Does Market Power Improve Corporate Carbon Efficiency? Based on Evidence from Listed Chinese Companies

Abstract

As a key component of China’s sustainable development strategy, the “dual-carbon” goal indicates the need to actively and steadily promote carbon peaking and carbon neutrality and strengthen resource conservation and environmental protection—a core research focus published in Sustainability. Existing research in environmental economics and corporate sustainability suggests that improving carbon efficiency is a key pathway to climate-resilient industrialization, but the role of corporate market forces in this context remains under-explored. Consistent with the interdisciplinary scope of sustainability in the environmental, economic and social dimensions. This paper takes the industrial sector from 2012 to 2021 as the research sample and empirically researches and analyzes the relationship between enterprise market power and carbon efficiency and its mechanism through theoretical derivation and a fixed effect model. We found that enterprise market power is a remarkable contributor to carbon efficiency. The mechanism test found that the promotion role is reflected in the improvement of profitability, and profitability plays a mediating role in market power and carbon efficiency. In the further heterogeneity analysis, this study found that the degree of environmental information disclosure and whether an enterprise is heavily polluting present notable differences in carbon efficiency. The positive correlation between firms’ market power and carbon efficiency is more significant when firms have a lower degree of environmental information disclosure and are non heavily polluting firms. Through the expansive analysis, it is found that there is a marked cohort effect on carbon efficiency, and the market power can effectively empower the industry cohort effect of carbon efficiency.

1. Introduction

The over-dependence and reckless utilization of industrial fossil energy has triggered a global ecological governance crisis. According to the statistics of the International Energy Agency (IEA), the average annual growth rate of CO₂ emissions from fossil fuel combustion during the period from 1990 to 2020 reached 1.7%, which has led to a continuous aggravation of the greenhouse effect and the cumulative effect of atmospheric pollutants, and has made the global climate governance system face severe challenges. China, the second-largest economy in the world and a responsible large nation, has long upheld the idea of the community of human destiny and actively contributed to the reform of the global climate governance framework. Through institutional and technological innovation, China has supported its low-carbon development plan and methodically built the “1 + N” policy structure since signing the Paris Agreement in December 2015. A new phase in China’s climate governance began in September 2020 when General Secretary Xi Jinping said at the 75th General Debate of the UN General Assembly that China would reach a carbon peak by 2030 and carbon neutrality by 2060. Specifically, the top-level design framework of the 14th Five-Year Plan includes “the coordinated promotion of pollution reduction and carbon reduction”. Its main goal is to build a multifaceted synergistic mechanism to enhance economic growth, avoid and regulate climatic hazards, and improve environmental quality. To optimize the combination of environmental–economic policy tools and strengthen synergistic governance frameworks, it is of great theoretical and practical significance to construct a thorough evaluation system for pollution and carbon reduction and analyze its spatial–temporal differentiation laws and driving mechanisms.

China’s carbon emission pattern shows prominent dynamic evolution characteristics and spatial heterogeneity. From the standpoint of temporal evolution, industrialization and urbanization are closely related to the average annual growth rate of total carbon emissions, which is 5.2% from 1990 to 2020. It is notable, meanwhile, that the carbon intensity—that is, the amount of carbon emissions per unit of GDP—shows a steady fall (see Figure 1), with a total decrease of 48.1% between 2005 and 2020, which confirms the positive trend of the gradual decoupling of economic growth and carbon emissions. At the provincial level, carbon emission spatial differentiation is notable: eastern coastal provinces have decreased carbon emission intensity by 4.3% annually through industrial structure upgrading and the diffusion of technological innovation. However, the resource-based provinces in central and western China are still facing the double pressure of economic growth and carbon emission reduction due to the rigid constraints of energy structure (coal consumption accounts for more than 60%) and the gap in technological efficiency. This composite feature of “co-existence of total control and intensity improvement, regional differentiation (see Figure 2) and path dependence” highlights the importance of building a differentiated policy system. At the micro level, industrial enterprises contribute more than 70% of the nation’s carbon emissions, and the primary cause of regional variations in carbon productivity is the difference in energy utilization efficiency. Therefore, we must concentrate on increasing the efficiency of micro-principal emission reduction and stimulating the vitality of the green technological innovation of enterprises through a combination of policy tools combining market incentives and regulatory limitations in order to resolve the “carbon lock-in” dilemma.

Research on carbon efficiency is relatively abundant, mainly focusing on macro factors such as the production level of industrial structure [1,2], economic development levels [3], and institutional environments [4], and micro-level factors such as return on assets under the DuPont analytical method [5] and technological innovations [6] were studied. As a kind of inherent competitiveness, enterprises’ market power influences some business choices. Prior research has found that enterprises with greater market power are more likely to achieve consistent cash flow and higher profitability [7,8], and their ability to undertake environmental and social responsibilities is also stronger. Enterprises with lower market forces have less ability to transfer risks to downstream enterprises or consumers when facing market shocks [9], and their ability to assume environmental and social responsibility is correspondingly weaker. Therefore, in the research process of the scientific calculation of carbon efficiency and the analysis of the influencing factors on it, whether the corporate market power can affect the efficiency of pollution reduction and carbon reduction is of great significance in finding the path to improve carbon efficiency. Given the inadequacies of the existing research, this paper theoretically analyzes the internal mechanism of the influence of market power on carbon efficiency, discusses its influence mechanism of profitability perspective, and investigates how it will affect the relationship between the two influences.

Based on the above considerations, this paper empirically examines the relationship between corporate market power and corporate carbon efficiency and its functioning mechanism by selecting industrial sectors from 2012 to 2021 (including mining, manufacturing, electricity, heat, gas, and water production and supply) as the research samples. According to the findings, enterprises’ market power and carbon efficiency are significantly positively correlated, meaning that enterprises with more market power also typically have higher carbon efficiency. The mechanism test found that this role is reflected in the improvement of profitability. This study’s additional heterogeneity analysis revealed that there are notable variations in carbon efficiency depending on the extent of environmental information disclosure and whether the enterprises are highly polluting. When the company is a non-heavy polluter and discloses less environmental information, the favorable relationship between the enterprise’s market power and the enterprise’s carbon efficiency is more apparent. In the extensive investigation, market power can effectively strengthen the industry cohort effect of carbon efficiency, and there is a strong cohort effect of carbon efficiency.

