Digital Transformation, Absorptive Capacity and Enterprise ESG Performance: A Case Study of Strategic Emerging Industries

Abstract

Digital transformation provides new drivers for economic performance growth in enterprises, but can it further improve ESG performance and support sustainable development? Based on the perspective of resources and capabilities, this article uses the relevant data of 1588 listed companies in strategic emerging industries from 2011 to 2021 to study the impact of digital transformation on enterprise ESG performance examines the intermediary role of absorptive capacity and the moderating role of regional digitalization level, and further analyzes the heterogeneity of property rights and industrial nature. The research results indicate that: firstly, digital transformation of enterprises can positively promote the improvement of ESG performance; secondly, absorptive capacity serves as a conduit through which digital transformation influences a company’s ESG performance; thirdly, the regional digitalization level positively moderates the promotion effect of digital transformation on enterprise ESG performance; fourthly, the impact of digital transformation on ESG performance of enterprises is significantly differentiated in the nature of enterprise property rights and industrial nature, and the ESG performance of state-owned enterprises and high-end equipment manufacturing enterprises is more sensitive to digital transformation. The research conclusion is based on a digital perspective, providing relevant insights for improving the ESG performance of strategic emerging industry enterprises and expanding their ESG development paths.

1. Introduction

Strategic emerging industries are industries that have an essential function in the long-term and holistic economic growth of the economy and society based on breakthroughs in cutting-edge technologies and major development needs. Moreover, strategic emerging industries, which are a deep integration of emerging technologies and industries, are important spaces for fostering the development of new technologies, goods, and driving forces as well as securing future competitive advantages in the global market [1]. The term “Enterprise Environmental, Social, and Governance Performance” (ESG) describes how well a corporation performs in these domains. The concept of ESG encourages organizations to prioritize sustainable development by emphasizing environmental friendliness, social responsibility, and corporate governance in their operations and development [2]. Strong ESG performance can boost market value and corporate performance [3], give companies a competitive edge in the marketplace, and assist companies in achieving sustainable development [4].

Digital transformation refers to the change in the way businesses create value for their customers by utilizing modern technology and communication methods [5]. Digital transformation is defined by the T/AIITRE 10001-2020 standard [6] as the process of adjusting to new industrial and technological developments and gradually incorporating IT technologies such as blockchain and AI. This procedure is intended to unleash the creative power of data, strengthen survival and advancement in the digital age, and accelerate corporate improvement, transformation, and refinement. To facilitate the creation, transfer, and acquisition of new value and to achieve both evolutionary and inventive advancement, it also seeks to restructure conventional processes and foster innovative growth drivers [6]. Its essence is the transformational behavior of improving enterprise resource allocation and reducing the impact of external uncertainty on enterprises through the efficient flow of data [7,8,9], which will inevitably affect business operations [10,11]. Research shows that digital transformation can maximize resource efficiency and promote the improvement of corporate economic benefits [12,13,14]; meanwhile, digital transformation can empower high-quality corporate development and enhance corporate market competitiveness [15,16]. Can businesses in strategic emerging industries (henceforth referred to as SEIEs) make the most of their unique advantages in the current industrial digitalization wave to enhance their ESG performance through digital transformation, acquire a sustained competitive edge, and eventually achieve sustainable growth and development?

Current studies have largely centered on the economic effects of digital transformation [9,17], with few studies exploring the non-economic effects brought about by digital transformation, such as the enhancement of enterprise ESG performance [18]. As societal attention to corporate ESG performance gradually increases, some scholars have begun to investigate how corporate ESG performance and digital transformation are related. Research indicates that digital transformation can enhance enterprise ESG performance [19,20,21,22], with mechanisms including promoting corporate green technology innovation [23], putting more pressure on external legitimacy [24], and reducing information asymmetry [24]; moderating effects include government subsidies and CEO professional experience [18]; heterogeneity includes property rights heterogeneity [18,23], heterogeneity of the technological level [18,23,25], lifecycle heterogeneity [23], market competition intensity heterogeneity [24], etc. In summary, the majority of the literature currently in publication examines how digital transformation affects corporate ESG performance from the viewpoints of information asymmetry and innovation capability within corporate dynamic capabilities [26]. Relatively little research has included the absorption capability within corporate dynamic capabilities into the research, and few studies have been conducted on the contextual mechanisms through which digital transformation affects corporate ESG performance [27].

