterprises listed between 2010 and 2021 as its research sample and applies a multi-period DID model to investigate the impact of the carbon emissions trading policy on these enterprises from a comprehensive perspective of environmental, social, and corporate governance. The findings provide valuable insights for China and other countries implementing low-carbon policies, helping to improve carbon trading markets by leveraging government and market mechanisms to optimize resource allocation, achieve long-term sustainable macroeconomic development, reduce costs, and enhance productivity at the microenterprise level. The main research conclusions are as follows: (1) The carbon emission trading policy has a significant positive impact on the ESG performance of heavily polluting enterprises, meaning that the implementation of CETP is an effective way to improve the ESG performance of such enterprises. This conclusion holds true even after robustness checks. (2) The mechanism analysis in this study shows that the carbon emission trading policy mainly enhances ESG performance through two channels: alleviating external financing constraints and stimulating corporate green technological innovation. (3) The heterogeneity analysis indicates that, in heavily polluting industries, the carbon emission trading policy has a more significant effect on the ESG performance of NSOEs than SOEs. However, the policy’s impact on the ESG performance of dual-role enterprises is less significant compared to split-role enterprises.

In addition to its conclusions, the research offers several notable contributions (1) It elucidates the mechanisms by which carbon emission trading policy influences the overall ESG performance of heavily polluting enterprises. By empirically testing the moderating effects of green technology innovation and financing constraints, the study expands the research perspectives on the micro-level effects of carbon emission trading policy and increases the focus on corporate performance across environmental, social, and governance dimensions. (2) By utilizing ESG performance ratings of Chinese listed companies provided by established rating agencies, the study conducts a detailed and in-depth examination of carbon emission trading policy’s effects, broadening the scope of research on enterprises’ comprehensive performance in environmental, social, and corporate governance aspects. (3) In light of the European Union’s Carbon Border Adjustment Mechanism (CBAM), which came into effect in May 2023, this study provides valuable guidance for heavily polluting enterprises in China on how to enhance their ESG performance, comply with the mechanism, and better integrate into global markets.

6.2. Policy Implications

Our findings provide some of the following policy recommendations: First, companies should prioritize and actively respond to low-carbon pilot policies by incorporating low-carbon transformation into their strategies and promoting high-quality development. During the policy implementation process, companies should pay attention to fiscal incentives and green finance policies in pilot areas to offset additional costs associated with energy saving and emission reduction, thus avoiding a vicious cycle of cost pressure and declining profitability.

Second, companies should enhance the capabilities of their management teams and broaden their strategic vision, planning sustainable development paths, optimizing resource allocation, seizing policy opportunities, phasing out outdated technologies and equipment, increasing research and development (R&D) investment, and improving innovation and production efficiency.

Third, local governments should formulate low-carbon transformation strategies based on industry characteristics, set transformation goals tailored to specific circumstances, and segment the strategies according to the level of difficulty in the transformation. They should encourage high-pollution industries to implement low-carbon transformations, guiding companies to fulfill their social responsibilities, enhance ESG performance, and promote high-quality development.