Digital Transformation and ESG Performance—Empirical Evidence from Chinese Listed Companies

Abstract

The rapid advancement and broad adoption of digital technologies have infused ESG practices with new dimensions and significance. Drawing on panel data from Chinese A-share listed companies spanning from 2012 to 2023, this paper aims to explain the impact of digital transformation on corporate ESG performance, explore its mechanisms and external regulatory effects, and provide systematic ideas and methods for improving corporate ESG performance from the perspective of digital transformation. The key findings of this study are summarized as follows: (1) Digital transformation (DT) has a significant positive effect on corporate ESG performance, and this association remains statistically robust following multiple robustness tests and a correction for potential endogeneity. (2) An analysis of the entire operational process reveals that DT improves ESG performance through enhancing environmental information disclosure quality, strengthening the integration of digital and physical industry technologies, and bolstering supply chain resilience. (3) The implementation of the “Broadband China” strategy exerts a positive moderating effect on the linkage between DT and ESG performance. (4) A heterogeneity analysis shows that the positive impact of DT on ESG performance is more significant and stable in non-state-owned enterprises, eastern regions, less-polluted areas, and growth stage enterprises. These findings offer theoretical and empirical insights for understanding ESG performance drivers. However, the focus on Chinese A-share firms and the use of Sino-Securities ratings may limit generalizability, warranting further improvement.

1. Introduction

Since the mid-20th century, global environmental challenges have become increasingly pronounced, posing unprecedented threats to sustainable development. As anthropogenic pressures on ecosystems intensify, environmental protection, social responsibility, and corporate governance have emerged as strategic imperatives for modern enterprises. In response, the environmental, social, and governance (ESG) framework has been adopted and has rapidly evolved into a critical instrument for international capital markets to assess long-term corporate value and risk [1,2,3]. Recent data suggest that more than USD 35 trillion in global assets now incorporate ESG factors into investment strategies, signifying a shift in ESG from a peripheral notion to a mainstream investment paradigm. The ESG framework has gained global recognition, with over 5370 institutions having joined the United Nations Principles for Responsible Investment (PRI) initiative. The ESG rating systems have also undergone progressive refinement, with more than 600 distinct ESG rating frameworks currently in existence worldwide, creating diversified evaluation approaches to meet the needs of various investors and stakeholders. DT serves as a critical enabler for enterprises to enhance their ESG performance. First, it effectively facilitates the establishment of a comprehensive environmental governance system [4]. This enables real-time monitoring of ecological changes during corporate operations. Second, DT drives improvements in corporate service quality, thereby strengthening social responsibility fulfillment. Third, DT facilitates interdepartmental data integration, enhancing governance efficiency and transparency [5].

As the world’s second-largest economy, China is navigating a distinctive development trajectory while contending with significant practical challenges in ESG implementation. On the one hand, the institutional and regulatory landscape is gradually advancing. For example, the Opinions on Comprehensively Promoting the Construction of Beautiful China, issued by the State Council in 2024, formally integrated ESG assessments into the national governance framework. On the other hand, notable deficiencies remain at the execution level. By 2024, only 40% of A-share listed firms had released ESG reports, with persistent problems such as “high costs and low motivation” and a tendency toward “form over substance” [6]. This institutional discrepancy underscores a critical question in China’s ESG evolution: how can a tailored ESG improvement pathway be constructed to align with the country’s unique governance context, enabling firms to sustain global competitiveness while meaningfully advancing their sustainability commitments within an accelerating global ESG landscape?

Digital transformation refers to the integration of digital technologies into business processes, where enterprise components and production stages are digitized to enhance productivity and optimize operations. It has emerged as a strategic imperative for enhancing competitiveness and advancing sustainable development [7,8]. Fueled by innovations such as big data, artificial intelligence, cloud computing, and block chain, DT has fundamentally restructured enterprise management and operational models, offering robust technical support for sustainable corporate growth [9]. Recent scholarship has progressively emphasized the impact of DT on corporate ESG performance. On the one hand, it enables firms to more precisely identify and mitigate environmental risks by optimizing resource allocation and reducing carbon emissions through data analytics [10,11]. On the other hand, digital technologies strengthen communication between enterprises and stakeholders, enhancing the effectiveness of social responsibility initiatives [12]. Additionally, digital tools improve the transparency and efficiency of corporate governance by reducing information asymmetry and streamlining management processes [13,14].

Digital transformation in enterprises has the potential to enhance ESG performance by improving information transparency, optimizing resource allocation, and increasing managerial efficiency [15,16,17]. Nevertheless, the existing literature identifies three key areas of contention regarding its impact on ESG outcomes: (1) Nonlinear effects: Empirical studies have observed an “inverted U-shaped” relationship between DT and ESG performance, suggesting a dual-edged impact wherein excessive digitalization may lead to diminishing returns [18]. (2) Adverse consequences: Wang et al. contend that during the digital transformation process, firms may encounter resource misallocation, rising managerial costs, and strategic short-termism, all of which can undermine ESG performance [19]. Moreover, digitalization may trigger issues such as privacy violations, labor exploitation, environmental degradation, monopolistic conduct, and algorithmic bias—phenomena that contradict the core principles of sustainable development [20]. (3) Transmission complexity: Some studies identify mediating mechanisms, such as green innovation and media attention, through which DT influences ESG outcomes [21]. Others highlight moderating effects from factors like common institutional ownership (CIO) and firms’ perceptions of economic policy uncertainty (SEPU) [22]. Despite these findings, the nuanced interdependencies and underlying mechanisms remain insufficiently examined and call for further empirical scrutiny.

Accordingly, drawing on panel data from China’s A-share listed firms over the period 2012–2023, this study applies a panel regression framework to examine the mechanisms through which DT affects corporate ESG performance. The primary innovations and anticipated contributions of this research are threefold: (1) This study is the first to empirically investigate the relationship between DT and ESG performance through three distinct mediating pathways—environmental information disclosure, the integration of digital and physical technologies, and supply chain resilience—framed within the full scope of enterprise operational processes. This framework provides novel insights into the intricate transmission mechanisms that link digital transformation with ESG outcomes. (2) It also represents the first empirical identification of a positive moderating effect of the “Broadband China” strategy on the relationship between DT and ESG performance, thereby offering fresh evidence for evaluating the spillover effects of digital economy policies. (3) Based on a rigorously constructed enterprise digital transformation index, this study reveals a strong positive association between DT and ESG performance, with more pronounced effects observed in eastern regions, non-state-owned enterprises, non-heavy-polluting industries, and growth stage enterprises. These empirical findings contribute to the growing body of literature on the non-financial impacts of the digital economy.

