ESG Performance and Digital Transformation: Evidence from Chinese A-Listed Companies

Abstract

Digital transformation is increasingly recognized as a key mechanism for enterprise upgrading, facilitating innovation, enhancing resource efficiency, and sustaining competitive advantage. This study investigates the influence of environmental, social, and governance (ESG) performance on the digital transformation of China’s new energy enterprises. Drawing on panel data from A-share listed firms from 2010 to 2023, the analysis assesses both the direct effect of ESG performance and the mediating role of dynamic capabilities, including absorptive, adaptive, and innovative capacities. The empirical results yield three key findings. First, superior ESG performance significantly advances digital transformation by expanding firms’ resource bases and supporting technological renewal. Second, the effect is more pronounced in mature firms and regions with supportive institutional environments and attenuated in early-stage or resource-constrained contexts. Third, the mediation analysis confirms that ESG fosters transformation by strengthening dynamic capabilities. These findings underscore the strategic role of ESG in enabling digital transformation and offer theoretical and practical insights for firms pursuing sustainability-oriented transformation.

1. Introduction

The global energy transition and carbon neutrality targets have elevated the strategic importance of the new energy industry. As a high-tech and capital-intensive sector, new energy enterprises are crucial for reducing dependence on fossil fuels, building low-carbon and efficient energy systems, and supporting sustainable development. In China, policies including the 14th Five-Year Plan for the Modern Energy System and the Carbon Peaking Action Plan before 2030 have emphasized technological innovation and the gradual establishment of a power system dominated by renewable energy. These policies position new energy firms as central to national sustainability goals, while simultaneously intensifying demands for technological breakthroughs and efficiency improvements. Digital transformation has emerged as a crucial pathway for new energy enterprises to enhance their competitiveness. For example, State Grid’s smart-grid projects demonstrate how digital integration enables large-scale renewable energy deployment while ensuring system stability [1,2]. Similarly, CATL has developed a digitalized battery management platform that supports real-time monitoring and intelligent scheduling, effectively extending battery life and improving energy efficiency [3,4]. Although many Chinese new energy enterprises have improved factor productivity under digital transformation, many new energy firms still face significant obstacles. These include financing constraints, heavy R&D expenditures, and uncertainties associated with rapidly evolving technologies [5,6]. Recent evidence highlights these bottlenecks. For instance, CATL’s 2024 annual report discloses multi-billion RMB investments in R&D, with its R&D-to-revenue ratio continuing to rise, underscoring sustained innovation pressure in the battery and storage sectors [7]. Moreover, the integration of smart-grid technologies is complicated by frequent technological iteration and evolving regulatory standards, which create long-term strategic uncertainty [2]. These cases collectively confirm that financing frictions, R&D intensity, and technological uncertainty jointly constitute the major barriers to digital transformation in the new energy industry. In this context, Environmental, Social, and Governance (ESG) practices are increasingly recognized as a strategic driver of digital transformation. Strong ESG performance enhances stakeholder trust, facilitates resource acquisition, and improves collaboration within industrial ecosystems [8]. For new energy enterprises, these advantages help alleviate barriers to digital transformation and support the integration of internal and external resources. Thus, ESG performance is expected to play a crucial role in enabling new energy firms to achieve digital transformation while advancing environmental sustainability.

However, current research perspectives remain limited. Most existing studies focus on digital transformation and innovation in the manufacturing and service sectors. These studies emphasize how digital technologies improve production efficiency, optimize service models, and enhance market competitiveness [9,10]. In contrast, the new energy sector has received relatively little systematic attention. This sector is characterized by high technological intensity and strong policy dependence. Compared with traditional industries, new energy enterprises face higher R&D investment requirements and greater exposure to uncertainty. At the same time, they carry the critical mission of contributing to the national “dual carbon” goals [11]. These distinctive features suggest that the relationship between ESG performance and digital transformation in the new energy sector may involve unique mechanisms. This issue merits further academic investigation.

To address these gaps, this study investigates the effect of ESG performance on digital transformation in Chinese new energy enterprises. Using panel data from A-share listed firms between 2010 and 2023, this study employs ESG ratings from the CNRDS database as the core explanatory variable. A two-way fixed effects model is applied to examine the relationship between ESG performance and digital transformation, controlling for both firm–year-specific unobserved heterogeneity. In addition, the study explores the mechanisms through which ESG influences digital transformation, focusing on absorptive capacity, adaptive capability, and innovation capability as dynamic channels. Theoretically, this study extends the resource-based and capability perspectives by conceptualizing ESG performance as a tool for enhancing firms’ resource capabilities. ESG practices strengthen the development of dynamic capabilities—such as absorptive capacity, adaptive capability, and innovation capability—that enable firms to integrate and reconfigure resources in response to environmental and technological challenges. Through this process, ESG becomes a strategic driver that facilitates digital transformation. This perspective provides a deeper understanding of how sustainability-oriented practices are converted into transformation.