The first potential additional contribution of this work to the current research is to supplement the studies on carbon efficiency at the micro-firm level. Numerous viewpoints on the variables influencing carbon efficiency have been thoroughly examined in the present literature, and matching strategic recommendations have been produced for these viewpoints. The majority of these studies concentrate on the macro level, examining the current situation and trends of carbon efficiency at the federal, state, and local levels overall. For example, Zhou and Nie’s [10] empirical evidence based on a nonparametric frontier found that the industrial carbon emission efficiency of China and its four major economic regions shows a growing trend, but the average level of the industrial carbon emission efficiency is low; there are also notable regional differences, with the efficiency in the east being considerably higher than that in the central-western and northeastern regions, while the gap between the last three regions becomes smaller. The middle and northeastern portions of the entire country only exhibit conditional convergence in terms of convergence features, whereas the east and west exhibit club convergence. Wang and Zhao [11] use the DEA model to evaluate and rank the carbon emission efficiency, and the results of the study show that the launch of the carbon trading market has a certain effect on the enhancement of the carbon emission efficiency, and the ranking of carbon emission efficiency in China’s pilot carbon trading areas remained unchanged or increased. From the perspective of the national average level, each region of China’s carbon emission efficiency is still low. Pure technological efficiency is the main reason for China’s provinces’ and regions’ poorer carbon emission efficiency compared to the national average. The degree of technical efficiency application is still in the medium-to-low end, with more space for improvement. In contrast, our thorough understanding of carbon efficiency at the micro level, particularly at the enterprise level, is limited by the relatively weak research, which primarily focuses on a single consideration of carbon emissions and fails to fully integrate economic indicators for a comprehensive assessment. As a result, carbon efficiency is rarely explored despite being a key indicator with both environmental and economic attributes. Secondly, past studies have generally centered on the implementation effect of carbon trading pilot polices [12,13] and the inhibition of carbon emissions by technological innovation [14,15], while few scholars have studied how market power plays a role in carbon emission efficiency. By constructing the analytical framework of “market power—carbon efficiency”, this paper introduces the market power under the theory of industrial organization into environmental economics for discussion, which enriches the connotation of the research on enterprises’ environment performance. The conclusions of this study provide a new theoretical perspective and research paradigm for the subsequent discussion of the interaction between market power and environmental governance. Lastly, in terms of practical application, the results of this study offer a decision-making reference for enterprises looking to optimize their carbon management plans. This aids them in precisely evaluating the dual role of market power in the low-carbon transformation process so as to formulate scientific approaches to increase carbon efficiency.

The rest of the paper is organized as follows (see Figure 3): the second part is the theoretical analysis and hypothesis formulation. The third part is the research design of this paper, including sample selection and data sources, model setting, and the definition of relevant variables. The fourth part is the analysis of empirical results, including the endogeneity test and robustness test. The fifth part contains further analysis, the mediation effect, heterogeneity analysis, and extension analysis. The sixth part is the conclusion and policy recommendations.

2. Literature Review and Theoretical Hypotheses

2.1. Literature Review

Under the background of the “dual carbon” target, the research field of carbon efficiency has attracted more and more attention from scholars. Carbon efficiency is used to reflect the economic performance of a unit of carbon dioxide emissions, that is, the level of productivity included at a given level of carbon emissions. There are five main methods to calculate carbon efficiency: the environmental/economic single ratio method, the index system method, data envelopment analysis (DEA), stochastic frontier analysis (SFA), and the factor decomposition method. Among them, the environmental/economic single ratio method applies to the calculation of carbon efficiency at the micro-enterprise level. The index system method is mostly used to construct carbon efficiency evaluation. The data envelopment analysis method, stochastic frontier analysis, and the factor decomposition method are commonly used for the macro carbon efficiency, such as the carbon efficiency of a certain country, region, or industry as a whole. Clarkson [16] and Yan et al. [17] utilized the environmental/economic single ratio method to measure carbon efficiency at the enterprise level, whose expression is operating income/carbon emissions, and the lower the carbon emissions, the higher the carbon efficiency level of the enterprise. Li and Guan [18] constructed a low-carbon efficiency evaluation system from both low-carbon environmental and economic benefits, which comprehensively reflected the whole process of enterprises’ carbon emissions and carbon control through the product life cycle. Wang [19] and Cheng et al. [20] measured the carbon efficiency of prefecture-level cities in Hunan Province, Jiangsu Province, and Zhejiang Province, respectively, using the SBM-DEA model.

Correlation studies primarily include micro-level factors that influence carbon efficiency from the viewpoints of financial performance and carbon information disclosure, as well as macro-level factor decompositions of carbon efficiency from provincial, regional, and industry input–output perspectives. Xi [21] took carbon emission as an input factor and relied on the input-oriented DEA model: energy consumption per unit of GDP affects the carbon efficiency in time, ownership structure affects the carbon efficiency in space, and the dependence degree of foreign trade and the level of economic development have the role in both effects. Wang [22] and Cao [23] found that the scale of tourism economy, urbanization quality, and R&D technology can promote the carbon emission efficiency of the tourism industry. Zhao [24] adopted the measurements based on the Modified Undesirable Epsilon to assess pollution reduction and carbon efficiency through spatial evolutionary analysis, and the results showed that green technology innovations significantly exalt pollution reduction and carbon efficiency, especially in midstream, low-carbon, and resource-poor cities, where formal and informal environmental regulations play substantive regulatory roles with varying efficacy. He [25] found a positive correlation between financial performance and carbon efficiency and a negative correlation between carbon information disclosure and carbon efficiency using the simultaneous equation model and the sample of S&P from 500 companies. Wang et al. [26] seek an effective path for cities to realize low-carbon development by sorting out the impact of digital economy development on urban carbon efficiency. The study’s findings demonstrate that the growth of the digital economy significantly raises Chinese cities’ levels of carbon efficiency, with technological innovation and public environmental awareness serving as the main mediating factors. Yu [27] took China’s large thermal power plants as the research objects and investigated the causal effect of the Carbon Emission Trading Scheme (CETS) on the carbon emission efficiency. Using a staggered installation of CETS pilots as a natural experiment, the study shows that carbon emission trading substantially improves the carbon efficiency of power plants, with a stronger carbon efficiency growth effect of factories that are non-state-owned, large-scale, or located in areas with a higher level of marketization.

Market power is an ability possessed by a firm to manipulate or influence market prices. Firstly, market power was defined as the ability of a firm to maintain its price above marginal cost [28]; subsequently, with further research, market power was defined as the capability of a firm to influence its competitors in the market or the market parameters such as price, promotions, etc. [29]. In 1989, Bresnahan [30] gave a similar definition to the one given by Lerner of the rate at which a firm’s pricing exceeds the marginal cost of its product in an imperfectly competitive market. When evaluating this capability, it is essential to know how to quantify an organization’s market power. As a result, numerous academics have put forth various methods to gauge market power, which are, thus far: the Herfindahl–Hirschman Index (HHI), the Entropy Index, the Industry Concentration Rate (CRn), the Lerner Index (Lerner), and others are used to quantify market power under the conventional “Structuralist School” SCP paradigm. After the introduction of industrial organization theory into game theory, the research scope of market power has been continuously expanding, and the measurement methods are also increasingly rich, with the emergence of New Empirical Industrial Organization (NEIO), including the Hall method [31], the residual demand elasticity model, BLP models [32], BH models [33], DLW methods [34], etc.