The organization of this study is outlined as follows. Section 1 outlines the research background, objectives, key innovations, and marginal contributions. Section 2 reviews the relevant literature and formulates the research hypotheses. Section 3 describes the research design, covering sample selection, data sources, and model specification. Section 4 reports the empirical analysis, including regression results and further tests of the model. Section 5 provides extended analysis and in-depth discussion. Finally, Section 6 summarizes the main findings, offers policy and managerial implications, and outlines the limitations of this study along with directions for future research.

2. Literature Review and Hypothesis Research

2.1. Digital Transformation and ESG Performance

In terms of research regions, the existing research on the relationship between digitalization and ESG performance demonstrates diverse regional characteristics. While the majority of studies focus on the Chinese context to analyze the characteristics of local enterprises and markets [18,19,21,22,23], international perspectives have expanded the analytical boundaries. Existing studies have validated the relationship using data from 39 ASEAN-6 banks [24], all companies listed on the Indonesian Stock Exchange [25], enterprises in EU member states [14,26], and further expand to 191 countries worldwide [27]. Due to the differences in sample selection and regions, the results of various studies vary. Most studies acknowledge the promotional effect of DT on corporate ESG performance, while some scholars believe that the relationships between ESG and digitalization variables are not statistically significant [28].

Existing research is mainly based on sustainability theory, information asymmetry theory, and stakeholder theory from a research perspective. Grounded in the theory of sustainable development, firms are expected to pursue economic performance while simultaneously advancing environmental and social objectives [29]. Digital transformation presents a strategic opportunity for companies to transition toward resource-efficient and environmentally sustainable business models [30]. By harnessing digital technologies, firms can accelerate green innovation, construct energy-efficient and low-emission production systems, and advocate for eco-friendly products and sustainable values [31]. Specifically, the substitution and integration of conventional production methods with intrinsically green digital technologies generate substantial “greening effects”, thereby reinforcing environmental governance capabilities [32].

Stakeholder theory provides a foundational framework for interpreting corporate social responsibility (CSR) [33]. Beyond maximizing shareholder wealth, firms are accountable to a broad range of stakeholders, including employees, customers, regulators, and communities [34]. The open, inclusive, and interconnected nature of digital technologies enables firms to build effective stakeholder engagement mechanisms [35], optimize resource distribution and operational workflows, and enhance the flow of information across platforms. Furthermore, DT strengthens analytical and decision-making capacities, allowing enterprises to better identify societal demands and devise innovative responses to complex social challenges.

From the perspective of information asymmetry theory, DT can effectively reduce information asymmetry and enhance transparency through both internal structural reorganization and external supervisory mechanisms. Internally, it facilitates a transition from rigid hierarchical models to flatter, network-oriented organizational structures [36]. This shift reduces bureaucratic layers, enhances communication efficiency, and enables leadership to track operational conditions in real time, thereby elevating governance effectiveness. Externally, increased scrutiny from analysts and the media has improved corporate transparency and accountability [37]. The deployment of digital tools also lowers the cost of ESG information validation for external evaluators, thereby strengthening stakeholder oversight [38].

In summary, DT enhances environmental governance by fostering green innovation and transforming production systems, facilitates stakeholder engagement in addressing social concerns, and improves governance effectiveness through both internal reforms and external monitoring. These integrated pathways collectively advance corporate ESG performance. Based on this reasoning, the following hypothesis is proposed:

H1: 

Digital transformation significantly enhances firms’ ESG performance.

2.2. Digital Transformation and Its Mechanism Effect on ESG Performance

A considerable body of literature has explored the mediating pathways by which DT impacts ESG performance. Drawing from diverse analytical perspectives, these mechanisms can be broadly categorized into three stages of enterprise operations: (1) Upstream operations, where studies focus on enterprise–supplier relationships, emphasizing the mediating roles of supply chain digitalization [39] and information transparency [40]. (2) Internal operations, which investigate intra-organizational processes, highlighting the influences of innovation capability [4], absorptive capacity for digital technologies [41], and strategic transformation [42]. (3) Downstream operations, which pertain to enterprise–customer interactions. For instance, Yin et al. [43] find that environmental information disclosure serves as a significant mediating factor between DT and ESG performance.

Collectively, the prior literature has covered the full range of enterprise operations. However, most existing studies tend to focus on isolated stages—particularly upstream or internal operations—while lacking a holistic examination that integrates the entire value chain. This piecemeal approach limits our ability to fully capture the intricate ways in which DT affects ESG outcomes. Given the interconnectedness across operational stages, adopting a comprehensive framework that spans the full scope of enterprise activities offers a more robust basis for understanding these mediating mechanisms.

According to Porter’s value chain theory, enterprise operations encompass procurement, production, sales, and after-sales services. Procurement reflects supplier relationships, production aligns with internal processes, and sales and after-sales services correspond to customer interactions. In supplier relationships, DT facilitates the collection, integration, processing, and use of supply chain data to support decision-making, thereby enhancing supply chain resilience and mitigating the bullwhip effect [40]. Within production, firms can adopt digital technologies to absorb advanced capabilities from the digital sector, enabling process reengineering that enhances product quality and resource efficiency [41]. In customer-facing processes, DT encourages environmental information disclosure through digital platforms and social media, thereby improving communication with customers [43]. CATL has significantly enhanced its innovation and risk management capabilities during its digital transformation process. It has improved its ESG performance through three mechanisms: supply chain resilience, the integration of digital and physical industry technologies, and environmental information disclosure. First, CATL has established a supply chain collaboration platform with automakers such as BMW and Tesla through a data sharing mechanism, enabling the real-time coordination of order forecasting, production scheduling, and logistics coordination, thereby enhancing supply chain resilience and optimizing inventory turnover by 25%. Second, CATL’s “AI Battery Institution” system uses deep learning algorithms to identify defects on battery surfaces, achieving a defect detection accuracy rate of 99.99% and reducing defect rates by nearly 30%. Third, CATL has established an IoT data analysis platform that effectively connects remote and local data centers via dedicated data lines, enhances transmission security, and improves the quality of information disclosure. Building on this foundation, this study identifies three core mediating pathways through which DT promotes ESG performance: supply chain resilience, the integration of digital and physical industry technologies, and environmental information disclosure.

First, supply chain resilience refers to a firm’s essential capability to sustain operational stability and pursue strategic advancement in the face of internal and external disruptions [44]. Rooted in dynamic capabilities theory, DT enhances supply chain resilience through two key mechanisms: (1) End-to-end digital platforms improve supply chain visibility and responsiveness, enabling real-time inventory and logistics management, and facilitating dynamic coordination between upstream and downstream flows [45]. (2) AI- and big data-driven decision support systems enhance risk prediction and response capabilities, thereby ensuring business continuity. The digitization of supply chains serves as a fundamental driver for enhancing corporate ESG performance: resilient operations ensure the consistent fulfillment of environmental commitments; emphasizing stakeholder protection during crises deepens social responsibility practices; and a stable operational environment coupled with effective risk management mechanisms further enhances corporate governance standards [46,47,48]. Based on the above analysis, the following hypothesis is proposed:

H2a: 

Digital transformation positively influences ESG performance by strengthening supply chain resilience.

