How Can We Improve the ESG Performance of Manufacturing Enterprises?—The Carbon Resilience Perspective

Abstract

In the context of low-carbon transformation, manufacturing enterprises are facing great pressures, and they need to improve their capability in order to successfully respond to these changes and achieve sustainable development. Based on the concept of organizational resilience, this paper proposed the concept of carbon resilience, representing the ability of an organization to keep stable, adapt, and evolve in the context of low-carbon transformation and tried to explore the role of carbon resilience in enterprises’ environment, social, and governance performance (ESG performance) and the conditional roles of coercive pressure, normative pressure, and the mimetic pressure between companies. Empirical research selected the data of Chinese A-share-listed manufacturing companies between 2012 and 2021 as research samples. Using the regression analysis method, the theoretical model was verified. The results show that carbon resilience can promote ESG performance. Moreover, coercive pressure and normative pressure promote the relationship between carbon resilience and ESG performance, whereas mimetic pressure inhibits the relationship. This study provides managerial implications for the government, society, and manufacturing enterprises, especially laying out a realistic approach to improving ESG performance from the perspective of carbon resilience.

1. Introduction

With the development of the global economy, global warming is a major concern threatening the living environment of humankind and restricting the sustainable growth of the world’s economy. The increase in carbon emissions is the main reason for global warming. To reduce the volume of carbon emissions, various carbon reduction actions have been carried out worldwide [1]. China proactively addresses environmental concerns and has set “dual carbon” target, aiming to achieve the goal of carbon peak by 2030 and carbon neutrality by 2060. According to data from the China Emission Accounts and Datasets from 2022, China’s manufacturing industry is responsible for 38.18% of total carbon emissions. Manufacturing enterprises are facing increasingly stringent low-carbon development requirements from the government, public media and competitors, etc. In this context, learning how to adapt to the changes brought by the low-carbon transition and how to promote manufacturing enterprises to improve sustainable development performance have become urgent needs for the development of manufacturing enterprises [2]. ESG performance, also known as the “triple bottom line” of the performance of an enterprise [3,4], represents a crucial criterion for measuring a company’s level of green and sustainable development [5]. ESG is becoming more and more popular since the introduction of the “dual carbon” target. Focusing on ESG performance allows enterprises to evaluate the balance between external environments and their internal governance and generate competitive advantages in turbulent environments. Especially for manufacturing enterprises, improving ESG performance serves as an important weapon to keep a good reputation and achieve sustainable development. Earlier studies have verified that high ESG performance is conducive to helping enterprises optimize their current strategies and improve their sustainability performance [6,7,8,9,10]. Given the importance of ESG performance, it is of great significance to find the influencing factors of ESG performance.

The “dual carbon” target will affect the concentration of labor, capital, and other factors. This impact will alter the inherent development path of the manufacturing industry. As a result, manufacturing enterprises will face new issues and challenges in low-carbon transformation [11]. Organizational resilience, the capacity of an organization to adjust and adapt in response to adverse environments, is critical for the sustainable development of an enterprise [12,13]. In this paper, we focused on the changes brought by low-carbon transformation and proposed the concept of carbon resilience. Grounded in the concept of organizational resilience [14,15,16,17], carbon resilience refers to the capability of an organization to keep stable, adapt and adjust, and evolve and transform under the changes brought by low-carbon transformation. Carbon resilience can be considered a special form of organizational resilience. Numerous studies have confirmed the impact of enterprises’ low-carbon behavior on various aspects of ESG performance, which includes carbon performance [18,19,20], sustainable performance [21], economic performance [22], environmental performance [23], corporate accountability [24], a firm’s brand value [25], and enterprises’ decision making [26]. Research on the relationship between enterprises’ low-carbon behavior and ESG performance further indicates that better carbon emission performance is positively associated with improved ESG performance [27]. Enterprises’ capability can shape corporate behavior and drive sustainable development. In the context of low-carbon transition, carbon resilience, as a capability that drives low-carbon transition and keeps companies running smoothly, is crucial for driving up ESG performance and sustainable development. However, existing studies lack the study of the relationship between carbon resilience and ESG performance.

Finally, it is worth mentioning that institutional pressures play an important role in the relationship between carbon resilience and ESG performance. Especially since the introduction of the “dual carbon” strategy, enterprises not only have to deal with policy and regulatory restrictions, but also face scrutiny from the public and the media, as well as intense industry competition [28,29,30,31]. These pressures influence the extent to which enterprises fulfill their ESG responsibilities in different ways. In the case of steel manufacturing companies, for example, those that fail to meet the carbon emission requirements set by the government are facing the forced closure of their production lines. At the same time, companies are increasingly focusing on pursuing a green image as the public and media pay more attention to their low-carbon transformation behaviors. In addition, the strategic action programs and achievements of the outstanding enterprises in the same industry have brought invisible pressures on the other enterprises in the industry. For example, HBIS Ltd. Co, for its part, has promoted its low-carbon transition by increasing investment in low-carbon innovation activities, relocating factories and optimizing management methods. This may lead to other companies blindly copying their practices and focusing too much on short-term economic benefits at the expense of long-term carbon reduction and ESG performance investments. Therefore, institutional pressures may serve as moderators in the relationship between carbon resilience and ESG performance.

Given this, grounded in dynamic capability theory, this study examines the impact of carbon resilience on ESG performance under different institutional pressures. To achieve the research goal, we utilize panel data from Chinese listed manufacturing enterprises spanning from 2012 to 2021. The results contribute to the existing literature in three ways. First, this study expands research on organizational resilience in the context of low-carbon transformation by introducing the concept of carbon resilience and developing a comprehensive measurement index. This offers new perspectives and approaches for manufacturing enterprises to cope with the demands of low-carbon transformation. Second, the research uncovers the relationship between carbon resilience and ESG performance, offering a realistic way for enterprises to improve ESG performance from the perspective of organizational resilience. This study broadens the understanding of the relationship between carbon resilience and ESG performance and extends our understanding of influential factors on ESG performance. Third, this study clarifies the moderating role of different kinds of institutional pressures between carbon resilience and ESG performance, addressing the call for research on organizational resilience across various contexts.

The remainder of this paper is organized as follows: Section 2 makes reviews according to the relevant literature. Section 3 develops the hypotheses. Section 4 comprises the design of our models and specification of the variables. Section 5 outlines the empirical results. Section 6 illustrates discussions and the theoretical and managerial implications.

2. Literature Review

2.1. ESG Performance

ESG performance, a comprehensive definition of environment, social, and governance performance, is a crucial criterion for the international community to assess the level of green and sustainable development achieved by enterprises. Enterprise environmental performance emphasizes that enterprises strive to minimize damage to the natural environment in all of their processes. Enterprise social performance focus on the relationship between enterprises and various stakeholders. Enterprise governance performance emphasizes the effectiveness of the internal and external governance of an enterprise. Previous studies have explored various factors influencing ESG performance, categorized into external and internal factors. The external factors include environmental uncertainty, environmental regulation, green finance reform, media oversight, and product competition. To be specific, enterprises are more likely to engage in ESG activities in uncertain economic conditions [32], and the implementation of China’s Environmental Protection Law has improved ESG performance in state-owned enterprises [33]. Green finance reform policies have also contributed to higher ESG performances [34]. Additionally, enterprises face regulatory pressures from society and the media, compelling them to fulfill ESG responsibilities [35]. However, intense product competition has been linked to lower ESG performance [36].