According to the dynamic capacities theory, businesses must constantly innovate, adapt, and learn to deal with changes in the outside marketplace [28]. Dynamic skills are essential to ensuring that businesses react to complicated and changing external contexts in the unpredictable and complex world of digital transformation. WANG and AHMED believe that dynamic capabilities include absorptive capacity, adaptive capacity, and innovative capacity [29]. Existing research has revealed the mediating role of innovative capacity within dynamic capabilities between digital transformation and corporate ESG performance [26]. However, there is still a lack of exploration into other sub-dimensions of dynamic capabilities. Among them, absorptive capacity allows enterprises to quickly identify and seize opportunities; scan, create, learn, share, and interpret resources in the external environment; and attempt to disperse organizational boundaries, absorb and integrate external knowledge and resources, and ultimately apply them to business practice [30,31]. Research has found that digital transformation can promote real-time and continuous exchange of data and information between enterprises and customers, suppliers, and within enterprises, expanding the scope of enterprise knowledge and improving the efficiency of knowledge creation, sharing, and utilization [32]. It is conducive to enterprises identifying business opportunities from a wide range of knowledge and perceiving changes in market demand, promoting the improvement of enterprise absorptive capacity [33], and thus helping enterprises better cope with the challenges brought about by social change, enhancing corporate performance in social responsibility, and obtaining long-term competitive advantage [34]. Therefore, it is valuable to investigate how absorptive capacity mediates the relationship between enterprise ESG performance and the digital transition to provide a thorough understanding of the company’s digitization process.

Secondly, drawing upon the resource dependence theory, the resources upon which an organization relies are distributed within its environment, and there is an interdependence between the organization and its environment [35]. The capacity of a firm to produce digital technology is largely contingent upon the local digital infrastructure and level of intelligence [36]. An enhanced level of regional digitization can provide firms with more comprehensive infrastructure and communication platforms, facilitating the smooth implementation of digital transformation [37]. Therefore, the level of regional digitization provides strong support for a firm’s digital transformation and is a key contextual mechanism by which the digital transformation of a firm influences its ESG performance. In light of this, this article discusses the scenario mechanism of digital transformation affecting the enterprise ESG performance from the perspective of the regional digitalization level, which has practical significance for promoting the enterprise digitization of enterprises and improving their ESG performance.

In summary, the questions to be explored in this article are: ① Can digital transformation promote the improvement of corporate ESG performance, and is a differential impact? ② What is the mechanism of digital transformation affecting corporate ESG performance? ③ What is the situational mechanism of digital transformation affecting corporate ESG performance? To address the above issues, this article uses a sample of 1588 SEIEs in A-share to conduct empirical research on the above issues using text analysis and panel two-way fixed effect models. The potential contributions of this article are as follows: 1. By introducing absorptive capacity, it broadens the research on the mechanisms through which digital transformation affects corporate ESG performance. 2. It introduces the level of regional digitization, deepening the contextual mechanisms by which digital transformation affects corporate ESG performance. 3. It conducts a heterogeneity analysis of strategic emerging industry sub-sectors, further enriching the research on the heterogeneity of the impact of digital transformation on corporate ESG performance. The remaining structure of this article is arranged as follows: Section 2 proposes the research hypotheses and provides theoretical analysis; Section 3 introduces the research design of this article; Section 4 is the empirical study section of this article; Section 5 is the discussion; Section 6 is the conclusion.