The core manifestation of the integration between digital and physical industry technologies lies in the dynamic process through which digital technologies are embedded into innovative applications across traditional industries. At its foundation, this integration involves leveraging technological innovation to reconstruct the industrial value chain [49]. Cross-industry technology sharing serves as the structural cornerstone of this integration, with emerging technologies—such as big data and artificial intelligence—deeply embedded within conventional industrial systems, thereby facilitating knowledge exchange and generating synergistic technological effects [50,51]. Through digital innovation platforms, enterprises engage in more frequent and intensive collaborations, enabling the absorption and transformation of complementary knowledge resources from both upstream and downstream actors in the industrial chain. These integrative dynamics facilitate joint breakthroughs in emerging and frontier technological domains, enhance the quality of firm-level innovation, and optimize operational processes using digital tools to reduce energy consumption intensity. Furthermore, digital–physical integration encourages greater investment in green technology R&D aimed at developing eco-friendly products and energy-efficient processes, while also contributing to employee well-being and promoting the construction of environmentally sustainable production systems [52]. Based on the above analysis, the following hypothesis is proposed:

H2b: 

Digital transformation positively influences ESG performance by deepening the integration of digital and physical industry technologies.

Digital transformation enhances both the quality and quantity of corporate environmental information disclosure (EID). On the one hand, firms with weaker environmental performance often face heightened regulatory scrutiny and tend to disclose more relevant information to mitigate reputational risks and demonstrate transparency [53]. By leveraging “ABCD” technologies—artificial intelligence, blockchain, cloud computing, and big data—enterprises can effectively reduce information asymmetry with stakeholders, facilitating the more efficient dissemination of environmental information and fostering a “herd effect” in ESG practices [43]. On the other hand, drawing on Information Processing Theory (IPT), an enterprise’s information processing capacity must align with its processing requirements to ensure effective decision-making [54]. Digitally transformed firms typically possess robust capabilities in data collection, analysis, and interpretation. These capabilities enable the efficient aggregation and management of environmental data through advanced analytics and data mining technologies, thereby improving the accuracy, completeness, and timeliness of disclosures. This, in turn, enhances corporate transparency and reinforces environmental and social accountability [55].

H2c: 

Digital transformation significantly improves ESG performance by enhancing corporate environmental information disclosure.

Furthermore, to deepen the understanding of how DT influences ESG performance, this study incorporates the moderating role of external institutional factors. Specifically, the “Broadband China” strategy—a national-level informatization initiative introduced in 2013—is examined as a moderating variable, given its potential to facilitate and shape firm-level digital transformation. The “Broadband China” policy supports the digital–ESG nexus through three primary mechanisms: (1) Top-down policy signaling: By advancing the national agenda of “digitally driven sustainable development”, the strategy sends strong institutional signals through high-level policy design. In regions with intensive policy implementation, enterprises are more inclined to prioritize digital adoption in ESG-related areas to gain institutional legitimacy. (2) Infrastructure externalities: As a shared public infrastructure, digital networks promote cross-regional knowledge spillovers, accelerate information exchange, and reduce technological diffusion barriers. These network effects help mitigate disparities in digital transformation resources and promote greater regional equity in ESG performance. (3) Technical standardization and diffusion: The policy contributes to the establishment of unified technical standards, lowers inter-organizational collaboration costs, and fosters the cross-industry diffusion of ESG management tools and best practices. Based on the above analysis, the following hypothesis is proposed:

H3: 

The “Broadband China” strategy positively moderates the impact of digital transformation on the ESG performance of Chinese enterprises.

Based on the theoretical mechanisms and hypotheses discussed above, a theoretical framework has been developed, as shown in Figure 1. The proposed model underpins the empirical analysis conducted in the subsequent sections.

3. Research Design

3.1. Sample and Date

Based on the data availability and integrity, this paper takes the panel data of A-share listed companies in China from 2012 to 2023 and processes them as follows: (1) excluding companies in the ST and *ST categories; (2) removing samples with significant anomalies in financial data; (3) eliminating samples with incomplete data for the statistical year; (4) to mitigate the potential influence of extreme values on test results, continuous variables were winsorized at the top and bottom 1%. Ultimately, a total of 38,416 sample observations were obtained. The ESG performance indicators are sourced from the Sino-Securities ESG Ratings, with the remaining data obtained from the CSMAR and Wind databases.

3.2. Definition of Variables

The explained variable: ESG performance. The research scope selected for this study consists of Chinese domestic listed companies. Therefore, the selection of ESG rating data must fully consider the representativeness of the data and its compatibility with the research content. Following the approach of Li et al. [56], Sino-Securities ESG rating data was chosen as the dependent variable. Additionally, in line with the method proposed by Tan et al. [57], the 9-level Sino-Securities ESG ratings, ranging from CCC to AAA, were assigned values from 1 to 9 to facilitate empirical analysis.

Explanatory variable: enterprise digital transformation. The construction of an evaluation index for enterprise digital transformation has been generally recognized in academia. This paper draws on the methods proposed by Gao et al. and Wu et al. [58,59], utilizing Python 3.13 crawlers combined with text recognition to construct a proxy variable for measuring enterprise digital transformation. First, a terminology dictionary for enterprise digital transformation is established, designed around three main dimensions: technological classification, organizational empowerment, and digital application. Additionally, five specific dimensions are constructed: artificial intelligence technology, big data technology, cloud computing technology, blockchain technology, and digital technology application. The dictionary includes a total of 314 word frequencies such as autonomous driving, data mining, cloud computing, blockchain, data platform, and digital infrastructure. The external measures of digital investment also usually include the transformation of innovation achievements such as investment in digital infrastructure and the number of digital technology patents, which is related to the frequency of words we choose, and indirectly indicates the validity of the dictionary.

Subsequently, Python crawlers are employed to extract textual data from the annual reports of Chinese listed companies spanning from 2012 to 2023. The corresponding word frequencies are then calculated, and after adding 1 to the frequency data, a logarithmic transformation is applied to normalize the values, resulting in the enterprise digital transformation index DT. A higher value of this index indicates a more advanced level of enterprise digital transformation.

Mediator variables. (1) Environmental information disclosure (EID): Following the approach proposed by Zhao et al. [60], this paper scores corporate performance across 27 indicators covering seven aspects, including environmental management, environmental regulation, and certification. The scores are aggregated and logarithmically transformed to derive the corporate environmental information disclosure index. (2) Digital–physical industry technology convergence (Tech-Conv): When a company applies for non-digital industry technology patents that cite digital industry technology patents, it is considered an instance of digital–physical industry technology convergence. Drawing on the method used by Huang et al. [49], this paper sums the annual occurrences of such instances and applies a logarithmic transformation to create an index reflecting the extent of a company’s digital–physical industry technology convergence. (3) Supply chain resilience (Resist): Following the method proposed by Yao et al. [61], we established a comprehensive evaluation index system to assess enterprise supply chain resilience across five key dimensions—predictive capability, resistance capability, recovery capability, human capital, and governmental influence. To determine the relative importance of each dimension and calculate the overall performance score of the sample, we employed the entropy weight-TOPSIS method. The final comprehensive score was then scaled by a factor of 100 for ease of interpretation.