The internal factors affecting ESG performance include financial status, management characteristics, executive equity incentives, green innovation, manager political preferences, dynamic capabilities, and organizational resilience. In particular, studies have shown a strong correlation between ESG performance and financial factors [37]. Firms led by female managers tend to exhibit superior ESG performances [38], and green innovation contributes to ESG improvements [39]. Additionally, the quality of an enterprise’s technological environment plays a crucial role [40]. Enterprises’ technology environment not only includes enterprises’ technology knowledge, but also includes enterprises’ technology management, innovation, and learning capabilities. Research also highlights that absorptive and adaptive capabilities support ESG strategy implementation [41]. Based on COR theory, organizational resilience has a complex impact on ESG performance, sometimes acting as a double-edged sword [42], while other studies suggest that its impact is entirely positive [43].

2.2. Carbon Resilience

2.2.1. Definitions of Carbon Resilience

Based on the concept of organizational resilience, this paper focuses on the impact of low-carbon transformation on manufacturing enterprises and proposes the novel concept of carbon resilience. Carbon resilience can also be regarded as a special kind of organizational resilience.

The word resilience originates from the Latin word “resilio”. In physics, resilience is used to describe the strain resistance of an object under strong pressure. Meyer first introduced resilience into the field of management [44]. Organizational resilience refers to the ability of a firm to adapt to and recover from challenges while maintaining or enhancing its functions [45]. Since its introduction, organizational resilience has gained increasing attention.

Currently, scholars mainly view organizational resilience from the following different perspectives: trait, outcome, capability, and process, as detailed in Table 1. From the trait perspective, organizational resilience is defined as the inherent characteristics or qualities that enable organizations to cope with crises and challenges [13,15]. Based on the outcome perspective, scholars have placed more emphasis on the results of organizational resilience, such as building new stakeholders and improving financial capacity [46,47]. The process perspective views organizational resilience as the process by which organizations respond to, recover from, and improve after crises [48,49]. Lastly, the capability perspective conceptualizes organizational resilience as the capacity to turn crises into opportunities, the ability to enable organizational growth, and the ability to adapt to change [50,51,52]. Overall, scholars mainly define organizational resilience from the perspective of capability.

We further analyzed the definition of carbon resilience from the capability perspective of organizational resilience. Existing research suggests that organizational resilience is a multifaceted dynamic framework composed of various interrelated capacities. Pettit identified three key dimensions of organizational resilience: anticipatory capacity, adaptive capacity, and recovery capacity [53]. Other studies highlight preparedness and adaptability as fundamental components of organizational resilience [54]. Additionally, scholars have categorized organizational resilience into three elements: cognitive resilience, behavioral resilience, and situational resilience [55]. Combining the above views, we believe that carbon resilience is a combination of stabilizing, adaptive, and evolutionary capabilities that can help organizations effectively respond to shocks from low-carbon transformation. Specifically, the ability to stabilize is the fundamental capacity of a firm to withstand low-carbon risks and preserve its own stability. It is expressed as the speed of a firm’s response to take emergency action when an external crisis occurs. The stronger the ability to stabilize, the less likely a company will be affected in the face of external low-carbon shocks. The ability to adapt is a company’s capacity to reallocate resources and adjust its actions. Manufacturing companies will be able to proactively react and adapt to low-carbon shocks and better avoid low-carbon hazards with higher adaptive capability. The ability to evolve is a crucial capability for manufacturing companies to achieve sustainable survival and long-term growth. In order to proactively fulfill the demands of the external low-carbon environment, enterprises need to learn from previous experiences, improve their crisis response process, and optimize their corporate governance structure. Ultimately, manufacturing companies should turn crises into opportunities and achieve green and sustainable development.

Table 1. Representative definitions of organizational resilience.

Perspective	Author	Definition
Trait	(Home III & Orr, 1997 [15])	Organizational resilience is an essential quality which can effectively respond to changes.
	(Shashi et al., 2020 [13])	Organizational resilience is a potential framework for dealing with crises.
Outcome	(Lengnick-Hall et al., 2011 [47])	Organizational resilience is developed from a combination of organizational-level and individual-level outcomes.
	(Bento et al., 2021 [46])	Organizational resilience is an outcome of identifying characteristics.
Process	(Allen & Toder, 2004 [48])	Organizational resilience is the process of recovery from damage.
	(McCarthy et al., 2017 [49])	Organizational resilience can be seen as an evolutionary process.
Capacity	(Fiksel, 2006 [53])	Organizational resilience is the ability of a company to survive, adapt, and grow in the face of turbulent change.
	(Valero et al., 2015 [56])	Organizational resilience is the ability of an organization to adapt or recover in the face of changes.
	(Ortiz-de-Mandojana & Bansal, 2016 [51])	Organizational resilience is the ability of an organization to anticipate and adapt to its environment.
	(Chowdhury et al., 2019 [57])	Organizational resilience is the ability to recover from adverse situations.
	(Sajko et al., 2021 [52])	Organizational resilience is the ability of an organization to anticipate, adjust, and adapt to external shocks.

Table 1. Representative definitions of organizational resilience.

Perspective	Author	Definition
Trait	(Home III & Orr, 1997 [15])	Organizational resilience is an essential quality which can effectively respond to changes.
	(Shashi et al., 2020 [13])	Organizational resilience is a potential framework for dealing with crises.
Outcome	(Lengnick-Hall et al., 2011 [47])	Organizational resilience is developed from a combination of organizational-level and individual-level outcomes.
	(Bento et al., 2021 [46])	Organizational resilience is an outcome of identifying characteristics.
Process	(Allen & Toder, 2004 [48])	Organizational resilience is the process of recovery from damage.
	(McCarthy et al., 2017 [49])	Organizational resilience can be seen as an evolutionary process.
Capacity	(Fiksel, 2006 [53])	Organizational resilience is the ability of a company to survive, adapt, and grow in the face of turbulent change.
	(Valero et al., 2015 [56])	Organizational resilience is the ability of an organization to adapt or recover in the face of changes.
	(Ortiz-de-Mandojana & Bansal, 2016 [51])	Organizational resilience is the ability of an organization to anticipate and adapt to its environment.
	(Chowdhury et al., 2019 [57])	Organizational resilience is the ability to recover from adverse situations.
	(Sajko et al., 2021 [52])	Organizational resilience is the ability of an organization to anticipate, adjust, and adapt to external shocks.

2.2.2. Measurement of Carbon Resilience

There is no consensus on the measurement of organizational resilience, which can be assessed using two main approaches: direct measurement and indirect measurement. The direct measurement approach involves developing scales based on specific definitions and traits of organizational resilience. For instance, scholars have measured organizational resilience using three key criteria: robustness, adaptability, and integrity [58]. Another model categorized organizational resilience into three factors: situational awareness, critical vulnerability management, and adaptive capacity [59]. Additionally, resilience has been measured using the following sets of three dimensions: anticipatory capacity, adaptive capacity, and resilience [53], as well as robustness, agility, and integrity [60].

The indirect measuring approach creates indicator indexes using publicly available financial data. For example, post-shock stock price volatility has been used to gauge organizational resilience [61]. Other studies have measured resilience by using long-term sales growth and financial volatility [62], or loss severity and recovery time [63]. Based on our research sample, it is necessary to develop a comprehensive measurement index based on the definition of carbon resilience.

2.3. Institutional Pressures

Institutional pressures are the combination of pressures such as rules, norms, social perceptions, and culture. They motivate firms to become legitimate subjects by impacting the rationalization of their structure, form, and behavior [64]. The early institutional analysis stage and the new institutional stage are two important components of institutional pressures. Early institutionalism considered that the behavior of enterprises was not the result of the rational economy, but that it was influenced by environmental factors in their actual operation [65]. Based on research about institutional analysis, Meyer and Rowan founded the school of neo-institutionalism [66]. They held that the environment in which a corporation is embedded affects its behavior. The environment encompasses not only the technological surroundings but also the institutional landscape. This institutional environment consists of the legal systems, social expectations, and cultural perceptions within which an entity is embedded [67].

