The Impact of Carbon Risk on Corporate Greenwashing Behavior: Inhibition or Promotion?

Abstract

Climate risks arising from carbon emissions have become a major global challenge, constraining economic development and exerting complex effects on firms’ operations at the micro level. This study examines A-share-listed companies from 2009 to 2023, calculating the extent of carbon risks and the degree of greenwashing. Our results show that carbon risks suppress greenwashing among these enterprises, with financing constraints and the quality of internal controls positively moderating this effect from both external and internal perspectives. This suggests that the pressure imposed by carbon risks encourages enterprises to prioritize environmental concerns and actively disclose relevant information. Additionally, governments and regulatory authorities should increase their policy and regulatory pressures on these enterprises, guiding them to confront environmental challenges, assume significant responsibility for ecological protection, and formulate sustainable development strategies to enhance their competitiveness and long-term viability.

1. Introduction

Global warming has caused a series of natural disasters and severe damage to the global ecosystem, leading to substantial economic losses worldwide [1]. China is the largest emitter of carbon dioxide globally, projections suggest it will account for the largest portion of anticipated increases in future carbon emissions. Although China’s economy accounts for approximately 18% of global GDP, its carbon emissions are approaching one-third of the global total. At the UN General Assembly in September 2020, China made a formal commitment to peak its carbon emissions and to attain carbon neutrality. The shift toward a low-carbon economic model generates emission-related uncertainties—commonly referred to as “carbon risks”—which manifest across physical, regulatory, and reputational domains. These risks have therefore become central considerations for firms when evaluating climate-related matters [2]. Against a backdrop of heightened international concern about climate change and sustainable development, consumers, investors, and public authorities increasingly scrutinize firms’ carbon emissions and environmental performance. As environmental regulations tighten, firms profiles face a heightened probability of environmental incidents that can produce material financial consequences—including regulatory fines, remediation expenses, and reputational losses that impair firm value [3]. Carbon risk not only affects investors’ decision-making [4] but also influences corporate financial decisions [5] and governance decisions [6]. Despite increasing attention to carbon-related risks, their influence on firms’ corporate social responsibility (CSR) decision-making remains inadequately studied. Empirical evidence and theoretical analyses detailing how carbon risk shapes CSR strategies, resource allocation, and stakeholder engagement are still limited. Corporations bear significant responsibility for environmental protection, and under the trend of sustainable development, the CSR movement has gained momentum. However, it is also noteworthy that a surge in false CSR behaviors has emerged. Specifically, due to regulatory loopholes, inadequate oversight, and the relative ease as well as low cost of greenwashing, such behaviors are intensifying and even spreading further.

The identification, assessment, and management of corporate carbon risks began with practice, with practical exploration preceding theoretical research. The well-known “Carbon Disclosure Project” (CDP) requires companies to assess carbon risks, but carbon information disclosure by Chinese companies currently follows a voluntary disclosure model, with no standardized content or methodology for carbon disclosures [7]. Beyond carbon information, companies, as key stakeholders in environmental protection, are expected to disclose a range of environmentally related information. However, amid opportunistic behavior and information asymmetry in the green market, companies may adopt greenwashing strategies to meet external demands and achieve higher returns, thereby embellishing their environmental performance [8]. Greenwashing is a means by which companies fulfill legitimacy demands and engage with stakeholders to build a positive corporate image [9]. Existing research has identified two primary drivers of corporate greenwashing. On one hand, external factors, such as government regulations [10,11], media and industry association pressures [12], and information asymmetry between consumers and companies [13], promote greenwashing. Media coverage, as an independent external pressure, can amplify or attenuate carbon-risk signals and, together with internal governance, influence firms’ greenwashing behavior [14]. On the other hand, internal characteristics, such as corporate performance, organizational management capabilities, incentive mechanisms [15,16], financial status [17,18], and the environmental impact of the company [19,20], also contribute to greenwashing behavior. Based on signaling theory, firms with high carbon risk may engage in greenwashing to conceal their environmental shortcomings, thereby signaling a falsely improved environmental image; in contrast, legitimacy theory argues that firms with high carbon risk face stronger stakeholder scrutiny and external constraints and thus are more likely to provide substantive disclosures and genuine mitigation efforts to avoid greenwashing. These two theories offer conflicting predictions about the carbon risk–greenwashing relationship, and this study aims to resolve this theoretical controversy through empirical analysis of A-share-listed firms.

Our study investigates the influence of carbon-related risk on corporate greenwashing by analyzing A-share-listed companies on the Shanghai (SSE) and Shenzhen (SZSE) stock exchanges between 2009 and 2023. Because environmental disclosure regimes remain underdeveloped and there is no uniform standard for reporting carbon emissions, carbon risk cannot be observed directly; consequently, the analysis employs carbon intensity as a proxy indicator. The empirical findings indicate an inverse association between carbon-related risk and corporate greenwashing, with higher levels of carbon risk corresponding to reduced greenwashing behavior. Moreover, financing constraints and internal control systems significantly moderate this relationship, acting, respectively, as external pressures and internal governance mechanisms that shape firms’ CSR choices in response to carbon risk. The analysis further reveals that media scrutiny matters: firms receiving fewer negative news reports are less likely to greenwash in response to carbon risk than firms that receive more adverse coverage.