Moderating variable. This paper adopts the approach of Wei et al. [62], using a dummy variable to represent cities selected as pilot sites for the “Broadband China” strategy by the Ministry of Industry and Information Technology and the National Development and Reform Commission in 2014, 2015, and 2016. For a city designated as a “Broadband China” pilot city in the year of approval and subsequent years, this variable takes the value of 1, while it is assigned 0 if the city is not selected.

Control variables. To address endogeneity issues arising from omitted variable bias, this study draws on prior research by scholars in similar fields and incorporates the following as control variables into the regression model: asset–liability ratio (Lev), firm size (Size), the shareholding percentage of the top 10 shareholders (Top10), return on equity (ROE), the proportion of independent directors (Indep), Tobin’s Q (TobinQ), whether the firm experienced a loss (Loss), years since listing (Listage), and cash flow ratio (CFR). Detailed definitions and measurement methods for these variables are provided in

Table 1. Variable definitions and measurement methods.

Type	Variables	Symbol	Measurement Methods
Explanatory variable	Digital Transformation	DT	The logarithmic transformation of the sum of keywords related to digital transformation in the company’s annual report.
Explained variable	ESG Performance	ESG	Scores assigned to each ESG rating by Sino-Securities, converted into values ranging from 1 to 9
Mediator variable	Environmental Information Disclosure	EID	Scores calculated across seven aspects, including environmental management, environmental supervision, and certification, with the total score subjected to logarithmic transformation.
	Digital–Physical Industry Technology Convergence	Tech-Conv	The logarithmic transformation of the number of times enterprises integrate digital and physical industry technologies.
	Supply Chain Resilience	Resist	The comprehensive score is determined using the entropy weight-TOPSIS model.
Control variable	Firm Size	Size	The natural logarithm of the company’s total assets.
	Asset–Liability Ratio	Lev	The ratio of the company’s total liabilities to its total assets.
	Percentage of Shares Held by the Top Ten Shareholders	Top10	The number of shares held by the top 10 shareholders as of year t divided by the total number of shares.
	Return on Equity	ROE	The ratio of net profit at the end of the period to net assets.
	Proportion of Independent Directors	Indep	The ratio of the number of independent directors to the total number of board members.
	Tobin’s Q	TobinQ	(The market value of tradable shares plus the number of non-tradable shares multiplied by the net asset value per share, plus the book value of liabilities) divided by total assets.
	Profitability Status	Loss	If the net profit for the year is less than 0, the value is set to 1; otherwise, it is set to 0.
	Years Since Listing	Listage	The current year minus the year the company went public.
	Cash Flow Ratio	Cashflow	Net cash flow generated from operating activities divided by total liabilities.
Moderating variable	“Broadband China” Strategy	DID	For a city approved as a “Broadband China” pilot city in the year of approval and thereafter, this variable is assigned a value of 1; if not selected, it is assigned a value of 0.

.

3.3. Model Specification

The research purpose of this paper is to explore the impact mechanism of DT on $ESG$. To achieve this, a model is constructed, incorporating time and industry fixed effects to develop a two-way fixed effects model. The specific equation is presented in Model (1):

$${ESG}_{i,t}={\alpha }_0+{\beta }_1{DT}_{i,t}+{\gamma }_1{Controls}_{i,t}+\mathsf{\Sigma }\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+\mathsf{\Sigma }\mathrm{I}\mathrm{n}\mathrm{d}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{y}+{\epsilon }_{i,t}$$

(1)

where the variable subscripts $i$ and t denote company and time, respectively. The explanatory variable ${ESG}_{i,t}$ is the $ESG$ performance of firm $i$ in year $t$; the core explanatory variable ${\mathrm{D}\mathrm{T}}_{\mathrm{i},\mathrm{t}}$ denotes the degree of DT of firm i in year t; and the collection of the following controls variables is represented by ${\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i},\mathrm{t}}$. Given the heterogeneous adoption levels of digital technologies among various industries, this study includes a series of fixed effects in the model. Specifically, $\mathsf{\Sigma }\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}$ and $\mathsf{\Sigma }\mathrm{I}\mathrm{n}\mathrm{d}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{y}$ represents time fixed effects and industry fixed effects, separately; the random disturbance term is represented by ${\epsilon }_{i,t}$. This paper focuses on the significance of the coefficient ${\mathsf{\beta }}_1$ of ${\mathrm{D}\mathrm{T}}_{\mathrm{i},\mathrm{t}}$ in Model (1). In accordance with the previous research hypothesis, a positive coefficient ${\mathsf{\beta }}_1$ is indicative of enhanced environmental, social, and governance (ESG) performance in corporate entities undergoing digital transformation.

4. Empirical Results and Analysis

4.1. Descriptive Statistics

The results of the descriptive statistics are shown in

Table 2. Descriptive statistics.

	(1)	(2)	(3)	(4)	(5)
Variables	N	Mean	sd	Min	Max
ESG	38,416	4.614	1.805	1	8
DT	38,416	5.269	0.105	3.655	5.987
EDI	38,416	0.414	0.348	0	1.778
Techconv	38,416	1.565	1.413	0	5.357
Resist	38,416	1.016	0.826	0.0135	26.48
Size	38,416	22.23	1.303	19.58	26.44
Lev	38,416	0.413	0.206	0.0349	0.925
ROE	38,416	0.0573	0.138	−0.962	0.414
Cashflow	38,416	0.0478	0.0681	−0.199	0.266
Loss	38,416	0.131	0.338	0	1
Indep	38,416	37.74	5.378	0	60
Top10	38,416	0.586	0.154	0.208	0.910
TobinQ	38,416	2.020	1.345	0.789	16.65
ListAge	38,416	2.062	0.935	0	3.434

. The observed values of the core explained variable ESG are distributed between one and eight, not reaching the AAA-level threshold. This indicates that no companies in the research sample have attained the highest level of the ESG rating system. Notably, the standard deviation of this variable is 1.805, revealing significant differences in ESG performance among the enterprises. For DT, the maximum value of the DT variable is 5.987, the minimum is 3.655, and the standard deviation is 0.105. These figures reflect that, despite some divergence among individual firms, the overall digital transformation exhibits low dispersion. This finding is encouraging, as it demonstrates that A-share listed companies have generally entered the digital transformation process. Most have surpassed the initial stage, achieving foundational outcomes in infrastructure construction and digital technology application [63]. Based on these statistical characteristics, it can be inferred that the sampled companies already possess the necessary technical accumulation and organizational resilience to address the future deepening of digital transformation. This provides a solid practical basis for enterprises to pursue long-term sustainability amid the rise in the digital economy.