The main focus of research on market power is on its determinants and measuring techniques, such as supply and demand dynamics, the state of competition, laws and regulations, technology aspects, and other aspects. For instance, through transmission mechanisms like economies of scale and scope, market resource synergy, R&D and innovation synergy, and management synergy, business mergers and acquisitions can increase an organization’s market power, and this boost is more pronounced downstream in the industrial chain [35]. Wang [36] estimated the impact of China’s carbon trading system on the market power on high-carbon enterprises employing the multiple difference method and mediated effects model. It can be found that the carbon emissions trading scheme (ETS) increases the internal operating costs of enterprises and changes the external development environment, which, in turn, affects the market power of high-carbon enterprises. China’s carbon market negatively affects the market power of the enterprises involved, primarily by reducing the level of horizontal integration without decreasing the level of vertical integration. In contrast, there are fewer studies on the economic consequences of market power, chiefly focusing on the effects of market power on firm performance, innovation activities, investment and financing behavior, and labor income shares. Zhang and Zhu [37] point out that market power and enterprise innovation could form a positive interaction, and put forward ideas and countermeasures to construct a benign interaction, while Chen and other scholars [38] propose the opposite conclusion, that is, market power contributes to the aggravation of “organizational inertia”, which inhibits the green innovation activities of the enterprise. However, market power has a positive impact on social responsibility, which, in turn, has an obvious masking effect on the relationship between market power and green innovation. Tinu Iype Jacob and Sunil Paul [39] examined the relationship between labor income shares and its main drivers, which are market power, capital intensity, and automation, and the results show that market power is significantly negatively related to labor income shares.

2.2. Theoretical Assumption

In the context of the “dual-carbon” goal, companies will have sufficient incentives to raise carbon efficiency in order to meet the demands of the low-carbon economy’s development while also lowering the cost of fines resulting from excessive carbon emissions in order to improve financial performance and increase profits [40]. As a sign of corporate competitiveness, market power influences business decision making to some extent and influences how much carbon efficiency a company develops.

Market power reflects the influence and control of enterprises in the market. Because of their brands and technologies, they have more resources and greater capacities to support carbon efficiency and emission reduction. Through dynamic capability development, market power—a byproduct of heterogeneous resource accumulation—forms the technical reconstruction path of the low-carbon transition. Specifically, market-dominant enterprises can effectively promote the diffusion of low-carbon technological innovations by relying on organizational learning mechanisms such as vertical integration in the supply chain and production process reengineering [41]. Furthermore, according to the theory of industrial organization, these companies use price leadership to pass on the marginal costs of environmental regulation (as determined by the Lerner Index) to customers, and the “buffer effect” of their capacity for risk-shifting considerably lessens the negative financial performance effects of environmental compliance expenses. As a result, a positive feedback loop of carbon-efficient investments is established [42]. However, it is more expensive for enterprises to take action to reduce carbon emissions and maximize carbon efficiency because of the high direct cost of carbon emissions, which is comparable to cleaner production equipment, and the high indirect cost, which is comparable to fines. For enterprises to sustain their environmental optimization efforts, they require both strong profitability and sufficient funds. From the perspective of profitability, companies with more market power have higher pricing power [7], which can help the enterprise to occupy the leading position in the market and obtain high profits. Their profitability is also comparatively higher, which offers a financial foundation for carbon efficiency optimization. For example, scholars such as Cormier and Magnan [43] found that highly profitable enterprises are faster and more efficient at addressing environmental problems. In addition, institutional theory reveals an externally driven logic: market leaders are under institutional isomorphic pressure and face a multidimensional governance system comprised of government regulation, media scrutiny [44], and ethical consumption demands, a compounding mechanism of pressure within the framework of stakeholder theory that drives enterprises’ environmental performance into the core of their strategies. Enterprises with higher market power tend to receive more attention due to their leadership position and resource advantages in the market, and this concern and regulation is not only reflected in the supervision of business activities, but in the strict requirements for firms’ environmental behavior and carbon emissions. Enterprises are more motivated to fulfill their environmental obligations and maintain or increase carbon efficiency when external oversight from stakeholders, such as the government, is in place. On the one hand, these enterprises will prioritize environmental protection inputs and carbon emission reduction measures to meet public and government expectations in order to preserve a positive business image and reputation. On the other hand, these enterprises will also aggressively enhance carbon efficiency to cope with ever-tougher environmental laws and regulations in order to prevent potential environmental hazards and compliance expenses. Based on the above research considerations, hypothesis H1a of this paper is proposed (see Figure 4).

H1a. 

Firms’ market power has a significant positive promotion effect on carbon efficiency.

However, in contrast to these facilitating mechanisms, the principal–agent theory reveals the potential negative effects of market power: monopoly positions may exacerbate the management trench effect, leading to X-efficiency losses. Growing agency costs may lead to resource mismatches as firms gain market dominance, resulting in “managerial slack” that stifles low-carbon innovation. Strong market power companies may not always make use of their resources and capabilities effectively, particularly when there is a lack of effective regulation and market competition. In these situations, an excessive concentration of resources may result in misalignment and a waste of resources as well as subpar financial performance, which impedes the advancement of carbon efficiency [45,46]. The paradoxical aspect of Schumpeter’s theory is that, while monopoly profits fund innovation, the absence of competition reduces the incentive for “creative destruction”. Additionally, path dependence theory explains how the established high-carbon technological system produces capability–grapthrough lock-in effects that cause firms to strategically exclude themselves from disruptive low carbon. Path dependence theory further demonstrates that established high-carbon technological systems create capability traps through lock-in effects, leading to the strategic exclusion of firms from disruptive low-carbon innovations, and that this technological path dependence may alienate market forces as a deterrent to the low-carbon transition. In particular, the economic foundation for enterprises to maximize carbon efficiency is provided by the financial advantage that comes with market dominance, which also lessens the pressure that enterprises face from competition. Enterprises’ “escape from competition effect” is weakened, and there is no need for enterprises to invest more resources and efforts in research and development and the promotion of more environmentally friendly and efficient technologies. They are likely to be satisfied with existing technologies and production patterns, thus inhibiting the advancement of carbon efficiency. Moreover, to maintain their existing technologies and production modes, enterprises with stronger market power take a conservative attitude towards emerging low-carbon technologies and hinder their promotion and application through market means. The technological conservatism and hindrance may limit the improvement of the carbon efficiency of the whole industry. Based on the above research considerations, hypothesis H1b of this paper is proposed.

H1b. 

Firms’ market power has a significant negative inhibitory effect on carbon efficiency.

3. Research Design

3.1. Sample Selection and Data Sources

In this paper, the industrial sectors (mining, manufacturing, electricity, heat, gas, water production and supply) from 2012 to 2021 are selected as the initial research samples, and the samples are screened as follows: (1) excluding the samples of companies that have been treated with special treatments, such as ST, PT, and so on, and (2) excluding other outliers and missing data samples. In order to control the effect of extreme values, continuous variables are winsorized up and down by 1% and, finally, 12,256 observations are obtained. The data in this paper come from the Cathay Pacific database, Wind database, National Bureau of Statistics, and China Statistical Yearbook.