The reasons for organizational convergence have been further discussed from the level of organizational relationships and organizational domains, and institutional pressures were divided into three dimensions: coercive pressure, normative pressure, and mimetic pressure [68]. Among them, coercive pressure refers to the coercive laws, policies, norms, and so on. Normative pressure refers to rules based on societal norms and moral principles. Mimetic pressure is the term used to describe the pressure brought by successful organizations in the same industries. It is believed that the institutional pressures consist of three levels: regulation, norms, and cultural cognition [69]. Therefore, this study classifies institutional pressures into three categories: coercive pressure, normative pressure, and mimetic pressure.

2.4. Theoretical Framework

Governments, the media, benchmark companies, and other entities have put significant emphasis on navigating the low-carbon revolution of manufacturing enterprises. Carbon resilience, which reflects an organization’s ability to remain stable, adapt, and evolve amid low-carbon transformation, is vital for influencing enterprises’ low-carbon behavior and driving the sustainable growth of manufacturing enterprises. The existing literature has extensively explored the relationship between low-carbon behavior and ESG performance. For instance, it has been discovered that enterprises’ carbon strategies can drive higher carbon performances [18]. Additionally, it has been shown that enterprises’ green innovation can enhance ESG performances [43]. However, it remains unclear how carbon resilience can influence ESG performance in the context of low-carbon transformation. Additionally, corporate operations are closely linked to institutional pressures from governments, the media, and benchmark enterprises. Especially against the background of low-carbon transformation, enterprises’ activities are restricted by more and more strict policies, voices from the media, and low-carbon technology innovation pressure from benchmark enterprises. These pressures motivate firms to become legitimate subjects by impacting the rationalization of their structure, form, and behavior [30]. In the broader context of climate change, institutional and stakeholder pressures have been shown to enhance organizational capabilities and corporate competitiveness [70]. Therefore, drawing on dynamic capability theory, this study examines the relationship between carbon resilience and ESG performance and the boundary effect of institutional pressures (coercive pressure, normative pressure, and mimetic pressure) between them. Figure 1 shows the theoretical model.

3. Research Hypothesis

3.1. Carbon Resilience and ESG Performance

Carbon resilience, the ability of an enterprise to remain stable, adjust and adapt, and evolve beyond challenges, helps firms anticipate, identify, and respond to crises in uncertain environments. It also enables them to leverage their strengths and adapt effectively after a crisis. Previous research suggests that a company’s resilience capability determines its ability to withstand shocks [12,42,60,71]. Accordingly, the higher the carbon resilience of a manufacturing enterprise, the more the enterprise can cope with the risks brought by the “dual carbon” target and lowering carbon emissions, ultimately leading to an improved ESG performance [27].

From an environmental perspective, carbon resilience, as an organizational capability of enterprises to cope with low-carbon shocks, enables enterprises to respond swiftly to shocks and utilize available resources to mitigate their impact. The improvement of enterprises’ low-carbon integration capability is conducive to improving enterprises’ environmental performance [72]. Moreover, carbon resilience fosters the development and promotion of low-carbon products and technologies, facilitating the low-carbon transformation of manufacturing enterprises.

From a social perspective, as awareness of environmental issues grows, stakeholders are increasingly focused on balancing economic benefits with environmental responsibility rather than pursuing financial performance alone [73,74]. Enterprises with high carbon resilience are concerned with green concepts, would like to increase investments into R&D activities, and are concerned with the fulfillment of corporate responsibility. This not only meets stakeholder expectations but also attracts investments from a broader range of stakeholders.

From a governance perspective, in order to achieve sustainable business goals, enterprises need to optimize their management methods and make appropriate strategies [9,63]. Manufacturing enterprises with high carbon resilience will make strategies accordingly and adjust their governance structures to stimulate the transmission of green development concepts across the whole company. The implementation of corporate carbon information disclosure strategies has also been shown to positively impact corporate brand value [25]. In sum, we argue that manufacturing companies with carbon resilience have better external perception capabilities, abilities to adjust and transform, and self-renewal capabilities, which are favorable in promoting ESG performance. Consequently, we propose the following hypothesis:

H1. 

Carbon resilience positively affects enterprises’ ESG performance.

3.2. The Moderating Role: Institutional Pressures

The institutional environment is crucial to a company’s ability to fulfill its ESG obligations. Research suggests that social, economic, and legal factors significantly shape responsible corporate behavior at the national level [75]. Additionally, the ESG performance of listed companies is often viewed as a strategic response to external institutional complexities [76]. Thus, institutional pressures may affect enterprises’ abilities to achieve high ESG performances. How, then, do institutional pressures impact on the connection between carbon resilience and ESG performance? Do different types of institutional pressures affect the relationship differently? The impact of types of institutional pressures on the relationship between carbon resilience and ESG performance was further investigated.

3.2.1. Coercive Pressure

Coercive pressure describes the influence exerted by powerful institutions like the government and state. These power institutions monitor organizational behavior by making laws and regulations [64,68]. The pressure comes in the form of environmental regulations, legal requirements, incentives, penalties, and pollution control policies [77]. These rules and standards constrain the social behavior of firms [78,79]. Since the implementation of the “dual carbon” policy, more and more policies have been published to restrict the volume of carbon emissions.

As a result, manufacturing enterprises face high coercive pressure, compelling them to align with these regulatory requirements by enhancing their carbon resilience. The stronger the coercive pressure, the more likely enterprises are to develop and strengthen their carbon resilience.

Accordingly, carbon resilience will lead to a series of changes in manufacturing enterprises, such as proactively increasing environmental investments, making green technology innovations, improving corporate environmental performance, and adopting environmentally friendly strategies to avoid the risk of being penalized. This, in turn, leads to an enhancement of corporate ESG performance. Therefore, we proposed the following hypothesis:

H2a. 

Coercive pressure positively moderates the relationship between carbon resilience and ESG performance.

3.2.2. Normative Pressure

Normative pressure originates from cultural identity, value realization, environmental perceptions, and educational attainment. Normative pressure includes professional standards, group perceptions, and pressures brought by organizations and groups. The attitudes, expectations, and standards of external stakeholders constrain the behavior of firms [68,80,81].

To meet stakeholder expectations, companies must adjust their strategies and make adjustments aligned with the accepted norms of behavior in the organizational field, leading to external recognition and support [82]. A trusting and cooperative relationship with stakeholders enhances a company’s competitive advantage and contributes to improved performance. Since the introduction of the “dual carbon” strategy, the media, environmental institutions, and the public have exerted increasing pressure on enterprises. On the one hand, ESG performance has become a major societal concern. For the media, they increasingly publish news on the public’s expectations of the fulfillment of ESG responsibility. To respond to the public’s expectations, enterprises will take measures to solve environmental problems, behave in ways aligned with the expectations of external stakeholders, and cultivate their carbon resilience. On the other hand, when enterprises demonstrate a high degree of carbon resilience, the media tend to highlight their success stories, helping them to build a positive corporate image. Therefore, normative pressure can encourage enterprises to act according to the expectations from external stakeholders. By shaping carbon resilience, enterprises will win reputation and obtain adequate resources, thus improving their ESG performance. Therefore, we proposed the following hypothesis:

H2b. 

Normative pressure positively moderates the relationship between carbon resilience and ESG performance.