4.2. Benchmark Regression

Table 3. Regression results of digital transformation on ESG.

	(1)	(2)	(3)
Variables	ESG	ESG	ESG
DT	1.223 ***	1.204 ***	0.748 ***
	(10.00)	(5.47)	(3.04)
Lev		−1.095 ***	−1.121 ***
		(−10.42)	(−13.36)
Size		0.115 ***	0.145 ***
		(6.08)	(6.53)
Top10		−0.062	0.096
		(−0.41)	(0.78)
ROE		1.011 ***	0.857 ***
		(6.17)	(5.23)
Indep		0.008 ***	0.007 ***
		(4.90)	(4.70)
TobinQ		−0.089 ***	−0.082 ***
		(−5.60)	(−5.12)
Loss		−0.074	−0.164 **
		(−1.04)	(−2.55)
ListAge		−0.254 ***	−0.246 ***
		(−6.95)	(−8.05)
Cashflow		−0.149	0.249
		(−0.67)	(1.42)
Constant	−1.827 ***	−3.412 ***	−1.749
	(−2.80)	(−3.55)	(−1.58)
Observations	38,416	38,416	38,416
R-squared	0.005	0.058	0.095
Industry-FE	NO	NO	YES
Year-FE	NO	NO	YES

Robust z-statistics in parentheses. *** p < 0.01, ** p < 0.05.

reports the baseline regression results. In Model (1), where neither control variables nor fixed effects are included, the coefficient of digital transformation (DT) is 1.223 and statistically significant at the 1% level, suggesting a strong positive relationship between DT and corporate ESG performance. Model (2) incorporates a set of control variables, and the DT coefficient remains significant at the 1% level, with an estimated value of 1.204. In Model (3), after further accounting for the time and industry fixed effects, the positive association between DT and ESG performance persists, with significance maintained at the 1% level. These results lend strong empirical support to Hypothesis H1. This finding partially supports the conclusions of Sun et al. [64] but diverges from the negative mechanism proposed by Wang et al. based on the “Solow Paradox” and IT evolution stage theory [19]. The analysis suggests that this divergence may stem from two factors: the positive impact of DT on ESG exhibits a lag effect as companies face higher adaptation costs during the early stages of digital transformation, making it challenging to balance technological investment with ESG responsibilities, thereby triggering self-interested behaviors and stakeholder neglect. By contrast, this study focuses on the long-term technological spillover effects of DT. Additionally, in recent years, Chinese enterprises’ DT has become increasingly advanced and deeply integrated into business operations, and its positive influence on ESG performance has grown more pronounced.

Analyzing the coefficient estimates of the control variables reveals that firm size generally has a positive impact on ESG performance, while the asset–liability ratio negatively affects ESG performance, with both significant at the 1% level. At the 1% significance level, Tobin’s Q and ESG are negatively correlated, which may be due to the fact that the Q value is affected by stock price volatility. If there is speculative sentiment in the market, it may cause Tobin’s Q to deviate from the true value of the company. Furthermore, if a company initially invests in ESG, it may increase current costs, causing the market value Tobin’s Q to temporarily fail to reflect its long-term benefits. Firm age also negatively impacts ESG performance. Moreover, the size of the corporate cash flow ratio has an insignificant impact on ESG performance.

4.3. Robustness Test

(1)

Replacement of Core Explanatory Variable

Following the approach of Feng et al. [65], this paper selects the digital transformation variables of corresponding enterprises from the CSMAR database as a new proxy variable DT*, replacing the original proxy variable DT for further regression. The results are shown in column (1) of

Table 4. Robustness test.

	(1)	(2)	(3)	(4)	(5)	(6)
Variables	ESG	ESG	ESG	E	S	G
DT*	0.780 ***
	(2.91)
L-DT		0.595 **
		(2.56)
DT			0.746 ***	0.758 ***	1.494 ***	0.233 ***
			(2.95)	(9.97)	(14.59)	(2.76)
Control	YES	YES	YES	YES	YES	YES
Industry-FE	YES	YES	YES	YES	YES	YES
Year-FE	YES	YES	YES	YES	YES	YES
Observations	38,416	32,193	34,799	38,416	38,416	38,416
R-squared	0.0926	0.0880	0.0928	0.1590	0.2403	0.1763

Robust z-statistics in parentheses. *** p < 0.01, ** p < 0.05.

. After introducing the control variables and simultaneously controlling for the industry and year fixed effects, DT* continues to exhibit a significant positive impact on corporate ESG performance, with a regression coefficient of 0.780.

(2)

One-period Lag

The implementation of corporate digital transformation policies typically requires time, as enterprises and organizations need to adapt to new technologies and processes. Consequently, the impact of digital transformation may only become evident after a certain period. Thus, this paper applies a one-period lag to the explanatory variable, with the results shown in column (2) of Table 4. The regression coefficient of the one-period lagged explanatory variable L-DT is 1.142, which is significant at the 1% level and consistent with the baseline regression results.

(3)

Changing the Sample Period

During the pandemic in 2020, China implemented relatively strict control policies, which significantly impacted the economy and society. All samples from 2020 were excluded, and a new regression analysis was conducted. The results are shown in column (3) of Table 4. Under the inclusion of the control variables and while still controlling for the industry and year fixed effects, the results remain largely consistent with the baseline regression.

(4)

Replacement of Explained variable

This study places the ESG ratings into three distinct dimensions—environmental (E), social (S), and governance (G)—to facilitate subsequent regression analysis. From an environmental perspective, digital transformation is associated with improved energy management and the optimization of energy consumption, thereby reducing carbon emissions per unit of income. From a social perspective, digital transformation necessitates that employees acquire new digital competencies. Corporate investment in employee training can enhance their technical capabilities and help them adapt to the rapidly evolving work environment. From a governance perspective, digital transformation is often accompanied by the adoption of information technologies, which enable more efficient energy management and lower overall energy consumption. The results are shown in columns (4) to (6) of Table 4. The regression coefficients of DT were 0.758, 1.494, and 0.233, respectively, all of which were significant at the 1% level and consistent with the baseline regression results.

4.4. Endogeneity Test

To address the potential endogeneity concerns stemming from the bidirectional relationship between DT and ESG performance, this study adopts the method proposed by Fang et al. [35], taking the mean DT level of other industry participants as an instrument for firm-level digital transformation. In terms of relevance, firms in the same industry face similar market environments, technological trends, and competitive pressures, leading to a certain synchronization in their digital transformation levels. Regarding exclusion, the digital transformation of other firms in the same industry should not directly affect the ESG performance of the focal firm. The digital transformation of peer firms primarily impacts the focal firm’s ESG performance through industry-wide spillover effects, such as knowledge diffusion or competitive mimicry, rather than through a direct causal channel. For instance, changes in other firms’ digital strategies do not directly modify the focal firm’s environmental management systems, social engagement initiatives, or governance structures. Furthermore, this indicator does not correlate with the control variables or random disturbance terms, satisfying the relevance and exclusion criteria for instrumental variables.