The structure of the paper is as follows: Section 2 reviews related literature and develops the theoretical framework. Section 3 introduces data sources, variable definitions, and methodology. Section 4 presents empirical results, robustness tests, heterogeneity analysis, and mechanism verification. Section 5 concludes with the main findings, theoretical and practical implications, limitations, and future research directions.

2. Literature Review and Theoretical Hypothesis

2.1. Existing Literature Review

The concept of ESG originates from the recognition of sustainable development, covering three dimensions: environmental, social, and governance responsibilities. It represents a concretization and institutionalization of Corporate Social Responsibility (CSR), extending its connotations into the domains of greening [12]. Existing studies highlight that ESG has become an important pathway for firms to obtain external resources and enhance competitiveness. The literature generally identifies three perspectives on the role of ESG.

First, ESG practice is regarded as a process of capability enhancement. Firms engaging in ESG activities strengthen their dynamic capabilities to adapt to environmental changes and improve competitiveness [13,14]. Second, ESG performance is viewed as a feedback mechanism that reflects the outcomes of practice, functioning as a corporate signal that enhances reputation, increases information transparency, and influences stakeholder decision-making [15,16]. Third, from the resource-based perspective, ESG performance is considered a valuable strategic resource. High ESG scores allow firms to accumulate reputation capital, attract talent, and access financing and policy support, thereby alleviating resource constraints for innovation activities [17,18,19].

Meanwhile, digital transformation is characterized by knowledge intensity, dependence on infrastructure, and strong spillover effects [20,21]. Successful digital upgrading requires firms to acquire and integrate multiple types of resources, including skilled human capital, advanced digital technologies and data, financial investment, policy incentives, and collaborative networks within the industrial ecosystem.

Existing research has linked ESG to firm value [22], general innovation [23,24], and overall digital transformation [25]. However, relatively few studies have examined how ESG affects the specific process of digital transformation, which differs from other forms of innovation due to its auxiliary and spillover characteristics. Moreover, the current literature mainly focuses on manufacturing and service industries, while the renewable energy sector has received limited attention. Given its high dependence on policy, substantial R&D intensity, and strict environmental constraints, the renewable energy industry provides a distinct context for understanding the role of ESG in digital transformation.

2.2. Theoretical Hypothesis

In new energy enterprises, digital transformation is not merely a technological upgrade but a strategic response to dynamic markets, policy requirements, and stakeholder expectations [8,9]. Achieving transformation requires sustained resource investment, supply chain coordination, and the ability to cope with uncertainties [13]. Corporate ESG performance can serve as a critical driver of dynamic capabilities, thereby facilitating digital transformation. First, strong ESG performance reflects a long-term commitment to environmental and social responsibility [26]. This strengthens stakeholder trust and improves access to key resources such as financing, talent, and partnerships, thus alleviating resource constraints and laying a foundation for digital upgrading. In addition, ESG engagement signals a proactive orientation toward sustainability, which aligns with the strategic requirements of digitalization in the new energy sector [27]. In this sense, ESG performance is not only an issue of compliance or reputation but also a capability-enhancing mechanism that enables firms to sense, seize, and transform in the digital era. Accordingly, Hypothesis 1 is formulated:

H1. 

Corporate ESG performance has a significantly positive impact on digital transformation.

Fulfilling ESG responsibilities requires firms to balance stakeholder demands across environmental, social, and governance dimensions. For new energy enterprises, this engagement fosters adaptive capability, enabling them to coordinate resources internally and respond flexibly to external pressures. On the one hand, ESG practices strengthen external linkages and social networks, which support firms in obtaining critical resources and knowledge for digital innovation [26,28]. On the one hand, ESG engagement shapes managerial cognition of resources and enhances firms’ ability to optimize internal resource utilization through improved processes and technological upgrading [23]. ESG orientation requires firms to adopt a long-term perspective and manage emerging social and environmental risks. By improving governance structures and operational processes, firms can adjust strategies in response to regulatory pressures such as carbon reduction and data governance [29]. This adaptive capability helps firms mitigate uncertainty, recognize new opportunities, and reallocate resources toward sustainable technological upgrading.

H2. 

ESG performance enhances adaptive capability, thereby promoting digital transformation in new energy enterprises.

High sensitivity to emerging social and environmental issues is a core feature of ESG responsibilities. For new energy enterprises, engagement in ESG practices stimulates interactions with multiple stakeholders and facilitates the exchange of knowledge and experiences, which strengthens absorptive learning capability [30]. Guided by ESG principles, firms place greater emphasis on external knowledge acquisition and internal knowledge assimilation, particularly concerning issues such as data ethics, privacy, and sustainability in digital improvement [24,30]. This absorptive capability enables firms to identify risks associated with digital innovation, expand their knowledge base, and adjust strategies in line with evolving technological demands and social expectations [29]. Consequently, new energy enterprises can transform market requirements for sustainability into drivers of digital solutions.

H3. 

ESG performance improves absorptive capability, thereby facilitating digital transformation in new energy enterprises.

