Corporate Digital Transformation and Environmental Accounting Information Disclosure: A Dual Examination of Internal Empowerment and External Monitoring

Abstract

Environmental accounting information disclosure is crucial for heavily polluting enterprises to strengthen environmental governance and realize sustainable development. However, some enterprises still suffer from weak disclosure awareness and low disclosure quality. Therefore, improving the quality of environmental accounting information disclosure in the digital era has become an urgent task to achieve China’s goal of a green and low-carbon economy. Using data from Shanghai and Shenzhen A-share listed companies in China’s polluting industries from 2013 to 2022, this study explores the impact and channels of influence of digital transformation and environmental accounting information disclosure. It has been found that digital transformation significantly impacts the quality of environmental accounting information disclosure. Further, based on the dual perspectives of internal empowerment and external monitoring, digital transformation improves environmental accounting information disclosure by promoting executive compensation incentives and enhancing analyst attention. Furthermore, the positive impact of digital transformation on environmental accounting information disclosure is more pronounced with the implementation of new environmental protection laws, high-quality audits and a high level of digital transformation, and non-state-owned enterprises. The findings provide theoretical support for the government to improve the environmental accounting information disclosure system and provide valuable policy insights to promote digitalization and green, low-carbon transformation paths for heavily polluting enterprises.

1. Introduction

The Opinions on Accelerating Comprehensive Green Transformation of Economic and Social Development, issued by the State Council of the Central Committee of the Communist Party of China on 31 July 2024, clearly states that leading the economic society towards a green and low-carbon transformation is a key strategic goal to achieve high-quality development in China and the basis for solving problems involving resources, the environment, and ecology. This underscores that environmental pollution control has become a priority in China’s development. Consequently, ensuring the authenticity and transparency of corporate environmental accounting information disclosure is vital for advancing green transformation and achieving sustainable development goals. Various factors influence the quality of environmental accounting information disclosure, yet some scholars have overlooked the impact of digital transformation in enterprises. The digital transformation of enterprises involves integrating existing organizational structures, management models, and other aspects, reshaping enterprises’ internal mechanisms and management methods, reducing agency costs, and enhancing efficiency in communication, collaboration, and decision making [1]. Through information integration and value reconstruction, digital transformation injects endogenous power into enterprises, helping propel them towards green and inclusive growth [2], which aids them in more effectively implementing digital economy strategies and achieving the “dual carbon” goal. In the context of the digital economy and the push for green and low-carbon development, improving the quality of corporate environmental accounting information disclosure has become a key focus in academia. This enhancement is crucial for promoting the long-term sustainable development of enterprises and maximizing their economic, environmental, and social benefits. This article examines the impact of digital transformation at the company level, specifically its role and internal mechanism in enhancing the quality of environmental accounting information disclosure. The study not only provides theoretical support for establishing a more robust institutional framework for such disclosures but also offers valuable policy guidance and insights for the digitalization and green transformation of heavily polluting enterprises.

Many scholars have conducted numerous studies on the economic impacts of enterprise digital transformation, but the conclusions remain inconsistent. Most scholars maintain a positive view of digital transformation. Digital transformation not only helps to optimize corporate structures and operation mechanisms [3], thus strengthening the reduction of executive opportunism [4], but it also effectively reduces the carbon intensity of suppliers [5] and promotes the green technological innovation of enterprises [6] to improve their ESG performance [7]. Moreover, digital transformation can reduce financing costs and ease financing constraints [8]. As a result, this transformation helps to reduce the risk of stock price crashes [9]. However, some scholars have reservations, primarily concerning the risk management implications that digital transformation may introduce. Specifically, the more “Internet+”-related information a company discloses, the higher the risk of a stock price crash, indicating that such a disclosure may lead to a strategic speculative impact [10]. Additionally, in an environment with high market volatility, digital transformation may not significantly enhance firms’ sustainability performance [11].

Environmental accounting information disclosure is a crucial method for reflecting and overseeing the environment-related activities of enterprises [12]. Research on the disclosure of environmental accounting information primarily falls into two categories: influencing factors and economic consequences. Regarding influencing factors, studies have examined the quality of environmental accounting information disclosure concerning factors like the nature of property rights and environmental regulations. Compared to non-SOEs, state-owned enterprises (SOEs) are expected to reveal a higher quality of monetary environmental accounting information due to the pressure and support stemming from government environmental regulations [13]. Environmental regulations urge manufacturers to be willing to protect the environment through green development practices [14,15], which prompts enterprises to pay more attention to environmental management issues to continuously improve the quality of environmental information disclosure [16]. At the same time, the closer the geographical distance between enterprises and ecological and environmental regulatory authorities, the more pressure and motivation enterprises have to act with environmental responsibility, and the quality of environmental information disclosure is improved [17,18]. Regarding economic consequences, the disclosure of environmental accounting information has the potential to lower the external financing costs of enterprises, enhance market stock liquidity, and mitigate financing challenges [19]. Environmental information disclosure helps enterprises establish a green reputation and good image, increase market demand, and thus improve their cash flow. Moreover, under the pressure of media public opinion, high-quality environmental information disclosure by enterprises helps to promote green innovation in heavily polluted industries [20], thus promoting urban carbon emission reduction [21]. However, few studies have explored the factors influencing the quality of environmental accounting information disclosure with digital transformation as an entry point.

Therefore, this paper constructs a comprehensive analysis framework that includes digital transformation, analysts’ attention, executive compensation incentives, and environmental accounting information disclosure. It aims to explore how the digital transformation of enterprises affects the quality of environmental accounting information disclosure. Compared with the existing studies, the marginal contribution of this paper is as follows. First, most of the existing studies focus on the impact of enterprise characteristics on the quality of environmental accounting information disclosure, such as enterprise size, ownership structure, and financial leverage, while there are relatively few studies on the impact of digital transformation on the quality of environmental accounting information disclosure. This paper discusses the impact of digital transformation on the quality of environmental accounting information disclosure, which not only expands the research on the factors affecting environmental accounting information disclosure but also enriches the related research on the economic consequences of digital transformation. Second, previous studies have examined the mechanism affecting environmental accounting information disclosure from a single dimension. From the perspective of the internal and external governance of enterprises, this paper incorporated executive compensation incentives and analysts’ attention into the research framework, revealing the process of digital transformation affecting the quality of environmental accounting information disclosure through these two mechanisms, opening the “black box” of causality between the two to a certain extent, and providing new ideas and perspectives for the study of digital transformation and environmental accounting information disclosure. Third, this paper integrates digital human assets into the practice of comprehensively measuring digital transformation to highlight the core role of digital talent in promoting enterprises’ digital transformation.