Additionally, to address the endogeneity issue caused by the omitted variables that are difficult to control, this paper constructs an interaction term between the historical telecommunications data of each province in 1984 and national information technology service revenue in 2022. After taking the natural logarithm, this interaction term serves as another instrumental variable for digital transformation, denoted as ln(IV). The 1984 telecommunications data belong to historical technological endowments established decades ago, which are not affected by the current ESG decisions of enterprises; while the macro industry trends reflected in the national information technology service revenue in 2022 are independent of the ESG behaviors of enterprises. Therefore, this interaction term is orthogonal to the omitted variables and random disturbance terms, enabling the isolation of the causal effect of digital transformation on ESG performance without interference from other factors.

Table 5. Endogeneity test.

	(1)	(2)	(3)	(4)
	First	Second	First	Second
Variables	DT	ESG	DT	ESG
IV1	0.002 ***
	(38.43)
Ln (IV2)			0.048 ***
			(74.61)
DT		1.064 **		2.979 ***
		(2.40)		(12.21)
Constant	5.298 ***	−0.991	4.827 ***	−11.091 ***
	(5165.89)	(−0.42)	(814.04)	(−8.63)
Control	YES	YES	YES	YES
Industry-FE	YES	YES	YES	YES
Year-FE	YES	YES	YES	YES
LM-test	985.721 ***		239.171 ***
	[0.000]		[0.000]
F-test	884.992		1871.174
Hausman Test	11.079 ***		11.491 ***
	[0.0009]		[0.0007]
Observations	35,175	35,175	30,350	30,350
R-squared	0.3646	0.1452	0.0771	0.1093

Robust z-statistics in parentheses. *** p < 0.01, ** p < 0.05.

reports the two-stage least squares (2SLS) regression results of the instrumental variables. Columns (1) and (2) present the regression results using the first instrumental variable, while columns (3) and (4) show the regression results using the second instrumental variable. The regression results confirm the validity of the two instrumental variables, demonstrating the robustness of the model after considering the endogeneity test.

4.5. Heterogeneity Test

Given that the mechanism through which digital transformation impacts corporate ESG performance may be influenced by variations in ownership structure, geographic location, and industry characteristics, conducting a heterogeneity test is theoretically essential to uncover the interaction between institutional context and technological effects. The detailed heterogeneity regression results are presented in

Table 6. Heterogeneity test.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
Variables	SOE	Non-SOE	Eastern	Central	Western	PE	Non-PE	Grow	Maturity	Decline
DT	0.295	1.152 ***	0.971 ***	0.116	0.352	0.109	0.937 ***	0.393 ***	0.191	0.272
	(1.32)	(6.32)	(5.80)	(0.32)	(0.96)	(1.61)	(6.04)	(2.68)	(0.49)	(0.94)
Constant	−1.342	−3.524 ***	−2.961 ***	0.201	−2.836	0.196	−3.236 ***	0.146	−0.047	−0.585
	(−1.20)	(−3.73)	(−3.43)	(0.11)	(−1.57)	(0.09)	(−4.10)	(0.22)	(−0.03)	(−0.44)
Control	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
Industry-FE	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
Year-FE	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
Observations	12,033	23,099	25,232	5433	4446	7658	27,474	26,451	4637	4044
R-squared	0.0608	0.0876	0.0666	0.0831	0.1210	0.0669	0.0710	0.0820	0.1260	0.1325

Robust z-statistics in parentheses. *** p < 0.01.

. Columns (1) and (2), respectively, show the regression results for state-owned enterprises (SOEs) and non-SOEs. It is evident that for non-SOEs, the level of corporate digital transformation (DT) has a significantly positive impact on ESG performance at the 1% significance level, with a regression coefficient of 1.152. For SOEs, while the regression coefficient remains positive, it does not reach statistical significance. Further analysis indicates that private enterprises, compared with SOEs, often face more severe market survival challenges and intense competitive pressures due to structural disadvantages stemming from a lack of institutional support [66]. Consequently, private enterprises tend to exhibit stronger innovation-driven traits and a greater propensity for proactive change. To address the legitimacy constraints in the capital market, private enterprises are inclined to enhance strategic flexibility through deeper digital transformation, thereby further improving their ESG performance.

In terms of regional heterogeneity, this study categorizes all enterprises into three regions—eastern, central, and western—based on their provincial locations. The regression results are displayed in columns (3)–(5). It is clear that only in the eastern region does DT significantly influence ESG, with a regression coefficient of 0.971 at the 1% significance level. Further analysis reveals that China’s eastern region, as the primary hub for listed private enterprises, benefits from its substantial economic scale and technology-intensive attributes, enabling stronger capital investment capabilities for digital transformation. Additionally, the pressing financing needs of these enterprises motivate them to optimize ESG performance to gain recognition in the capital market and alleviate financing constraints. Moreover, the eastern region implements a stricter environmental regulation system, complemented by a comprehensive framework of green incentive policies. This drives enterprises to incorporate environmental protection and green innovation into their long-term strategic development paradigms, effectively strengthening the role of DT in improving corporate ESG performance during dynamic evolution.

Furthermore, this paper classifies firms into heavy-polluting and non-heavy-polluting categories according to their pollution profiles and performs an industry-level heterogeneity analysis. As presented in columns (6) and (7) of Table 6, the estimated coefficient of digital transformation for non-heavy-polluting enterprises is 0.937 and is statistically significant at the 1% level. This finding implies that greater digital transformation among non-heavy-polluting enterprises is positively linked to improved ESG outcomes. Enterprises in non-heavy-polluting industries generally demonstrate higher baseline ESG ratings and adopt development models marked by low resource dependency (particularly traditional resources such as minerals), often concentrated in technology- and innovation-intensive sectors. From a resource-based view, these firms utilize advanced digital capabilities and manage dynamic financing constraints to create ESG premium effects through strategic digital transformation. This dual mechanism, driven by technology and capital, enables them to communicate sustainable development signals to the capital market by optimizing ESG performance, thereby building differentiated competitive advantages.

In terms of the heterogeneity of enterprise age, this study categorizes enterprises into three life cycle stages: growth, maturity, and decline. The regression results are reported in columns (8) through (10) of Table 6. The findings reveal that digital transformation (DT) exerts a statistically significant effect on ESG performance only for firms in the growth phase, with a coefficient of 0.393, which is significant at the 1% level. Further analysis indicates that enterprises in the growth stage are typically undergoing a period of rapid development, characterized by significant growth potential and operational flexibility. However, they also encounter challenges such as limited resources, unstable market positions, and underdeveloped management systems. Digital transformation plays a crucial role during this phase, as it enables firms to streamline operations, strengthen competitiveness, and ultimately enhance their ESG performance.