Engagement in ESG practices contributes to the development of firms’ innovation capability. For new energy enterprises, ESG responsibilities require continuous investment in R&D and the exploration of sustainable technologies, which stimulates the creation of new products, processes, and business models [29]. Strong ESG performance signals long-term commitment to sustainability, which motivates firms to allocate more resources to innovation-oriented activities [31]. This improves the efficiency of R&D utilization and encourages the transformation of environmental and social requirements into technological solutions. By enhancing innovation capability, ESG performance enables firms to transform external pressures into internal drivers of digital transformation [30].

H4. 

ESG performance enhances innovation capability, thereby fostering digital transformation in new energy enterprises.

In summary, Hypotheses H1–H4 constitute the theoretical framework of this study, which is illustrated in Figure 1.

3. Methodology Adopted

3.1. Data Collection and Sample Selection

The data are obtained from the “New Energy” concept sector of the Wind Financial Terminal, which provides an initial sample of 149 listed new energy companies in China. The annual reports of these firms are then examined to refine the sample based on their main business or core products. Firms classified as ST or *ST, as well as those with abnormal financial conditions such as a leverage ratio exceeding 80 percent, are excluded. To reduce the influence of outliers on regression results, the main continuous variables are winsorized at the 1% on both tails. Data on corporate digital transformation is collected from annual reports of listed companies and is available through the CSMAR (China Stock Market & Accounting Research) database. ESG data are obtained from the ESG ratings of listed companies published in the CNRDS (China National Research Data Service Platform) database. Financial and governance data are collected from both the CSMAR database and the Wind Financial Terminal databases. All statistical analyses are conducted using Stata/SE 17.0 (StataCorp LLC, College Station, TX, USA).

3.2. Definition of Variables

ESG Performance (ESG): The ESG data are obtained from the ESG ratings in the CNRDS database, which consist of three components: environmental, social, and governance. The database also discloses ESG scores and rankings. This study selects the ESG score and applies a logarithmic transformation after adding one, forming the main explanatory variable ESG.

Digital Transformation (DT): This study measures DT through text analysis, following established approaches in the literature [32,33]. Keywords, including big data, information technology, intelligence, robotics, Internet of Things, blockchain, automation, digitalization, and cloud computing, are identified and used to calculate the frequency of occurrence in corporate reports. The resulting frequency counts are adjusted by adding one and subsequently transformed into their natural logarithm to construct DT.

To accurately assess the impact of corporate ESG performance on digital transformation, this study controls for macroeconomic factors and firm characteristics that may affect enterprise digital transformation. On the macroeconomic side, factors such as industry characteristics (Industry) and time trends (Year) are controlled for their potential impact. In the regression analysis, this study controls for firm-specific characteristics as follows. Firm size (Size) is measured by the natural logarithm of total assets. Firm age (lnAge) is calculated as the natural logarithm of the number of years since establishment. Profitability (ROA) is represented by the ratio of net income to total assets, reflecting a firm’s ability to generate returns from its resources. Financial leverage (Lev) is measured by the ratio of total liabilities to total assets, which captures financial risk and flexibility. Ownership concentration (Top1) is defined as the shareholding ratio of the largest shareholder, reflecting potential influence on governance and strategic decisions. The number of independent directors (lninde) is measured by the natural logarithm of the number of independent directors, indicating the level of board independence. Enterprise value (TobinQ) is measured by the ratio of the firm’s market value to its total asset value, reflecting market expectations of future growth and investment opportunities.

3.3. Model Design

To evaluate the impact of corporate ESG performance on digital transformation, this study constructs the following model for benchmark regression analysis:

$${\mathit{D}\mathit{T}}_{\mathit{i}\mathit{t}}={\mathit{a}}_0+{{\mathit{a}}_1\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i}\mathit{t}}+{\mathit{\alpha }}_2{\sum \mathit{C}\mathit{o}\mathit{n}\mathit{t}\mathit{r}\mathit{o}\mathit{l}\mathit{s}}_{\mathit{i}\mathit{t}}+{\mathit{\gamma }}_{\mathit{i}}+{\mathit{\mu }}_{\mathit{t}}++{\mathit{\epsilon }}_{\mathit{i}\mathit{t}}$$

(1)

In the regression equation, $i$ denotes the firm, and $t$ denotes the year. The variable, ${\mathit{D}\mathit{T}}_{\mathit{i}\mathit{t}}$ represents the DT level for company $i$ in year $t$. ${\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i}\mathit{t}}$ indicates the ESG performance of firm $i$ in year $t$. ${Controls}_{it}$ denotes a set of control variables. The terms, ${\mathit{\gamma }}_{\mathit{i}}$ and ${\mathit{\mu }}_{\mathit{t}}$ capture firm-specific and time-fixed effects, respectively, while ${\mathit{\epsilon }}_{\mathit{i}\mathit{t}}$ is the random disturbance term, and $a_0$ is the constant term. The key coefficient is $a_1$, which captures the impact of ESG on DT. A significantly positive ${\alpha }_1$ indicates that a better ESG is associated with higher DT level, thereby providing empirical support for the study’s theoretical framework.

