The Impact of Environmental Information Disclosure in the “Carbon Trading Pilot” Project on the Financial Performance of Listed Enterprises in China

Abstract

Environmental policy has long been regarded as the key to achieving sustainable growth goals. Because China is one of the most energy-consuming and carbon-emitting countries globally, its carbon reduction actions have received worldwide attention. This study aims to simultaneously focus on the impact of environmental disclosure and the level of environmental disclosure on enterprise performance. Thus, we use China’s 2013 “Carbon Trading Pilot” policy as an exogenous shock and adopt the DID (difference-in-differences) method to examine the impacts of policy-related disclosure and the disclosure level on the financial performance of listed enterprises from 2009 to 2020. The results are as follows: (1) The “Carbon Trading Pilot” policy-related environmental disclosure negatively affects enterprise financial performance; however, the environmental disclosure level is positively correlated with enterprise financial performance, and both impacts are heterogeneous. (2) The impact of the “Carbon Trading Pilot” project-related environmental disclosure level on enterprise financial performance has a threshold effect, where its impact is enhanced when the environmental disclosure index reaches 10.074. (3) Further exploration of mechanisms reveals that total liabilities play an action mechanism role in the above two relationships. Studying the impact of environmental policies on enterprise financial performance is of paramount significance for economic sustainability.

1. Introduction

The challenges of global sustainable development caused by climate change have long been a concern because of the serious damage to human well-being, economic prosperity, and the natural environment [1,2]. Many initiatives and actions, such as the Paris Agreement, the Global Climate Action Summit, and Sustainable Development Goals, have been proposed to alleviate climate change. The Chinese government has also devoted efforts to address climate change, such as promoting energy conservation, reducing emissions, reducing renewable energy, and committing to achieving peak carbon emissions, to address climate change. Enterprises are an important driving force for China’s economic growth but are also major contributors to environmental pollution. Given their dual nature, enterprises have a unique opportunity and responsibility to contribute to economic sustainability. This involves balancing economic growth with environmental protection and social welfare. Thus, the Chinese government has long attempted to balance the trade-off between the financial performance of and pollution generated by enterprises. The carbon peak and a series of environmental policies, such as the 14th Five-Year Plan for Ecological Protection Supervision Plan, put additional demands on maintaining this balance. Will the proposal and further implementation of environmental policies have a negative impact on enterprise financial performance? How long will this impact last? How should the Chinese government respond to these challenges? Examining this series of issues is beneficial for providing a reference for the Chinese government, as well as the governments of other major polluting countries, to better formulate and implement environmental policies.

In October 2011, the Chinese government officially published a “Carbon Trading Pilot” policy and approved pilot projects on carbon trading in seven provinces or cities, including Beijing, Shanghai, Tianjin, Chongqing, Hubei, Guangdong, and Shenzhen. As a part of this policy, the government issued a certain number of carbon emission quotas to enterprises participating in the carbon trading pilots so that they could buy and sell on the trading market. Enterprises that exceed their allotted emissions are required to purchase additional quotas, and enterprises that have excess quotas can sell them. The goal of the policy is to incentivise enterprises to reduce carbon emissions through the carbon trading market mechanism, achieve low-cost carbon reduction targets, and promote low-carbon economic transformation and economic sustainability. The pilot regional carbon markets were gradually rolled out in the second half of 2013. Although this policy promotes energy conservation and emission reduction in China, it poses certain challenges to the Chinese economy and environment. China’s economic growth largely relies on traditional industries, such as steel, cement, and coal, which are large energy consumers and pollution generators. If this policy is implemented, it may cause enormous losses to the Chinese economy. Moreover, transforming traditional high-polluting industries into green industries requires significant financial and technological support, and SMEs (small and medium-sized enterprises) often face financing difficulties and technological bottlenecks. Therefore, it is necessary to study the impact of policies on enterprise financial performance. This is highly important for maintaining a balance between the two.

Although numerous studies have discussed the possible impacts of environmental policies on enterprise financial performance, most have only focused on the shock of environmental disclosure or the level of environmental disclosure [3,4,5]. Few studies have focused on the effects of both on enterprise financial performance. Additionally, very few studies have explored any threshold effects. In addition, existing research has not paid sufficient attention to the heterogeneity of and mechanism through which environmental policies impact enterprise financial performance. Insufficient research may result in scholars being unable to fully understand the impact of environmental policies on enterprise financial performance, thus failing to provide effective references for urban planners and policy-makers.

On the basis of literature analysis, our aims are the following: (1) simultaneously focus on the impact of environmental disclosure and the level of environmental disclosure on enterprise financial performance, (2) analyse differences in the impacts of environmental disclosure on enterprise financial performance and the level of environmental disclosure on enterprise financial performance, and (3) detect potential mechanisms through which environmental disclosure and its levels affect enterprise financial performance.

To address the existing research, we use China’s “Carbon Trading Pilot” policy as a quasi-natural experiment and a DID method to analyse the impact of environmental disclosure on the financial performance of listed Chinese enterprises. This policy is a mandatory policy. The enterprises that disclosed the policy form the treatment group, while those that did not disclose it form the control group. We also use a panel two-way fixed effect model to identify the impact of the level of environmental disclosure on enterprise financial performance and the threshold effect of this impact. Furthermore, we tested the heterogeneity of this impact and its underlying mechanisms.

The contribution of our paper is that we study the impacts of both environmental disclosure and the level of environmental disclosure on enterprise financial performance and investigate the corresponding heterogeneity and underlying mechanisms. Our research advances the understanding of environmental policy impacts on enterprise financial performance and enriches the empirical literature. Our conclusions can provide a reference for policy-makers and stakeholders in conducting reasonable risk management and strategic deployment during the green transformation process.