5. Further Discussion

5.1. Mediating Effect

To examine the mechanism proposed in the earlier theoretical analysis regarding how enterprise digital transformation influences ESG performance, this study draws on the perspective of Jiang Ting [67] and constructs an X-M model, refocusing the research emphasis on enhancing the credibility of the causal identification between X and Y. The impact of M on Y is informed by a comprehensive literature review, on which a recursive mediation model (2) is developed.

$$M_{i,t}={\alpha }_1+{\beta }_2{DT}_{i,t}+{\gamma }_2{Control}_{i,t}+\Sigma Year+\Sigma Industry+{\mathsf{\epsilon }}_{i,t}$$

(2)

This study uses $M_{i,t}$ to denote the mediating variables that will subsequently be incorporated into the model. If the explanatory variable significantly influences the mediating variable, it can be concluded that a mediating effect exists.

In line with the fundamental logic of the two-step approach, the detailed regression outcomes are reported in

Table 7. Mechanism test.

	(1)	(2)	(3)	(4)	(5)
Variables	ESG	EID	Tech-Conv	Resist	ESG
DT	0.765 ***	0.654 ***	3.591 ***	0.825 ***	0.268 *
	(5.44)	(16.94)	(23.56)	(18.57)	(1.91)
DID					−0.892 *
					(−1.75)
DID × DT					0.170 *
					(1.75)
Constant	−2.602 ***	−5.188 ***	−20.738 ***	−9.217 ***	0.303
	(−3.62)	(−31.36)	(−33.08)	(−45.21)	(0.45)
Control	YES	YES	YES	YES	YES
Firm-FE	NO	NO	NO	NO	YES
Industry-FE	YES	YES	YES	YES	YES
Year-FE	YES	YES	YES	YES	YES
Observations	35,132	35,132	35,132	35,132	38,416
R-squared	0.0702	0.3350	0.4600	0.4234	0.0930

Robust z-statistics in parentheses. *** p < 0.01, * p < 0.1.

. Column (1) presents the estimates without considering the mediation effect, indicating that the coefficient of DT is 0.765 and statistically significant at the 1% level.

Column (2) shows the regression results for the mediator EID, with the regression coefficient of DT being 0.654, which is significant at the 1% level, demonstrating that corporate DT can enhance environmental information disclosure levels, positively influencing corporate ESG performance. Specifically, there is a significant correlation between the market fairness of ESG ratings and the level of environmental information disclosure. The quality of corporate environmental information disclosure directly impacts stock price information content [68]. High-ESG-rated companies, aiming to strengthen their information transmission effect to boost market valuation, often proactively pursue digital transformation strategies. This digital empowerment not only optimizes the effectiveness of information disclosure mechanisms but also creates a positive feedback loop for ESG rating improvement by increasing information transparency and disclosure frequency.

Column (3) presents the regression results for Tech-Conv, where the regression coefficient of DT is 3.591, the largest among the coefficients, indicating that the mediating effect of the integration of digital and real economy technologies in DT and ESG performance is the most pronounced. As an endogenous outcome of corporate digitalization, the core attributes of the integration of digital and physical industry technologies play a pivotal role in advancing sustainability and promoting responsible business conduct within the real economy [14]. On the one hand, digital technologies constitute a necessary path dependency for enterprises to achieve excellent ESG performance evaluations. Traditional industries bridge the technological gap in digital transformation by adopting and applying advanced technologies from the digital sector, further improving operational efficiency and reducing information transmission costs. The resulting cost efficiency offers enterprises practical room to reallocate strategic resources, allowing them to transform marginal gains into focused investments across ESG dimensions, including environmental management, social responsibility implementation, and the enhancement of corporate governance structures.

Finally, combining the research conclusions of Zhang et al. [69] with the results shown in column (4), it is evident that enterprise digital transformation performance can significantly enhance supply chain resilience. Specifically, firms with a higher degree of digital transformation can construct more effective supply chain management frameworks by broadening their network of suppliers and customers, thereby mitigating supply chain concentration. This diversification not only reduces operational risks but also enhances the firm’s overall integration capabilities and strengthens its position within the supply chain ecosystem, ultimately contributing to greater supply chain resilience. In addition, digital transformation facilitates the reduction in information asymmetry between upstream and downstream operations across the supply chain. By leveraging big data analytics and blockchain technologies, transparent information platforms can be built, enhancing collaboration traceability while diversifying business partnerships. This breaks traditional relationship-based cooperation path dependencies and curbs opportunistic behaviors that erode ESG investments.

5.2. Moderating Effect

To examine whether the “Broadband China” strategy has a moderating effect, this study builds on the multi-period DID model by introducing moderating variables and constructing a moderating effect model, as detailed in Model (3).

$${ESG}_{i,t}={\alpha }_2+{\beta }_3{DT}_{i,t}+{\theta }_1{DT}_{i,t}\times {DID}_{i,t}+{\phi }_4{DID}_{i,t}+{\gamma }_5{Controls}_{i,t}+\Sigma Year+\Sigma Industry++\Sigma Firm{+\epsilon }_{i,t}$$

(3)

Model (3) introduces an interaction term between enterprise digital transformation (DT) and the “Broadband China” indicator variable (DID) to assess how this strategy moderates the influence of digital transformation on ESG performance.

Column (5) of Table 7 presents the interaction term between the dummy variable for the “Broadband China” strategy and the degree of enterprise digital transformation. The regression coefficient of the interaction term is 0.170 and is significant at the 10% level. This clearly demonstrates that the “Broadband China” strategy, as a key national information infrastructure initiative, exhibits significant moderating advantages in its interaction with enterprise digital transformation. By improving infrastructure such as high-speed networks and cloud computing, this strategy reduces the technical barriers and data circulation costs associated with digital transformation. In particular, by enhancing network coverage in remote areas, it alleviates regional disparities in digital development, enabling more enterprises to access digital technologies at reduced costs and laying the foundation for the broader accessibility of ESG management. In this process, the “Broadband China” strategy and digital transformation form a two-way synergy of “policy empowerment—technology implementation”. For instance, the integration of 5G networks with industrial internet platforms not only accelerates the R&D cycle for green technologies but also provides systematic support for ESG practices by optimizing data collaboration across the entire supply chain.

Figure 2 presents the dynamic policy effects of the “Broadband China” policy. The graph illustrates that the policy effect exhibits a lagged impact. It started to manifest in 2016, peaked in 2019, and subsequently declined gradually following the outbreak of the pandemic.

Table 8. Heterogeneity test of “Broadband China” strategy effect.