3.2.3. Mimetic Pressure

Mimetic pressure refers to the imitation approach used by firms to simulate successful or similar organizations [68,83]. Unlike coercive pressure, mimetic pressure is a non-mandatory behavior adopted voluntarily by firms [84]. On the one hand, when enterprises are under great mimetic pressure, they may refuse to increase investments to avoid increased costs and decreased incomes. This reluctance can weaken the link between carbon resilience and ESG performance. The Schumpeter effects states that innovation is driven by higher expected profits and that competition will reduce innovation profits [85]. When faced with fierce competition, enterprises with carbon resilience may reduce investments in innovative activities to decrease the loss of innovation profits, which eventually inhibits ESG performance. On the other hand, when the pressure from the market is excessively high, enterprises need to screen a lot of uncertain information from their external channels. This leads to high time costs and weakens the benefit which carbon resilience can have on ESG performance. All in all, in environments with high uncertainty, enterprises would like to decrease their investments and make strategies in consistent with their competitors to avoid losing competitive advantages, which in turn hinders the ESG performance of enterprises. High mimetic pressure will diminish the relationship between carbon resilience and ESG performance. Therefore, we proposed the following hypothesis:

H2c. 

Mimetic pressure negatively moderates the relationship between carbon resilience and ESG performance.

4. Method

4.1. Sample and Data Collection

The research sample of this paper was Chinese A-share-listed manufacturing companies, spanning from 2012 to 2021. We selected our research sample of Chinese manufacturing companies from 2012 to 2021 mainly for two reasons.

First, since 2012, the Chinese government have realized that the traditional development mode could not satisfy the long-term requirements of China’s economy. As a result, more and more policies have been published to guide manufacturing companies to lower their carbon emissions and achieve green transformation [86]. Second, ESG is not only highly consistent with the green manufacturing and high-quality development strategies of China’s manufacturing industry, but also coincides with China’s “dual carbon” strategy [87]. The manufacturing industry accounts for 38.18% of total carbon emissions. Guiding manufacturing enterprises to disclose their environmental information will stimulate them to adjust their current business activities and decrease carbon emissions, which will be helpful to achieve the goal of carbon neutrality by 2060.

The primary sources of data for this paper were the annual reports of China’s manufacturing companies and some statistical databases. Data to measure carbon resilience came from the CSMAR database and the annual reports of listed corporations. Data relevant to ESG performance were sourced from the Wind database. The other variables included in our work were obtained from the China Environment Yearbook, China Research Data Service Platform (CNRDS), and CSMAR database. To obtain a more accurate result, the following measures were used in our paper: (1) companies with ST and *ST were deleted, and (2) enterprises with serious missing data were deleted. At the 1% and 99% levels, Winsorization was applied to all the continuous variables in order to mitigate the impact of extreme values. We used Stata 16.0 to analyze the data, test our research model, and conduct a series of robustness tests.

4.2. Variable Design

4.2.1. Dependent Variable: ESG Performance

Referring to pervious research [88,89], we used CSI ESG rating to measure ESG performance.

4.2.2. Independent Variable: Carbon Resilience

Based on previous clarification, carbon resilience can be understood as the ability of an organization to stabilize, adapt, and evolve in the face of changes in low-carbon transformation.

The ability of an organization to stabilize refers to its capacity to keep operating steadily under adverse circumstances. In the face of low-carbon shocks, enterprises with advanced low-carbon technologies can launch timely responses to external changes [90]. Additionally, the availability of skilled talent, financial resources, and asset reserves guarantees the stability of an organization [91]. Therefore, we chose carbon technology integration capability, financial resource integration capability, human resource integration capability, and asset integration capability as the first-level indicators of the ability to stabilize.

Enterprises with fewer carbon emissions and high carbon technology application levels can be less affected by external changes [92]. Therefore, carbon emission intensity and carbon technology application level were selected as the second-level indicators of carbon technology integration capability. Using Formulas (1) and (2), the carbon emission intensity of enterprises was calculated, referring to Chapple [93] and Shen [94], with the CO₂ conversion factor sourced from standard 2.493 of the Xiamen Energy Conservation Center. The degree of carbon technology use was determined by dividing the total number of words in the annual reports of listed companies by the number of related keywords [95]. Referring to the relevant research [96], we selected green power, technological innovation, fuel substitution, green process, and efficient operation of coal power as the keywords. For financial resource integration ability, financial position and operating results can be an important signal of the stability ability of a company. Green credit policies have the potential to offer financial assurances for businesses engaging in green innovation, ultimately enhancing both environmental and financial performance [97]. Additionally, long-term growth and financial volatility are indicators of organizational resilience [62]. Therefore, we selected industry classification under green credit restrictions and return on assets (ROA) as the second-level indicators of financial resource integration capability.

For human resource integration capability, talents are fundamental driving forces that support the long-term development of enterprises [98]. One of the second-level indications of talent resource integration capacity was the ratio of R&D personnel. For asset integration capability, adequate assets play an important role when faced with emergencies [99]. The inventory percentage was selected as a second-level indicator of asset consolidation capability.

The ability of an organization to adapt is its capacity to modify its structure and resources in a shock and revert to its initial state. The more adaptive a firm is, the more it can proactively adjust and adapt to new changes [100]. Therefore, we chose adjustment capability and transformation capability as the first-level indicators of the ability to adapt. For adjustment capability, enterprises’ adjustment activities are driven by executives’ decisions. The degree of corporate green innovation increases with the executive team’s attention to environmental protection measures [98,101]. Therefore, the green perception of executives was selected as a second-level indicator of adjustment capability. Referring to relevant research [102], the proportion of keyword terms to total words in listed firms’ annual reports was used in order to gauge executives’ opinions about the environment. We chose environmental protection concept, environmental education, environmental protection inspector, and environmental protection work as our keywords. For transformation capability, the development of an industry will be directly impacted by the introduction of new products and the emergence of new technology [103]. Emphasizing innovation in green products and technologies can help manufacturing enterprises to quickly adapt to external environmental changes. Therefore, the quantity of green patent applications was selected as a second-level indicator of transformation capability.

The ability to evolve refers to an organization’s capacity to focus on long-term goals and ultimately achieve green transformation. By learning and reflecting on their crisis response process, manufacturing enterprises can optimize their internal structure, change their production model, and adjust their development strategy, thus promoting a better green and low-carbon transition [104]. Therefore, we chose strategic change capability and innovation capability as the first-level indicators of the ability to evolve. For strategic change capability, the proactive adoption of low-carbon strategies enables manufacturing enterprises to gain a competitive advantage [105]. Therefore, environmental strategic planning was selected as a second-level indicator of strategic change capability. Corporate environmental strategy was measured by the ratio of the number of words containing keywords related to the strategy to the total number of words in the listed companies’ annual reports [102]. For innovation capability, innovation serves as a key driver of long-term development [90]. Therefore, we selected R&D input intensity as a second-level indicator of innovation capability.

Ultimately, an assessment framework was developed to gauge carbon resilience, incorporating 8 first-level indicators and 10 second-level indicators, as shown in

Table 2. Representative indicators of carbon resilience.