The remainder of this paper is organised as follows. In Section 2, we present a literature review to identify gaps in the literature and emphasise our contributions. After introducing the data sources and methods, in Section 4, we analyse the spatial–temporal evolution of the financial performance of listed Chinese enterprises from 2009 to 2020. The empirical results, discussion, and conclusions are presented in Section 5, Section 6 and Section 7, respectively.

2. Literature Review

2.1. Factors Influencing Enterprise Performance

Enterprise performance is an important indicator that is widely considered and valued [6] because it is strongly associated with the market competitiveness, profitability, employee welfare, and social responsibility of an enterprise [7]. Improvements in enterprise performance can increase national economic growth [8], international competitiveness, employment opportunities, and social welfare. To date, many studies have highlighted the multifaceted nature of factors that can positively or negatively influence enterprise performance.

Numerous studies underscore the pivotal role of effective management in bolstering enterprise performance. Management effectiveness includes diverse aspects such as institutional ownership structure [9], managerial conative competencies [10], intellectual capital [11], conducive communication climates [12], optimising supply chain management practises [13], cross-border acquisitions [14], and social responsibility initiatives [15,16,17]. Additionally, fostering innovation behaviour has been identified as a key driver of enterprise performance, as evidenced by research conducted by Kraśnicka et al. [18], Gao et al. [19], Li et al. [20], and Peng and Tao [21].

Several scholars have analysed some negative factors that can influence enterprise performance. Ni et al. [22] found that an increase in the minimum wage reduces organisational performance, including the employment rate and productivity. Brown et al. [23] reported that the human and political capital of home-grown entrepreneurs but not that of returnee migrant entrepreneurs does not. Zhao [24] and Iwasaki et al. [25] reported that state-dominated financial institutions can harm enterprise financial performance; however, the presence of domestic outside and foreign investors as owners contributes to enterprise financial performance.

2.2. Impact of Environmental Policies on Enterprise Performance

Thus far, the impact of environmental policies on enterprise performance is twofold. First, some articles directly explore the influence of environmental policies on enterprise performance. Most of these studies focused only on a certain industry or area. By combining statistical and semistructured interview data, Kagan et al. [26] reported that strict environmental policies resulted in substantial improvements in the environmental performance of pulp manufacturers. Zhang and Wang [3] used samples of 51 A-share listed energy enterprises in Shenzhen and Shanghai from 2015 to 2019 and adopted multiple regression analysis to investigate the positive role of command-and-control policies on energy enterprise financial performance by stimulating technological innovation. However, Zhang and Vigne [27] reported a different opinion in that environmental policies have a “punishment” effect on pollution-intensive enterprise financial performance by expanding financial constraints, especially for state-owned enterprise performance, on the basis of data on firms in Jiangsu Province. Using 194 listed enterprises in the mining industry, Hao et al. [28] reported that environmental policies positively impact short-term enterprise performance but negatively impact long-term enterprise financial performance. Second, the literature has also paid substantial attention to the potential impact of the level of environmental disclosure required by environmental policies on enterprise financial performance [29]. Mohammad and Wasiuzzaman [30] use the ESG disclosure data of 661 listed firms in Bursa Malaysia and find consistent evidence that the level of environmental disclosure improves enterprise financial performance and is moderated by a firm’s competitive advantage. Similarly, Carnini Pulino et al. [31] suggest that enterprises that disclose more comprehensive information about their environmental activities are likely to achieve greater financial performance. Nevertheless, Duque-Grisales and Aguilera-Caracuel [32] and Saygili et al. [33] demonstrate a negative effect of the level of environmental disclosure on enterprise financial performance.

The action mechanism by which environmental policies affect enterprise financial performance may include the following pathways. Some studies have explored the positive influence of environmental policies on enterprise green technology innovation [34,35], the regulation of enterprises’ environmental behaviour [4,5], future market benefits [36,37], and customer recognition, which may increase enterprise financial performance. On the other hand, a few scholars have reported signs of a negative association between environmental policies and enterprise financial performance. Thus, they argue that environmental policies can significantly reduce enterprise innovation [38]; they can also cause environmental disclosure costs to decrease and cause immediate cash flow [32] and employees to decrease [7]; additionally, the export duration can be shortened for highly polluting enterprises [39], hindering their financial performance.

2.3. Gaps in the Literature

Four gaps exist in the literature. First, while some articles have directly studied the impact of environmental policies on enterprise financial performance, none have explored the impact of environmental policies on all types of enterprises at the national level. In fact, a comprehensive analysis at the industry or country level is important. Additionally, the conclusions of these articles are inconsistent, making it difficult to reach a consensus. Second, the literature has paid less attention to heterogeneity. Research can help better identify heterogeneity in the influence of different industries and enterprise characteristics, thus providing a theoretical foundation for industry management and policy formulation. Third, to our knowledge, only one empirical study has examined the mechanism by which environmental policies affect enterprise financial performance. Finally, the majority of scholarly research has emphasised only the impact of the level of policy environmental disclosure on enterprise financial performance, and the threshold effect remains difficult to observe and analyse.

The “Carbon Trading Pilot” policy may initially reduce enterprise financial performance by increasing costs and exposing potential problems, but as the level of environmental disclosure increases, there are advantages for firms in terms of stimulating technological innovation [35,40], regulating environmental behaviour [4,5], and improving managerial competence [41]. We therefore propose our first hypothesis.

H1: 

The “Carbon Trading Pilot” policy disclosure has a negative effect on enterprise financial performance, but the disclosure level improves enterprise financial performance.