2. Theoretical Analysis and Research Hypotheses

2.1. Digital Transformation and Environmental Accounting Information Disclosure

To build a strong environmental image, companies often engage in greenwashing behaviors, which involve giving misleading or vague information about their environmental practices. This creates an information gap between the company and external stakeholders, potentially misleading analysts and auditors in their evaluations. As a result, this can harm investor decision making and negatively affect the efficient allocation of resources within the capital market. Digital transformation can reduce information asymmetry, inhibit corporate greenwashing behavior, and improve the quality of environmental accounting information disclosure through emerging technologies such as blockchain, big data, cloud computing, and the Internet of Things. Specifically, in terms of information quality, digital technology can store and export real-time data throughout the company’s business operations as standardized and structured information that is easy to understand. Blockchain technology, as a decentralized distributed ledger database, is highly secure and tamper-proof [22]. It has increased the reliability and timeliness of information and reduced irregular and false information. These have enabled analysts to obtain more accurate and practical information, reducing information asymmetry and improving the quality and depth of corporate environmental accounting information disclosure. In terms of the amount of information, digital transformation enables enterprises to collect and process a large amount of data, including information related to the environment, which can provide analysts with more reliable sources of information to help assess the environmental performance and sustainability of enterprises, thereby enhancing the breadth of environmental accounting information disclosure. In terms of information transmission, digital transformation can break down information barriers between external stakeholders, change how data are transmitted, shorten the spatial distance of information transmission, improve the efficiency of information transmission, and improve the information environment. The digital transformation of enterprises not only improves the availability of information in terms of quality and quantity but also improves the efficiency of information transmission and reduces the cost of information acquisition and processing. Therefore, analysts and other stakeholders will pay special attention to enterprises implementing digital transformations. Under such external pressure, enterprises must pay attention to environmental pollution, increase investments in environmental governance [21,23], and truthfully disclose corporate environmental accounting information to promote green transformation. Therefore, Hypothesis 1 (H1) is proposed based on the above statements.

H1. 

Implementing digital transformation by enterprises can significantly improve the quality of environmental accounting information disclosure.

2.2. Mechanisms of Digital Transformation Affecting the Quality of Environmental Accounting Disclosure

2.2.1. Analyst Attention

From the perspective of external supervision, the “spotlight” effect of digital transformation can often attract more analysts’ attention. Analysts rely on their professional ability to assess the authenticity of their company’s environmental information to improve the quality of environmental accounting information disclosure. Enterprises disclose environmental information in a standardized way through digital technology, effectively reducing the cost of analysts’ obtaining and processing information, attracting analysts’ attention [24], and prompting analysts to conduct on-site research to gain a deeper understanding of the actual situation of the enterprise [25], in order to improve the accuracy of environmental accounting information disclosure. Digital transformation brings high levels of risk, uncertainty, and system complexity. In this case, enterprises with high levels of digital transformation will face stricter external supervision and attract more analysts and media attention to environmental issues. Analysts’ optimistic predictions about the environmental status of a company can help demonstrate its efforts in environmental management [26], thereby enhancing its corporate image. According to the reputation theory, enterprises will also be more proactive in disclosing detailed environmental accounting information. Analysts play a key role in monitoring companies’ environmental performance and sustainability. Digital technology enables companies to provide more detailed and accurate environmental data, helping analysts better understand the company’s performance. At the same time, data processing and analysis have become more efficient and convenient, accelerating the acquisition of environmental accounting information and improving the effectiveness of supervision. In addition, management may use digital technology to hide or distort environmental accounting information to influence investor decisions, increase operational uncertainty, and increase fraud risks [27]. However, analysts with professional knowledge have keen judgment, can discern long-term profit opportunities and abnormal situations, and can effectively restrict the opportunistic behavior of management. The external supervision function of analysts enables management to focus more on the company’s sustainable development when facing choices between benefits and risks, avoiding short-sighted behavior. Therefore, Hypothesis 2 (H2) is proposed based on the above statements.

H2. 

Analysts’ attention can enhance the positive impact of digital transformation on improving the quality of environmental accounting information disclosure.

2.2.2. Executive Compensation Incentives

From the perspective of the principal–agent problem, digital transformation often incurs high costs and risks. Currently, executives tend to be wary of digital transformation due to risk aversion and their desire for income stability. Moreover, the transformation and upgrading of enterprises require executives to be prepared to accept a short-term decline in performance and even bear the enormous pressure brought about by the failure of a transformation [28]. Reasonable executive compensation incentives can mitigate executives’ risk aversion and encourage a focus on the long-term development of enterprises. This, in turn, promotes the degree of digital transformation within organizations and effectively reduces agency costs and internal opportunism [29]. As a result, it improves the quality of environmental accounting information disclosure [30]. According to the “salary defence hypothesis”, when executives receive higher salaries, it is easy to attract the attention and doubts of the capital market and the public [31]. To alleviate this pressure, executives are motivated to demonstrate that their job performance justifies their substantial pay. In this scenario, the goals of executives and shareholders often align, which reduces the incentive for executives to conceal or postpone the disclosure of negative information. Consequently, this alignment contributes to the enhanced quality of environmental accounting information disclosure. Furthermore, digital transformation aids enterprises in establishing effective environmental management and internal control systems. Driven by executive compensation incentives, enterprises will pay more attention to the digital construction of environmental management and the substantive content of environmental disclosure [32,33], such as detailed environmental governance measures. Through digital tools and platforms, real-time monitoring, analyses, and reports of environmental data are carried out by enterprises to improve the quality of environmental accounting information. Therefore, Hypothesis 3 (H3) is proposed based on the above statements.

H3. 

Executive compensation incentives can enhance digital transformation’s positive role in improving the quality of environmental accounting information disclosure.

3. Research Design

3.1. Sample and Data

This study selects the Shanghai and Shenzhen A-share listed companies in China’s most polluting industries from 2013 to 2022 as its research object and handles their data as follows. Firstly, samples in the financial industry are excluded. Secondly, samples with missing data on research indicators are excluded. Thirdly, samples listed as subject to special treatment (ST), those marked as having potential risk (PT), those that were suspended, and those that were delisted during the sample period are excluded. Fourthly, all continuous variables are winsorized at the 1st and 99th percentiles to reduce the impact of outliers on empirical results. Finally, 5876 sample observations are obtained. Company annual report data are sourced from the official websites of the Shanghai and Shenzhen Stock Exchanges, while other data are retrieved from the China Stock Market and Accounting Research Database (CSMAR). Data processing for this study is conducted using Stata 17.