4. Results

4.1. Descriptive and Correlation Analysis

Table 1. Results of descriptive statistics.

Variable	N	Mean	Median	SD	Min	Max
DT	2331	0.0390	0.0200	0.0650	0	0.500
ESG	2331	26.72	24.83	10.24	3.316	76.63
lnAge	2331	2.394	2.565	0.767	0	3.526
Size	2331	22.78	22.65	1.363	18.35	27.25
ROA	2331	0.0180	0.0250	0.107	−2.555	0.863
Lev	2331	0.530	0.540	0.208	0.0270	3.262
Top1	2331	0.324	0.308	0.148	0	0.761
Lninde	2331	0.853	0.405	0.638	0	2.485
SOE	2331	0.428	0	0.495	0	1
Duality	2331	0.238	0	0.426	0	1
TobinQ	2331	1.677	1.377	1.044	0.670	14.57

presents the descriptive statistics for the main variables, reporting the mean, median, standard deviation, minimum, and maximum values. The dataset consists of 2331 firm-year observations from an unbalanced panel covering Chinese listed firms.

For the dependent variable, the mean and median values of DT are 0.039 and 0.020, respectively, with a minimum of 0 and a maximum of 0.500. The standard deviation of 0.065 indicates that digital innovation output remains relatively low overall, with limited variation across firms. The key explanatory variable, ESG, has a mean value of 4.100, a median of 4, and ranges from 1 to 7, with a standard deviation of 0.978. This suggests substantial heterogeneity in ESG practices across firms, reflecting differences in the extent to which ESG principles are adopted and implemented.

The control variables include firm age (lnAge), firm size (Size), profitability (ROA), financial leverage (Lev), ownership concentration (Top1), board independence (Lninde), state ownership (SOE), CEO duality (Duality), and firm value (TobinQ). These variables show distributions consistent with theoretical expectations. Overall, the descriptive statistics confirm that the dataset is well-structured and does not exhibit severe outliers, providing a reliable basis for subsequent regression analysis.

4.2. Benchmark Regression

Table 2. Results for ESG and DT.

Variable	(1)	(2)	(3)
	DT	DT	DT
ESG	0.034 ***	0.010 ***	0.006 ***
	(11.94)	(3.84)	(2.88)
lnAge			−0.068
			(−0.86)
Size			0.336 ***
			(6.84)
ROA			−0.121
			(−0.59)
Lev			−0.423 **
			(−2.15)
Top1			0.174
			(0.68)
Lninde			−0.108 *
			(−1.70)
SOE			−0.066
			(−0.53)
Duality			−0.092
			(−1.55)
TobinQ			0.010
			(0.46)
_cons	1.814 ***	1.831 ***	−4.654 ***
	(19.67)	(23.75)	(−4.67)
Fixed effect	No	Yes	Yes
N	2394	2394	2285
adj. R2	0.138	0.366	0.713

Note: The values in parentheses represent the corresponding t-values for each coefficient. ***, **, * indicate significance at the 0.1, 0.05, and 0.01 levels, respectively. The same convention applies to the following tables.

reports the regression results of ESG on DT using different model specifications. Columns (1) and (2) present benchmark results without and with fixed effects, respectively. In both cases, the coefficient of ESG remains significantly positive at the 1% level, indicating that higher ESG scores are associated with greater DT output. Column (3) incorporates a two-way fixed effects specification, controlling for both firm- and year-level unobserved heterogeneity. The results show that ESG continues to exert a positive and significant impact on DT, with a coefficient of 0.006 at the 1% significance level. This provides strong evidence that the positive relationship between ESG and DT is robust even after accounting for potential confounding effects. Overall, these findings suggest that ESG performance plays an important role in promoting firms’ digital transformation activities and serves as a driving force for sustainable digital improvement.

4.3. Endogeneity Test

Omitted variables and reverse causality often cause endogeneity problems. These issues may weaken the reliability of regression results. To address this concern, this study applies three methods: instrumental variable (IV) estimation, the SYS-GMM model, and the propensity score matching (PSM) method. The results of the robustness tests are summarized in

Table 3. Endogeneity test results.

	(1)	(2)	(3)
Variable	First-ESG	Second-DT	DT
Peer	0.417 ***
	(21.875)
ESG		0.013 **	0.031 *
		(2.491)	(0.018)
L.DT			0.914 ***
			(0.103)
Constant	−18.294 ***	−1.735 ***
	(−4.683)	(−3.876)
Controls	Yes	Yes	Yes
FE	Yes	Yes	Yes
Cragg-Donald Wald F	242.657 {16.38}		—
Hansen J	p < 0.01		—
AR(1) Test	—	—	2098
AR(2) Test	—	—	0.001
Hensen Test	—	—	0.116
N	2134	2134	0.900
adj. R2	—	0.101	—

Note: The values in parentheses represent the corresponding t-values for each coefficient. ***, **, * indicate significance at the 0.1, 0.05, and 0.01 levels, respectively. The same convention applies to the following tables.