Many articles state that the impact of policy on enterprises may vary depending on specific characteristics [42,43]. While some enterprises may benefit financially from increased transparency regarding their carbon trading activities, others may not experience the same positive effects or may even face negative consequences. Different types of enterprises may affect an enterprise’s operational strategy and financial performance in multiple dimensions, leading to inconsistent results. Specifically, industrial enterprises often face greater environmental pressures [44] and experience greater financial burdens as a result of increased compliance costs and penalties for exceeding the emission limit. Enterprises with post-2003 listings may be fragile because they have experienced a relatively shorter period of market exposure and maturity. For enterprises, a relatively long history in the public markets can help them become more mature and stable [45]. Enterprises with less than four major committees may face financial vulnerabilities because of ineffectiveness when it comes to internal decisions [46]. Enterprises with high ownership concentration may lack work enthusiasm and innovation due to an over-controlling working environment [47], leading to decision-making bias when facing shock. Thus, when analysing the impacts of the “Carbon Trading Pilot” policy on enterprise financial performance, it is necessary to consider the differences in the characteristics of these enterprises. The second hypothesis can be stated as follows:

H2: 

The impacts of the “Carbon Trading Pilot” policy disclosure and its disclosure level on enterprise financial performance are heterogeneous.

Total liabilities, including short-term liabilities and long-term liabilities, are key indicators for measuring an enterprise’s liabilities. An enterprise’s total liabilities have a significant effect on its financial performance. A moderate liability level can increase shareholder returns through leverage, but a high level of liability can increase financial risk, the interest burden, and profit volatility and limit an enterprise’s investment ability and growth potential. The sudden impact of the “Carbon Trading Pilot” policy environment disclosure can exacerbate enterprise costs and result in higher levels of liability, thereby dampening financial performance. However, as the level of disclosure increases, the high level of liability faced by enterprises gradually eases, and the leverage effect can also increase shareholder returns and enhance enterprise financial performance [48]. Thus, we propose a third hypothesis for this purpose.

H3: 

In the early stage of the “Carbon Trading Pilot” policy regarding environmental disclosure, high total liabilities hinder financial performance; as the disclosure level increases, the total liabilities are alleviated, which is beneficial for enterprise financial performance.

3. Method and Data

3.1. Method

3.1.1. DID Method

The “Carbon Trading Pilot” policy is a useful exogenous shock for our research; it is conducive to overcoming endogeneity and spurious relationship problems and is better at identifying causality. The basic DID model is as follows:

$${EFP}_{it}={\beta }_0 +{\beta }_1{ DID}_{it} +{\beta }_2{\mathit{X}}_{\mathit{i}\mathit{t}}+{\epsilon }_{it}$$

(1)

where ${EFP}_{it}$ represents the financial performance of listed enterprises $i$ in year $t$.${ DID}_{it}$ is the core explanatory variable of the “Carbon Trading Pilot” policy shock. The sample listed enterprises are categorised into treatment and control groups on the basis of whether they are influenced by the policy. We are interested in the coefficient ${\beta }_1$, which captures the actual net impact of the pilot policy on enterprise financial performance. The constant ${\beta }_0$ represents the intercept term, ${\mathit{X}}_{\mathit{i}\mathit{t}}$ is a series of control variables, ${\beta }_2$ is the correlation coefficient capturing the impact of control vector ${\mathit{X}}_{\mathit{i}\mathit{t}}$ on enterprise financial performance, and ${\epsilon }_{it}$ is the random disturbance term.

3.1.2. Panel Two-Way Fixed Effect Model

The environmental disclosure required by the “Carbon Trading Pilot” policy provides an opportunity for studying the implementation phase of the pilot projects. In general, greater environmental disclosure reduces management costs, as enterprises voluntarily follow an external set of measured objectives [49] that may be positively correlated with better implementation of environmental policies. Next, we use a panel two-way fixed effect model to measure the impact of the level of environmental disclosure required by the “Carbon Trading Pilot” policy on enterprise financial performance. The specific formula is as follows:

$${EFP}_{it }={\alpha EDL}_{it}+\mathsf{\beta }{\mathit{X}}_{it} +{\theta }_i +{ \gamma }_t +{ \epsilon }_{it}$$

(2)

where ${EFP}_{it}$ is the financial performance of enterprise i in year $t$, $\alpha$ is the coefficient of the explanatory variables reflecting the impact of the “Carbon Trading Pilot” policy environmental disclosure level on enterprise financial performance, and β is the coefficient of the control vector that reflects the impact of a nonexperimental factor on enterprise financial performance. ${\theta }_i$ and ${\gamma }_t$ measure individual and time fixed effects, respectively, whereas ${\epsilon }_{it}$ is the remainder disturbance for enterprise i in year $t$. ${\mathit{X}}_{\mathit{i}\mathit{t}}$ is a series of control variables, and ${\beta }_2$ is the correlation coefficient capturing the impact of control vector ${\mathit{X}}_{\mathit{i}\mathit{t}}$ on enterprise financial performance. ${\epsilon }_{it}$ is the random disturbance term.

3.1.3. Threshold Effect Model

Many economic phenomena are not simply linear relationships. For example, the pilot policy may be effective in achieving its intended goals under certain conditions. However, if those conditions change, the policy’s effectiveness may diminish or even reverse.

Considering that traditional linear models may not be able to fully explain complex data relationships, we adopt a threshold effect model. The threshold effect refers to the phenomenon of a change in the direction or quantity of an economic parameter caused by another economic parameter reaching a certain threshold value [50]. The threshold effect model can better capture complex features in data and improve the explanatory power of the model by introducing threshold variables [51]. We introduce threshold variables based on Equation (2) and obtain Formula (3) for the threshold effect model.

$${EFP}_{it}={\theta }_i+{\alpha }_{1}EDL\left(EDL\le \tau \right)+{\alpha }_{2}EDL\left(EDL>\tau \right)+\mathsf{\beta }{\mathit{X}}_{it}+r_t+{\epsilon }_{it}$$

(3)

where $\tau$ is the threshold value that indicates that when the “Carbon Trading Pilot” policy environmental disclosure level reaches τ, the direction or intensity of its impact on enterprise financial performance (${EFP}_{it}$) changes. ${\alpha }_1$ and ${\alpha }_2$ are the action coefficients of the policy environmental disclosure level on enterprise financial performance when the threshold value has not been reached or has been exceeded, respectively.