3.2. Model Construction

To examine how enterprise digital transformation influences the quality of environmental accounting information disclosure, this study designs the following baseline model:

$${{\displaystyle \mathrm{EAID}}}_{it}={{\displaystyle {{\displaystyle \alpha }}_0+{{\displaystyle \alpha }}_{1}DT}}_{it}+{\displaystyle \sum {{\displaystyle \alpha }}_{\mathrm{n}}}{{\displaystyle Controls}}_{it}+{\displaystyle \sum Ind}+{\displaystyle \sum Year}+{{\displaystyle \epsilon }}_{it}$$

(1)

In Formula (1), the dependent variable is the environmental accounting information disclosure (EAID), which measures the quality of information. The core independent variable is enterprise digital transformation (DT), and Controls represent the selected control variables. Ind is the industry dummy variable, Year is the year dummy variable, and ε is the random disturbance term.

3.3. Variable Definitions

3.3.1. Dependent Variable

Referring to Kong et al. (2021) [34] and Nie et al. (2018) [35], in this paper, the content analysis method is used to classify the quality of environmental accounting information disclosure according to environmental liabilities, costs, inputs, and performance; quantify the subdivided projects; sum the scores of all target projects; and divide these by the best score of environmental accounting information disclosure projects, which is the quality index of environmental accounting information disclosure, to measure the quality of the disclosure. The specific scoring items of environmental accounting information disclosure are shown in

Table 1. Environmental accounting information disclosure index scoring criteria.

Disclosure Type	Disclosure Item	Scoring Criteria
Environmental liabilities	Wastewater emissions	Quantitative and qualitative description: 2 points Qualitative description only: 1 point None: 0 points
	COD emissions
	SO2 emissions
	CO2 emissions
	Smoke and dust emissions
	Industrial solid waste emissions
Environmental costs	Greening cost, maintenance cost, and hazardous waste disposal cost
	Environmental fines, lawsuits, and compensation
Environmental investment	Environmental protection investment and environmental protection, R&D investment, and
	government subsidies
Environmental performance	Waste emission reduction and treatment
	Waste gas emission reduction governance
	Dust and fume management
	Solid waste utilization and disposal
	Noise, light pollution, radiation, and other governance
	Implementation of cleaner production

.

3.3.2. Independent Variable

Digital transformation involves more than just technological changes; it requires a holistic shift in employee skills, management approaches, and organizational culture. In this process, enterprises must build a professional, talented team with digital literacy and technical capabilities, which is the core foundation for successful transformation. First of all, digital talent with advanced technical capabilities and innovative thinking can provide technical support and solutions for enterprises to help optimize business processes and improve efficiency. Secondly, a high level of digital literacy enables employees to quickly adapt to new technologies and tools, promoting the digital transformation of organizational culture. In addition, digital talent is able to provide a scientific basis for corporate decision making and enhance market competitiveness through data analysis and intelligent applications. Therefore, investing in digital human assets is not only an inevitable choice for the digital transformation of enterprises, but also a strategic move to realize sustainable innovation and growth. Thus, drawing on the practices of Wu et al. and Ho et al. (2011) [36,37], this paper aims to integrate indicators of digital human assets to provide a more comprehensive measurement of enterprise digital transformation. The measurement of digital human assets refers to the research of Yu et al. (2022) [38] and He et al. (2024) [39]. The relevant data are derived from the CSMAR database, as shown in

Table 2. Indicator system for enterprise digital transformation.

Comprehensive Indicator	Primary Indicator	Secondary Indicator	Measurement Method
Digital transformation	Digital investment	Digital hardware assets	The ratio of the year-end balance of fixed assets related to digitization to total assets
		Digital software assets	The ratio of the year-end balance of intangible assets related to digitization to total assets
		Digital human assets	The ratio of digital technology employees to total employees
	Digital technology application	Fundamental digital technology	The number of keywords related to digital technology in the annual report text
		Depth of digital technology application	The number of keywords related to digital business in the annual report text

.

3.3.3. Control Variables

To minimize the estimation bias caused by omitted variables, this study builds on the existing literature by controlling for relevant firm characteristics and corporate governance factor variables that affect environmental accounting information disclosure as much as possible. These include firm age (ListAge), financial leverage (Lev), profitability (ROA), firm growth (Growth), board size (Board), percentage of independent directors (Indep), integration of two roles (Dual), whether the firm is loss-making (Loss), year (Year), and industry (Ind), as shown in

Table 3. Definition of control variables.

Variables	Symbol	Definition
Firm age	ListAge	Add 1 to the number of years the company has been listed and take the natural logarithm
Financial leverage	Lev	The ratio of total liabilities to total assets at the end of the period for a company
Profitability	ROA	The ratio of net profit to the average balance of total assets
Firm growth	Growth	The ratio of this year’s operating income to the previous year’s operating income minus 1
Board size	Board	The natural logarithm of the number of board members
Percentage of independent directors	Indep	The proportion of independent directors to the total number of directors on the board of directors
Integration of two roles	Dual	When the same person holds the position of board of directors and general manager, it is assigned a value of 1; otherwise, it is assigned a value of 0
Whether the firm is loss-making	Loss	If the net profit for the year is less than 0, take 1; otherwise, take 0
Year	Year	Dummy variable
Industry	Ind	Dummy variable

.

4. Empirical Results and Discussion

4.1. Descriptive Statistics

The descriptive statistical results of the main variables are shown in

Table 4. Descriptive statistical results of main variables.