2. Theoretical Analysis and Research Hypotheses

2.1. The Impact of Digital Transformation on ESG Performance of SEIEs

Firstly, modern production and lifestyle practices have been significantly impacted by the digitization of the real economy. The new era’s core principles are balanced growth and environmentally friendly, sustainable development. In addition to improving corporate economic performance, corporate digital transformation adds value to non-economic performance areas like the environment, society, governance, and corporate culture [38]. Particularly against the backdrop of the digital economy, various stakeholders have higher expectations for corporate social responsibility fulfillment capabilities and performance, which, in turn, forces companies to innovate. Digital technology innovation can also improve corporate social responsibility fulfillment capabilities and performance [39]. After undergoing digital transformation, the support of digital technology can help enterprises more efficiently refine and enhance their green image, customer reputation, and product quality, thereby bringing growth in orders and profits for the enterprises [40]. Meanwhile, digital transformation can also bring about changes in organizational structure and internal management [41], reducing the expected costs for enterprises to undertake green transformations and activities, and even altering their profit models. While reducing corporate costs, it can also create more employment opportunities for society, thus promoting sustainable economic growth [42]. Therefore, this article proposes the first research hypothesis H1:

H1: 

Digital transformation can positively enhance corporate ESG performance.

2.2. The Impact Channels of Digital Transformation on ESG Performance of SEIEs

In the current environment of rapid development of the digital economy, SEIEs need to facilitate a comprehensive incorporation of cutting-edge digital tech with their operations to accelerate their digitalization process. Therefore, how to use digital advances to ameliorate the quality of internal operations and innovation capability has become the key to enterprises carrying out digital transformation. Using digital technology enables companies to efficiently reduce the threshold for obtaining innovative resources [43], providing them with more knowledge and resources, thereby enhancing the enterprise’s ability to identify and absorb knowledge and opportunities, that is, absorptive capacity, ultimately helping enterprises create their value and achieve better development [44]. Meanwhile, industry practices and empirical research indicate that corporate R&D innovation is a necessary technical prerequisite for enterprises to implement production transformation [45], and the enhancement of absorptive capacity can reduce the cost of corporate digital transformation and exert a stable positive impact on corporate ESG responsibility performance [46]. Therefore, this article proposes the second research hypothesis H2:

H2: 

Digital transformation promotes the improvement of ESG performance by enhancing the absorption capacity of enterprises.

2.3. The Impact Mechanism of Digital Transformation on ESG Performance of SEIEs

2.3.1. The Regulatory Role of Regional Digitalization Level

Enterprises are embedded in regional environments, and regional digitalization represents a new external environment. This implies that the effectiveness of enterprise digital transformation will be constrained by the level of regional digitalization [47]. On the one hand, the better the level of regional digitalization is, the more sound the regional digital infrastructure will be, laying a better foundation for enterprise digitalization. This accelerates the dissemination and sharing of information and knowledge among enterprises, facilitates interconnectivity between enterprises, reduces the cost of obtaining information, and is conducive to enterprise digital transformation [48]. On the other hand, as regional digitalization accelerates, the region’s digital technology becomes relatively advanced, and with strong government support for digital transformation, enterprises are better able to efficiently utilize the new knowledge they have absorbed and convert it into new knowledge and capabilities, thus better conducting ESG practices [49]. Given this, this article proposes the following sub-hypothesis H3a within the third research hypothesis:

H3a: 

The level of regional digitalization plays a positive moderating role between digital transformation and corporate ESG performance.

2.3.2. The Heterogeneous Impact of Digital Transformation on ESG Performance of SEIEs

Digital transformation can impact corporate ESG performance through channels such as enhancing corporate information transparency. Existing literature indicates that this impact exhibits a certain degree of heterogeneity, varying across different industries [50], regions [51], and organizational sizes [52]. This article further analyzes whether there is a heterogeneous effect of digital transformation on corporate ESG performance under different ownership properties and industrial properties.

• Ownership Characteristics

State-owned enterprises (SOEs) occupy a significant position in China’s economic development and play a crucial role in supporting national sustainable development strategy goals. Therefore, SOEs have an external motivation to further enhance their ESG performance to comply with ESG-related policies issued by the government and regulatory agencies [53]. Moreover, compared to private enterprises, SOEs possess more abundant financial resources and more stable human and material resources, which facilitate digital transformation and ensure corporate ESG practices [54,55]. Through digital transformation, SOEs can improve operational efficiency, reduce resource consumption, better manage supply chains, and enhance the monitoring and control of environmental impacts, thereby achieving higher levels of corporate social responsibility. Therefore, this article proposes the subdivision hypothesis H3b of the third research hypothesis:

H3b: 

There are property rights differences in the driving effect of digital transformation on enterprise ESG performance.

2.