3.2. Model Design and Variable Selection

In order to test the impact of individual firms’ market power on carbon efficiency, this paper constructs the following regression model:

$${Log\left(CEE\right)}_{i,t}={\beta }_0+{\beta }_1{lerner}_{i,t}+{\beta }_2{Gr\_in}_{i,t}+{\beta }_3{Analyst}_{i,t}+{\beta }_4{ER}_{i,t}+{\beta }_5{Size}_{i,t}+{\beta }_6{Lev}_{i,t}+{\beta }_7{Growth}_{i,t}+\mathsf{\epsilon }$$

(1)

$${Log\left(CEE\right)}_{i,t}={\beta }_0+{\beta }_1{lerner}_{i,t}+{\beta }_2{{ROA}_{i,t}+{\beta }_{3}Gr\_in}_{i,t}+{\beta }_4{Analyst}_{i,t}+{\beta }_5{ER}_{i,t}+{\beta }_6{Size}_{i,t}+{\beta }_7{Lev}_{i,t}+{\beta }_8{Growth}_{i,t}+\mathsf{\epsilon }$$

(2)

where CEE represents the magnitude of the carbon efficiency of the explanatory variables, measured by the annual gross operating income of the enterprise/the annual carbon emissions of the enterprise, β₀ is the intercept, and β₁ is the coefficient of the core explanatory variable market power. This paper mainly focuses on the symbols of β₁ as well as its significance. If β₁ is positive, it indicates that the market power has a positive impact on carbon efficiency. The main variables included in model (1) are defined as follows.

3.2.1. Explained Variable: Carbon Efficiency

The explained variable in this paper is carbon efficiency (CEE). Given the reliability of data at the micro enterprise level and the confidentiality of firm-caliber data, this paper draws on the approach of Clarkson [16] and Yan et al. [17] to measure carbon efficiency at the enterprise level by using a single ratio method of economy/environment, that is, Carbon Efficiency = total annual operating income of the enterprise /annual carbon emissions of the enterprise, and the larger the value, the higher the carbon efficiency level. Since fewer enterprises in China publicly disclose data related to carbon information, considering the availability of data, this paper estimates enterprises’ carbon emissions through industry carbon emissions with the help of total operating costs, which are calculated by multiplying the energy consumption of the published industries in the China Statistical Yearbook by the energy carbon emission coefficients in the 2019 IPCC Guidelines for National Greenhouse Gases. Carbon efficiency is calculated as shown in Equation (3).

$$\mathrm{Carbon}\mathrm{efficiency}={\displaystyle \frac{\mathrm{O}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{i}\mathrm{n}\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{p}\mathrm{r}\mathrm{i}\mathrm{s}\mathrm{e}}{\frac{\mathrm{Sec}\mathrm{toral}\mathrm{carbon}\mathrm{emissions}}{\mathrm{Industry}\mathrm{main}\mathrm{operating}\cos \mathrm{ts}}\times \mathrm{Enterprise}\mathrm{operating}\mathrm{costs}}}$$

(3)

3.2.2. Core Explanatory Variable: Market Power

The core explanatory variable of this paper is market power, which is measured by the Lerner Index (Lerner), drawing on the research of scholars such as Zhou [46]. The specific theoretical measurement formula is shown in Equation (4):

$${Lerner}_{it}={\displaystyle \frac{p_{it}-{mc}_{it}}{p_{it}}}$$

(4)

where Lerner is the market power, P is the product price, and MC is the marginal cost. Given the availability of marginal cost, we use Equation (5) to measure the Lerner Index.

$$\mathrm{Lerner}={\displaystyle \frac{\mathrm{operating}\mathrm{income}-\mathrm{operating}\mathrm{costs}-\mathrm{sales}\mathrm{expense}-\mathrm{overhead}}{\mathrm{operating}\mathrm{income}}}$$

(5)

3.2.3. Control Variable

The size of carbon efficiency is affected by other factors in addition to market power, and the model enterprise’s control variables are commonly used in the previous literature concerning the research of scholars such as Zhou et al. [4] This paper selects firm size (Size), environmental regulation intensity (ER), green technology innovation (Gr_in), analysts focus (Analyst), growth capacity (Growth), and gearing ratio (Lev) as control variables in this paper (see

Table 1. Variable definitions.

Variable Type	Variable Name	Variable Symbol	Variable Interpretation
Explained variable	Carbon efficiency	CEE	See (3)
Core explanatory variables	Market power	Lerner	Individual Stock Lerner Index
	Enterprise size	Size	Log (total assets)
	Intensity of environmental regulation	ER	Industrial pollution control completion/Gross regional product (industry)
	Growth capacity	Growth	Operating profit growth rate
	Green technology innovation	Gr_in	Logarithmic number of green patent applications
	Gearing	Lev	Total liabilities/total assets
	Analysts focus	Analyst	Followed by Analysts
Intermediary variable	Profitability	ROA	Return on assets
Moderator variable	Environmental information Disclosure index	EDI	Environmental disclosure Score/Best disclosure score
	Heavily polluting enterprises	High_p	1 if a heavy polluter, 0 otherwise

).

4. Analysis of Empirical Results

4.1. Descriptive Statistics

Table 2. Descriptive statistical analysis.

Variable	N	Mean	SD	Median	Min	Max.
Lcee	12,256	3.8979	1.2963	4.254	0.817	6.106
Lerner	12,256	0.1341	0.0987	0.122	−0.189	0.495
Markup	12,256	0.0461	0.0573	0.035	0.010	4.082
ROA	12,256	0.0457	0.0603	0.044	−0.838	0.786
Gr_in	12,256	0.0429	0.1455	0.000	0.000	2.560
Analyst	12,256	1.3743	1.1792	1.386	0.000	4.205
ER	12,256	0.1844	0.1349	0.150	0.009	0.714
Size	12,256	21.4805	0.9389	21.380	19.570	24.461
Lev	12,256	0.3808	0.1756	0.373	0.057	0.858
Growth	12,256	0.2021	0.3914	0.117	−0.632	2.697
EDI	12,256	0.2316	0.1825	0.184	0.000	0.921
High_p	12,256	0.3642	0.4812	0.000	0.000	1.000

presents the results of descriptive statistics for all variables. Among them, the minimum and maximum values of market power are −0.189 and 0.495, respectively, with a polar deviation of 0.684 and a mean value of 0.1341. Meanwhile, the minimum and maximum values are 0.817 and 6.106, with a polar deviation of 5.289, and the mean value of the log of carbon efficiency of enterprises is 3.8979, with a standard deviation of 1.2963, respectively, implying that there are large differences in the carbon. This vital differentiation of carbon efficiency of enterprises essentially maps out the differentiated allocation ability of environmental management costs of enterprises with different market power. The mean value of ROA is 0.0457, and among the control variables, the mean value of Gr_in is 0.0429, the mean value of ER is 0.1844, the mean value of Analyst is 1.3743, the mean value of Growth is 0.2021, and the mean value of lev is 0.3808. The mean value of EDI is 0.2316, which suggests that the companies’ environmental information disclosure is poor, and the mean value of High_p is 0.3642, indicating that nearly half of the observations in the sample were heavily polluting firms.