Variables	N	Mean	Sd	Min	Median	Max
EAID	5876	0.2449	0.1891	0.0000	0.2188	0.7500
DT	5876	0.0118	0.0123	0.0024	0.0091	0.1519
ListAge	5876	2.6008	0.5478	0.0000	2.7081	3.4012
Lev	5876	0.4331	0.2001	0.0463	0.4286	0.9203
ROA	5876	0.0404	0.0647	−0.3343	0.0362	0.2395
Growth	5876	0.1383	0.3392	−0.5249	0.0849	2.5923
Board	5876	2.1524	0.1890	1.6094	2.1972	2.7081
Indep	5876	37.1575	5.1389	28.5700	33.3300	60.0000
Dual	5876	0.2013	0.4010	0.0000	0.0000	1.0000
Loss	5876	0.1193	0.3242	0.0000	0.0000	1.0000

. The average quality of environmental accounting information disclosure (EAID) of the listed companies in China’s most polluting industries is 0.2449, the median is 0.2188, the maximum value is 0.75, and the minimum value is 0. This indicates significant differences in environmental accounting information disclosure quality among different enterprises. The mean and standard deviation of digital transformation (DT) are 0.0118 and 0.0123, respectively, with a maximum value of 0.1519 and a minimum value of 0.0024, indicating that the overall level of digital transformation is relatively low, and there are significant differences in the degree of digital transformation among the listed companies. The descriptive statistical results of the other control variables are within a reasonable range, so these are not elaborated on here.

4.2. Benchmark Regression

Based on the empirical model (1), this study explores the impact of digital transformation on the quality of the environmental accounting information disclosure of the listed companies in heavily polluting industries. The regression analysis results are displayed in

Table 5. Benchmark regression results.

	(1)	(2)	(3)
Variables	EAID	EAID	EAID
DT	1.3909 ***	1.0911 ***	0.6916 ***
	(6.97)	(4.78)	(3.02)
ListAge			0.0214 ***
			(4.63)
Lev			0.1667 ***
			(12.13)
ROA			0.2096 ***
			(4.16)
Growth			−0.0329 ***
			(−4.72)
Board			0.1498 ***
			(10.09)
Indep			0.0015 ***
			(2.89)
Dual			−0.0313 ***
			(−5.56)
Loss			−0.0339 ***
			(−3.63)
Constant	0.2285 ***	0.2383 ***	−0.2816 ***
	(67.14)	(20.57)	(−5.95)
Observations	5876	5876	5876
R-squared	0.008	0.086	0.154
Year FE	NO	YES	YES
Ind FE	NO	YES	YES

Note: *** denote 1% significance levels, respectively, and t-values adjusted for robust standard errors are shown in parentheses.

. In column (1), the year, industry-fixed effects, and control variables were not incorporated. Notably, the regression coefficient for DT stands at 1.3909, surpassing the 1% level of statistical significance. Transitioning to column (2) with the inclusion of both the year and industry-fixed effects from column (1), the core explanatory variable DT continues to exhibit significant positivity at the 1% level. Subsequently, in column (3), integrating the control variables based on column (2), the regression coefficient for DT is 0.6916, remaining significant at the 1% level. Economically, a 1% increase in digital transformation corresponds to a 0.6916 rise in environmental accounting information disclosure quality by heavily polluting listed companies. This underscores how digital transformation facilitates the dissemination of high-quality environmental accounting information, reinforcing hypothesis 1 (H1) in this study. The regression outcomes for control variables like firm age (ListAge), financial leverage (Lev), and firm growth (Growth) align closely with prior research findings.

5. Robustness Tests

5.1. Replacement of Core Explanatory Variables

Referring to Yuan et al. (2021) [40], a dictionary of enterprise digitalization terms is constructed based on the semantic system of national policies. Through text analysis, the cumulative frequencies of terms related to enterprise digitalization are tallied, and the total frequency of words serves as a measure of the enterprises’ digital transformation level after the natural logarithm transformation (Ln_DT). In

Table 6. Robustness test results.

	(1)	(2)
Variables	EAID	EAID
Ln_DT	0.0058 *
	(1.76)
DT		0.4982 **
		(2.16)
ListAge	0.0225 ***	0.0031
	(4.71)	(0.68)
Lev	0.1675 ***	0.1359 ***
	(12.03)	(9.87)
ROA	0.2161 ***	0.0819
	(4.23)	(1.62)
Growth	−0.0328 ***	−0.0388 ***
	(−4.64)	(−5.67)
Board	0.1558 ***	0.1342 ***
	(10.31)	(9.09)
Indep	0.0017 ***	0.0011 **
	(3.27)	(2.20)
Dual	−0.0323 ***	−0.0211 ***
	(−5.66)	(−3.81)
Loss	−0.0331 ***	−0.0300 ***
	(−3.50)	(−3.28)
ATO		0.0455 ***
		(7.33)
Inst		0.1233 ***
		(11.24)
Constant	−0.3119 ***	−0.2784 ***
	(−6.43)	(−6.02)
Observations	5725	5858
R-squared	0.154	0.181
Year FE	YES	YES
Ind FE	YES	YES

Note: ***, **, and * denote 1%, 5%, and 10% significance levels, respectively, and t-values adjusted for robust standard errors are shown in parentheses.

, examining column (1) reveals a positively significant coefficient for enterprise digital transformation at the 10% level. This suggests that fostering digital transformation within enterprises notably enhances the quality of environmental accounting information disclosures, reinforcing the robustness of this study’s findings.

5.2. Adding Control Variables

To address the issue of omitted variables, this study introduces control variables—total asset turnover (ATO) and institutional investor shareholding (Inst)—into the regression model following the baseline regression. Examining column (2) of Table 6, the coefficient associated with the level of digital transformation exhibits significant positivity at the 5% level. This reaffirms that digital transformation is vital in enhancing environmental accounting information disclosures for heavily polluting entities, thereby validating the solidity of the benchmark regression outcomes.

5.3. Endogeneity Test

5.3.1. Instrumental Variable Approach

To address potential endogeneity due to the impact of digital transformation on environmental accounting information disclosure in the listed heavy pollution companies, this study employs an instrumental variable approach. Drawing from Wang et al. (2024) [41], the one-period lagged degree of digital transformation (L.DT) functions as an instrumental variable in a two-stage regression analyzing corporate environmental accounting disclosure quality. Conceptually, digital transformation within enterprises occurs over the long term with a time lag, influencing current digitization levels and meeting instrumental variable relevance criteria. Additionally, the lagged digital transformation level does not directly influence enterprise environmental accounting information disclosure, satisfying exogeneity requirements. The analysis in

Table 7. Endogeneity test results.