.

Firstly, the instrumental variable method. The lagged industry–year average of ESG ratings (Peer) is selected as the instrument. Columns (1) and (2) in Table 3 report the IV results. The first-stage regression shows that the instrument is valid. In the second stage, the impact of ESG on DT remains significant. The Cragg-Donald statistic value is greater than the critical value 16.38, which rules out the problem of weak instruments. The Hansen test supports the exogeneity of the instrument. These results suggest that the conclusion has strong causal explanatory power.

Secondly, the SYS-GMM model. As shown in column (3), the sample passes the AR(1), AR(2), and Hansen tests, satisfying the assumptions of first-order autocorrelation, no second-order autocorrelation, and no over-identification. This confirms the validity of the model specification. The regression coefficient of ESG remains significantly positive under the SYS-GMM dynamic panel estimation, which is consistent with the baseline regression results.

Thirdly, the PSM estimation. The treatment group is defined as firms with higher ESG performance, while the control group consists of firms with lower ESG performance. The propensity scores are estimated through a logit model, with all the covariates in Model 1, while controlling for industry and year effects. Figure 2 reports the standardized mean differences in covariates before and after matching, showing that most covariates approach zero after matching.

Table 4. Results of PSM estimation.

Variable	Sample	Treated	Controls	Difference	S.E.	t-Statistic
lndt	Unmatched	2.7894	2.6531	0.1363	0.0452	3.02
lndt	ATT	2.7867	2.7102	0.0765	0.0457	1.68

reports the average treatment effect on the treated (ATT). Before matching, the difference in digital transformation between the treatment and control groups was 0.136, which was significant at the 1% level. After matching, the ATT remained positive at 0.076, with a t-statistic of 1.68, indicating that firms with higher ESG performance exhibit a higher level of digital transformation compared to their matched counterparts. Although the level of significance is reduced after matching, the positive ATT supports the baseline findings and confirms that the effect of ESG on digital DT is robust after addressing potential sample selection bias.

In addition, to ensure the scientific validity of variable selection, the following procedures were conducted: Firstly, the time window was adjusted. To mitigate the potential impact of reverse causality on the regression results, the lagged value of ESG was used as the explanatory variable to re-estimate Model (1). Secondly, the core explanatory variable was replaced. Given that ESG rating systems vary across different agencies and that differences exist in the weighting of indicators, the Huazheng ESG rating system for Chinese enterprises was adopted for re-examination. Both the mean (HZH_mean) and median (HZH_median) values of Huazheng ESG ratings were included in the regression models. Thirdly, the dependent variable was replaced to address potential concerns regarding measurement validity. In the baseline analysis, DT is measured with the logarithmic frequency of digitalization-related terms in annual reports of listed firms. Recognizing that text-based proxies may have limitations in precision and contextual interpretation, Digitalrate is introduced as the alternative measurement, defined as the ratio of digital intangible assets to total intangible assets. The results based on this alternative specification remain consistent in direction and significance, confirming the robustness of the baseline findings. The robustness test results are reported in

Table 5. Robustness test results.

	(1)	(2)	(3)	(4)	(5)
Variable	DT	DT	DT	DT	Digitalrate
L.ESG	0.009 ***
	(4.08)
L2. ESG		0.007 **
		(2.52)
HZH_mean			0.058 *
			(1.92)
HZH_median				0.058 **
				(2.18)
ESG					0.001 **
					(2.52)
_cons	−4.520 ***	−4.514 ***	−3.151 ***	−4.673 ***	−5.712 ***
	(−4.18)	(−4.16)	(−3.30)	(−2.74)	(−2.91)
FE	Yes	Yes	Yes	Yes	Yes
N	Yes	Yes	Yes	Yes	Yes
adj. R2	2220	2220	1741	2008	1823

Note: The values in parentheses represent the corresponding t-values for each coefficient. ***, **, * indicate significance at the 0.1, 0.05, and 0.01 levels, respectively. The same convention applies to the following tables.

. The findings are consistent with the baseline regression results, further confirming the reliability of the benchmark model.

4.4. Mechanism Channels Analysis

The study elaborates on the theoretical mechanism underlying the relationship between ESG and DT, and empirical analysis was conducted to test its impact. Building on this foundation, the mechanism analysis further explores the pathways through which ESG facilitates DT. During the practice of ESG, different dimensions of dynamic capabilities can be enhanced, thereby promoting DT. Following the two-step approach proposed, Model (2) is constructed as follows:

$${Mechanism}_{it}={\beta }_0+{{\beta }_{1}ESG}_{it}+{\beta }_2{\sum Controls}_{it}+{\gamma 2}_i+{\mu 2}_t++{\epsilon 2}_{it}$$

(2)

where ${Mechanism}_{it}$ represents absorptive capacity (Absorb), adaptive capacity (Adapt), and innovation efficiency (Innovation). The other parameters remain consistent with the baseline regression settings. The test results are summarized in

Table 6. Results of Mechanism tests.