4.2. Benchmark Regression Analysis

In order to test the impact of firms’ market power on firms’ carbon efficiency, this paper utilizes model (1) to conduct a fixed effects panel analysis on the sample data of industrial industries (mining, manufacturing, and electricity, heat, gas, and water production and supply) for the period of 2012–2021, and the results of the empirical regression are shown in

Table 3. Market power and carbon efficiency.

	(1)	(2)	(3)	(4)
Lerner	0.7429 ***	1.0188 ***	0.5861 ***	0.9927 ***
	(0.0626)	(0.0620)	(0.0703)	(0.0732)
Gr_in			−0.0305	−0.0514 **
			(0.0244)	(0.0247)
Analyst			0.0395 ***	0.0031
			(0.0058)	(0.0061)
ER			0.1117 ***	−0.0433
			(0.0370)	(0.0362)
Size			−0.0263 ***	0.0068
			(0.0076)	(0.0078)
Lev			−0.0670 *	−0.0106
			(0.0344)	(0.0345)
Growth			−0.0093	−0.0150
			(0.0126)	(0.0125)
_cons	3.7984 ***	3.7614 ***	4.3376 ***	3.6314 ***
	(0.0089)	(0.0096)	(0.1562)	(0.1610)
Industry	Yes	Yes	Yes	Yes
Year	No	Yes	No	Yes
N	12,255	12,255	12,255	12,255
R-sq	0.765	0.772	0.766	0.772

Note: Standard deviation in parentheses; ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

. When industry fixed effects are taken into account, control variables are added, and the 1% level is applied, the results in Columns (1) and (3) demonstrate that the influence coefficients of market power on carbon efficiency are 0.7429 and 0.5861, respectively. From an economic point of view, carbon efficiency rises by 58.61% on average for every unit increase in market power. Columns (2) and (4) show that the influence coefficients of firms’ market power on carbon efficiency are 1.0188 and 0.9927, respectively, controlling the industry and time fixed effects as well as the condition of control variables, and both are significantly positive at the 1% level. This suggests that the stronger the market power, the more capable it will be of reducing costs, internalizing the cost of environmental management into the prices of products, and mitigating the erosion of profits by the “green premium”, and, thus, use abundant cash flow to support clean technology and reduce the financial constraints of emission reduction. In summary, the stronger the market power, the more willing and able enterprises are to assume environmental responsibility and improve carbon efficiency. The hypothesis H1 has been preliminarily verified.

4.3. Robustness Test

4.3.1. Substitution of Explanatory Variables

In order to strengthen the reliability of the empirical regression results, in this section, we conduct a robustness test to adjust the core explanatory variables in this study. Drawing on the practice of scholars such as Shi et al. [47], the average value of the sum of the total sales proportion of the top five customers and the total purchase proportion of the top five suppliers is used as a proxy variable. The larger the total sales proportion of the top five customers and the total purchase proportion of the top five suppliers, the smaller their corresponding market power, for which we take the reciprocals. This indicator’s design has dual economic implications. Firstly, supply chain concentration is negatively correlated with enterprises’ market power; as the proportion of total sales for the top five customers and the proportion of total purchases for the top five suppliers tends to rise, it suggests that enterprises’ bargaining power over upstream and downstream is diminished. When the top five customers’ total sales percentage, the top five customers’ total purchase percentage, and the top five suppliers’ total buy percentage all tend to rise, it shows enterprises have less negotiating power over upstream and downstream. The enterprise’s bargaining power over the upstream and downstream is weaker when the sum of the total sales proportion of the top five customers, the total purchase proportion of the top five customers, and the total purchase proportion of the top five suppliers tend to increase. This can be turned into a positive indicator by using the inverted form. Secondly, the use of the harmonic mean instead of the arithmetic mean is more in line with the asymmetric characteristics of the supply chain’s power; a high degree of concentration at either end (customer or supplier) will remarkably weaken the enterprise’s market power. This is calculated as follows.

Markup = 2/(Total sales proportion of the top five customers + Total purchase proportion of the top five suppliers)

The regression results are shown in

Table 4. Market power and carbon efficiency: substitution of explanatory variables.

	(1)	(2)	(3)	(4)
Markup	0.6501 ***	0.2574 ***	0.6509 ***	0.2668 ***
	(0.1823)	(0.0891)	(0.1927)	(0.0938)
Gr_in			−0.0454 *	−0.0590 **
			(0.0246)	(0.0250)
Analyst			0.0522 ***	0.0362 ***
			(0.0054)	(0.0056)
ER			0.0701 *	−0.0336
			(0.0374)	(0.0366)
Size			−0.0272 ***	−0.0025
			(0.0076)	(0.0078)
Lev			−0.2002 ***	−0.2106 ***
			(0.0332)	(0.0328)
Growth			−0.0035	−0.0064
			(0.0127)	(0.0126)
_cons	3.8681 ***	3.8862 ***	4.4462 ***	3.9802 ***
	(0.0102)	(0.0071)	(0.1557)	(0.1598)
Industry	Yes	Yes	Yes	Yes
Year	No	Yes	No	Yes
N	12,255	12,255	12,255	12,255
R-sq	0.763	0.767	0.766	0.769

Note: Standard deviation in parentheses; ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

.

The results show that, even if the explanatory variables are replaced and the control variables are added under the control fixed effects model, the regression coefficients are all positive and significant at the 1% level. The market power reconfiguration indicator essentially captures the pivotal position of firms on the supply chain ecosystem, whereby firms with strong bargaining power can bind upstream and downstream through green terms, such as requiring their suppliers to share carbon footprint data with their customers or to adopt cleaner production processes. This robustness feature can be explained from the perspective of value chain governance. Second, the mechanism of “de-vulnerability” is revealed by the inverse measure of supply chain dependence. When enterprises are not dependent on a single supplier or customer, they have more strategic freedom and can devote more resources to longer-cycle carbon-neutral technology research and development. Furthermore, the mathematical properties of the harmonic mean amplify the “barrel effect”, suggesting that firms need to establish competitive advantages on both the client and supply side to gain substantial market power to drive carbon efficiency, which is highly consistent with the theoretical expectation of “complementary assets” in the resource-based view. Therefore, the double robustness test not only verifies the reliability of benchmark results, but reveals the boundaries of the role of market power in influencing carbon efficiency through the reconstruction of supply chain power.

4.3.2. Sample Selection

Considering the outbreak of the COVID-19 pandemic since 2020, the production and operation of many regions and industries have been restricted to varying degrees. For this reason, in this paper, we consider excluding the sample data from 2020 to 2021 and then conducting the regression again. The regression results are as shown in

Table 5. Market power and carbon efficiency: changing sample ranges.