This article extends understanding of carbon risk and corporate greenwashing along three distinct dimensions. First, it shifts the research focus from the traditional examination of the economic consequences of carbon exposure to firm-level CSR outcomes. In light of China’s “dual carbon” targets and the ongoing green transition, enhancing the recognition and governance of carbon risk has become a core component of corporate sustainability strategies. This study systematically reveals how carbon exposure influences firms’ CSR disclosure. This extension enriches the carbon-risk literature and contributes to the theoretical advancement of environmental and climate economics. Second, existing studies typically classify the drivers of corporate greenwashing into external pressures (such as government or regulatory requirements) and internal firm characteristics (such as firm size). For example, Delmas and Burbano highlight the influence of external regulatory and internal organizational factors, while Testa et al. examine social pressures from media and industry associations [12,16]. As an emerging form of risk, carbon risk operates not only through institutional and regulatory channels but also through financial and reputational pressures, thereby forming an integrated external–internal driving force that shapes firms’ strategic behavior and CSR engagement. By integrating both the internal and external dimensions of carbon risk, this study broadens the analysis of greenwashing motivations among A-share-listed companies. Third, in addition to examining the direct impact of carbon risk on corporate greenwashing behavior, we further investigate the moderating effects of internal control and financing constraints, as well as the heterogeneity of this relationship under varying intensities of negative media coverage. This not only deepens academic understanding of the interaction between carbon risk and corporate CSR information disclosure, but also provides valuable insights for regulators seeking to curb greenwashing and optimize carbon regulation policies.

2. Literature Review and Hypotheses Development

2.1. Literature Review

Carbon risk captures the range of uncertainties linked to climate change and to firms’ persistent reliance on fossil fuels [21]. Labat and White conceptualize carbon risk as comprising three interrelated yet distinct dimensions [22]. Regulatory risk derives from existing and prospective carbon regulations—such as tighter emission standards or carbon trading requirements—that can raise compliance costs or impose new market obligations, thereby affecting firms’ financial performance. Physical risk denotes climate-induced hazards that endanger production facilities and disrupt supply chains. Business risk encompasses reputational, legal, and competitive pressures at the firm level: firms judged environmentally irresponsible may suffer reputational damage, increased litigation risk, and competitive disadvantage, which can precipitate operational disruptions and weaken both market position and financial stability. Unlike broader environmental risks such as pollution or resource scarcity, carbon risk specifically arises from the transition toward a low-carbon economy and the tightening of carbon governance mechanisms [23]. It reflects firms’ direct exposure to carbon pricing, emission quotas, and decarbonization requirements, which exert unique pressures distinct from traditional environmental management challenges.

The linkage between carbon risk and firms’ financial outcomes continues to attract academic attention, with empirical evidence producing mixed results [2]. Some studies indicate that carbon risk significantly affects corporate performance and value. Future climate-related policies and regulatory measures may expose firms to significant downside risks [24]. The uncertainty associated with forthcoming policy implementation can impose substantial strain on corporate financial outcomes—particularly in fossil fuel-dependent industries—ultimately contributing to deteriorations in financial performance [25]. However, other research has failed to find such correlations [26,27]. From an external perspective, as low-carbon transition concepts gain traction, corporate carbon risk will significantly influence external stakeholders’ evaluations of firms, particularly drawing the attention of investors. Consequently, this will directly affect companies’ performance in capital markets, including company valuations, stock price levels, and stock returns. From a positive viewpoint, Oestreich et al. report that carbon exposure may give rise to a “carbon premium” embedded in stock valuations, whereas Bolton observes that higher-emitting firms have tended to earn superior stock returns [4,28]. Additionally, Monasterolo et al. discovered that firms with low carbon emissions face lower systematic risks [29]. Conversely, some studies have demonstrated the negative impacts of carbon risk, for example, through elevated stock return co-movement and a deterioration of price-discovery mechanisms in capital markets [30]. In China, carbon risk is deeply intertwined with the evolution of national carbon governance. Since 2013, regional carbon trading pilots and the launch of the national ETS in 2021 have progressively refined quota allocation, sectoral coverage, and verification systems, reshaping firms’ carbon exposure and risk perception. Facing both regulatory and market pressures, industries have responded through technological innovation [31], energy efficiency improvement [32], and carbon asset management. Overall, China’s carbon trading system has become a key institutional driver of corporate low-carbon transition [33], highlighting the dynamic interplay between regulatory carbon markets and firm-level strategic adaptation.

Recent research has explored how regulatory environments influence firms’ voluntary disclosure of carbon-related information, and the majority of findings indicate a positive association between regulatory stringency and disclosure behavior [34,35,36,37,38]. However, a more comprehensive assessment of corporate greenwashing should examine not only firms’ disclosed carbon information but also their actual carbon-emission performance [39]. As ESG regulations continue to proliferate [40], companies tend to develop more environmentally friendly and responsible products and services. Nonetheless, cost considerations may discourage firms from fully honoring their environmental commitments, resulting in vague or misleading sustainability claims and giving rise to the phenomenon known as greenwashing [41]. The scholarly conceptualization of greenwashing has evolved from a singular dimension to a more comprehensive perspective. Delmas and Burbano define greenwashing from the angle of communicating false information as “poor environmental performance but positive communication regarding environmental performance” [16]. The limitation of this definition lies in confining greenwashing to firms engaged in positive communication, neglecting the potential for greenwashing in firms that avoid communication altogether. Lyon and Maxwell and Lyon and Montgomery subsequently enriched the concept of greenwashing, incorporating scenarios where insufficient communication and the use of imagery convey a false green image [19,42]. Today, the phenomenon of greenwashing is increasingly observed in climate information disclosure; for instance, some companies superficially fulfill emission reduction targets by purchasing carbon credits and renewable energy certificates while failing to reduce actual carbon emissions, thus falling within the realm of greenwashing.