	Instrumental Variable		PSM	Heckman Two-Stage
	(1)	(2)	(3)	(4)	(5)
Variables	DT	EAID	EAID	DT_Dum	EAID
DT		0.6900 ***			0.7275 ***
		(2.74)			(3.17)
L.DT	0.9761 ***
	(42.00)
Mean_DT				0.2175 *
				(1.74)
Treat_DT			0.0160 ***
			(3.33)
IMR					0.4046 ***
					(4.87)
ListAge	0.0001	0.0248 ***	0.0191 ***	−0.1972 ***	−0.0313 ***
	(0.87)	(4.47)	(4.28)	(−5.36)	(−2.64)
Lev	0.0006	0.1663 ***	0.1715 ***	0.0141	0.1689 ***
	(1.45)	(10.92)	(12.62)	(0.14)	(12.32)
ROA	0.0028 *	0.2240 ***	0.2402 ***	1.8218 ***	0.6607 ***
	(1.69)	(4.10)	(4.63)	(4.91)	(6.38)
Growth	0.0003	−0.0309 ***	−0.0456 ***	−0.0500	−0.0456 ***
	(0.90)	(−3.84)	(−7.01)	(−0.92)	(−6.19)
Board	0.0013 **	0.1470 ***	0.1086 ***	0.5813 ***	0.2967 ***
	(2.57)	(9.13)	(7.41)	(5.30)	(8.94)
Indep	0.0000 *	0.0015 ***	−0.0001	0.0140 ***	0.0050 ***
	(1.78)	(2.63)	(−0.26)	(3.62)	(5.66)
Dual	−0.0000	−0.0280 ***	−0.0355 ***	0.0622	−0.0155 **
	(−0.45)	(−4.54)	(−6.40)	(1.45)	(−2.41)
Loss	−0.0001	−0.0317 ***	−0.0380 ***	−0.0590	−0.0523 ***
	(−0.48)	(−3.12)	(−4.16)	(−0.87)	(−5.22)
Constant	−0.0039 **	−0.1373 ***	−0.1301 ***	−2.7182 ***	−1.2894 ***
	(−2.46)	(−2.64)	(−2.74)	(−7.60)	(−6.12)
Kleibergen–Paap rk LM statistic		73.791
Kleibergen–Paap rk Wald F statistic		1763.82
Observations	5102	5102	5865	5876	5876
Pseudo R2/R-squared	0.8402	0.144	0.152	0.0626	0.157
Year FE	YES	YES	YES	YES	YES
Ind FE	YES	YES	YES	YES	YES

Note: ***, **, and * denote 1%, 5%, and 10% significance levels, respectively, and t-values adjusted for robust standard errors are shown in parentheses.

, specifically column (1), indicates the positive coefficient of the instrumental variable (L.DT) in the initial regression stage at the 1% significance level, meeting instrumental variable correlation test standards. Moreover, the Kleibergen–Paaprk rank Wald F-statistic surpasses the critical value for the Stock–Yogo weak instrumental variable identification F-test at the 10% level, dismissing weak instrumental variable possibilities. The Kleibergen–Paaprk LM statistic, significant at the 1% level, rejects hypotheses of insufficient instrumental variable identification. These outcomes affirm the appropriateness of the instrumental variables utilized. The subsequent analysis in column (2) reveals a coefficient of 0.6900 for the degree of digital transformation in the second-stage regression, passing the 1% significance threshold, thereby being consistent with the baseline regression findings and ensuring the robustness of this study’s conclusions.

5.3.2. PSM

To mitigate endogeneity stemming from sample self-selection bias, this study employs the propensity score matching (PSM) technique. Initially, the sample is stratified into a high digital transformation group (treatment) and a low digital transformation group (control) based on the median digital transformation level. Subsequently, all designated control variables are utilized as covariates in a 1:1 nearest neighbor matching approach with replacements, facilitated by a Logit model. Following this, regression analyses are conducted on the matched sample. The results in Table 7, specifically column (3), demonstrate a continued positive relationship between the degree of digital transformation post-matching and environmental accounting information disclosure at the 1% significance level. This reaffirms the robustness of the study’s findings after the mitigation of the sample self-selection bias.

5.3.3. Heckman Two-Stage Method

To mitigate sample selection bias, this study implements a Heckman two-stage model. Firstly, drawing on Kugler et al. (2020) [42], the average degree of digital transformation (Mean_DT) of other firms in the same industry was selected, and all control variables were used as exogenous variables for the Probit model regression. Subsequently, the inverse Mills ratio (IMR) derived from the regression is incorporated as a control variable for the second stage of regression testing. As shown in column (5) of Table 7, the regression coefficient of the IMR ratio is significantly positive at the 1% level, which indicates that the sample has an endogeneity problem. Simultaneously, the regression coefficient for enterprise digital transformation and environmental accounting information disclosure stands at 0.7275, which passes the significance level of 1% and is in line with the results of the baseline regression, and the findings of this study are robust.

6. Further Research

6.1. Mechanism Test

According to the previous theoretical analysis, this paper constructs the following mediation effect model to further examine the role of and mechanism by which enterprise digital transformation affects environmental accounting information disclosure.

$${{\displaystyle M}}_{it}={{\displaystyle \beta }}_0+{{\displaystyle \beta }}_1{{\displaystyle DT}}_{it}+{\displaystyle \sum {{\displaystyle \beta }}_n}{{\displaystyle Controls}}_{it}+{\displaystyle \sum Ind}+{\displaystyle \sum Year}+{{\displaystyle \epsilon }}_{it}$$

(2)

$${{\displaystyle EAID}}_{\mathrm{it}}={{\displaystyle \gamma }}_0+{{\displaystyle \gamma }}_1{{\displaystyle DT}}_{it}+{{\displaystyle \gamma }}_2{{\displaystyle M}}_{it}+{\displaystyle \sum {{\displaystyle \gamma }}_n}{{\displaystyle Controls}}_{it}+{\displaystyle \sum Ind}+{\displaystyle \sum Year}+{{\displaystyle \epsilon }}_{it}$$

(3)

In Formula (2), ${{\displaystyle M}}_{\mathrm{it}}$ are the mechanism variables, representing analyst attention (Analyst) and executive compensation incentives (Excu), respectively. They are expected to have significantly positive coefficients. If only the coefficient ${{\displaystyle \gamma }}_2$ is significant in Formula (3), it is demonstrated that the mechanism acts as a full mediator, and if ${{\displaystyle \gamma }}_1$ and ${{\displaystyle \gamma }}_2$ are both significantly positive, and if the absolute value of ${{\displaystyle \gamma }}_1$ is less than the absolute value of ${{\displaystyle \alpha }}_1$, the transmission mechanism acts as a partial mediator. The remaining variables are consistent with the baseline regression model.