	(1)	(2)	(3)	(4)
Lerner	0.8963 ***	1.1589 ***	0.7522 ***	1.1614 ***
	(0.0454)	(0.0439)	(0.0504)	(0.0499)
Gr_in			−0.0009	−0.0036
			(0.0194)	(0.0196)
Analyst			0.0281 ***	−0.0070
			(0.0043)	(0.0043)
ER			0.0609 **	−0.0521 **
			(0.0278)	(0.0255)
Size			−0.0215 ***	0.0102 *
			(0.0056)	(0.0055)
Lev			−0.0844 ***	−0.0432 *
			(0.0259)	(0.0250)
Growth			−0.0181 *	−0.0230 **
			(0.0095)	(0.0090)
_cons	3.8117 ***	3.7792 ***	4.2713 ***	3.6025 ***
	(0.0067)	(0.0068)	(0.1155)	(0.1137)
Industry	Yes	Yes	Yes	Yes
Year	No	Yes	No	Yes
N	8477	8477	8477	8477
R-sq	0.907	0.914	0.908	0.914

Note: Standard deviation in parentheses; ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

. Column (1) and Column (3) show that, with controlling the fixed effects of industries and gradually adding control variables, the influence coefficients of the market power of enterprises on carbon efficiency are 0.8963 and 0.7522, respectively, and both of them are significantly positive at the 1% level. Columns (2) and (4) indicate that the influence coefficients of firms’ market power on carbon efficiency are 1.1589 and 1.1614, with the gradual addition of control variables under the control of industry and year-fixed effects, and both are also significantly positive at the 1% level. This suggests that firms with stronger market power are more willing and able to take environmental responsibility and improve carbon efficiency and, therefore, the robustness of the hypothesis results is again verified.

4.3.3. Elimination of Endogenous Interference

After the initial test of the relationship between individual firms’ market power and its impact on carbon efficiency, to exclude the impact of issues such as omitted variables and measurement errors on the findings, we use the reference of Zhou [46], who adopted the lagged one-period market power as an instrumental variable to overcome the endogeneity problems. In order to meet the requirements of instrumental variable relevance, the design is based on the dynamic adjustment mechanism in the theory of industrial organization. On the one hand, the formation of enterprise market power has path-dependent characteristics, and the historical market position continues to influence the current business decision making through economies of scale, brand barriers, and other sunk cost effects. On the other hand, the technological diffusion of carbon emission efficiency improvement has a time lag, and it is difficult for current carbon efficiency to inversely influence the market structure of the previous period to safeguard the exclusivity constraint. The two-stage empirical results are shown in

Table 6. Market power and carbon efficiency: instrumental variables.

	First	Second
Variables	Lerner	Lcee
L. Lerner	0.7334 ***
	(0.0109)
Lerner		0.8632 ***
		(0.1252)
Gr_in	−0.0070	−0.0355
	(0.0041)	(0.0507)
Analyst	0.0113 ***	0.0060
	(0.0006)	(0.0079)
ER	−0.0006	−0.0240
	(0.0047)	(0.0603)
Size	−0.0008	0.0074
	(0.0008)	(0.0091)
Lev	−0.0513 ***	−0.0177
	(0.0041)	(0.0478)
Growth	0.0034 *	−0.0206
	(0.0020)	(0.0187)
Industry/Year	Yes	Yes
N	8993	8993

Note: Standard deviation in parentheses; ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

. L.lerner passes the test of weak instrumental variable (F-value of 9744.094). The enterprises’ market power is significantly and positively related to the carbon efficiency of firms, and the regression results in Column (2) after controlling the endogeneity problems still support the previous conclusions, which is that enterprises’ market power can sizably improve carbon efficiency.

5. Further Study

5.1. Mechanism Analysis

The Mediating Effect of Profitability

One of the important mediators of the effect of vendor market power on carbon efficiency is the profitability of firms. In order to gain a dominant position in the market and generate substantial profits during their operations, organizations with greater market power will have greater pricing power and market share. The earnings give enterprises the capacity to handle the burden of environmental legislation and carbon emission reduction, in addition to giving them additional funds to assist their green technical innovation and industrial structural upgrading. Therefore, this paper hypothesizes that market power acts on carbon efficiency by affecting firms’ profitability.

The results in Column (1) of

Table 7. Market power, ROA, and carbon efficiency.

	(1)	(2)
Variable	ROA	Lcee
Lerner	0.3501 ***	0.7342 ***
	(0.0079)	(0.0871)
ROA		0.7384 ***
		(0.1266)
Gr_in	0.0107 ***	−0.0592 **
	(0.0022)	(0.0247)
Analyst	0.0072 ***	0.0022
	(0.0004)	(0.0062)
ER	−0.0031	−0.0410
	(0.0031)	(0.0361)
Size	0.0022 ***	0.0052
	(0.0006)	(0.0077)
Lev	−0.0561 ***	0.0308
	(0.0033)	(0.0353)
Growth	−0.0069 ***	−0.0099
	(0.0011)	(0.0125)
_cons	−0.0348 ***	3.6571 ***
	(0.0118)	(0.1608)
Industry/Year	Yes	Yes
N	12,255	12,255
R-sq	0.473	0.773

Note: Standard deviation in parentheses; ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

show that market power has a significantly positive effect on profitability. The second column shows that the coefficient of ROA is significantly positive at the 1% level, indicating that market power can remarkably boost the profitability of enterprises. Meanwhile, the Bootstrap test found that ROA plays a mediating role in the effect of market power on carbon efficiency (the 95% confidence intervals of 0.396 and 0.041), suggesting that profitability is a fully mediating factor for the firms’ market power to improve carbon efficiency. This finding shows that firms with high market power are able to enhance their carbon efficiency by increasing their profitability.

5.2. Heterogeneity Analysis

5.2.1. The Impact of Environmental Information Disclosure on Market Power and Carbon Efficiency

In the above studies, we have demonstrated that market power has a positive effect on carbon efficiency improvement. The publication of environmental information will unavoidably have an impact on the degree to which the established intensity of market power can increase carbon efficiency. On the one hand, the advancement of environmental information disclosure promotes the transparency of information disclosure, which not only reduces the information asymmetry between enterprises and external investment, but also transmits positive and green information to investors, attracts investors and other external stakeholders to invest external funds, and optimizes the financing environment. The financial advantage brought by market power has been replaced, which, to some extent, weakens the positive effect of market power on carbon efficiency enhancement to a certain extent. On the other hand, the positive effect of environmental information disclosure on the improvement of carbon efficiency will be reduced. Meanwhile, the external supervision brought by environmental information disclosure, such as government and media, encourages enterprises to take more initiative to assume environmental responsibility and optimize carbon efficiency. Therefore, we predict that environmental information disclosure has a negative moderating effect on market power and carbon efficiency.

Moderating variable: environmental information disclosure index. Drawing on the methods of Tang et al. [48] and Xie [49], the environmental information disclosure index is constructed. The specific operation is as follows: According to the environmental disclosure score criteria of listed companies (see

Table 8. Environmental information disclosure index score criteria.