Dimensionality	First-Level Indicators	Second-Level Indicators	Data Source	Property
The ability to stabilize	Carbon technology integration capability	The carbon emission intensity	Formula method	Negative
		The carbon technology application level	Text analysis	Positive
	Financial resource integration capability	Whether enterprises belong to the green credit-restricted industries	Green Credit Guidelines	Negative
		The return on assets	CSMAR	Positive
	Human resource integration capability	The ratio of the number of R&D personnel	CSMAR	Positive
	Asset integration capability	The inventory percentage	CSMAR	Positive
The ability to adapt	Adjustment capability	The green perception of executives	Text analysis	Positive
	Transformation capability	The number of green patent applications	CSMAR	Positive
The ability to evolve	Strategic change capability	Environmental strategic planning	Text analysis	Positive
	Innovation capability	The R&D input intensity	CSMAR	Positive

.

$$Enterprise carbon emission intensity={\displaystyle \frac{Enterprise carbon emissions}{Enterprise main revenue}}$$

(1)

$$Enterprise carbon emission={\displaystyle \frac{Totalindustryenergyconsumption\times {CO}_{2}conversionfactor}{Industrymaincost}}$$

(2)

In order to reduce the artificial influence factors, the entropy weight method (EWM) was utilized to obtain a comprehensive result of carbon resilience. The calculation procedures were as follows:

The first step was to conduct a standardization treatment of the indicators. For positive indicators, we used Equation (3) to achieve the standardization result; for negative indicators, Equation (4) was used to obtain the result.

$$X_{ij}^{\prime }={\displaystyle \frac{X_{ij}-min\left\{X_j\right\}}{max\left\{X_j\right\}-min\left\{X_j\right\}}}$$

(3)

$$X_{ij}^{\prime }={\displaystyle \frac{max\left\{X_j\right\}-X_{ij}}{max\left\{X_j\right\}-min\left\{X_j\right\}}}$$

(4)

where X′_ij represents the value of the jth indicator of the ith enterprise; min{X_j} represents the minimum value of the jth indicator; max{X_j} represents the maximum value of the jth indicator; and X’_ij represents the standardization result of the jth indicator of ith enterprise.

The second step was to calculate the weight of $j$th indicator of the $i$th enterprise in total enterprises:

$${\omega }_{ij}={\displaystyle \frac{X_{ij}^{\prime }}{{\sum }_{j=1}^m{X}_{ij}^{\prime }}}$$

(5)

The third step was to calculate the entropy value:

$$e_j=-{\displaystyle \frac{1}{\ln m}}{\sum }_{j=1}^m{\omega }_{ij}\times \ln {\omega }_{ij}$$

(6)

The fourth step was to calculate the coefficient of difference:

$$d_j=1-e_j$$

(7)

The fifth step was to calculate the weights of each indicator. The larger the index coefficient is, the higher the weight is:

$${\phi }_j={\displaystyle \frac{d_j}{{\sum }_{j=1}^m{d}_j}}$$

(8)

Finally, a comprehensive index of carbon resilience was obtained according to Equation (9).

$${CR}_i={\sum }_{j=1}^m{\phi }_j\times {\omega }_{ij}$$

(9)

4.2.3. Moderating Variable: Institutional Pressures

Referring to previous research [68], we chose coercive pressure, normative pressure, and mimetic pressure as moderating variables.

Government laws and regulations serve as a key source of coercive pressure for manufacturing enterprises. When more regulations are enacted in a particular region, firms experience stronger coercive pressure [106]. Drawing on previous research [107], we used the quantity of environmental regulations issued by local governments as a proxy for coercive pressure.

The public, customers, and communities are the primary sources of normative pressure. The more attention an enterprise receives from the external media, the greater the perceived normative pressure [108]. Based on prior studies [109], we measured normative pressure by the total number of headline and content appearances in the news of the public companies.

Mimetic pressure arises from the pressure from benchmark firms. We adopted the 1-Herfindahl index to measure the mimetic pressure. The market is less competitive with a high Herfindahl index. The formula is as follows:

$$COM=1-{\sum }_i^a\left({\displaystyle \frac{X_i}{X}}\right)$$

(10)

X_i denotes the current operating revenue of enterprise i, X denotes the total current operating revenue of all the enterprises in the industry, X_i/X denotes the market share of enterprise i, and a represents the sum of the number of enterprises in the industry.

4.2.4. Control Variables

In accordance with earlier research [110,111,112], the following control variables were added to the regression model: (1) the equity ratio (ER), measured by the ratio of total liabilities to total equity; (2) the separation of powers ratio (SR), measured by the difference between the separation of rights in ownership and control; (3) the asset–liability ratio (Lev), measured by the ratio of total liabilities to total assets; (4) the cash flow ratio (Cashflow), measured by the ratio of surplus before interest income tax depreciation and amortization to annual interest expense; (5) the ratio of independent directors (Indep), measured by the number of independent directors to the total number of directors; (6) the shareholding ratio of the largest shareholder (Top1), measured by the number of shares held by the largest shareholder to the total number of shares; (7) firm age (Age), measured by the logarithm of the number of years from the date of incorporation to the observation date plus one; and (8) corporate growth (Growth), measured by the logarithm of the difference between current-year operating income and prior-year operating income minus one.

5. Results

5.1. Model Construction

5.1.1. Main Effect Model

To test the impact of carbon resilience on the ESG performance of manufacturing enterprises, the following research model was constructed:

$${\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i}\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{C}\mathrm{R}}_{\mathrm{i}\mathrm{t}}+\mathsf{\phi }\mathrm{Z}+{\mu }_i+{\delta }_n+{\epsilon }_{\mathrm{i}\mathrm{t}}$$

(11)

where ESG_it represents ESG performance; CR_it represents carbon resilience; Z represents the control variables; β₀ represents the intercept term; μ represents the industry fixed effect; δ represents time fixed effect; and ε_it represents a random error term.

5.1.2. Moderating Effect Model

To test the moderating effect of institutional pressures, the following model was constructed:

$${\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i}\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{C}\mathrm{R}}_{\mathrm{i}\mathrm{t}}+{\beta }_2{M}_{\mathrm{i}\mathrm{t}}+{\beta }_3{\mathrm{C}\mathrm{R}}_{\mathrm{i}\mathrm{t}}\times M_{\mathrm{i}\mathrm{t}}+\mathsf{\phi }\mathrm{Z}+{\mu }_i+{\delta }_n+{\epsilon }_{\mathrm{i}\mathrm{t}}$$

(12)

where M_it is the moderating variable and CR_it × M_it represents the interaction term between carbon resilience and the moderating variable.

5.2. Descriptive Analysis

The results of the descriptive statistics analysis are displayed in

Table 3. Descriptive analysis and correlations.

	Mean	SD	ESG	CR	POL	MEDIA	COM	ER	SR	Lev	Cashflow	Indep	Top1	FirmAge	Growth
ESG	6.3041	1.043	1
CR	0.0178	0.016	0.064 ***	1
POL	5.0972	0.594	−0.019 **	0.144 ***	1
MEDIA	4.5077	1.101	0.091 ***	0.189 ***	−0.155 ***	1
COM	0.8930	0.083	0.00900	0.042 ***	−0.055 ***	0.032 ***	1
ER	0.7658	0.669	−0.00300	0.00900	−0.027 ***	0.137 ***	−0.073 ***	1
SR	4.5034	7.400	0.064 ***	0.00500	0.00600	0.019 *	−0.036 ***	0.098 ***	1
Lev	0.3727	0.179	0.0150	−0.00900	−0.026 ***	0.142 ***	−0.078 ***	0.926 ***	0.094 ***	1
Cashflow	0.0549	0.064	0.104 ***	−0.064 ***	0.00100	0.047 ***	−0.0100	−0.178 ***	0.061 ***	−0.187 ***	1
Indep	0.3785	0.054	−0.0100	−0.020 **	−0.031 ***	0.038 ***	0.00700	−0.00200	−0.079 ***	0.00300	−0.00200	1
Top1	0.3273	0.138	0.108 ***	0.025 ***	0.00400	−0.023 **	−0.071 ***	−0.028 ***	0.178 ***	−0.041 ***	0.122 ***	0.070 ***	1
FirmAge	2.9179	0.285	0.076 ***	−0.236 ***	−0.137 ***	0.00700	−0.037 ***	0.095 ***	0.065 ***	0.105 ***	0.039 ***	−0.017 *	−0.043 ***	1
Growth	0.1812	0.347	0.022 **	−0.017 *	0.0120	0.100 ***	0.060 ***	0.021 **	−0.0160	0.050 ***	0.029 ***	−0.00500	0.00500	−0.068 ***	1

Note. * p < 0.1, ** p < 0.05, *** p < 0.01.