Most literature suggests that higher levels of regulation help reduce corporate greenwashing behaviors [16,43]. Regulatory pressure may prompt companies to disclose carbon information, thereby increasing the likelihood of them substantively adopting socially responsible green initiatives [44]. In this context, Delmas and Burbano demonstrate that compulsory environmental disclosure serves as an effective mechanism to curb corporate greenwashing [16]. However, carbon risk is not merely a form of regulatory pressure; it encompasses regulatory risks, reputational risks, and technical risks associated with low-carbon transitions, affecting firms more broadly. Under China’s “dual carbon” (carbon peaking and carbon neutrality) strategy, enterprises are increasingly confronted with complex, multi-dimensional carbon risks arising from the integration of regulatory, market-based, and reputational factors. Yet, how these risks jointly influence firms’ greenwashing behaviors within the unique institutional and policy environment of China remains insufficiently explored and warrants further empirical investigation.

2.2. Hypothesis Development

The theories of voluntary disclosure and signaling provide a theoretical basis for understanding how firms manage external communication under carbon risk. Signaling theory emphasizes that firms can reduce information asymmetry by voluntarily releasing credible information, thereby conveying their underlying value to external stakeholders [45]. Voluntary disclosure theory further posits that managers choose whether and how to disclose information based on strategic considerations, using disclosure to improve stakeholders’ understanding of corporate behavior, mitigate information imbalances, and alleviate principal–agent conflicts [46,47].

In the context of carbon risk, firms with relatively low carbon exposure are more capable of providing transparent and comprehensive carbon-related information. By doing so, they use credible disclosure as a positive signal to highlight their superior environmental performance and to distinguish themselves from high-emission peers [48]. In contrast, firms facing higher carbon risk encounter greater regulatory, reputational, and financial pressure. For these firms, full and honest disclosure may expose substantial environmental weaknesses and trigger adverse market reactions. As a result, they are more likely to adopt selective, strategic, or embellished disclosure practices, including greenwashing, to obscure their true carbon profile while maintaining a favorable public image. Under voluntary disclosure and signaling frameworks, stakeholders largely depend on the information released by firms and often lack the capacity to fully verify its authenticity, which provides room for such opportunistic behavior. Building on this logic, we propose the following hypothesis:

H1a. 

Carbon risk promotes corporate greenwashing behavior.

Stakeholder theory and legitimacy theory shed light on how firms adjust their disclosure strategies in response to external pressures [49]. Stakeholder theory posits that firms are accountable to a broad set of stakeholders beyond shareholders, including consumers, employees, communities, regulators, and media audiences. Because these groups hold diverse and sometimes conflicting interests, firms need to release information that responds to heterogeneous informational demands and demonstrates responsibility toward multiple constituencies [50]. Legitimacy theory emphasizes that corporate survival depends on maintaining consistency between organizational behavior and socially accepted norms and values [51]. When firms perceive a potential legitimacy gap, they tend to adopt strategies that signal compliance with societal expectations in order to secure continued support.

From stakeholder and legitimacy perspectives, environmental and carbon-related disclosure can be used not only as a channel for information transmission, but also as a governance and reputation tool [52]. On the one hand, complying with disclosure regulations and relevant codes of conduct reduces the likelihood and severity of negative media coverage, enhances public trust, and helps protect long-term firm value [53]. On the other hand, for firms with higher carbon exposure, accurate and timely disclosure of carbon risk can function as a proactive response to external scrutiny. When such firms sense rising skepticism or criticism, they may choose to strengthen disclosure transparency to demonstrate environmental responsibility, reduce suspicion of concealment, and stabilize their legitimacy in the eyes of stakeholders [54,55]. By truthfully disclosing carbon risk information, firms are seen as honoring their social contract and responding constructively to external expectations [49]. Based on this reasoning, we propose the following hypothesis:

H1b. 

Carbon risk mitigates corporate greenwashing behavior.

Lenders perceive elevated carbon emissions as a material credit risk and therefore require higher risk premiums from firms with greater carbon risk to compensate for this exposure [56]. Firms facing tight financing constraints find it more difficult to obtain external funds and lack sufficient internal cash flow to buffer bankruptcy and distress risks. As discussed earlier, carbon risk induced by environmental regulation heightens external scrutiny and encourages firms to assume environmental responsibility, improve the accuracy and completeness of carbon disclosure, and curb greenwashing. Under conditions of stronger financing constraints, resource availability is more limited, and high-carbon firms face greater external pressure. To ease these constraints, such firms have stronger incentives to adopt more credible ESG strategies and enhance environmental information disclosure in order to attract institutional investors, reduce financing costs, and alleviate funding pressures [57]. Building on this logic, we propose the following hypothesis:

H2. 

Financing constraints moderate the relationship between carbon risk and corporate greenwashing.

Frameworks such as COSO, the Sarbanes–Oxley Act (SOX), and China’s Basic Internal Control Standards consistently stress that internal control systems are intended to standardize managerial processes, strengthen risk management capabilities, and protect market order and public interest [58]. As carbon risk constitutes an important source of systemic risk, sound internal controls can help firms enhance risk awareness, maintain sensitivity to emerging risks, and implement effective monitoring and early warning mechanisms. The checks-and-balances embedded in internal control structures help constrain managerial opportunism, improve corporate governance quality, and reduce the likelihood of misreporting or opportunistic disclosure. Existing evidence shows that strong internal control systems can significantly curb firms’ risk-taking tendencies [59,60]. They can also improve financial performance by enhancing the reliability of internal information and strengthening oversight mechanisms [61]. In this context, firms with higher-quality internal controls are better positioned to absorb the financial pressure associated with carbon risk, comply with disclosure requirements, and avoid resorting to symbolic or misleading environmental communication. Accordingly, we propose the following hypothesis:

H3. 