Industrial Nature

High-end equipment manufacturing, also known as advanced manufacturing, refers to industries that produce high-tech, high-value-added advanced industrial facilities and equipment [56]. This industry is characterized by high product value, high technology intensity, and strong growth potential, offering significant competitive advantages and development prospects. Under the guidance of relevant policies, high-end equipment manufacturing has become a key sector for achieving the “dual carbon” goals. Compared to other industries, high-end equipment manufacturing is more willing to respond to national calls and actively fulfill corporate social responsibility [57].

Additionally, due to the special nature of their products and technologies, high-end equipment manufacturing enterprises typically possess high technical research and development capabilities and a strong sense of innovation. They play an active role in energy conservation, and green manufacturing. These companies are more inclined to acknowledge the significance of environmental conservation and social accountability for their enduring growth. Consequently, they exhibit a greater propensity to allocate resources and exert efforts towards a range of initiatives aimed at mitigating environmental effects and elevating the extent to which they fulfill their social duties [58]. Therefore, this article proposes the subdivision hypothesis H3c in the third research hypothesis:

H3c: 

The promotion of corporate ESG performance by digital transformation exhibits heterogeneity across sub-industries in strategic emerging industries.

3. Research Design

3.1. Model Setting

To examine the influence of digital transformation on ESG performance and the impact mechanism, this article sets a benchmark regression model as shown in Equation (1):

$$ES{G}_{\mathrm{it}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_{1}D{T}_{\mathrm{it}}+{\displaystyle \sum _{\mathrm{j}}{\mathsf{\beta }}_{\mathrm{j}}Controls}+{\mathsf{\lambda }}_{\mathrm{i}}+{\mathsf{\mu }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{it}}$$

(1)

where ESG is the explained variable—enterprise ESG performance, DT is the explanatory variable—enterprise digital transformation, Controls is the control variable, ${\mathsf{\lambda }}_i$ is the individual fixed effect, ${\mathsf{\mu }}_t$ is the time fixed effect, and ${\mathsf{\epsilon }}_{it}$ is the random error term.

Most of the existing literature has used the step-by-step method proposed by Baron and Kenny (1986) to test the mediating effect [59]. However, Jiang (2022) pointed out that the main problem of the current mediating effect analysis is the abuse of the stepwise test of mediating effect grafted from psychology, which leads to errors in the mediating effect test [60]. Meanwhile, we observed that most of the literature has used the same control variables as the benchmark regression in the mediating effect test process, resulting in logical defects in the mediating effect test. Thus, we refer to the suggestion proposed by Jiang (2022) [60], set different control variables for different mediating variables, and constructed the mediating effect test model as shown in Equation (2):

$$A{C}_{\mathrm{it}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_{1}D{T}_{\mathrm{it}}+{\displaystyle \sum _{\mathrm{j}}{\mathsf{\beta }}_{\mathrm{j}}Controls+{\mathsf{\lambda }}_{\mathrm{i}}+{\mathsf{\mu }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{it}}}$$

(2)

Similarly, the moderating mechanism test method proposed by Jiang Ting (2022) [60] was adopted to test the moderating effect of regional digitalization level between digital transformation and enterprise ESG performance. The regression model is shown in Equation (3):

$$ES{G}_{\mathrm{it}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_{1}D{T}_{\mathrm{it}}+{\mathsf{\beta }}_{2}RD{L}_{\mathrm{it}}+{\mathsf{\beta }}_{3}D{T}_{\mathrm{it}}\times RD{L}_{\mathrm{it}}+{\displaystyle \sum _{\mathrm{j}}{\mathsf{\beta }}_{\mathrm{j}}Controls+{\mathsf{\lambda }}_{\mathrm{i}}+{\mathsf{\mu }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{it}}}$$

(3)

Among them, the mediating variable is the absorptive capacity (AC), and the moderating variable is the regional digitalization level (RDL).

3.2. Variable Setting

3.2.1. Explained Variable: ESG Performance (ESG)

In this article, ESG score data from the selected companies spanning the years 2011–2021 were utilized for the analysis. The ESG scores were categorized into nine tiers, ranging from C to AAA, with each tier assigned a score from 1 to 9 and assessed quarterly. The mean score from the four quarterly evaluations was taken as the dependent variable, with higher scores indicating superior ESG performance for the respective companies.