6.1.1. External Monitoring Effect

In the era of integrating the digital economy and green technology, low-carbon listed companies in heavily polluting industries that implement digital transformation often attract the attention of investors. Nevertheless, due to investors’ knowledge constraints, they frequently encounter challenges in accurately interpreting the informative signals emitted by corporate behavior. Conversely, analysts with specialized expertise and discernment tend to closely monitor such enterprises, acquiring more thorough and intricate environmental insights. Under the external supervision of analysts, firms actively fulfil their environmental protection responsibilities, thus improving the quality of environmental accounting information disclosures. Therefore, this study empirically scrutinizes the pathway and impact of analysts’ attention in the nexus linking enterprises’ digital transformation and environmental accounting information disclosure.

Referring to Huang et al. (2021) [43], this study employs the natural logarithm of the number of analyst teams within the tracked firms plus one as a metric for analyst attention (Analyst). The regression findings analyzing the analyst attention mechanism are detailed in

Table 8. Mechanism analysis results: analysts’ attention.

	(1)	(2)	(3)
Variables	EAID	Analyst	EAID
DT	0.6969 ***	10.3947 ***	0.4024 *
	(3.04)	(7.66)	(1.80)
Analyst			0.0283 ***
			(12.49)
ListAge	0.0207 ***	−0.3293 ***	0.0300 ***
	(4.46)	(−12.08)	(6.49)
Lev	0.1653 ***	0.6926 ***	0.1457 ***
	(12.02)	(8.68)	(10.65)
ROA	0.2103 ***	9.3289 ***	−0.0540
	(4.17)	(27.51)	(−0.99)
Growth	−0.0342 ***	−0.0629	−0.0324 ***
	(−4.88)	(−1.44)	(−4.65)
Board	0.1540 ***	0.9907 ***	0.1259 ***
	(10.33)	(12.23)	(8.42)
Indep	0.0015 ***	0.0168 ***	0.0011 **
	(2.98)	(5.81)	(2.08)
Dual	−0.0307 ***	0.1086 ***	−0.0338 ***
	(−5.45)	(3.23)	(−6.05)
Loss	−0.0328 ***	0.2578 ***	−0.0401 ***
	(−3.52)	(5.29)	(−4.38)
Constant	−0.2910 ***	−0.9578 ***	−0.2638 ***
	(−6.13)	(−3.74)	(−5.62)
Observations	5858	5858	5858
R-squared	0.154	0.286	0.175
Year FE	YES	YES	YES
Ind FE	YES	YES	YES

Note: ***, **, and * denote 1%, 5%, and 10% significance levels, respectively, and t-values adjusted for robust standard errors are shown in parentheses.

. Column (1) illustrates the impact of firms’ digital transformation on the quality of environmental accounting disclosure; column (2) delves into the influence of digital transformation on analyst attention. The outcomes reveal a positive coefficient for digital transformation (DT) at the 1% significance level, indicating a substantial boost in analyst attention due to digital transformation. In column (3), the coefficients of analysts’ attention (Analyst) and digital transformation (DT) are positive at the 1% and 10% significance levels, respectively, which indicates that the increase in analysts’ attention, as one of the external monitoring mechanisms, has a positive effect on the quality of environmental accounting disclosure due to enterprises’ digital transformation and plays a partially mediating role, thus supporting hypothesis 2 (H2).

6.1.2. Internal Governance Effect

According to Yi et al. (2010) [44], the internal governance level of a company exerts a notable impact on information disclosure quality. As one of the internal corporate governance mechanisms, executive compensation incentives can effectively curb management’s short-sighted behavior. Managers can focus on environmental management practices [45], improve the utilization of energy and natural resources, and actively disclose environmental accounting information [46]. Based on this, this paper explores how digital transformation enhances the quality of environmental accounting information disclosure through executive compensation incentives. Drawing on the methodology of Tang and Sun (2014) [47], this paper utilizes the natural logarithm of the average total compensation of the top three executives as a proxy for executive compensation incentives (Excu). Higher executive compensation reduces the likelihood of management disregarding the company’s and shareholders’ interests for personal gain, thereby indicating a higher level of management governance.

Table 9. Mechanism analysis results: executive compensation incentives.

	(1)	(2)	(3)
Variables	EAID	Excu	EAID
DT	0.6969 ***	4.6347 ***	0.5418 **
	(3.04)	(5.27)	(2.46)
Excu			0.0335 ***
			(9.31)
ListAge	0.0207 ***	0.0493 ***	0.0190 ***
	(4.46)	(2.80)	(4.13)
Lev	0.1653 ***	0.1341 ***	0.1608 ***
	(12.02)	(2.61)	(11.70)
ROA	0.2103 ***	3.5956 ***	0.0900 *
	(4.17)	(17.64)	(1.74)
Growth	−0.0342 ***	−0.0905 ***	−0.0312 ***
	(−4.88)	(−3.06)	(−4.47)
Board	0.1540 ***	0.3221 ***	0.1432 ***
	(10.33)	(6.35)	(9.66)
Indep	0.0015 ***	0.0048 ***	0.0014 ***
	(2.98)	(2.66)	(2.70)
Dual	−0.0307 ***	−0.0044	−0.0306 ***
	(−5.45)	(−0.22)	(−5.45)
Loss	−0.0328 ***	−0.0009	−0.0328 ***
	(−3.52)	(−0.03)	(−3.52)
Constant	−0.2910 ***	11.7936 ***	−0.6856 ***
	(−6.13)	(72.76)	(−11.11)
Observations	5858	5858	5858
R-squared	0.154	0.271	0.166
Year FE	YES	YES	YES
Ind FE	YES	YES	YES

Note: ***, **, and * denote 1%, 5%, and 10% significance levels, respectively, and t-values adjusted for robust standard errors are shown in parentheses.

presents the regression analysis results for the executive compensation incentive mechanism. Column (1) demonstrates the impact of corporate digital transformation on the quality of environmental accounting information disclosure. Column (2) examines the effect of digital transformation on executive compensation incentives, revealing that the coefficient for digital transformation (DT) is significantly positive at the 1% level, suggesting that implementing digital transformations in enterprises can effectively boost executive compensation incentives. In column (3), the coefficients for both executive compensation incentives (Excu) and digital transformation (DT) are significantly positive at the 1% and 5% levels, respectively, indicating that executives motivated by compensation can facilitate the positive impact of corporate digital transformation on the quality of environmental accounting information disclosure, partially playing a mediating role, thereby supporting Hypothesis 3 (H3).