Internal controls moderate the relationship between carbon risk and corporate greenwashing.

Media coverage helps firms build public legitimacy by transmitting information to stakeholders [56]. In China’s largely voluntary ESG-disclosure environment, information asymmetry and uneven enforcement keep the credibility of high-pollution firms’ ESG claims under close media watch. Unfavorable exposure signals a legitimacy shortfall that can threaten survival [57], and stakeholders lacking verifiable data often lean on media reports to judge conduct [58]. Although coverage can discipline firms [56,59], intense scrutiny also raises the immediacy and visibility of legitimacy threats. Because genuine decarbonization is costly and slow to materialize, managers facing tight media timelines may substitute lower-cost, highly salient symbolic actions for substantive reforms, using selective disclosure and impression management as “quick fixes” to stem reputational damage [60]. In such settings, the disciplining channel of carbon risk is crowded out by the signaling payoff of symbolic claims, so firms are more prone to greenwashing despite high carbon risk.

H4. 

The impact of carbon risk on corporate greenwashing may be heterogeneously influenced by negative media coverage.

3. Data and Empirical Design

3.1. Sample Selection and Data Sources

We select A-share companies listed on the SSE and SZSE from 2009 to 2023 as the initial research sample. The sample is subsequently refined according to the following criteria: (1) firms in the financial and insurance industries are excluded; (2) firms with extensive missing data that cannot be reliably imputed are removed; and (3) firms designated as ST or *ST due to abnormal operations or delisting risks are also excluded. To mitigate the influence of outliers, all continuous variables are winsorized at the 1% and 99% levels. The final dataset comprises 33,458 firm-year observations, covering a total of 3395 listed firms.

Firm-level financial information is primarily obtained from the China Stock Market and Accounting Research (CSMAR) database. Additional information was manually collected from multiple sources, including corporate annual reports, the China Research Data Service (CNRDS) platform, disclosures from the SSE and SZSE, as well as official statistical yearbooks.

3.2. Measurement of Carbon Risk

In this study, corporate carbon risk is assessed through carbon emission intensity. Specifically, the total carbon emissions of listed firms are composed of four components: combustion and fugitive emissions, process-related emissions, emissions from waste treatment, and those arising from land-use changes (e.g., conversion of forests into industrial land) [6,62]. To mitigate the impact of firm size, total carbon emissions are scaled by operating revenue (in millions of RMB), yielding a measure of carbon intensity [4]. This indicator effectively captures both the efficiency of carbon emissions in production activities and the degree of exposure to environmental risk.

3.3. Measurement of Greenwashing Behavior

Following Hu et al., we count the frequency of these environment-related keywords (e.g., “green,” “low-carbon,” “environmental protection”) appearing in the Management Discussion and Analysis (MD&A) section of each firm’s annual report [63]. If a firm’s word frequency in a given industry-year exceeds the corresponding industry median, its oral green publicity dummy variable (Oral) is coded as 1; otherwise, it equals 0.

The measure of actual environmental performance (Actual) is based on whether the firm is penalized for environmental violations within the fiscal year. If the firm has received any regulatory or administrative environmental sanctions (such as fines or rectification orders), Actual = 1; otherwise, Actual = 0.

As shown in Model (1), we then construct a binary indicator Gw to capture direct greenwashing behavior. A firm-year observation is classified as greenwashing if the company exhibits strong green rhetoric (Oral = 1) but poor environmental performance (Actual = 1):

$${\mathrm{G}\mathrm{w}}_{\mathrm{i},\mathrm{t}}=\left\{\begin{array}{c}1, \mathrm{if} {\mathrm{O}\mathrm{r}\mathrm{a}\mathrm{l}}_{\mathrm{i},\mathrm{t}}=1 \mathrm{and} {\mathrm{A}\mathrm{c}\mathrm{t}\mathrm{u}\mathrm{a}\mathrm{l}}_{\mathrm{i},\mathrm{t}}=1\\ 0,\mathrm{otherwise}\end{array}\right.$$

(1)

A value of one indicates a clear inconsistency between a firm’s verbal emphasis on environmental responsibility and its actual conduct—representing the typical “say–do gap.”

3.4. Control Variables

Following the studies of Roulet and Touboul and Zhang et al., this study selects several conventional variables that may influence corporate greenwashing as control variables [18,64]. These include company Size (Size), leverage (Lev), return on equity (Roe), total asset turnover (Ato), board Size (Board), and the proportion of independent directors (Indep). The definitions of these variables are provided in

Table 1. Definition of main variables.

Variable Definitions	Symbols	Measurements
Greenwashing	Gw	Dummy variable indicating excessive green disclosure with poor environmental performance
Carbon Risk	Cr	Total carbon emissions divided by operating revenue (in millions).
Company Size	Size	Natural log of total assets.
Financial Leverage	Lev	Ratio of total liabilities to total assets
Return on Equity	Roe	Net operating income deflated by equity
Assets Turnover	Ato	Net sales deflated by total assets
Board Size	Board	Natural logarithm of the total number of directors on the board
The Proportion of Independent Directors	Indep	Natural logarithm of the total number of independent directors on the board

.