3.2. Variables and Data Sources

The dependent variable in our paper is enterprise financial performance, which is measured via the total comprehensive income of the enterprise. Total comprehensive income represents the details of all recognised revenues and expenses of the enterprise and is viewed as a critical measure of overall company performance [52,53]. Gazzola and Amelio [54] further state that compared with net income, total income provides more information for the evaluation of enterprise financial performance. The independent variables include the “Carbon Trading Pilot” policy environmental disclosure and its disclosure level. The former is measured via policy shocks, and the latter is measured via the environmental disclosure index in the official environmental, social, and governance (ESG) database.

The listed enterprise data are from the China Stock Market Accounting Research Database (CSMAR) and Huazheng. There were a total of 4140 listed enterprises in the A-share market in 2020. The industry distribution of listed enterprises is diverse and is concentrated in multiple fields, such as information technology services, manufacturing, machinery, and equipment.

Table 1. Descriptive statistics of variables.

	Variables	Definition	Units	Obs	Mean	Std. Dev.	Min	Max
CTP	Carbon trading pilot	/	/	53,958	0.330	0.470	0	1
EFP	Enterprise financial performance	Comprehensive performance	CNY 1,000,000	36,079	927.662	8694.050	−79,661.422	326,532
NI	Net income	/	CNY 1,000,000	53,277	701.201	7304.075	−68,742.562	317,685
OP	Operating profit	/	CNY 1,000,000	53,562	869.204	9173.461	−71,550.906	391,382
ES	Enterprise size	The natural log of the total assets owned	/	53,565	21.787	1.506	10.842	31.138
NSM	Number of senior managers	/	People	47,990	6.271	2.415	0	64
IAR	Intangible assets ratio	The proportion of intangible assets to total assets	%	51,254	0.044	0.063	−0.033	0.938
GAE	G&A expense	General and administration expense	CNY 100,000,000	52,688	3.177	19.206	−3.556	905.640
AIL	Asset impairment loss	The loss caused by the recoverable amount of assets being lower than its book value	CNY 1,000,000	40,159	144.502	2896.249	−202,668	161,594
OE	Operating expense	/	CNY 1,000,000	52,479	5758.179	46,697.567	−25.347	2488,852
TCA	Total current assets	/	CNY 1,000,000	52,955	5571.974	32,767.224	0	1,577,630.100
LLR	Long-term liability rate	The proportion of noncurrent liabilities to total assets	%	48,284	0.074	0.109	−0.123	6.830
FE	Financial expense	/	CNY 1,000,000	52,734	102.624	548.791	−4845.546	27,816
TL	Total liabilities	/	CNY 1,000,000	53,568	35,368.853	586,614.740	−2.033	30,435,544
CPDR	Cash paid for debt repayment	/	CNY 1,000,000	47,445	3957.958	35,623.760	−235,175	1,512,900
EDL	Environmental disclosure level in the ESG index system	/	/	9325	8.867	7.909	0	54.264
NE	Number of employees	/	People	31,204	6300.767	24,313.67	2	552,810
TAR	Tangible asset ratio	Net fixed assets plus net inventory assets divided by total assets	%	31,345	0.353	0.183	0	0.971
NPCC	Net profit cash cover	Actual income situation of an enterprise	/	27,808	1.871	34.528	−1640.051	4887.906
ALR	Asset–liability ratio	The proportion of total asset to total liability	%	31,231	0.454	1.146	−0.195	178.345
TCD	Total compensation of directors	/	CNY 1,000,000	31,176	2.527	3.227	0	82.755

Note: Because the GAE coefficient in the later regression is very small, we use units of CNY one hundred million instead of CNY one million.

shows the statistical descriptions of relevant enterprise variables for all listed enterprises from 2009 to 2020, which are used to analyse the impact of environmental policy disclosure and the disclosure level on enterprise financial performance.

4. Spatial–Temporal Evolution

4.1. Industrial and Nonindustrial Categories

Industrial enterprise financial performance declined until 2015 and then gradually improved (Figure 1). In contrast, the financial performance of nonindustrial enterprises exhibited a fluctuating but generally upwards trend. Industrial enterprises are usually resource dependent and environmentally unfriendly. The “Carbon Trading Pilot” policy may include reducing pollutant emissions, improving energy efficiency, and limiting resource use, which directly affect the operation and production methods of industrial enterprises and reduce employment and productivity. Therefore, industrial enterprises initially experience a decrease in financial performance. In the long run, because industrial enterprises inevitably change their production methods to adapt to new rules, their financial performance gradually recovered after 2015.

Spatially, nonindustrial enterprise financial performance increased slightly in 2020 (Figure 2), whereas industrial enterprise financial performance mildly diminished. Specifically, the financial performance of industrial enterprises in Beijing was quite high in 2009, whereas that of those in other provinces remained relatively low. This may have occurred because, at that time, the Chinese government adopted a large-scale economic stimulus plan to manage the pressure of an economic downturn to minimise the negative impact of the 2008 global financial crisis. These stimulus measures include tax cuts, infrastructure investments, and expanded credit, which provided a favourable development environment for the manufacturing industry, especially in Beijing, where implementation is more effective under the central government. In 2020, nonindustrial enterprises in eastern China exhibited the most effective financial performance. Over the past decade, environmental issues have gradually become the main factor restricting sustainable economic development. Thus, the Chinese government has enacted strict environmental policies such as the “Carbon Trading Pilot” policy, which abates industrial enterprise pollution emissions and activities [39] and encourages the advancement of nonindustrial enterprises. Eastern China has relatively advanced technology and high financial support [55], which effectively promote the healthy development of nonindustrial enterprises.