	(1)	(2)
Variables	PE	Non-PE
DT	1.122 ***	0.081
	[3.40]	[0.53]
DID	−0.173	−1.100 *
	[−0.15]	[−1.94]
DID × DT	0.013	0.216 **
	[0.06]	[2.00]
Constant	−3.466 **	1.161
	[−2.23]	[1.56]
Control	YES	YES
Firm-FE	YES	YES
Industry-FE	YES	YES
Year-FE	YES	YES
Observations	7658	27,474
R-squared	0.0755	0.1009

Robust z-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

presents the heterogeneity regression between heavily polluting and non-heavily polluting industries. The results, shown in columns (2) of Table 8, indicate that the regression coefficient of the interaction term is 0.216 and is significant at the 5% level. This explanation is consistent with what was said above. Non-heavily polluting enterprises typically attain higher ESG ratings and tend to rely on technological innovation rather than traditional resource inputs. These firms are often found in technology-intensive industries. From the perspective of resource dependence, they utilize advanced technologies and flexible financing mechanisms to enhance ESG value through digital strategies. The synergistic effect of technology and capital enables them to improve ESG performance, demonstrate long-term sustainability to the market, and thereby establish a distinct competitive advantage.

6. Conclusions and Recommendations

6.1. Conclusions

This study employs panel data from China’s A-share listed companies from 2012 to 2023 to empirically examine the multi-dimensional relationship between digital transformation and corporate ESG performance. By analyzing the full spectrum of business operations, it investigates the complex mediating mechanisms through which digital transformation influences ESG outcomes. Additionally, it explores—for the first time—the moderating role of the “Broadband China” strategy in this relationship. The findings offer key insights with both theoretical significance and practical relevance.

A robust and substantial positive association is identified between digital transformation and ESG performance. This relationship remains consistent across a range of robustness checks and endogeneity treatments, indicating that digital transformation is a vital driver of corporate sustainable development. This challenges the traditional view that digital technologies primarily impact financial outcomes, thereby enriching the literature on the non-financial implications of the digital economy.

Digital transformation improves ESG performance through three distinct yet parallel pathways: supply chain resilience, the integration of digital and physical industrial technologies, and environmental information disclosure. A quantitative comparison of these mechanisms is presented within a unified analytical framework for the first time, offering actionable guidance for optimizing the allocation of digital strategic resources. Furthermore, empirical evidence shows that the “Broadband China” initiative significantly amplifies the positive impact of digital transformation on ESG performance. This suggests that digital infrastructure policies not only promote economic development but also yield substantial environmental, social, and governance spillover benefits. These results contribute new insights into the comprehensive value of digital economy policies and provide empirical support for aligning national digital strategies with sustainable development goals.

The impact of digital transformation on ESG performance also exhibits considerable heterogeneity. It is notably stronger in eastern regions, non-state-owned enterprises, non-heavy-polluting industries, and growth stage enterprises, while weaker or statistically insignificant in central and western regions, state-owned enterprises, heavy-polluting sectors, and non-growth stage enterprises. This pattern reveals structural disparities in China’s ESG landscape, emphasizing the differentiated effects of digital transformation across institutional and regional settings. It offers a fresh perspective for understanding the uneven pathways of ESG advancement in China.

6.2. Recommendations

The empirical results of this study offer valuable implications for promoting the integrated advancement of digitalization and ESG performance within Chinese firms. Drawing on these findings, the following strategic policy suggestions are put forward.

At the government level, establishing a comprehensive institutional framework has become an urgent priority. Policymakers should prioritize the development of digital infrastructure, particularly through targeted investments in underdeveloped regions, and provide industry-specific support for pollution-intensive sectors to address regional disparities. The existing “Broadband China” policy framework requires significant improvements to incorporate advanced digital application scenarios and implement differentiated incentive mechanisms tailored to the varying levels of technological readiness across different industrial sectors. A standardized ESG disclosure system should be established at the national level to ensure the verifiability and auditability of data. Fiscal policy tools need to be carefully calibrated to incentivize digital–ESG integration, including progressive tax incentives linked to quantifiable improvements in sustainability performance. Additionally, establishing dedicated research funding programs can drive innovation in digital solutions for key ESG challenges, particularly through public–private research collaboration models.

Industry associations should play a more proactive role in promoting cross-industry knowledge sharing and technology transfer. There is an urgent need to develop industry-specific digital transformation guidelines to enhance ESG performance, which requires collaboration among multiple stakeholders. Creating a digital platform for the entire industry will facilitate the faster dissemination of best practices and effective implementation methods. It is particularly important to establish standard measurement systems and assessment tools to accurately evaluate the impact of digital transformation on ESG performance, enabling meaningful comparisons and the tracking of results. Industry organizations should provide comprehensive training programs, including technical workshops, management training, and peer exchange opportunities, to help members better understand and apply new technologies. Cross-industry partnerships can also promote innovation by bringing together technology companies and traditional enterprises to jointly develop customized ESG solutions.

Corporate organizations should implement a more systematic strategy for embedding digital transformation within their ESG management systems. This requires systematically incorporating digital elements into core ESG strategy development processes, supported by dedicated organizational units with clear accountability mechanisms. From an investment perspective, priority should be given to implementing comprehensive monitoring systems to track key ESG metrics across the entire value chain. These systems should leverage advanced technological solutions such as IoT networks, remote sensing capabilities, and predictive analytics platforms. Workforce development strategies must be strengthened through targeted skill enhancement programs and specialized recruitment initiatives to cultivate digital–ESG integrated capabilities. Executive compensation schemes should incorporate specific digital ESG performance metrics to ensure the alignment of incentive mechanisms. Companies can accelerate progress by establishing innovation incubators and strategic technology partnerships. These initiatives can facilitate piloting and scaling new digital solutions addressing ESG challenges. Most importantly, organizations must establish robust impact measurement frameworks to quantitatively demonstrate the ESG value generated by digital transformation investments, thereby building a strong business case for continuing this strategic integration.

6.3. Limitations and Future Research

Although this study provides empirical evidence on the relationship between digital transformation and ESG performance for Chinese A-share listed companies, several limitations must be acknowledged. First, this study may suffer from sample selection bias. On a broader scale, there may be no significant positive correlation between digital transformation and ESG performance [27]. Therefore, this study is primarily applicable to the Chinese capital market context, and its generalizability to other emerging or developed economies requires further validation. Second, ESG data primarily relies on third-party rating agencies, and differences in indicator weights and assessment methods among these agencies may introduce measurement errors. Although measurements based on the MSCI ESG Index also show a positive correlation between digital transformation and ESG performance in Chinese companies [70], the ESG metrics in these reports may be subject to “green washing” [71]. Therefore, future research could expand to cross-country comparative analyses and utilize multiple ESG data sources for measurement. Employing natural experiment methods could further strengthen causal inference. These improvements will help build a more universally applicable theoretical framework.