3.5. Model Specification

To test our hypotheses, we construct the following linear probability model with firm and year fixed effects:

$${{\mathrm{G}\mathrm{w}}_{\mathrm{i},\mathrm{t}}=\beta }_0+{\beta }_1{\mathrm{C}\mathrm{r}}_{\mathrm{i},\mathrm{t}}+{\beta }_{\mathrm{j}}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i},\mathrm{t}}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{F}\mathrm{E}+\mathrm{F}\mathrm{i}\mathrm{r}\mathrm{m}\mathrm{F}\mathrm{E}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(2)

$${{\mathrm{G}\mathrm{w}}_{\mathrm{i},\mathrm{t}}=\beta }_0+{\beta }_1{\mathrm{C}\mathrm{r}}_{\mathrm{i},\mathrm{t}}+{\beta }_2{{\mathrm{W}\mathrm{w}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\mathrm{C}\mathrm{r}}_{\mathrm{i},\mathrm{t}}\times {\mathrm{W}\mathrm{w}}_{\mathrm{i},\mathrm{t}}+\beta }_{\mathrm{j}}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i},\mathrm{t}}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{F}\mathrm{E}+\mathrm{F}\mathrm{i}\mathrm{r}\mathrm{m}\mathrm{F}\mathrm{E}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(3)

$${{\mathrm{G}\mathrm{w}}_{\mathrm{i},\mathrm{t}}=\beta }_0+{\beta }_1{\mathrm{C}\mathrm{r}}_{\mathrm{i},\mathrm{t}}+{\beta }_2{{\mathrm{I}\mathrm{c}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\mathrm{C}\mathrm{r}}_{\mathrm{i},\mathrm{t}}\times {\mathrm{I}\mathrm{c}}_{\mathrm{i},\mathrm{t}}+\beta }_{\mathrm{j}}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i},\mathrm{t}}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{F}\mathrm{E}+\mathrm{F}\mathrm{i}\mathrm{r}\mathrm{m}\mathrm{F}\mathrm{E}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(4)

Model (2) assesses H1a and H1b by quantifying carbon risk’s direct influence on firms’ propensity to engage in greenwashing. Model (3) incorporates an interaction term to test H2, assessing whether financing constraints alter the impact of carbon risk on firms’ propensity to engage in greenwashing. Model (4) incorporates an interaction term to test H3, evaluating whether the quality of internal controls moderates the effect of carbon risk. In the equations, i and t index firm and year; ${Cr}_{i,t}$ denotes corporate carbon risk; ${Gw}_{i,t}$ denotes corporate greenwashing; ${Ww}_{i,t}$ denotes financing constraints; YearFE and FirmFE denote year fixed effects and firm fixed effects; ${Ic}_{i,t}$ denotes internal-control quality; and ${\epsilon }_{i,t}$ is the stochastic disturbance (error) term. To address potential within-group autocorrelation, standard errors are clustered at the firm level across all regression analyses.

4. Empirical Results and Discussions

4.1. Descriptive Analysis

Table 2. Summary of descriptive statistics.

Variables	Obs	Mean	SD	Min	Median	Max
Gw	33,458	0.1667	0.3727	0.0000	0.0000	1.0000
Cr	33,458	0.0003	0.0002	0.0000	0.0003	0.0153
Size	33,458	22.2567	1.3237	19.3129	22.0351	26.4523
Lev	33,458	0.4187	0.2053	0.0278	0.4115	0.9343
Roe	33,458	0.0570	0.1459	−2.1749	0.0699	0.4179
Ato	33,458	0.5921	0.3616	0.0475	0.5236	2.6445
Board	33,458	2.1228	0.1964	1.6094	2.1972	2.7081
Indep	33,458	0.3758	0.0537	0.2500	0.3636	0.6000

provides an overview of the descriptive statistics for the variables employed in this study. Based on 33,458 firm-year observations, the mean value of Gw is 0.1667 with a standard deviation of 0.3727, suggesting that greenwashing is present but varies notably among firms. The average Cr is 0.0003 (SD = 0.0002), indicating that firms’ carbon risk is broadly similar and at an acceptably low level. To some extent, this suggests that China’s recent efforts to advance the low-carbon transition have yielded encouraging results.

4.2. Pearson Correlation Analysis

Table 3. Results of Pearson correlation analysis.

Variables	Gw	Cr	Size	Lev	Roe	Ato	Board	Indep
Gw	1.0000
Cr	−0.0706 ***	1.0000
Size	0.2979 ***	−0.0449 ***	1.0000
Lev	0.1568 ***	0.0086	0.5197 ***	1.0000
Roe	0.0021	−0.0385 ***	0.0770 ***	−0.2137 ***	1.0000
Ato	0.0370 ***	−0.0239 ***	0.0066	0.0985 ***	0.1379 ***	1.0000
Board	0.0274 ***	0.0224 ***	0.2564 ***	0.1591 ***	0.0355 ***	0.0254 ***	1.0000
Indep	−0.0009	−0.0065	0.0071	−0.0065	−0.0181 ***	−0.0364 ***	−0.5448 ***	1.0000

Note: *** indicates significance at the level of 1%.

reports the correlation matrix of the main variables. The correlation between Cr and Gw is −0.0706 and statistically significant at the 1% level, indicating a negative association: firms with higher carbon risk tend to engage less in greenwashing. Nevertheless, this bivariate result is only suggestive; a more rigorous conclusion requires multivariate regressions that control for relevant covariates and fixed effects.

4.3. Baseline Regression

The baseline regression outcomes are displayed in

Table 4. Baseline regression results.