Kind	Project	Scoring Criteria	Top Score
Environmental management disclosure	Environmental protection concept; environmental protection goal; environmental protection management system; environmental protection education and training; environmental protection special action; environmental incident emergency response mechanism; environmental protection honors or awards; “Three simultaneous system”	Disclosure assigned to 1, otherwise 0	8
Environmental regulation and certification disclosure	Whether pollutant emission meets standards	Achievement of the standard is assigned the value of 1, otherwise it is 0	1
	Whether passed the ISO14001 [50] certification and ISO9001 [51] certification	Assign the value of 1 by authentication, otherwise 0	2
Environmental information disclosure vehicle	Annual reports of listed companies, social responsibility reports, environmental reports	Disclosure assigned to 1, otherwise 0	3
Disclosure of environmental liabilities	Wastewater emissions; COD emissions; SO2 emissions; CO2 emissions; soot and dust emissions; industrial solid waste generation	Quantitative descriptions are assigned a value of 2, qualitative descriptions are assigned a value of 1, and no descriptions are assigned a value of 0	12
Environmental performance and governance disclosure	Waste gas emission reduction and governance; waste water emission reduction and governance; soot and dust governance; solid waste utilization and disposal; noise, light pollution, radiation and other governance; cleaner production implementation	Quantitative descriptions are assigned a value of 2, qualitative descriptions are assigned a value of 1, and no descriptions are assigned a value of 0	12

), the score is summarized and divided by the optimal disclosure score, which, in this paper, is 38, and the calculation formula is EDIi = ∑EDIi/MEDIi.

The samples are separated into five groups according to the environmental information disclosure index to examine the effect of the degree of environmental information disclosure on the relationship between enterprises’ power and carbon efficiency. The group with the lowest degree of environmental information disclosure is the top fifth of the environmental information index, while the group with the highest degree of environmental information disclosure is the bottom two fifths. The differences between the two groups are examined for intergroup differences after the middle fifth has been eliminated.

Table 9. Heterogeneity analysis.

	(1)	(2)	(3)	(4)
	Low-Level Environmental Information Disclosure	High-Level Environmental Information Disclosure	Non Heavily Polluting Enterprises	Heavily Polluting Enterprises
Lerner	1.1991 ***	0.6074 ***	1.1430 ***	0.6274 ***
	(0.0821)	(0.1489)	(0.0584)	(0.1711)
Gr_in	0.0242	−0.1292 ***	−0.0342	−0.3068 ***
	(0.0293)	(0.0478)	(0.0220)	(0.1135)
Analyst	−0.0667	0.0183	−0.0033	0.0151
	(0.0070)	(0.0118)	(0.0048)	(0.0142)
ER	−0.1003 **	0.0127	−0.0596 *	0.0855
	(0.0415)	(0.0702)	(0.0314)	(0.0713)
Size	0.0089	0.0061	0.0132 **	−0.0051
	(0.0098)	(0.0146)	(0.0061)	(0.0171)
Lev	0.0155	−0.0656	−0.0260	0.0855
	(0.0411)	(0.0691)	(0.0281)	(0.0737)
Growth	−0.0270 **	0.0088	−0.0382 ***	0.0419
	(0.0135)	(0.0301)	(0.0100)	(0.0425)
_cons	3.8014 ***	3.4204 ***	4.0170 ***	2.9557 ***
	(0.2042)	(0.3068)	(0.1274)	(0.3548)
Industry/Year	Yes	Yes	Yes	Yes
N	5346	4729	7791	4462
R-sq	0.844	0.707	0.834	0.590
b0-b1	0.592		0.516
p-value	0.001		0.000

Note: Standard deviation in parentheses; ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

reports the results of the above tests, which reveals that the market power has a positive effect on carbon efficiency at both low and high levels of environmental information disclosure and is significant at the 1% level, while the test for the differences in group coefficients between the two groups is significant at the 1% level.

5.2.2. The Effect of Pollution Levels on Market Power and Carbon Efficiency

Due to the difference in enterprise nature, there is often considerable variation in the extent of their environmental pollution, which, in turn, may further influence the impact of firms’ market power on carbon efficiency. Non heavily polluting firms produce relatively low carbon emissions in their production processes, so they face less pressure on carbon emission. This may result in firms having greater flexibility and freedom to improve carbon efficiency and, thus, being more susceptible to market power. In contrast, enterprises that emit a lot of pollutants are subject to more stringent environmental rules and policy constraints. As a result, they may be more motivated to increase their carbon efficiency due to external pressures like environmental regulations and policy requirements. These enterprises might be more concerned with enhancing carbon efficiency to satisfy legal requirements than with factors based only on market dominance. Moreover, additional external variables, such as financial strains and technological bottlenecks, may limit their efforts to increase carbon efficiency, lessening the impact of market power on carbon efficiency.

In order to further investigate this issue, this paper introduces the variable of whether the firm is a heavy polluter (High_p) in model (1), which is assigned a value of 1 when the firm is a heavy polluter and 0 otherwise. The regression results illustrated in Columns (3) and (4) of Table 9 show that the coefficient of market power is significantly different for non heavily polluted enterprises and heavily polluted enterprises, and the effect of enterprise market power on carbon efficiency is greater in non heavily polluted enterprises compared with heavily polluted enterprises.

5.3. Extended Analysis: The Cohort Effect of Carbon Efficiency

The cohort effect refers to the fact that the behavior and decision making of individuals in a particular group will vary due to the behavior and decision making of other individuals in the same group and, thus, the cohort effect also widely exists in the behavior and decision making of enterprises [52]. The optimal course of action for the enterprise is to mimic and learn from the firms in the cohort, since the relationships and interactions among the various enterprises within the cohort will crucially influence the decisions or behaviors of the member enterprises to converge [53]. With the country’s deep concern for carbon emission reduction, enterprises in various industries and regions, according to their own needs and capabilities, competitively begun to competitively pool the resources they possess in the major task of reducing pollution and carbon and improving the efficiency of carbon emission. Due to the diversity of enterprises both within the same industry and across the region, the carbon emission efficiency varies as well. If the market enterprises want to increase their carbon efficiency, they can study and learn from other enterprises with outstanding carbon emission efficiency in this group. It is evident that the actions taken by enterprises to increase their carbon efficiency may have a cohort effect. In order to test the above inference, this paper measures the carbon efficiency cohort effect at the industry level, based on which, the following model is constructed concerning Cheng [51] to test the existence of the carbon efficiency cohort effect.

$${LCEE}_{i,t}=\alpha 0+\alpha 1{Coh\_LCEE}_{i,t}+\alpha 2{Controls}_{i,t}+\mathsf{\epsilon }$$

(6)

In model (6), Coh_LCEE denotes the cohort effect of industry carbon efficiency, including both industry (Coh_LCEE_ind) and province(Co h_LCEE_pro), and the control variables contain the variables in model (1). As can be seen from

Table 10. The cohort effect of carbon efficiency.