	(1)	(2)	(3)
Variables	Absorb	Adapt	Innovation
ESG	0.003 *	0.001 *	0.001 **
	(1.78)	(1.75)	(2.27)
_cons	−0.540	0.067	−0.315 ***
	(−0.74)	(1.50)	(−3.54)
Controls	Yes	Yes	Yes
FE	Yes	Yes	Yes
N	2272	2272	2154
adj. R2	0.505	0.718	0.508

Note: The values in parentheses represent the corresponding t-values for each coefficient. ***, **, * indicate significance at the 0.1, 0.05, and 0.01 levels, respectively. The same convention applies to the following tables.

.

4.4.1. Absorbative Capability

To examine the mediating role of absorptive capacity, this study introduces Absorb as the mediating variable. The ratio of firms’ R&D investment to operating revenue is used as a proxy to measure absorptive capacity. A higher value of this indicator reflects stronger absorptive capacity.

The regression results in Column (1) of Table 6 show that ESG has a significant positive effect on Absorb, with a coefficient of 0.003, significant at the 10% level. This indicates that firms with higher ESG performance tend to have stronger absorptive capacity, reflected in a higher ratio of R&D investment to operating revenue. Li et al. (2024) [30] demonstrated that absorptive capacity plays a positive role in promoting digital transformation, and the present findings further confirm its role as a mediating channel. Specifically, firms with stronger ESG performance place greater emphasis on long-term development goals and are more likely to allocate resources to sustainable transformation activities [34]. At the same time, ESG-oriented firms are generally more sensitive to technological changes and external knowledge, which enables them to better identify, absorb, and utilize new technologies. Therefore, ESG performance enhances firms’ absorptive capacity, thereby laying a solid foundation for digital transformation.

4.4.2. Adaptive Capability

This study introduces adaptive capacity as a mediating variable. The five-year moving standard deviation of firms’ profit margins is used as a proxy for adaptive capacity, denoted as Adapt. A higher value of this indicator reflects a stronger ability to adapt to external environmental changes.

Column (2) of Table 6 reports that ESG has a significantly positive effect on Adapt, with a coefficient of 0.001, significant at the 10% level. Stronger adaptive capacity enables firms to adjust strategies and reallocate resources promptly in response to technological upgrades and rapid shifts in the industrial environment, thereby reducing risks associated with uncertainty [35]. For digital transformation, an improvement in adaptive capacity allows firms to adopt and apply new technologies more efficiently and to adjust organizational processes accordingly, ensuring the smooth progress of digital transformation [30]. Thus, enhancing adaptive capacity strengthens firms’ resilience while providing sustained momentum and support for digital transformation.

4.4.3. Innovation Capability

To examine the mediating role of innovation capability, this study introduces Innovation as the mediating variable. The ratio of the natural logarithm of the total number of invention, utility model, and design patent applications plus one to firms’ R&D expenditure is used as a proxy to measure innovation capability. A higher value of this indicator reflects stronger innovation capability.

The regression results in Column 3 of Table 6 show that ESG has a significant positive effect on Innovation, with a coefficient of 0.001, significant at the 5% level. This result indicates that firms with higher ESG performance tend to have stronger innovation capability, measured by patent output relative to R&D expenditure. Previous studies suggest that innovation capability is an important driver of digital transformation [36,37]. It allows firms to convert R&D investment into technological achievements and organizational improvements [6]. The findings of this study further confirm its role as a mediating channel. Firms with stronger ESG performance are more likely to pursue sustainable innovation, increase the efficiency of R&D use, and produce higher quality patents [38]. The innovation outcomes enhance firms’ technological competitiveness and provide the technological basis for digital transformation.

4.5. Heterogeneity Test

To further test whether firm differences affect the baseline results, this study conducts heterogeneity analysis. Corporate strategies and transformation paths are shaped by development stage and geographic location. Firms at different stages vary in resources, governance, and innovation capability, which may lead to different effects of ESG on DT. Geographic location reflects institutional environment, policy support, and market conditions. These regional differences may also influence how ESG performance affects digital transformation. Based on this, heterogeneity tests are conducted from the perspectives of development stage and geographic location.

4.5.1. Heterogeneity in Development Stage

To test the heterogeneity of the impact of ESG on DT at different stages of firm development, the sample is divided into three groups: growth stage (Life = 0), maturity stage (Life = 1), and decline stage (Life = 2). The classification of firm development stages follows the approach of Dickinson [39]. The regression results are reported in Columns (1)–(3) of

Table 7. Heterogeneity in development stage.

	(1)	(2)	(3)
Variables	Life = 0	Life = 1	Life = 2
ESG	0.014	0.004 **	0.005
	(1.28)	(2.08)	(1.11)
_cons	1.423	−2.558	−5.766 *
	(0.39)	(−1.13)	(−1.74)
Controls	Yes	Yes	Yes
FE	Yes	Yes	Yes
N	306	936	712
adj. R2	0.666	0.660	0.698

Note: The values in parentheses represent the corresponding t-values for each coefficient. **, * indicate significance at the 0.05 and 0.01 levels, respectively. The same convention applies to the following tables.