6.2. Heterogeneity Analysis

The positive impact of digital transformation on environmental accounting disclosure has been verified in the previous section. However, the effect of digital transformation on the quality of environmental accounting information may differ when the audit quality and the nature of property rights are different. Different levels of digital transformation have differential effects on the existence of digital transformation and environmental accounting information disclosure. Moreover, the impact of digital transformation of enterprises on environmental accounting information disclosure can also be affected by environmental regulations to different degrees. Therefore, this part conducts a group regression test on the entire sample based on the above analysis to deeply explore the heterogeneous impact of corporate digital transformation on the quality of environmental accounting information disclosure.

6.2.1. Audit Quality

Digital transformation plays a crucial role in enhancing the verifiability and credibility of environmental accounting information. Companies that embrace digital transformation tend to disclose higher-quality environmental accounting information, particularly those audited by the top four international accounting firms categorized as having a superior audit quality (denoted as one), contrasting with those identified as having a lower audit quality (denoted as zero). Columns (1) and (2) in

Table 10. Heterogeneous results based on audit quality and the nature of property rights.

	(1)	(2)	(3)	(4)
Variables	EAID	EAID	EAID	EAID
DT	0.6916 ***	0.4331 *	0.6916 ***	2.7280 ***
	(3.02)	(1.91)	(3.02)	(7.55)
Big4		0.1014 ***
		(7.95)
Dig_big		1.0520 *
		(1.81)
SOE				0.0768 ***
				(10.69)
Dig_soe				−2.7533 ***
				(−6.37)
ListAge	0.0214 ***	0.0155 ***	0.0214 ***	0.0073
	(4.63)	(3.39)	(4.63)	(1.54)
Lev	0.1667 ***	0.1606 ***	0.1667 ***	0.1542 ***
	(12.13)	(11.77)	(12.13)	(11.23)
ROA	0.2096 ***	0.1592 ***	0.2096 ***	0.2257 ***
	(4.16)	(3.16)	(4.16)	(4.53)
Growth	−0.0329 ***	−0.0302 ***	−0.0329 ***	−0.0316 ***
	(−4.72)	(−4.40)	(−4.72)	(−4.57)
Board	0.1498 ***	0.1465 ***	0.1498 ***	0.1315 ***
	(10.09)	(10.15)	(10.09)	(8.86)
Indep	0.0015 ***	0.0014 ***	0.0015 ***	0.0013 ***
	(2.89)	(2.69)	(2.89)	(2.61)
Dual	−0.0313 ***	−0.0290 ***	−0.0313 ***	−0.0235 ***
	(−5.56)	(−5.20)	(−5.56)	(−4.14)
Loss	−0.0339 ***	−0.0329 ***	−0.0339 ***	−0.0311 ***
	(−3.63)	(−3.57)	(−3.63)	(−3.38)
Constant	−0.2816 ***	−0.2634 ***	−0.2816 ***	−0.2525 ***
	(−5.95)	(−5.73)	(−5.95)	(−5.44)
Observations	5876	5876	5876	5876
R-squared	0.154	0.179	0.154	0.170
Year FE	YES	YES	YES	YES
Ind FE	YES	YES	YES	YES

Note: *** and * denote 1% and 10% significance levels, respectively, and t-values adjusted for robust standard errors are shown in parentheses.

present heterogeneous outcomes predicated on external audit quality. The analysis reveals that the interaction coefficient between enterprise digital transformation and audit quality (Dig_big) is 1.0520, surpassing the 10% significance threshold. This underscores that the higher the audit quality (Big4), the more pronounced the promotional impact of digital transformation on the quality of the environmental accounting information disclosed by the listed companies within heavily polluting industries.

6.2.2. Nature of Property Rights

Non-state-owned enterprises confront heightened market competition pressures, with their viability and progress hinging more on market reputation, investor confidence, and stakeholder backing than state-owned enterprises. Digital transformation has the potential to assist non-state-owned enterprises in presenting their environmental accounting information with greater transparency, thereby facilitating the establishment of a favorable corporate image in the market and attracting increased investor and partner interest. Consequently, this study designates the nature of property rights (SOE) as a dummy variable, denoted as one for state-owned enterprises and zero for non-state-owned enterprises. Columns (3) and (4) in Table 10 are predicated on the outcomes concerning the heterogeneity of the enterprise’s nature. The interaction term between digital transformation and firm nature (Dig_soe) is recorded at 2.7533, exhibiting significant negativity at the 1% level. This suggests that digital transformation has a more pronounced positive impact on enhancing the quality of environmental accounting disclosure in non-state-owned firms as opposed to state-owned entities.

6.2.3. Digital Transformation

Different levels of digital transformation will also have different effects on the disclosure of environmental accounting information. Therefore, according to the median of digital transformation, this paper divides the whole sample into two sub-samples (a high digital transformation level and a low digital transformation level) and carries out a grouping test. The regression results are shown in columns (1) and (2) of

Table 11. Heterogeneity results based on digital transformation and new environmental protection law.

	(1)	(2)	(3)	(4)
	High Digital Transformation Level	Low Digital Transformation Level	After Implementing the Nepl	Before Implementing the Nepl
Variables	EAID	EAID	EAID	EAID
DT	0.4564 *	−2.5127	0.6904 ***	0.7378
	(1.80)	(−1.40)	(2.92)	(0.85)
ListAge	0.0354 ***	0.0107 *	0.0203 ***	0.0218 ***
	(4.90)	(1.75)	(3.28)	(3.13)
Lev	0.1547 ***	0.1773 ***	0.1725 ***	0.1430 ***
	(7.31)	(9.80)	(10.89)	(5.12)
ROA	0.1830 **	0.2099 ***	0.2384 ***	0.0573
	(2.51)	(3.02)	(4.21)	(0.54)
Growth	−0.0326 ***	−0.0313 ***	−0.0340 ***	−0.0229 *
	(−3.15)	(−3.33)	(−4.21)	(−1.70)
Board	0.1284 ***	0.1665 ***	0.1539 ***	0.1289 ***
	(6.31)	(7.50)	(8.99)	(4.47)
Indep	0.0003	0.0026 ***	0.0016 ***	0.0006
	(0.40)	(3.42)	(2.77)	(0.66)
Dual	−0.0293 ***	−0.0340 ***	−0.0273 ***	−0.0471 ***
	(−3.76)	(−4.16)	(−4.13)	(−4.75)
Loss	−0.0431 ***	−0.0271 **	−0.0300 ***	−0.0470 **
	(−3.01)	(−2.19)	(−2.80)	(−2.55)
Constant	−0.2084 ***	−0.3236 ***	−0.2901 ***	−0.1502 *
	(−3.25)	(−4.36)	(−5.33)	(−1.66)
Observations	2938	2938	4697	1179
R-squared	0.151	0.159	0.134	0.148
year FE	YES	YES	YES	YES
Ind FE	YES	YES	YES	YES

Note: ***, **, and * denote 1%, 5%, and 10% significance levels, respectively, and t-values adjusted for robust standard errors are shown in parentheses.