	(1)	(2)	(3)
	LCEE	LCEE	LCEE
Coh_LCEE_ind	1.001 ***		1.000 ***
	(0.001)		(0.001)
Coh_LCEE_pro		0.061 **
		(0.025)
Lerner			1.012 ***
			(0.021)
Coh_LCEE_ind_lerner			0.011 **
			(0.005)
Controls	YES	YES	YES
_cons	0.173 ***	3.679 ***	−0.153 ***
	(0.026)	(0.192)	(0.015)
N	12,243	12,243	12,243
R-sq	0.99	0.77	1.00

Note: Standard deviation in parentheses; ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

, the cohort effect coefficients at the industry and province level are all significantly positive at the 5% level, indicating that the carbon efficiency improvement will be affected by the carbon efficiency improvement of the cohort enterprises. At the same time, in terms of the degree of impact, industry-level carbon efficiency has a greater influence on the level of carbon efficiency improvement. Based on this, the cohort enterprise with similar backgrounds is more valuable for learning and copying the focal enterprise’s carbon efficiency improvement.

On this basis, this paper constructs the following model to test the influence of market power on the industry cohort effect of carbon efficiency.

$${LCEE}_{i,t}=\alpha 0+\alpha 1{Coh\_LCEE\_ind}_{i,t}+\alpha 2{Lerner}_{i,t}+\alpha 3{Coh\_LCEE\_ind\_lerner}_{i,t}+\alpha 4{Controls}_{i,t}+\mathsf{\epsilon }$$

(7)

In model (7), Coh_LCEE_ind_lerner is the cross-multiplier of the carbon efficiency of industry enterprises and market power; α3 indicates the influence of market power on the carbon efficiency industry cohort effect. If α3 is significantly positive, it indicates that the market power is capable of empowering the industry cohort effect of carbon efficiency. As can be seen from Table 10, the coefficient of Coh_LCEE_ind_lerner is significantly positive at the 5% level, indicating that market power can strengthen the carbon efficiency industry cohort effect and has a strong facilitating effect. Market power creates favorable conditions for enterprises to improve carbon efficiency by optimizing resource allocation, transferring risk, etc. Meanwhile, it also makes enterprises increase their motivation to learn from high-quality enterprises in the same industry with higher carbon efficiency. It can be seen that market power can promote enterprises in the same industry to improve carbon efficiency together.

6. Conclusions and Policy Recommendations

6.1. Conclusions

In the framework of China’s “Dual Carbon” goals, which seek to attain carbon peak and carbon neutrality by encouraging green manufacturing and sustainable development, this study offers new insights into the connection between market power and carbon efficiency. Using a comprehensive dataset of industrial enterprises (including mining, manufacturing, and utilities sectors) from 2012 to 2021, we empirically examine the mechanisms through which market power influences carbon efficiency. According to the test, enterprise market power significantly contributes to carbon efficiency, and profitability mediates the relationship between market power and carbon efficiency. This study’s additional heterogeneity analysis revealed that there are notable variations in carbon efficiency depending on how much environmental information an enterprise discloses and whether it pollutes. The positive correlation between firms’ market power and carbon efficiency is more significant when firms have a lower degree of environmental information disclosure and are non heavily polluting firms. In the expansive analysis, it is found that there is a significant cohort effect of carbon efficiency, and the market power can effectively empower the industry cohort effect of carbon efficiency.

6.2. Discussion

The strength of this relationship varies considerably across different firm characteristics. Notably, the positive effect of market power on carbon efficiency is more pronounced among firms with lower environmental information disclosure levels and non-heavy-polluting enterprises. This variation implies that the relationship between market power and carbon efficiency may be moderated by public scrutiny and regulatory forces. The marginal gains of market power on carbon efficiency seem to decrease for big polluters and companies with high environmental transparency, maybe as a result of stricter regulatory compliance and better baseline environmental performance. These findings contribute to the ongoing debate about the role of market structure in environmental governance. While our results support the potential for market leaders to drive carbon efficiency improvements, they also highlight the importance of considering firm-specific factors in designing climate policies. The observed heterogeneity underscores the requirements for differentiated regulatory approaches that account for variations in environmental disclosure practices and pollution intensity across industries.

6.3. Policy Recommendations

This study provides a new perspective for understanding the relationship between market power, profitability and carbon efficiency while revealing the heterogeneous performance of high and low environmental disclosure and heavy polluters in this process. These findings are of great significance for companies to formulate carbon reduction strategies and improve carbon efficiency. In view of this, the following are recommendations from an international perspective, combined with the latest global carbon market construction trends.

First of all, the international community should strengthen the regulatory framework of the global carbon market and promote the establishment of a unified carbon pricing mechanism and strict emission penalty standards for non-compliance to ensure the global fairness, transparency, and efficient operation of the carbon market. Governments can encourage enterprises to increase investment in R&D and innovation of low-carbon technologies through international tax coordination, green financial support, technology transfer incentive mechanisms, and other means of transnational cooperation to enhance global carbon efficiency. In addition, they should actively participate in and promote the in-depth integration of the international carbon trading market and strengthen international cooperation and exchanges in carbon capture and storage technologies, green financial instruments, policy synergies, etc. to jointly build a global climate governance system and effectively respond to the challenges of climate change.

Secondly, to increase public knowledge and engagement in environmental protection, the international community should work together to promote and educate people about low-carbon environmental protection concepts, and multilateral platforms and international media should also be utilized. An internationally recognized green product certification system should be established to enhance the competitiveness and recognition of green products in the global market. At the same time, the establishment of a global carbon emission information disclosure platform and social supervision mechanism should be promoted, the international media, non-governmental organizations, and the public should be encouraged to conduct cross-border supervision and report on carbon emission behaviors, and a favorable atmosphere for global participation in carbon emission reduction should be created.

Last but not least, enterprises should be based on an international vision, R&D investment in low-carbon technologies should be increased, international cooperation platforms should be utilized to accelerate technology introduction, digestion, absorption, and re-innovation, and energy utilization efficiency and carbon emission reduction capacity should be enhanced in the global supply chain. By improving the added value of their products and their differentiation strategies, enterprises can exalt their competitiveness and profitability in the global market, increase the international market share of low-carbon and environmentally friendly products, and actively modify their product structure. Additionally, enterprises should adhere to international standards to establish a sound carbon management system, strengthen the monitoring, accounting, reporting, and third-party verification mechanisms for carbon emissions, and apply advanced management tools to achieve the scientific management of carbon emissions and effective emission reduction so as to contribute to the improvement of global carbon efficiency.

6.4. Research Prospects

Future research can further explore how market power positively affects carbon efficiency by influencing the profitability and resource allocation efficiency of enterprises. Moreover, the differences and commonalities of enterprises in different industries and sizes in this process can be analyzed. The various low-carbon policies issued by the government can be assessed and analyzed to explore their specific impact on enterprises’ carbon efficiency, market power, and profitability. Furthermore, targeted policy recommendations can be put forward to provide a reference for the government to formulate more scientific and reasonable low-carbon policies.