4.2. Listing Date

We divide the sample into enterprises listed after 2003 and those listed before 2003. Figure 3 shows that enterprises listed after 2003 exhibited a roughly upward trend in financial performance from 2009 to 2014; however, after 2014, they experienced a sharp decline lasting for three years. Moreover, enterprises listed before 2003 developed steadily prior to 2019 because of the long listing time. Therefore, enterprises became more familiar with the fluctuations and risks of the capital market and adapted to them by establishing a sound risk management system and investor relationships, enabling them to better respond to future market fluctuations and risks.

Figure 4 shows that only enterprises listed after 2003 in Beijing performed particularly well in 2009. This is perhaps because there are many policies to attract investments in Beijing, bringing in eligible, newly listed enterprises and more financial support for development. In 2020, more enterprises in Beijing, Guangdong, and Fujian showed improved financial performance, including both those listed after 2003, as well as those listed before 2003. In short, while both types of enterprises increased between 2009 and 2020, the increasing trend for enterprises listed before 2003 was greater. From a long-term perspective, enterprises that have been listed for more years have greater capital accumulation and greater development potential, especially in East China, which has a good economic foundation, and in Guizhou, which has long-established brands such as Moutai.

4.3. More or Less than Four Major Committees

Considering that having more than four major committees within an enterprise may lead to superior management and decision-making structure and thus can contribute to better performance, we compare the temporal evolution characteristics of enterprises with more than versus less than four major committees (Figure 5). Both types of enterprises show a roughly upwards trend in performance. However, enterprises with more than four major committees have smaller fluctuations in financial performance and are more stable. This is perhaps because these enterprises usually have more standardised internal evaluation, decision-making, and management systems; thus, they can improve their investment [56,57] and financial performance.

Spatially, the financial performance of enterprises nationwide is generally unsatisfactory (Figure 6). Compared with enterprises with fewer than four major committees, those with more than four major committees experienced more obvious financial performance growth in multiple provinces across the country in 2020. This finding proves that having more than four major committees within an enterprise is likely related to mature internal management that can better respond to external competition and enhance financial performance. We also find that enterprises with more than four major committees have driven financial performance in western China, especially in Yunnan, Chongqing, Guizhou, and Shaanxi. This is perhaps because these enterprises are capable of responding to the inferior and inefficient market environment in western China.

4.4. Ownership Structure

Previous studies have confirmed that ownership structure has an effect on enterprise financial performance [25]. We further divide the sample enterprises into two types: enterprises with higher and lower ownership concentrations. Enterprises with higher ownership concentration clearly tended to improve their financial performance from 2009 to 2014, with a downwards fluctuation in financial performance in the following three years (Figure 7). Before 2019, enterprises with lower ownership concentrations experienced a fluctuating increase in their financial performance. Under diversification, enterprises with a higher ownership concentration are typically characterised as having inadequately protected minority shareholders’ rights [58], ultimately influencing overall stability and financial performance.

The financial performance of both types of enterprises has increased to varying degrees (Figure 8). However, the number of enterprises with higher ownership concentrations in 2020 was significantly lower than that in 2009, possibly because of delisting and bankruptcy. Concentrated ownership, especially for underdeveloped enterprises in emerging markets, may increase conflicts of interest between controlling and minority shareholders or inefficient activities such as maximising market share or engaging in technological leadership [59]. Therefore, concentrated ownership could cause financial performance to decline and potentially lead to bankruptcy.

5. Results and Analysis

5.1. Parallel Trend Test

Before conducting the DID regression, we performed parallel trend testing on the data to ensure the effectiveness and accuracy of the method. Figure 9 shows the results of the parallel trend test, where 2013 is the policy shock year. There is no significant difference in the fluctuating trends in enterprise financial performance between the treatment and control groups before 2013; however, after the implementation of the pilot policy, the enterprise financial performance of the processing group is significantly affected. The results indicate that the “Carbon Trading Pilot” policy can be viewed as an exogenous shock that affects enterprise financial performance.

5.2. Impact of the “Carbon Trading Pilot” Policy Environmental Disclosure and Its Environmental Disclosure Level on Enterprise Financial Performance

The coefficient of ${ DID}_{it}$ in Model 1 in

Table 2. Impact of the “Carbon Trading Pilot” environmental disclosure policy and its disclosure level on enterprise financial performance.

	DID			The Panel Two-Way Fixed Effect Model			IV
Variables	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7
CTP × Post	−21.998 ** (10.179)	−15.097 ** (6.410)	−16.631 ** (8.391)
EDL				0.106 *** (0.034)	0.111 *** (0.034)	0.095 ** (0.038)	0.197 *** (0.056)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Province-fixed				Yes	Yes	Yes	Yes
Time-fixed				Yes	Yes	Yes	Yes
Constant				6.951	−72.636	−51.339
N	28,650	33,123	33,123	5618	5709	5535	4881
R2	0.670	0.680	0.690	0.388	0.454	0.355
KP Wald F							567.624

Note: Robust standard errors are shown in parentheses; *** and ** represent p < 0.01 and p < 0.05, respectively.

is negatively significant, meaning that the “Carbon Trading Pilot” policy inhibits improvements in enterprise financial performance. This finding is consistent with that of the study by Zhang and Wang [3]. Intensive and stringent environmental policies tend to harm employment and production activities [60], curb the expansion of enterprises’ scale, and thus negatively affect enterprises’ financial performance. Considering that some scholars have used net income [61] and operating profit to represent enterprise financial performance, we further introduce the indicators of enterprise net income (Model 2) and operating profit (Model 3) to replace total income. The results show that enterprise financial performance is negatively affected by the pilot policy shock, whether measured in terms of total income, net income, or operating profit. Institutional theory can provide a more comprehensive understanding of the potential impact of the “Carbon Trading Pilot” policy on financial performance. When facing environmental policies, enterprises face pressure from regulations and social norms [62]. This institutional pressure makes enterprises want to transform. However, in the short term, enterprises may not have adjusted their operational strategies or fully utilised policy support.