Variables	(1)	(2)	(3)
	Gw	Gw	Gw
Cr	−87.8770 ***	−27.8666 ***	−25.0911 ***
	(−5.6545)	(−5.2068)	(−4.4659)
Size	0.0906 ***	0.0722 ***	0.0776 ***
	(24.8506)	(19.7215)	(10.5398)
Lev	−0.0151	0.0587 ***	−0.0525 **
	(−0.7781)	(3.0961)	(−2.0194)
Roe	−0.0765 ***	−0.0004	0.0203
	(−3.9640)	(−0.0190)	(1.1921)
Ato	0.0400 ***	0.0458 ***	−0.0130
	(3.9792)	(4.7797)	(−0.8614)
Board	−0.1457 ***	−0.0336 *	−0.0159
	(−7.4191)	(−1.7163)	(−0.5753)
Indep	−0.3090 ***	−0.2020 ***	−0.0406
	(−4.4183)	(−2.9955)	(−0.4905)
_cons	−1.4119 ***	−1.3356 ***	−1.4746 ***
	(−16.0118)	(−15.6288)	(−8.2590)
YearFE	No	Yes	Yes
FirmFE	No	No	Yes
F	128.3776	108.0389	22.0612
R2	0.0978	0.1493	0.4298
N	33458	33458	33458

Note: *, **, and *** indicate significance at the level of 10%, 5%, and 1%, respectively; T-statistics for the regression coefficients are in parentheses.

. Column (1) includes only the control variables, while Columns (2) and (3) progressively add year and firm fixed effects to control for unobservable heterogeneity.

The coefficient of the core explanatory variable, Cr, is significantly negative at the 1% level across Columns (1), (2), and (3). These results demonstrate that the negative association between Cr and Gw is robust to different model specifications, confirming Hypothesis H1b. In other words, higher carbon risk significantly suppresses greenwashing behavior, indicating that companies facing higher carbon exposure are more likely to enhance the authenticity of their environmental disclosures to mitigate regulatory and reputational risks.

4.4. The Moderating Role of the Financial Constraints and Internal Controls

We use the Ww index to proxy firms’ financing constraints; higher Ww values indicate greater difficulty in obtaining external finance. Column (1) of

Table 5. The moderating effects of financial constraints and internal controls.

Variables	(1)	(2)
	Gw	Gw
Cr	−457.4147 **	83.0264 *
	(−2.1803)	(1.8842)
Ww	0.2416 **
	(2.4396)
Cr × Ww	−436.9225 **
	(−1.9854)
Ic		0.0088
		(1.4897)
Cr × Ic		−40.5829 ***
		(−2.5974)
Size	0.0664 ***	0.0891 ***
	(7.3963)	(9.9388)
Lev	−0.0528 *	−0.0386
	(−1.9041)	(−1.2328)
Roe	0.0415 **	0.0289
	(2.3465)	(1.5789)
Ato	−0.0214	−0.0058
	(−1.3013)	(−0.3327)
Board	−0.0146	0.0139
	(−0.4877)	(0.4293)
Indep	−0.0572	0.0442
	(−0.6573)	(0.4635)
_cons	−0.9801 ***	−1.8360 ***
	(−4.9099)	(−8.4682)
YearFE	Yes	Yes
FirmFE	Yes	Yes
F	15.8712	15.7357
R2	0.4331	0.4704
N	28,642	26,121

Note: *, **, and *** indicate significance at the level of 10%, 5%, and 1%, respectively; T-statistics for the regression coefficients are in parentheses.

presents the test for hypothesis H2. The interaction term Cr × Ww in the table has a coefficient of −436.9225, significant at the 5% level, indicating that financing constraints amplify the inhibitory effect of carbon risk on corporate greenwashing. A comparison of the fitted slopes in Figure 1a shows that the absolute value of the slope for the “High Ww” line is larger than that for the “Low Ww” line; therefore, under tighter financing constraints the suppressive effect of carbon risk on greenwashing is stronger, which further supports hypothesis H2. Companies with high financing constraints face more restrictions and higher costs during financing activities, making them more eager to address information asymmetry issues by cultivating a positive corporate image to gain investor trust. Consequently, firms exposed to carbon risk tend to enhance the disclosure of environmental information and green initiatives, which in turn reduces their likelihood of engaging in greenwashing.

Internal control data are obtained from the DIB (Dibo) Database. Based on the internal control index scores, the internal control quality of listed firms is classified into the following grades in descending order: AAA, AA, A, BBB, BB, B, C, and D, which are assigned numerical values from 8 to 1, respectively. The empirical findings for hypothesis H3 are presented in Column (2) of Table 5. The interaction term Cr × Ic in the table has a coefficient of −40.5829, significant at the 1% level, indicating that higher-quality internal controls further strengthen the negative effect of carbon risk on corporate greenwashing. In Figure 1b, the fitted-slope comparison shows that the absolute value of the slope for the “High Ic” line exceeds that for the “Low Ic” line; hence, when internal-control quality is high, the suppressive effect of carbon risk on greenwashing is stronger, which further supports hypothesis H3. Companies with high-quality internal controls possess stronger organizational structures, better-qualified personnel, and well-established operational procedures, which reduce their likelihood of engaging in greenwashing. On one hand, such companies possess better risk management capabilities and can prevent and manage risks from multiple perspectives; on the other hand, high-quality internal controls foster better ethical standards, reducing the likelihood of non-compliant fraudulent activities.