.

The results show that ESG has a significant positive effect on DT during the maturity stage, while the effects in the growth and decline stages are not statistically significant. This suggests that firms in the maturity stage are better positioned to translate ESG performance into digital transformation outcomes. One possible explanation is that mature firms have more stable resources, stronger governance structures, and clearer strategic orientations, which enhance their ability to integrate ESG practices with digital innovation. By contrast, firms in the growth stage face resource constraints, and firms in the decline stage face survival pressures, both of which may limit the role of ESG performance in driving digital transformation.

4.5.2. Heterogeneity in Geographic Location

Regional characteristics constitute an important external factor shaping corporate activities. Firms located in regions with more favorable conditions are more likely to attract capital inflows and benefit from a supportive business environment. In this study, the variable Area is defined according to whether a firm is registered in a Yangtze River Economic Belt city. Firms located within the Yangtze River Economic Belt are assigned a value of 1, and those outside the region are assigned a value of 0. In addition, the level of digital infrastructure is considered, as it provides critical support for digital transformation. Cities designated as part of the “Broadband China” pilot program are identified as having relatively advanced digital infrastructure and are assigned a value of 1, while all other cities are assigned a value of 0. This variable is denoted as Digital. Based on these classifications, grouped regressions are conducted using Model 1, and the results are presented in

Table 8. Heterogeneity in geographic location.

	(1)	(2)	(3)	(4)
Variable	Area = 1	Area = 0	Digital = 1	Digital = 0
ESG	0.010 ***	0.003	0.004	0.009 **
	(3.11)	(0.80)	(1.56)	(2.32)
_cons		0.007 **
		(2.52)
Controls	Yes	Yes	Yes	Yes
FE	Yes	Yes	Yes	Yes
N	950	1270	1468	751
adj. R2	0.712	0.713	0.723	0.683

Note: The values in parentheses represent the corresponding t-values for each coefficient. ***, ** indicate significance at the 0.1 and 0.05 levels, respectively. The same convention applies to the following tables.

.

Table 8 reports the heterogeneity tests based on geographic location. Columns (1) and (2) present the results for firms located inside and outside the Yangtze River Economic Belt. The coefficient of ESG is significantly positive for firms within the Yangtze River Economic Belt, with a value of 0.010 at the 1% level, while the effect is insignificant for firms outside the region. This suggests that the Yangtze River Economic Belt provides a favorable institutional and market environment that allows ESG practices to be more effectively transformed into DT. The concentration of policy support, industrial clusters, and financial resources in this region may enhance firms’ capacity to integrate ESG performance with digital transformation.

Columns (3) and (4) report the results for cities with different levels of digital infrastructure. For firms located in Broadband China pilot cities, the effect of ESG on DT is positive but not significant. In contrast, for firms in non-pilot cities, the coefficient of ESG is significantly positive at the 5% level, with a value of 0.009. This indicates that in regions where digital infrastructure is less developed, firms rely more on ESG performance to obtain external resources, thereby promoting digital transformation.

5. Conclusions and Discussion

This study examines the impact of ESG on DT in Chinese listed new energy enterprises from 2010 to 2023. The following conclusions are drawn: First, strong ESG significantly promotes DT. Specifically, firms with better ESG demonstrate stronger resource capabilities, which provide better conditions for their digital transformation and technological upgrading. These findings highlight the important role of ESG as a strategic driver of digital transformation in the new energy sector. The above results are generally consistent with the findings of Cheng et al. (2025) [33]. Second, the impact of ESG on DT varies across different stages of development and regional contexts. Firms in the maturity stage of development and those located in regions with favorable institutional and market conditions show a stronger positive relationship between ESG and DT. In contrast, the effect is weaker in firms in the growth and decline stages, as well as those in regions with limited external resources. Third, by applying the dynamic capabilities perspective, this study further explains the mechanisms through which ESG influences digital transformation. ESG enhances firms’ adaptability, absorptive capacity, and innovation capabilities, which in turn improve DT.

In summary, this study draws on the perspective of dynamic capabilities to conceptualize ESG performance as a tool for enhancing resource capabilities. Existing studies adopt two approaches when considering the relationship between ESG and corporate digital transformation: one treats ESG outcomes as the research subject, focusing on the role of ESG in resource acquisition and capability enhancement during digital transformation [40]. The other considers the long-term and dynamic nature of ESG performance, viewing it as a dynamic process [41]. Regarding the pathway to achieving ESG and DT, this study supports the second view. Prior research has often positioned ESG within institutional and signaling theory frameworks [42,43]. In contrast, this study adopts the dynamic capabilities perspective, highlighting ESG’s role as a crucial channel for enhancing capabilities and achieving digital transformation goals. This approach deepens the understanding of ESG’s contribution to corporate transformation. In the traditional Resource-Based View, ESG is seen as a heterogeneous resource [26]. It acts as a link between internal and external networks, helping firms gain a competitive advantage during the transformation process. However, this view does not account for the dynamic nature of ESG. This study redefines ESG through the lens of dynamic capabilities, emphasizing it not only as an opportunity for enhancing resource capabilities but also as a process of “learning by doing.” This perspective offers a more comprehensive explanation of ESG’s role in corporate transformation.