. The mean and standard deviation of ESG performance are 66.3041 and 1.043, showing that there are substantial differences in the level of ESG performance between various enterprises. The average value of carbon resilience (CR) is 0.0178, meaning that most manufacturing companies have low levels of carbon resilience. The average value of coercive pressure (POL) is 5.0972, showing that enterprises are facing high coercive pressures. The average value of normative pressure (MEDIA) is 4.5077, showing that enterprises face high normative pressures. The average value of mimetic pressure (COM) is 0.893, which means that enterprises are facing low mimetic pressure. The results of Spearman’s correlation demonstrate a significant positive relationship between carbon resilience and ESG performance at the 0.1% level, meaning that ESG performance is strongly associated with carbon resilience. The variance inflation factors (VIFs) of all predictors ranged from 1.03 for growth to 7.1 for the asset–liability ratio. All the VIFs were less than 10, showing that there is no significant multicollinearity in this paper.

5.3. Main Effect Test

Table 4. Effect of carbon resilience on ESG performance and moderating effects of institutional pressures.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
	ESG	ESG	ESG	ESG	ESG	ESG	ESG
CR	8.9061 ***	8.9352 ***	8.6268 ***	8.4685 ***	7.3365 ***	8.9086 ***	9.0190 ***
	(8.5730)	(8.5953)	(8.1832)	(8.1282)	(6.3015)	(8.5750)	(8.6794)
POL		−0.0143	−0.0194
		(−0.7796)	(−1.0441)
CR × POL			1.7817 *
			(1.7535)
MEDIA				0.0525 ***	0.0558 ***
				(4.7324)	(4.9853)
CR × MEDIA					1.5275 **
					(2.1771)
COM						−0.1627	0.0433
						(−0.4472)	(0.1171)
CR × COM							−22.8155 ***
							(−3.0352)
ER	−0.2251 ***	−0.2252 ***	−0.2260 ***	−0.2284 ***	−0.2272 ***	−0.2253 ***	−0.2257 ***
	(−5.7303)	(−5.7326)	(−5.7530)	(−5.8190)	(−5.7900)	(−5.7338)	(−5.7467)
SR	0.0038 ***	0.0038 ***	0.0038 ***	0.0038 ***	0.0038 ***	0.0038 ***	0.0038 ***
	(2.7539)	(2.7692)	(2.7204)	(2.7658)	(2.7937)	(2.7551)	(2.7620)
Lev	0.9857 ***	0.9853 ***	0.9889 ***	0.9475 ***	0.9408 ***	0.9862 ***	0.9820 ***
	(6.6313)	(6.6286)	(6.6526)	(6.3715)	(6.3261)	(6.6344)	(6.6084)
Cashflow	1.7136 ***	1.7206 ***	1.7175 ***	1.6671 ***	1.6633 ***	1.7127 ***	1.7065 ***
	(10.5419)	(10.5686)	(10.5500)	(10.2473)	(10.2253)	(10.5347)	(10.4997)
Indep	−0.2355	−0.2394	−0.2321	−0.2747	−0.2785	−0.2352	−0.2363
	(−1.2734)	(−1.2941)	(−1.2547)	(−1.4854)	(−1.5066)	(−1.2719)	(−1.2783)
Top1	0.6186 ***	0.6168 ***	0.6200 ***	0.6314 ***	0.6312 ***	0.6186 ***	0.6200 ***
	(8.1899)	(8.1624)	(8.2034)	(8.3624)	(8.3622)	(8.1898)	(8.2107)
FirmAge	0.3066 ***	0.3055 ***	0.3051 ***	0.3069 ***	0.3065 ***	0.3068 ***	0.3063 ***
	(8.2015)	(8.1646)	(8.1534)	(8.2166)	(8.2085)	(8.2049)	(8.1961)
Growth	0.0056	0.0059	0.0064	−0.0074	−0.0093	0.0057	0.0065
	(0.1900)	(0.2003)	(0.2173)	(−0.2512)	(−0.3182)	(0.1960)	(0.2239)
_cons	4.8056 ***	4.8780 ***	4.9638 ***	4.5925 ***	4.6043 ***	4.9463 ***	4.8648 ***
	(13.4523)	(13.2153)	(13.3325)	(12.7680)	(12.8017)	(10.3898)	(10.2064)
N	10,226	10,226	10,226	10,226	10,226	10,226	10,226
industry	Yes	Yes	Yes	Yes	Yes	Yes	Yes
year	Yes	Yes	Yes	Yes	Yes	Yes	Yes
F	19.6137	19.2085	18.8762	19.7132	19.4083	19.1991	19.0062
R2	0.0814	0.0815	0.0818	0.0834	0.0839	0.0814	0.0823

Note. * p < 0.1, ** p < 0.05, *** p < 0.01.

shows the empirical results of carbon resilience and ESG performance and the moderating effect of institutional pressures. Table 4’s column (1) shows that, at the 1% level, the coefficient of carbon resilience on ESG performance is 8.9061. This suggests that there is a substantial positive correlation between carbon resilience and ESG performance. Therefore, H1 is supported. The reason for this is that carbon resilience can help manufacturing companies react quickly under low-carbon transformation conditions so that they are able to adjust their strategies and make full use of the resources at hand. Organizations with carbon resilience are concerned with the development and promotion of low-carbon products and technologies in enterprises. These actions are conducive to improving ESG performance by enhancing green protection, upholding social responsibility, and strengthening corporate governance. To avoid the endogeneity problem that may be caused by reverse causality, based on relevant research [113], we chose data of carbon resilience lagging by one period as our instrumental variable and used the two-stage least square method to conduct regression analysis. As seen in columns (1) and (2) of

Table 5. Regression results of two-stage least squares model.

	(1)	(2)
	CR	ESG
L.CR	0.571 ***
	(97.506)
CR		14.536 ***
		(7.192)
ER	0.000	−0.254 ***
	(0.007)	(−5.405)
SR	0.000	0.005 ***
	(0.175)	(3.072)
Lev	0.001	1.095 ***
	(0.971)	(6.078)
Cashflow	−0.004 ***	1.932 ***
	(−3.772)	(9.827)
Indep	−0.001	−0.256
	(−0.640)	(−1.162)
Top1	−0.000	0.688 ***
	(−0.538)	(7.531)
FirmAge	0.000	0.382 ***
	(0.276)	(8.184)
Growth	−0.000	0.021
	(−1.235)	(0.586)
_cons	0.041 ***	4.553 ***
	(20.443)	(10.753)
Anderson canon. corr. LM statistic		4251.691
Cragg–Donald Wald F statistic		9507.329 [16.38]
N	7654	7654
Industry	Yes	Yes
Year	Yes	Yes
F	874.154	16.733
R2	0.838	0.091

Note. *** p < 0.01.

, the results are both significant at the 1% level, showing that an endogeneity problem does not exist in this paper.