The panel two-way fixed effect model regression results are reported for Models 4–6 in Table 2. After the “Carbon Trading Pilot” policy is enacted, the benchmark regression (Model 4) shows that enterprises that disclose more environmental information have better financial performance. This is because enterprises with high levels of environmental disclosure can reduce downside risk [63] and increase the short-term cumulative returns of stocks [64] and other assets, thereby facilitating improved overall enterprise financial performance. Correspondingly, to verify the results, we also use indicators of enterprises’ net income and operating profit in place of comprehensive income (Models 5–6). Stakeholder theory suggests that managers, employees, and others responsible for making decisions for the enterprise need to make decisions that balance ethics and performance [65], which will effectively enhance their environmental reputation. This helps them maintain greater resilience in the face of potential legal risks and social pressures.

The DID method estimates the net effect of policies by comparing the differences between the treatment group and the control group before and after policy implementation. The parallel trend hypothesis holds, which can alleviate endogeneity issues to some extent. Unlike the DID method, the panel two-way fixed effects model may have obvious endogeneity issues. Thus, we add IV (instrumental variable) methods and use a lagged one period approach in the panel two-way fixed effects model (see Model 7). The results of Model 7 are still significant, indicating that even after endogeneity and other potential confounding factors are considered, the net impact of the policy on the treatment group is still significant. Additionally, the result of the KP Wald F statistic is greater than the corresponding critical value, and the null hypothesis of “weak instrumental variables” can be rejected. This is a positive result that supports the conclusion about policy effectiveness.

Given the inclusion of significant economic events such as the COVID-19 pandemic in the original study period (2009–2020), it was a prudent decision to exclude data from 2020 to minimise the potential distortion of results caused by this exceptional event. Conducting benchmark, robust, and heterogeneous regression analyses again, excluding the 2020 data, helped to ensure the validity and reliability of the findings. The fact that the regression results remained significant despite excluding the data from 2020 indicates that the relationships identified in the study are robust and not solely driven by the COVID-19 pandemic. This suggests that the findings have broader applicability and can potentially provide valuable insights into the economic dynamics and trends in the period under investigation, without the confounding influence of the pandemic. Owing to space limitations, we do not display all the regression results here.

5.3. Robustness Test

We conducted a series of tests to ensure the robustness of our results, including a placebo test, lagging the policy implementation time, and a two-sided tail shrinking test.

First, to validate the main effect of the “Carbon Trading Pilot” policy on enterprise financial performance, we referred to the study of La Ferrara et al. [66] and initially adopted a placebo test. We first constructed pseudo-treatment and control groups by randomly selecting enterprises from the samples, assuming that they were influenced by the policy. To improve the accuracy of the placebo trial identification, we repeated random sampling 1350 times. Figure 10 shows the probability density distribution of the estimated coefficient of policy on enterprise financial performance on the basis of random samples. The estimated coefficients are normally distributed at approximately zero, indicating that the real policy impact is significantly different from the placebo test results and that other random factors do not interfere with the results.

Second, we re-estimated the influence of the “Carbon Trading Pilot” policy on enterprise financial performance by assuming that the pilot policy changed in 2017, preventing our result from being influenced by other policies (Model 1 of

Table 3. Robustness test.

Variables	Model 1	Model 2	Model 3	Model 4	Model 5
CTP × Post	−101.533 (66.040)	−6.859 * (4.038)	−4.820 *** (1.568)
EDL				0.101 *** (0.034)	0.073 ** (0.033)
Control variables	Yes	Yes	Yes	Yes	Yes
Constant				1.215	9.859
Province-fixed			Yes		Yes
Time-fixed			Yes		Yes
N	28,650	28,650	28,650	5618	5618
R2	0.670	0.570	0.490	0.383	0.369

Note: Robust standard errors are shown in parentheses; ***, **, and * represent p < 0.01, p < 0.05, and p < 0.1, respectively.

).

Third, for the impact of the “Carbon Trading Pilot” policy on environmental disclosure and the disclosure level on enterprise financial performance, we processed the sample data by truncating the tail by 1% (Models 2 and 4) and 5% (Models 3 and 5), respectively, to reduce the impact of the sample outliers on the baseline regression result.

5.4. Heterogeneity Analysis and Threshold Effect

We conducted a heterogeneity analysis of different enterprises and threshold effects, and the results are shown in

Table 4. Heterogeneity analysis and threshold effect.

Variables	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7
CTP × Post	−36.483 *** (12.184)	−34.479 *** (13.018)	−97.266 ** (48.787)	−63.278 *** (20.858)
EDL					0.143 *** (0.043)	0.112 *** (0.037)
EDL (EDL < 10.074)							0.020 (0.053)
ED (EDL > 10.074)							0.100 ** (0.040)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Heterogeneity description	Industrial enterprises	Enterprises listed after 2003	Enterprises with fewer than four major committees	Enterprises with higher ownership concentrations	Enterprises listed before 2003	Enterprises with more than four major committees
Constant					62.863	16.794	0.939
Province-fixed					Yes	Yes
Time-fixed					Yes	Yes
N	19,042	16,331	2861	12,646	3192	4981	1210
R2	0.710	0.700	0.790	0.710	0.385	0.394	0.407

Note: Robust standard errors are shown in parentheses; *** and ** represent p < 0.01 and p < 0.05, respectively.

. Owing to space limitations in the table, we only display the significant heterogeneity results.