3.2.2. Explanatory Variable: Digital Transformation (DT)

In this article, referring to the method of Wu Fei et al. (2021), the ratio of word frequency of six categories of digital transformation, AI, big data, cloud computing, blockchain, and digital application to the total number of words in the annual report [61] were adopted as the original data of digital transformation. Additionally, this article refers to the method of Lu Ming and Chen Zhao (2004), adding 0.00000001 to the word frequency ratio and then taking the natural logarithm to serve as the explanatory variable. Among these, the word frequency ratio of DT was used as the core explanatory variable [62].

3.2.3. Mechanism Variable: Absorptive Capacity (AC)

Inspired by scholar Xiao Jing et al. (2023), this article introduces absorptive capacity as a mechanism variable between digital transformation and enterprise ESG performance [34].

3.2.4. Moderating Variable: Regional Digitization Level (RDL)

Based on the practice of Xiao Jing et al. (2023), this article introduces the regional digitalization level as the moderating variable between digital transformation and enterprise ESG performance [34]. Among them, the regional digitization level was measured by the comprehensive index of regional digitization level, which was calculated by weighting five indicators, including the digital output, fixed rate of telephone penetration, mobile telephone penetration rate, internet broadband penetration rate, and number of web pages per capita.

3.2.5. Control Variable

To guarantee the consistency of the research results, we selected enterprise size (Size), age (Age), operating income growth rate (Growth), asset-liability ratio (Lev), cash flow ratio (Cash Flow), profitability (ROA), ownership concentration index (1%) (TOP1), board size (Board), independent director ratio (Indep), dual (Dual), executive shareholding ratio (M Share), and executive team size (TMT Size) as control variables. The main variable definitions in this article are shown in

Table 1. Variable definition.

Variable Type	Name	Symbol	Definition
Explained Variable	ESG Performance	ESG	Sino-Securities ESG rating data
Explanatory Variable	Digital Transformation	DT	The proportion of digital transformation word frequency in the total word number of the annual report
	Artificial Intelligence Technology	AI	The proportion of word frequency of artificial intelligence technology in total word number of annual report
	Big Data Technology	BD	The proportion of word frequency of big data technology in total word number of annual report
	Cloud Computing Technology	CC	The proportion of cloud computing technology word frequency in total annual report word number
	Blockchain Technology	BC	The proportion of blockchain technology word frequency in the total number of annual reports
	Digital Technology Application	ADT	The proportion of word frequency of digital technology application in total word number of annual report
Mechanism Variable	Absorptive capacity	AC	Annual R&D expenditure/operating income
Regulating Variable	Regional Digitization Level	RDL	Using the entropy weight method, the digital output, fixed telephone penetration rate, mobile telephone penetration rate, Internet broadband penetration rate, and number of web pages per capita were weighted to calculate the comprehensive index of regional digitalization level
Control Variable	Enterprise Scale	Size	The natural log of total assets
	Enterprise Age	Age	Ln (Year—year of listing + 1)
	Revenue Growth Rate	Growth	(Revenue growth/total revenue of last year) × 100%
	Asset-liability Ratio	Lev	Total liabilities/total assets × 100%
	Cash Flow Ratio	Cash Flow	Net cash flow from operating activities/ending current liabilities
	Profitability	ROA	(Net profit/average total assets) × 100%
	Ownership Concentration Index 1(%)	TOP1	The proportion of the largest shareholder
	Board Size	Board	Ln (Number of Directors)
	Proportion of Independent Directors	Indep	(Number of independent directors/Number of directors) × 100%
	Dual Function	Dual	The combination of chairman and general manager is 1, otherwise, it is 0
	Executive Ownership Ratio	M Share	Number of shares held by executives/total shares
	Executive Team Size	TMT Size	Ln (Number of executives)

.