Moreover, to mitigate the endogeneity problem caused by sample self-selection, following a relevant study [7], this study employed propensity score matching (PSM) to re-evaluate the impact of carbon resilience on ESG performance. PSM is primarily employed to address endogeneity arising from sample selection bias. By constructing comparable treatment and control groups, PSM mitigates the influence of confounding variables, approximates a randomized experimental design, and thereby enhances the accuracy of causal effect estimation between variables. First, based on the annual median of the carbon resilience indicator, companies with carbon resilience higher than the median were classified as the treatment group (Treat = 1), and companies with carbon resilience lower than the median were classified as the control group (Treat = 0). Then, all the control variables in the regression model were selected as matching variables, and a Logit model was employed to regress the variable (Treat) to estimate the propensity score values. Finally, one-to-two-nearest-neighbor matching was performed on the treatment group and the control group, and the matched samples were substituted into the baseline regression model for re-analysis. As can be seen from the results in column (1) of

Table 6. Regression results of PSM.

	(1)
	ESG
CR	9.1563 ***
	(6.3180)
ER	−0.2006 ***
	(−3.8004)
SR	0.0030
	(1.5986)
Lev	0.8455 ***
	(4.2376)
Cashflow	1.7249 ***
	(7.7762)
Indep	−0.2294
	(−0.9197)
Top1	0.7705 ***
	(7.4818)
FirmAge	0.3193 ***
	(6.3507)
Growth	0.0048
	(0.1197)
_cons	4.7756 ***
	(26.2224)
N	5652
Industry	Yes
Year	Yes
F	26.4226
R2	0.0869

Note. *** p < 0.01.

, the coefficient of carbon resilience is significantly positive and passes the significance test at the 1% level. This result indicates that, after PSM, there is still a significant positive correlation between carbon resilience and ESG performance, indicating that the study results are reliable to a certain extent.

5.4. Moderating Effect Test

In columns (2) and (3) of Table 4, the coefficient between carbon resilience and coercive pressure is positive and statistically significant at the 10% level. Therefore, H2a is supported. The reason for this may be that when coercive pressure is high, governments tend to publish more strict policies to restrict the production activities of manufacturing enterprises. Under such circumstances, manufacturing companies will attempt to adjust their strategies to reduce their carbon emissions and invest more into green technologies and products, thus improving corporate ESG performance. In columns (4) and (5) of Table 4, the interaction coefficient between carbon resilience and normative pressure is positive and statistically significant at the 5% level. Therefore, H2b is supported. The reason for this may be that when normative pressure is high, the attitudes and standards of firms’ external stakeholders have a stronger binding effect on firms. Carbon resilience can significantly promote corporate environmental information disclosure. It aids firms in enhancing their performance in environmental protection, thus improving ESG performance. In columns (6) and (7) of Table 4, the interaction coefficient between carbon resilience and mimetic pressure is negative and statistically significant at the 1% level. Therefore, H2c is supported. The reason for this may be that when mimetic pressure is high, companies face a more competitive environment with more uncertainties. In these situations, firms might prioritize their own short-term success over enhancing their ESG performance.

5.5. Robustness Test

Three methods were used to assure the robustness of the results. First, considering that there may be a time lag in the relationship between carbon resilience and ESG performance, we replaced carbon resilience with data lagging by one period. As shown in column (1) of Table 4, the main effect of carbon resilience and ESG performance is significant at the 99% confidence intervals. Thus, H1 is supported. As shown in columns (2), (3) and (4) of Table 4, the moderating effect of coercive pressure is significant at the 90% confidence interval and the moderating effects of normative pressure and mimetic pressure are significant at the 99% confidence interval. Thus, H2a, H2b, and H2c are supported. Second, we changed the time intervals to 2012–2020 and 2013–2021 to check if the results still remained robust. As shown in columns (5) and (6) of Table 4, the main effect of carbon resilience and ESG performance is significant at the 99% confidence intervals. Third, to prevent biased results caused by omitting province-level variables, province fixed effects were set, correspondingly. The results are shown in column (7) of Table 4. The estimation results in

Table 7. Robustness tests.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
	ESG	ESG	ESG	ESG	ESG	ESG	ESG
L.CR	8.3062 ***	7.9243 ***	7.0829 ***	8.3195 ***
	(7.1651)	(6.7080)	(5.9617)	(7.1791)
CR					8.0683 ***	8.9062 ***	7.4518 ***
					(7.6278)	(8.5588)	(7.2173)
POL		−0.0108
		(−0.4809)
CR × POL		2.9059 *
		(1.8303)
MEDIA			0.0547 ***
			(4.1438)
CR × MEDIA			3.0420 ***
			(3.4805)
COM				0.2701
				(0.5778)
CR × COM				−31.6860 ***
				(−2.7514)
ER	−0.2540 ***	−0.2550 ***	−0.2580 ***	−0.2542 ***	−0.2341 ***	−0.2263 ***	−0.2534 ***
	(−5.3841)	(−5.4057)	(−5.4764)	(−5.3919)	(−5.7324)	(−5.7585)	(−6.4848)
SR	0.0051 ***	0.0050 ***	0.0051 ***	0.0051 ***	0.0035 **	0.0038 ***	0.0044 ***
	(3.0737)	(3.0581)	(3.0877)	(3.0786)	(2.4177)	(2.7345)	(3.1965)
Lev	1.1080 ***	1.1123 ***	1.0702 ***	1.1025 ***	0.9932 ***	0.9912 ***	1.0910 ***
	(6.1291)	(6.1533)	(5.9222)	(6.1011)	(6.4523)	(6.6645)	(7.3799)
Cashflow	1.8775 ***	1.8854 ***	1.8335 ***	1.8696 ***	1.6899 ***	1.7131 ***	1.7109 ***
	(9.5155)	(9.5456)	(9.2855)	(9.4775)	(9.9866)	(10.5324)	(10.5636)
Indep	−0.2663	−0.2586	−0.3117	−0.2751	−0.3112	−0.2427	−0.3397 *
	(−1.2050)	(−1.1693)	(−1.4109)	(−1.2456)	(−1.6174)	(−1.3122)	(−1.8509)
Top1	0.6846 ***	0.6850 ***	0.6936 ***	0.6858 ***	0.5887 ***	0.6216 ***	0.6315 ***
	(7.4634)	(7.4650)	(7.5689)	(7.4789)	(7.5130)	(8.2268)	(8.3939)
FirmAge	0.3833 ***	0.3815 ***	0.3786 ***	0.3830 ***	0.2932 ***	0.3062 ***	0.2908 ***
	(8.1727)	(8.1303)	(8.0828)	(8.1675)	(7.5987)	(8.1885)	(7.8192)
Growth	0.0181	0.0187	0.0043	0.0186	0.0131	0.0061	0.0130
	(0.4928)	(0.5088)	(0.1173)	(0.5074)	(0.4310)	(0.2098)	(0.4519)
_cons	5.1513 ***	5.2852 ***	4.8587 ***	5.0977 ***	4.9322 ***	5.3511 ***	5.1496 ***
	(12.9365)	(12.6974)	(12.0791)	(8.9175)	(13.7776)	(15.9052)	(14.4885)
N	7654	7654	7654	7654	9374	10,216	10,226
Industry	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Province	No	No	No	No	No	No	Yes
F	16.7076	16.0799	16.5579	16.1728	16.8273	20.0380	16.6918
R2	0.0899	0.0904	0.0928	0.0909	0.0751	0.0814	0.1111

Note. * p < 0.1, ** p < 0.05, *** p < 0.01.

were consistent with those in Table 4, indicating that our results were robust.

6. Discussion

Based on our model design, this paper examines the relationship between carbon resilience and ESG performance in the context of low-carbon transformation.