First, we examined the heterogeneity between the “Carbon Trading Pilot” policy environmental disclosure and enterprise financial performance. The results of Model 1 demonstrate that the policy impact focuses on industrial and not nonindustrial enterprises. This is because most industrial enterprises are usually carbon intensive, resulting in many stranded assets, such as coal mines, oilfields, and gas lines [44], after pilot policy shocks. Therefore, costs increase, and enterprise financial performance deteriorates. A plausible reason is that industrial enterprises need more funds to manage pollutant emissions, resulting in increased costs and decreased financial performance. The results of Model 2 show that the pilot policy has a significant negative effect on enterprise financial performance for those firms listed after 2003, whereas enterprises listed before 2003 are almost unaffected. A plausible reason is that enterprises that have been listed for a long time are more mature [45] and have stronger risk-resistance capabilities. Moreover, they can promptly incorporate policy and the market environment considerations into their strategic decisions, thereby choosing more environmentally friendly business strategies. This strategy can facilitate competitive advantages in product differentiation and enhance enterprise financial performance [67].

Model 3 illustrates that environmental disclosure under the “Carbon Trading Pilot” policy has a negative effect on enterprises with fewer than four major committees but not on those with more than four committees. Recent research has shown that four major committees are helpful for policy-makers’ and researchers’ assessments and can positively affect enterprises’ return on equity, net profit margin, and investment decisions [46,56,57]. Therefore, enterprises with more than four major committees can suppress the side effects of policies, whereas those with fewer than four major committees fail to effectively manage the side effects of such policies. Model 4 shows that environmental disclosure under the “Carbon Trading Pilot” policy causes substantial damage to the financial performance of enterprises with higher but not lower ownership concentrations. In a high ownership concentration context, controlling shareholders tend to be overzealous in their monitoring and ratification roles over subordinates, causing work enthusiasm and innovation to decline [47,68]. Therefore, when policy shocks occur, enterprises with higher ownership concentrations may be unable to absorb diversification suggestions because of a sense of oppression by controlling shareholders, possibly creating decision-making errors and affecting financial performance.

In addition, we conducted the same heterogeneity test as above to examine the heterogeneity of the impact of the level of environmental disclosure under the “Carbon Trading Pilot” policy on enterprise financial performance. Compared with enterprises listed after 2003, those listed before 2003 (Model 5) may have had more time to develop and mature. Therefore, a higher environmental disclosure level has a significant positive effect on their financial performance. Additionally, having more than four major committees (Model 6) has a positive moderating effect on the impact of the “Carbon Trading Pilot” policy environmental disclosure on enterprise financial performance, in contrast to enterprises with fewer than four major committees. Having four major committees, such as broad nominating committees and audit committees, may play a key role in ensuring adequate governance, which may contribute to enterprise financial performance [69,70]. However, we could not find evidence that the effect of the “Carbon Trading Pilot” policy environmental disclosure level on enterprise financial performance differs between industrial and nonindustrial enterprises and between different ownership concentration levels.

The threshold effect results are reported via Model 7 and indicate a single threshold for the relationship between the “Carbon Trading Pilot” policy environmental disclosure level and enterprise financial performance. When the disclosure level is less than 10.074, the two are not correlated; when the disclosure level is greater than 10.074, for every single point increase in the environmental disclosure level, enterprise financial performance correspondingly increases by 0.1 point. This finding indicates that only after a certain level of environmental disclosure does it have a positive effect on enterprise financial performance by strengthening environmental awareness and promoting green transformation and development.

5.5. Mechanism Analysis

To clarify the mechanism through which the “Carbon Trading Pilot” policy and disclosure level affect enterprise financial performance, we further explored this mechanism. The regression results are shown in

Table 5. Mechanism analysis.

Variables	Model 1	Model 2	Model 3	Model 4
CTP×Post	67.571 ** (33.753)
EDL		−0.058 * (0.030)		0.040 ** (0.016)
TL			0.770 ** (0.028)
Control variables	Yes	Yes	Yes	Yes
Constant		−822.493	68.912	−130.205
N	33,123	28,650	21,028	5709
R2	0.930	0.234	0.486	0.822

Note: Robust standard errors are shown in parentheses; ** and * represent p < 0.05 and p < 0.1, respectively.

.

Model 1 shows that the environmental disclosure requirement under the “Carbon Trading Pilot” policy increases total liabilities, and Model 2 indicates that greater total liabilities can hinder enterprise financial performance. Total liabilities act as an action mechanism between the pilot disclosure policy and enterprise financial performance. From a short-term perspective, enterprises need to invest more funds in environmental facility upgrades, research and technological innovation, compliance testing and certification, and information disclosure expenses in order to meet stricter environmental standards and disclosure requirements. These additional investments and expenses directly lead to an increase in the total liabilities of the enterprise. When the total liabilities of an enterprise sharply increase in the short term, the conflict of interest between shareholders and management often intensifies, leading to a significant increase in agency costs [71] and a decrease in financial performance. In the short term, excessive total liabilities may also pose a greater risk of loss of control for the enterprise [71], as it is unable to effectively manage its debt and operational activities, which can have a negative impact on financial performance.

Models 3–4 show that a higher “Carbon Trading Pilot” policy disclosure level increases total liabilities, which contributes to enterprise financial performance. From a long-term perspective, enterprises that disclose more environmental information are more likely to obtain bank loans [72]. This is because banks consider an enterprise environmental performance and social responsibility performance when assessing loan risk. Enterprises that actively disclose environmental information and demonstrate good environmental management capabilities are often considered to have lower default risk and more stable development prospects, making it easier to obtain bank loans. Liabilities can positively enhance enterprise financial performance [48]. The premise of this viewpoint is that enterprises can provide appropriate financing for their liabilities and share the debt repayment pressure over a longer period. In this case, liabilities can be viewed as a financial leverage that helps businesses expand their business scale, increase market share, or engage in technological innovation. When the returns from these investments exceed the liability cost, the overall financial performance of the enterprise will improve.