3.3. Sample and Data Source

For the research sample, this article uses pertinent data from 1588 representative A-share companies that are part of the China Strategic Emerging Industries Comprehensive Index (000891) for the years 2011–2021. The following is how the data were handled: Initially, ST samples and delisted samples were eliminated; thereafter, all continuous variables were winsorized at the top and bottom 1% to lessen the extreme values’ impact on the results. With STATA 17.0, data processing and regression analysis were carried out. In the end, 11,682 valid data entries encompassing nine industries, including new-generation information technology, were acquired from the 1588 strategic emerging industry-listed enterprises. The CSMAR database provided the financial information utilized in this article, the data related to the business digital transition from the Mark Data Network, the enterprise ESG rating data from the WIND database, and other original data from the annual reports of listed companies, the National Bureau of Statistics of China’s official website, and the CSMAR database.

4. Empirical Results and Analysis

4.1. Descriptive Statistics

This article utilizes Stata 17 software to conduct descriptive statistics on variables. The results, as shown in

Table 2. Descriptive statistics for variables.

Variable	Obs	Mean	Std. Dev.	Min	Max
ESG	11,682	4.0316	1.0973	0.5000	7.7500
DT	11,682	−12.3013	4.0074	−18.4207	−6.0397
AC	11,682	0.0689	0.0630	0.0000	0.3643
RDL	11,682	0.3687	0.1397	0.0933	0.6528
Size	11,682	21.9823	1.1605	19.8575	25.6465
Age	11,682	2.8346	0.3333	1.7918	3.4657
Growth	11,682	0.2102	0.3871	−0.4782	2.3539
Lev	11,682	0.3689	0.1888	0.0441	0.8157
CashFlow	11,682	0.0449	0.0647	−0.1362	0.2384
Roa	11,682	0.0513	0.0673	−0.2298	0.2497
Top1	11,682	0.3138	0.1373	0.0794	0.6873
Board	11,682	2.1064	0.1887	1.6094	2.5649
Indep	11,682	37.8100	5.3249	33.3300	57.1400
Dual	11,682	0.3355	0.4722	0.0000	1.0000
MShare	11,682	0.1791	0.2084	0.0000	0.7049
TMTSize	11,682	2.8115	0.1944	2.3979	3.3322

, indicate that ESG had a mean of 4.03, with a maximum of 7.75 and a minimum of 0.5, suggesting significant variations in ESG performance among different companies, with an overall tendency towards the lower end. The core explanatory variable had a mean of −12.3 after natural logarithm transformation, indicating that a considerable portion of companies have yet to engage in digital transformation, with a maximum of −6.03966 and a minimum of −18.42068, suggesting a generally low level of digital transformation across companies. Additionally, descriptive statistics for control variables show a high level of consistency with relevant literature [23].

4.2. Benchmark Regression

This article utilizes Stata 17 software and employs the OLS method for coefficient estimation of Equation (1) (baseline regression), looking into how the digital transition affects corporate ESG performance.

Table 3. Benchmark regression results (core explanatory variables).

Variable	(1)	(2)	(3)	(4)
	ESG	ESG	ESG	ESG
DT	0.0201 ***	0.0160 ***	0.0180 ***	0.0133 ***
	(7.9729)	(3.8532)	(7.2885)	(3.3056)
Size			0.2644 ***	0.3053 ***
			(24.7852)	(9.3811)
Age			0.0296	−0.2475
			(0.9623)	(−1.4990)
Growth			−0.1073 ***	0.0287
			(−4.0747)	(1.1640)
Lev			−0.3582 ***	−0.2765 **
			(−5.5434)	(−2.2707)
CashFlow			0.4510 ***	0.2048
			(2.7045)	(1.1489)
Roa			1.2754 ***	−0.5636 ***
			(7.1594)	(−2.8428)
Top1			0.2707 ***	0.7857 ***
			(3.7502)	(3.3929)
Board			0.0107	−0.3782 **
			(0.1379)	(−2.5029)
Indep			0.0127 ***	0.0047
			(5.5940)	(1.2369)
Dual			−0.0930 ***	−0.0587
			(−4.3241)	(−1.5552)
MShare			0.3244 ***	0.7547 ***
			(6.1248)	(5.1429)
TMTSize			0.3095 ***	0.2939 **
			(4.5873)	(2.2992)
Constant	4.2794 ***	4.0319 ***	−3.0586 ***	−2.3772 ***
	(130.9129)	(54.0189)	(−10.7682)	(−3.0019)
Observations	11,682	11,682	11,682	11,682
R-squared	0.0054	0.0284	0.0966	0.0610
code fe	no	yes	no	yes
year fe	no	yes	no	yes
Number of id	1588	1588	1588	1588