This study provides practical implications for both policymakers and enterprise managers seeking to foster ESG-driven digital transformation in the new energy sector.

For policymakers, there are three suggestions. First, an ESG-oriented governance framework tailored to the specific characteristics of the new energy industry should be developed. Unlike traditional manufacturing sectors, new energy enterprises face distinct challenges such as high policy dependence, rapid technological evolution, and volatile production cycles. ESG standards and disclosure frameworks should therefore reflect these sectoral characteristics. Specifically, regulators may require firms to disclose indicators such as carbon emission reduction per kilowatt-hour, localization rate of key components, digital management level of energy storage systems, and the proportion of renewable energy utilized. These metrics capture both environmental and digital dimensions and can help evaluate firms’ innovation capacity and technological integration.

Second, a targeted capacity-building mechanism should be established to assist firms—particularly those in growth or transition phases—in formulating and implementing ESG strategies. This could include ESG training programs, pilot demonstration projects, and third-party advisory support, focusing on areas such as carbon accounting, green finance integration, and smart manufacturing practices.

Third, although not a central focus of this study, the heterogeneity analysis reveals that ESG’s positive impact on digital transformation is less pronounced in central and western regions. These areas often face limitations in resource endowment and digital infrastructure. Accordingly, context-sensitive policy instruments such as regional ESG guidance funds, land and energy access for R&D, and public investment in green infrastructure may help reduce structural constraints. These recommendations, while supplementary, reflect spatial disparities in ESG effectiveness and support more inclusive transformation outcomes. Empirical evidence indicates that firms with higher ESG performance often exhibit stronger capabilities in resource integration, technological absorption, and organizational innovation. Therefore, beyond policy intervention, strategic responses at the enterprise level constitute a critical factor in facilitating effective transformation. Incorporating ESG considerations into strategic planning processes can facilitate the pursuit of low-carbon objectives while simultaneously improving operational efficiency and environmental governance through digital means. It exhibits not only regional heterogeneity but is also shaped by firms’ technological capabilities and stages of development. Enterprises should consider advancing ESG-oriented digital tools in key application areas such as energy storage optimization, smart grid operations, and carbon asset management. The development of digital ESG performance indicators and multi-dimensional evaluation frameworks can help better align corporate sustainability efforts with technological transformation goals. Technologies such as blockchain and big data analytics may further improve the accuracy, transparency, and reliability of ESG information disclosure, which is essential for enhancing external credibility and stakeholder engagement.

Moreover, the technological complexity, extended supply chains, and policy sensitivity characteristic of the new energy sector necessitate differentiated approaches to ESG capability-building. Case evidence from leading firms suggests that industry best practices offer replicable frameworks. For example, CATL has implemented an internal mechanism for dynamically adjusting technology strategies in response to policy changes under an ESG-oriented framework [4]. LONGi has developed a structured knowledge management system to facilitate employee learning in green and digital technologies [44,45].

Finally, the institutionalization of ESG practices requires formal governance structures within firms, including the appointment of dedicated roles such as Chief ESG Officers and the integration of ESG criteria into digital transformation performance management systems. The arrangements will enhance internal accountability and position ESG as a core strategic function in enterprise-level innovation.

While this study provides valuable insights into the long-term impact of ESG on DT, there are some limitations that should be considered. First, this research primarily focuses on the long-term effects of ESG performance and does not address the potential short-term impacts that may arise from unexpected external shocks, such as public health crises, geopolitical conflicts, or economic disruptions. These short-term disruptions could influence how firms manage their ESG obligations and allocate resources during periods of uncertainty. Another limitation of this study is that, while we have focused on exploring the relationship between ESG performance and digital transformation, particularly in the context of new energy enterprises, we have not extended the analysis to the economic and environmental consequences of ESG performance and digital technology innovation. These aspects, including economic impacts such as profitability, productivity, and market performance, as well as environmental outcomes such as carbon emission reduction and resource efficiency improvement, are crucial for a comprehensive understanding of the broader effects of ESG practices.

In future research, it is suggested that the economic and environmental consequences of ESG performance and digital innovation be further explored. Such analysis would contribute to a more holistic understanding of how digital transformation, driven by ESG initiatives, influences not only corporate performance but also the broader societal and environmental impacts. These additional dimensions would provide valuable insights into the sustainability of digital transformation and the long-term benefits of ESG adoption. Furthermore, future studies could build upon this research by investigating the short-term effects of ESG performance under external shocks, offering a more comprehensive understanding of how firms balance their ESG commitments with the immediate demands of digital transformation. Examining how firms in different industries or regions respond to such external shocks could provide a more nuanced understanding of ESG’s role in crisis management and its impact on digital transformation in volatile environments.