6. Discussion

Our verification confirms H1. That is, both the “Carbon Trading Pilot” policy of environmental disclosure and its disclosure level impact enterprise financial performance. The “Carbon Trading Pilot” policy may cause short-term shocks and damage enterprise financial performance; however, as the level of environmental disclosure increases, enterprise financial performance improves. The initial shock and potential damage to financial performance caused by the policy’s introduction highlight the need for careful planning and adaptation by enterprises. Our research not only validates the conclusion of Zhang and Vigne [27] regarding the potential negative impact of environmental disclosure on enterprise financial performance but also provides a more comprehensive and insightful perspective by expanding the scope to include listed companies nationwide. We also delved into the positive effects of environmental disclosure levels on enterprise financial performance, which serves as a valuable addition to existing research.

Additionally, there is greater heterogeneity in the impact of environmental policy disclosure on enterprise financial performance, whereas improvements in the disclosure level are heterogeneous for only the two types of enterprises. This observation validates H2, highlighting the varied responses of different enterprises regarding increased financial performance. Specifically, enterprises that are engaged in industry, are listed after 2003, have fewer than four major committees, and have high ownership concentration tend to suffer more from the “Carbon Trading Pilot” policy, which hurts their enterprise financial performance. In comparison, we find that for enterprises listed before 2003, having at least four committees plays a positive regulatory role at the environmental disclosure level and in enterprise financial performance. Additionally, there is a significant positive effect on enterprise financial performance when an enterprise’s environmental disclosure level is greater than 10.074. This threshold may represent a tipping point where the benefits of transparency and sustainability outweigh the initial costs and disruptions associated with the policy.

Furthermore, our findings corroborate H3, emphasising the pivotal role of liabilities as a key mechanism in mediating the effects of the pilot policy environmental disclosure and its disclosure level on enterprise financial performance. This result supplements the existing literature on the underlying mechanisms, which has focused on factors such as enterprise green technology innovation [34,35] and enterprises’ environmental behaviour [4,5].

Our research findings have certain policy implications for policy-makers. First, considering the impact of the “Carbon Trading Pilot” policy on enterprise financial performance, the Chinese government is expected to provide more financial assistance for the development of new carbon intensive asset decarbonisation technologies and create favourable conditions for enterprises, especially more vulnerable industrial or underdeveloped enterprises. This financial assistance can take the form of tax incentives, grants, or low interest loans to encourage private sector investment in green technologies. Second, the heterogeneity of enterprises indicates that more targeted policies should be formulated and implemented to reduce the impact of policies on enterprises. For example, valuable insights and feedback can be provided through negotiation and dialogue with businesses, especially those industrial enterprises that may be most affected. Third, by keeping liabilities at a manageable level, an enterprise can maintain a healthier financial position, with more flexibility to invest in growth initiatives, respond to market changes, and weather economic downturns. This, in turn, can lead to improved financial performance and stronger long-term prospects for the enterprise.

This study has several limitations. First, we analysed only enterprise financial performance from 2009 to 2020 because of limited data. A longer period would allow for a more complete relationship between the “Carbon Trading Pilot” policy and enterprise financial performance to be observed. Second, on the basis of the data currently available to us, we can use only total comprehensive income as a measure of financial performance. However, the use of total comprehensive income as the sole measure of financial performance has its limitations. The inclusion of additional metrics would provide a more comprehensive analysis. Third, our sample is listed enterprises, but SMEs not only play an important role in economic growth and employment but also play a key role in promoting innovation and social development. Therefore, in policy-making and economic research, it is necessary to fully consider their characteristics and needs. Fourth, our sample is distributed more in eastern China, where the economy is more developed, whereas the economy in the western China has relatively lagged behind. Excessive attention to eastern China may overlook the development needs of western China, thereby exacerbating regional economic disparities. Finally, we examined the mechanism of action only in terms of liabilities. In fact, there is more than one mechanism through which environmental policies affect enterprise financial performance, which is worth studying in the future.

7. Conclusions

Environmental policy impacts have received increasing attention in recent decades. We first identified the impact of the “Carbon Trading Pilot” policy environmental disclosure and its disclosure level on enterprise financial performance via DID and a panel two-way fixed effects model, respectively. We then examined the heterogeneity of these impacts and the underlying mechanisms. Considering that threshold values may exist for the environmental disclosure level, we also conducted a threshold effect analysis. The main findings are as follows.

Different types of enterprises exhibit distinct spatial–temporal evolution characteristics. First, owing to environmentally friendly production methods and industrial structures, both nonindustrial enterprises and industrial enterprises, especially those in East China, have exhibited increasing financial performance. Second, enterprises listed before 2003 seem more stable and efficient because they have established mature risk management systems and investor relationships. Third, enterprises with more than four major committees exhibit stable growth trends in financial performance around the country, which may be attributed to superior internal evaluation, decision-making, and management systems. Finally, enterprises with lower ownership concentrations exhibit more efficient financial performance because this type of enterprise can represent a diverse environment and can protect minority shareholders’ rights, thereby inhibiting creativity and financial performance.

The “Carbon Trading Pilot” policy initially caused short-term disruptions but ultimately improved financial performance as the level of environmental disclosure increased. Heterogeneity was evident in policy impacts, with industrial enterprises, post-2003 listings, enterprises with fewer than four committees, and enterprises with high ownership concentrations suffering more initially. Conversely, pre-2003 listings and firms with at least four committees benefited from enhanced disclosure levels, reinforcing positive financial outcomes. A threshold effect emerged, with financial performance significantly increasing when the disclosure level exceeded 10.074. We further identified total liabilities as an action mechanism through which the pilot policy’s disclosure and its level influence enterprise financial performance.