While existing research on organizational resilience has primarily focused on crises such as COVID-19, the financial crisis, and digital transformation [63,71,103,114], our study takes a different approach. We specifically investigate the disruptions caused by low-carbon transformation, extending prior research on organizational resilience in distinct scenarios [63,100,115,116,117,118].

In addition, our paper addresses the influence of carbon resilience on ESG performance. Our empirical findings demonstrate a significant positive relationship between carbon resilience and ESG performance, which broadens the research on the factors impacting ESG performance and offers manufacturing companies a new strategy for enhancing ESG performance from a carbon resilience perspective.

Furthermore, our findings highlight the crucial role of institutional pressures in shaping the relationship between carbon resilience and ESG performance. In particular, when under high coercive pressure, manufacturing companies will proactively take actions to shape carbon resilience, which in turn contributes to higher ESG performance. Under high normative pressure, carbon resilience can significantly promote corporate environmental disclosure, which promotes ESG performance. When under high mimetic pressure, in order to lower the decrease in innovation profits, manufacturing firms may choose to decrease investments into innovative activities, which in turn inhibits ESG performance. Significant theoretical and practical implications have been disclosed from these discoveries.

6.1. Implications for Theory

Theoretically, this research contributes to the existing literature to some extent.

First, by focusing on the disruptions caused by low-carbon transformation, this study clarifies the concept of carbon resilience and develops a comprehensive measurement framework. This contributes to the broader field of organizational resilience by providing a structured approach to assessing carbon resilience. Even though research on organizational resilience has attracted more and more attention in the last decade [50], there remains a need to further explore its concepts, characteristics, and impact mechanisms in specific contexts. Building on the literature related to organizational resilience and the low-carbon transformation challenges faced by manufacturing enterprises [25,46,119], this study extends the concept of organizational resilience within the framework of low-carbon transformation. By doing so, it offers a new perspective on organizational resilience in the context of low-carbon transition and presents an alternative approach for manufacturing enterprises to adapt to external changes.

Second, this study extends the literature on ESG performance by exploring the positive effect of carbon resilience on ESG performance. By revealing this relationship, this study suggests that the improvement of enterprises’ capabilities towards low-carbon transformation can help enterprises achieve sustainability, and enterprises can achieve a balance between low-carbon development and sustainable development. Previous studies have emphasized internal and external factors that affect ESG performance. From the perspective of internal factors, studies have shown that enterprises’ low-carbon behaviors play a crucial role in shaping enterprises’ ESG performances [18,19,20,21,22,23,24,26]. Under the context of low-carbon transformation, manufacturing enterprises are facing stricter criteria. As a specific kind of organizational resilience, carbon resilience is vital for driving enterprises’ low-carbon behavior and guiding the sustainable development of manufacturing companies. Overall, this study introduces a new perspective on enhancing ESG performance through carbon resilience.

Third, compared with previous studies, this study deeply discloses the relationship between carbon resilience and ESG performance in the context of different sources of institutional pressures. Investigations into the boundary conditions of carbon resilience’s impact remain limited, and research on carbon resilience within the context of low-carbon transformation is still in its early stages. In light of low-carbon transformation, especially since the introduction of the “dual carbon” strategy, enterprises are facing increasing pressures from governments, the media, and benchmark enterprises. By examining the boundary conditions of three kinds of institutional pressures—coercive pressure, normative pressure, and mimetic pressure—this study comprehensively responds to the conditions under which the influence of carbon resilience and ESG performance can be promoted or inhibited.

6.2. Implications for Practice

Our findings also provide managerial implications for the government, society, and manufacturing enterprises.

First, government departments should improve the system of environmental laws and regulations and strengthen corporate supervision and guidance in order to enhance corporate carbon resilience and promote low-carbon sustainable development. Government departments can formulate relevant laws and regulations according to the current situation of China’s economic development; form a monitoring and incentive mechanism for environmental, social, and governance information disclosure; and guide enterprises to emphasize the improvement of environmental, social, and governance performance. Policies such as setting up special R&D funds and providing tax incentives can also be used to encourage enterprises to increase their investment in carbon resilience, shaping and improving their ability to cope with low-carbon transition.

Second, the public and the media, as an important part of public opinion, should play a proper monitoring role. The media has a profound impact on public perception, investor decision making, and consumer choice. Therefore, the media should play an active and responsible role in promoting carbon resilience and ESG performance. The media need to increase reports related to ESG responsibility fulfillment so that investors, consumers, and other stakeholders can convey their special needs through social media, thus promoting enterprises to improve ESG performance.

Third, manufacturing enterprises should take the cultivation of carbon resilience as an important direction for enterprise transformation and upgrading. Manufacturing enterprises should deeply recognize the positive impact of carbon resilience on ESG performance and actively explore effective ways to enhance carbon resilience in order to meet the challenges of low-carbon transformation and achieve sustainable corporate development. At the same time, manufacturing companies need to pay close attention to changes in environmental regulations and policies and keep abreast of the government’s latest requirements on environmental protection and carbon emissions, so that they can quickly adjust their corporate strategies and business models. They should also face and properly respond to normative pressures from the media and the public, and enhance their image and reputation through effective low-carbon practices, thus laying a solid foundation for their sustainable development. Finally, manufacturing companies should correctly view market competition in the industry as a driving force for their own development and progress.

6.3. Limitations and Suggestions for Future Research

Even though much attention has been paid to carbon resilience, limitations still exist and need to be further discussed in subsequent research.

First, this study only broadens the research scope of organizational resilience within the particular context of low-carbon transformation. In the future, scholars can further explore organizational resilience in other contexts. Additionally, since certain indicators of carbon resilience are derived from text analysis, its measurement may involve a certain degree of subjectivity. Future research can improve this measurement by incorporating multiple data sources to enhance objectivity and reliability.

Second, the influence of carbon resilience on overall ESG performance is the only thing our study examines, and ESG performance is measured solely using the CSI ESG rating. In the future, scholars can enhance the measurement of ESG performance by incorporating multiple ESG rating frameworks or analyzing the sub-dimensions individually to provide a more comprehensive evaluation. Furthermore, considering the availability of data, this paper only takes a sample interval from 2012 to 2021. Scholars can extend sample intervals to achieve more robust and generalizable results. In addition, we limited our discussion to the connection between listed manufacturing companies’ carbon resilience and ESG performance. Future research can concentrate on small and medium manufacturing enterprises to better clarify the interactions between them.

Third, this study explores institutional pressures as a moderating factor but acknowledges that other contextual variables may also influence the relationship between carbon resilience and ESG performance. Thus, the relationship between carbon resilience and ESG performance can be explored more deeply. Although this study employs two-stage least square analysis and PSM to address endogeneity concerns, there may still be unobserved endogeneity issues. Future research should pay greater attention to this issue to better detect and mitigate potential endogeneity.

7. Conclusions

ESG performance is receiving growing attention. However, how carbon resilience affects firm ESG performance remains understudied. Based on the idea of organizational resilience, we proposed the notion of carbon resilience, representing the ability of an organization to keep stable, adapt, and evolve under the context of low-carbon transformation. Our study further examines the effects of carbon resilience on ESG performance. The results show that carbon resilience can strongly enhance ESG performance. In addition, institutional pressures play important roles in the relationship between carbon resilience and ESG performance. The positive relationship between carbon resilience and ESG performance is stronger under high coercive pressure, stronger under high normative pressure, and stronger under low mimetic pressure. These findings add to the organizational resilience literature by extending its context to low-carbon transformation and uncovering the relationship between carbon resilience and ESG performance and the boundary conditions between them.