4.5. Robustness and Endogeneity Checks

4.5.1. Alternative Measure of Carbon Risks

For robustness, we replace the emissions-based carbon metric with the firm-level carbon disclosure score from the CNRDS (denoted as Cr2). A higher score denotes greater managerial attention to carbon issues and hence higher carbon-related exposure. In the year- and firm-fixed-effects specification reported in the table, the coefficient on Cr2 is −0.0012, significant at the 1% level.

In

Table 6. Results of robustness and endogeneity checks.

Variables	(1)	(2)	(3)	(4)	(5)
	Gw	Cr	Gw	Gw	Gw
Cr			−168.7523 **	−32.2019 ***	−460.8562 **
			(−2.2658)	(−2.7287)	(−2.2844)
Cr2	−0.0012 ***
	(−2.7104)
Cr_Mean		0.6850 ***
		(4.6978)
Size	0.0681 ***	−0.0001 ***	0.0702 ***	0.0862 ***	0.6846 ***
	(9.2434)	(−6.7392)	(8.4336)	(9.3320)	(8.6638)
Lev	−0.0369	0.0000 *	−0.0489 *	−0.0584 *	0.8042 ***
	(−1.4575)	(1.7222)	(−1.8657)	(−1.6762)	(2.8619)
Roe	0.0221	−0.0000 **	0.0163	0.0269	0.3327 *
	(1.3004)	(−2.2121)	(0.9208)	(1.1830)	(1.9397)
Ato	−0.0169	−0.0001 ***	−0.0218	−0.0220	−0.0787
	(−1.1301)	(−6.2388)	(−1.3856)	(−1.1896)	(−0.5242)
Board	−0.0224	0.0000 **	−0.0099	−0.0009	0.0163
	(−0.8232)	(2.4334)	(−0.3546)	(−0.0243)	(0.0570)
Indep	−0.0841	0.0001 *	−0.0293	0.0176	0.2199
	(−1.0270)	(1.6934)	(−0.3513)	(0.1578)	(0.2625)
_cons	−1.2520 ***	0.0011 ***	N/A	−1.7031 ***	N/A
	(−7.0524)	(7.4484)	N/A	(−7.5675)	N/A
YearFE	Yes	Yes	Yes	Yes	Yes
FirmFE	Yes	Yes	Yes	Yes	Yes
F	15.0589	9.7176	18.7245	16.0866	N/A
R2/Pseudo R2	0.4078	0.2281	0.0007	0.4911	0.3418
N	32,333	33,370	33,370	16,474	18,564
Kleibergen–Paap rk LM			21.336 ***
Cragg–Donald Wald F			680.648 [16.38]

Note: *, **, and *** indicate significance at the level of 10%, 5%, and 1%, respectively; T-statistics for the regression coefficients are in parentheses; and [16.38] represents the 10% Stock–Yogo critical value for the weak identification test.

, when the firm-level carbon disclosure score is used instead of the carbon risk measure, the coefficient is −0.0012, significant at the 1% level. This is consistent with the baseline findings.

4.5.2. Instrumental Variable (IV) Approach

Although several control variables potentially affecting corporate greenwashing have been included, unobserved factors may still be correlated with the dependent or independent variables, raising endogeneity concerns due to omitted variables. To address this, we employ the industry-year average carbon risk (Cr_Mean, excluding the firm itself) as an instrumental variable. Regarding relevance, Cr_Mean reflects the carbon intensity of firms within the same industry. Due to commonalities in core business and production patterns, firms in the same sector generally face similar carbon risks. Regarding exogeneity, this variable captures only industry-level carbon characteristics and has no direct deterministic relationship with the greenwashing behavior of individual firms.

As shown in Column (2) of Table 6, Cr_Mean is significantly positively correlated with the firm’s own carbon risk, indicating that industry-level carbon risk affects individual firms. Column (3) shows that after instrument variable adjustment, the coefficient of Cr on Gw is −168.7523, significant at the 5% level. Further diagnostics support the validity of the instrument: the under-identification test yields a significant Kleibergen–Paap rk LM statistic at the 1% level, rejecting the null hypothesis of insufficient identification; the weak instrument test shows that the Cragg–Donald Wald F-statistic exceeds the 10% critical value, rejecting the weak instrument null hypothesis. Overall, these results confirm the reliability of Cr_Mean as an instrument, and the main findings remain robust even after accounting for potential endogeneity.

4.5.3. Propensity Score Matching (PSM) Method

In practice, firms with varying levels of carbon risk may exhibit systematic differences, such as corporate strategies or production and operational models, which could lead to sample self-selection bias in previous analyses. To address this issue, we employ the PSM method to re-examine the relationship between corporate carbon risk and greenwashing. Specifically, firms with carbon risk above the sample mean in a given year are assigned to the treatment group, while the remaining firms form the control group. A Logit model is then used to estimate the likelihood of high carbon risk for each firm, with the same set of control variables as in the main analysis.

Column (4) of Table 6 reports the baseline regression results using the matched sample. The findings indicate that Cr remains significantly negatively associated with Gw at the 1% level, suggesting that even after accounting for potential sample self-selection bias, the main hypothesis of this study continues to hold.

4.5.4. Replacement of the Regression Model

Given that the dependent variable, Gw, is a binary indicator, it is arguably more appropriate to employ a Logit model for estimation.

Column (5) of Table 6 reports the results after replacing the original specification with a Logit model. In this case, the coefficient of Cr on Gw is −460.8562 and remains statistically significant, although the significance level decreases from 1% in the linear probability model to 5%. This finding continues to support our earlier conclusions and indicates that the main result is robust under a more rigorous and suitable regression framework.