. The study found that when the level of digital transformation is high, the digital transformation of heavily polluting enterprises has a positive correlation with environmental accounting information disclosure, and it is significant at the level of 10%. When the level of digital transformation is low, the regression coefficient of digital transformation cannot pass the significance test. It can be concluded that when the level of digital transformation is higher, the positive effect of digital transformation of heavily polluting enterprises on the quality of environmental accounting information disclosure is more pronounced.

6.2.4. The New Environmental Protection Law

In order to effectively curb environmental pollution and ecological damage, China formally implemented the new Environmental Protection Law (Nepl) on 1 January 2015. The new Environmental Protection Law puts forward precise legal requirements for environmental accounting information disclosure, and enterprises that violate the regulations will be punished according to the law. Such legal constraints urge enterprises to pay more attention to environmental accounting information disclosure and promote the application of digital transformation in environmental accounting information disclosure. This paper divides the research samples into two groups, according to whether or not they implement the new environmental protection law. After implementing the new environmental protection law, the samples are assigned a value of one. Otherwise, they are assigned a value of zero, and then a grouping regression is carried out to explore whether implementing the new Environmental Protection Law will affect the relationship between digital transformation and the quality of environmental accounting information disclosure. The regression results are shown in columns (3) and (4) of Table 11. It can be seen that after the implementation of the new Environmental Protection Law, the regression coefficient of enterprise digital transformation is 0.6904, which has passed the significance test at 1%. Before implementing the new Environmental Protection Law, there is no significant relationship between digital transformation and environmental accounting information disclosure. This shows that after implementing the new Environmental Protection Law, the digital transformation of heavily polluting enterprises has a more significant role in promoting the quality of environmental accounting information disclosure.

7. Conclusions and Implications

7.1. Conclusion and Implications

This study conducted an empirical analysis using a two-way fixed-effects model and mediated-effects model for companies listed on Shanghai and Shenzhen A-shares in heavily polluting industries in China between 2013 and 2022, aiming to explore the impact of enterprises’ digital transformation on the quality of their environmental accounting disclosure and its underlying mechanism. The findings reveal that enterprises’ digital transformation can notably enhance the quality of their environmental accounting information disclosure. Specifically, companies augment digital investments (encompassing fixed assets, intangible assets, and human assets), expedite the application of digital technologies in depth and breadth, aid in optimizing the internal governance structure of corporate executives, facilitate the bidirectional flow of information within and outside the enterprise, attract external analysts’ attention regarding the environmental concerns of the enterprise, and consequently elevate the quality of their environmental accounting information disclosure. Furthermore, the promotion effect of enterprise digital transformation on environmental accounting information disclosure is mainly achieved through analyst attention and executive compensation incentives. In addition, the impact of digital transformation on environmental accounting information disclosure is affected by internal and external factors. Specifically, the promotion effect is more significant in enterprises with a higher audit quality, those with a high level of digital transformation, and non-state-owned enterprises. Moreover, implementing the new Environmental Protection Law significantly contributes to the positive effect of digital transformation on environmental accounting information disclosure.

Building upon the conclusions drawn in this study, the following subsequent strategies are recommended. First, heavily polluting listed enterprises are advised to increase investments in fixed assets and digital-related intangible assets, emphasizing the nurturing of digital green composite talent through initiatives like collaborative programs between universities and businesses. This approach aims to reinforce the pivotal role of digital technology in internal management, enhancing the precision and timeliness of environmental accounting information disclosures. Second, enterprises are encouraged to integrate digital transformation with management compensation incentives, fostering widespread digital technology integration to ensure high-quality environmental accounting information disclosures. Enterprise leadership is anticipated to expedite digital transformation efforts within their organizations to secure elevated compensation, diminish agency costs, and ensure accurate environmental accounting disclosures. External analysts are urged to intensify the level of oversight, monitoring the environmental stewardship practices of heavily polluting listed enterprises to deter any attempts to manipulate environmental accounting disclosures for unwarranted benefits. This vigilance is crucial for elevating the credibility of environmental accounting information disclosures. Third, the government should increase the amount of attention given to and governance of enterprises and state-owned enterprises with a low audit quality, increase the penalties for environmental violations by heavily polluting enterprises, encourage external stakeholders such as analysts and auditors to participate in supervision and governance actively, strictly review heavily polluting listed enterprises that adhere to green preferential policies, such as by conducting regular return visits, and vigorously develop professional environmental supervision, including third-party environmental supervision, certification institutions, and environmental audits.

7.2. Research Limitations and Perspectives

There are certain limitations in this study on the impact of corporate digital transformation on environmental accounting information disclosure. First, this paper mainly analyzes the impact of digital transformation on environmental accounting information disclosure from two independent perspectives, namely, internal governance (e.g., executive compensation incentives) and external monitoring (e.g., analysts’ attention). However, it does not delve deeper into the interactions between these mechanisms. Future research will delve into the interactions between executive compensation incentives and analysts’ attention and analyze how they jointly affect firms’ environmental accounting information disclosure behavior. Second, this paper does not fully consider the impact of other potential mechanisms. Future research could analyze how external pressures (e.g., institutional investors, customers, and NGOs) moderate the relationship between digital transformation and environmental accounting information disclosure. Finally, this paper focuses on the facilitating effect of digital transformation on environmental accounting information disclosure, but the two may have an alternating relationship in terms of resource allocation. Digital transformation and environmental accounting information disclosure may compete for limited funds and managerial attention due to limited corporate resources, especially when capital budgets are constrained. Future research can introduce corporate financial constraint variables to analyze the relationship between digital transformation and environmental accounting information disclosure under different resource conditions.