5. Further Investigations

Negative media coverage can erode market participants’ favorable perceptions of a company, weakening its competitive position, valuation, and financial performance [65], and prompting firms to adopt impression-management-oriented governance responses [66]. Using CNRDS, we measure firm-level counts of negative reports and split the sample at the overall mean.

Table 7. Heterogeneity test based on negative media coverage.

Variables	(1)	(2)
	Less Negative News	More Negative News
	Gw	Gw
Cr	−30.3949 ***	−19.2379 **
	(−3.0202)	(−2.5713)
Size	0.0885 ***	0.0778 ***
	(8.8179)	(8.0896)
Lev	−0.0025	−0.0839 **
	(−0.0674)	(−2.4557)
Roe	0.0372	0.0194
	(1.4220)	(0.8795)
Ato	−0.0138	−0.0061
	(−0.6619)	(−0.3127)
Board	0.0209	−0.0391
	(0.5895)	(−1.0429)
Indep	0.1220	−0.1613
	(1.1362)	(−1.4051)
_cons	−1.8423 ***	−1.4181 ***
	(−7.8269)	(−5.8056)
YearFE	Yes	Yes
FirmFE	Yes	Yes
F	15.6626	12.9777
R2	0.4999	0.4844
N	16081	15922

Note: ** and *** indicate significance at the level of 5% and 1%, respectively; T-statistics for the regression coefficients are in parentheses.

shows that for firms with fewer negative reports (Column 1), the carbon-risk coefficient is −30.395 and significant at the 1% level; for firms with more negative reports (Column 2), the coefficient is −19.238 and significant at the 5% level. The shift from −30.395 to −19.238 indicates that the inhibitory effect of carbon risk on greenwashing weakens as adverse coverage intensifies. Put differently, carbon risk suppresses greenwashing more strongly when firms face little negative media than when scrutiny is high, which supports H4. A reputational channel offers a consistent explanation: reputation is a critical intangible asset, and negative press threatens legitimacy and can lead to severe consequences, including restructuring or even failure [67]. Under heavier negative coverage, firms are more likely to adopt symbolic actions and cosmetic disclosure to defend their image, which offsets the disciplining role of carbon risk. By contrast, where negative coverage is limited and reputational capital remains intact, firms tend to respond to carbon risk with more credible carbon disclosure and substantive adjustments, reinforcing the reduction in greenwashing.

6. Research Conclusions, Implications and Limitations

6.1. Research Conclusions

This study conceptualizes carbon risk as an external regulatory and reputational signal. Using data on A-share-listed companies in China from 2009 to 2023, we examine how carbon risk affects corporate greenwashing. The findings indicate that: (1) Higher levels of carbon risk are associated with less corporate greenwashing. When firms face greater carbon-related pressures, they tend to strengthen compliance and engage in more substantive environmental disclosure rather than relying on concealment or superficial green claims. (2) Firms with stronger financing constraints and higher internal control quality are less likely to engage in greenwashing under carbon risk. (3) Further analysis reveals that the negative relationship between carbon risk and greenwashing is more pronounced among firms with fewer negative media reports.

6.2. Managerial and Policy Implications

First, governments should recognize the crucial role of carbon regulation in curbing corporate greenwashing. They should therefore continue to firmly advance the dual-carbon goals while improving the guidance and supervision of corporate CSR information disclosure. Enterprises should be urged to disclose truthful and reliable non-financial information to break down information barriers between firms and external stakeholders. Second, regulatory authorities should pay particular attention to high-carbon firms that face relatively low financing constraints but possess weak internal controls, implementing differentiated oversight and stricter disclosure reviews for such firms, as they are more likely to engage in greenwashing when confronted with carbon emission risks. Third, governments should employ media supervision as an important means to regulate corporate information disclosure, yet they must also guide the media to report objectively and fairly. Excessive negative coverage may discourage firms from genuinely engaging in CSR disclosure, potentially driving them toward greenwashing and superficial image management.

6.3. Potential Limitations and Directions for Future Research

While this study offers meaningful insights, several limitations suggest directions for future work: First, the data are drawn primarily from Chinese firms, and China’s carbon-reduction pathways, policy instruments, and ESG disclosure regimes have distinct characteristics; consequently, the findings may have limited generalizability to firms operating under different national institutional settings. Future research should refine and extend these analyses across countries with varying regulatory and policy environments. Second, this study only considers heterogeneity arising from media scrutiny intensity; due to data constraints we were not able to fully examine possible industry-level heterogeneity in the carbon risk–greenwashing relationship—particularly differences between carbon-intensive and non-carbon-intensive sectors. Future scholars should investigate these industry dynamics, which may yield richer and potentially divergent conclusions. Third, carbon risk itself is a complex and multidimensional concept, typically divided into regulatory, physical, and reputational risks. However, existing studies, including ours, have not yet developed a mature and practical framework that can reliably distinguish and quantify these different components. Future research could make valuable progress by constructing dimension-specific, fine-grained measures of carbon risk, which would allow for a more nuanced examination of the heterogeneous drivers and economic consequences associated with each type of risk. Finally, the measurement of greenwashing used in this study has certain shortcomings. On the one hand, it relies on a binary indicator, which cannot capture differences in the severity or intensity of greenwashing across firms. On the other hand, due to potential gaps in regulatory enforcement, some firms that exaggerate their environmental performance but avoid sanctions may still be incorrectly classified as non-greenwashing. Future research could address these issues by developing more rigorous and fine-grained measures of greenwashing that more accurately reflect firms’ true disclosure behavior.