Standard errors are in parenthesis. *** p < 0.01, ** p < 0.05.

presents the baseline regression results of this article, with columns (1) and (2) indicating the baseline regression and two-way fixed effects regression results after only incorporating the primary variable used to explain digital transformation. Columns (3) and (4) represent the baseline regression and two-way fixed effects regression results after incorporating control variables. The empirical results, above all, indicate, at a 10% confidence level, a positive and significant function of digital transformation in enhancing corporate ESG performance. This validates the research hypothesis H1.

Additionally, this study further refines the explanatory variables into five sub-indicators such as artificial intelligence technology. These five sub-indicators were introduced into the regression analysis, which is presented in

Table 4. Benchmark regression results (subdivided explanatory variable).

Variable	(1)	(2)	(3)	(4)	(5)
	ESG	ESG	ESG	ESG	ESG
AI	0.0032
	(0.0044)
BD		0.0099 *
		(0.0040)
CC			0.0150 ***
			(0.0044)
BC				0.0081
				(0.0074)
ADT					0.0059
					(0.0040)
_cons	−2.5282 **	−2.3658 **	−2.2894 **	−2.4490 **	−2.5115 **
	(0.8020)	(0.7976)	(0.7943)	(0.8032)	(0.7934)
N	11682	11682	11682	11682	11682
r2	0.0597	0.0604	0.0612	0.0598	0.0599
F	23.2358	23.3905	23.3942	23.2236	23.3029
p	0.0000	0.0000	0.0000	0.0000	0.0000

Standard errors are in parenthesis. *** p < 0.01, ** p < 0.05, * p < 0.1.

, indicating that among the five sub-indicators, the effects of big data technology and cloud computing technology on enhancing corporate ESG performance were more pronounced.

4.3. Endogeneity Problem and Robustness Test

4.3.1. Endogeneity Problem

To tackle endogeneity concerns, this research utilizes PSM alongside multi-period DID methods. Initially, companies were categorized into two cohorts: those that had undergone digital transformation and those that had not. For PSM, control variables served as matching criteria. Various matching techniques, including nearest neighbor, radius, and kernel methods, were applied to identify suitable control group counterparts for the treatment group. Post balance checks, regression analysis was conducted on the matched samples, with findings detailed in columns (1) to (3) of

Table 5. Endogeneity test results.

Variable	PSM			DID
	(1)	(2)	(3)	(4)
	ESG	ESG	ESG	ESG
DT	0.0138 **	0.0133 **	0.0138 **
	(0.0043)	(0.0040)	(0.0043)
DID				0.0742 **
				(2.5400)
controls	YES	YES	YES	YES
code fe	YES	YES	YES	YES
year fe	YES	YES	YES	YES
_cons	−2.1427 *	−2.3757 **	−2.1427 *	−2.6264 ***
	(0.8353)	(0.7924)	(0.8353)	(−3.3422)
N	9276	11,672	9276	11,682
r2	0.0609	0.0611	0.0609	0.0604

Standard errors are in parenthesis. *** p < 0.01, ** p < 0.05, * p < 0.1.

. The regression results indicate a significant positive effect of digital transformation on corporate ESG performance, supporting the robustness and credibility of the core conclusion of this article.

Furthermore, the study treats the temporal progression of digital transformation within the sample as a quasi-experimental setup and applies the multi-period DID approach for analysis. The pertinent regression models are delineated by Equations (4) and (5):

$$ES{G}_{\mathrm{it}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_{1}d{u}_{\mathrm{it}}+d{t}_{\mathrm{it}}+{\displaystyle \sum _{\mathrm{j}}{\mathsf{\beta }}_{\mathrm{j}}Controls+{\mathsf{\lambda }}_{\mathrm{i}}+{\mathsf{\mu }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{it}}}$$

(4)

$$ES{G}_{\mathrm{it}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_{1}d{u}_{\mathrm{it}}+d{t}_{\mathrm{it}}\times D{T}_{\mathrm{it}}+{\displaystyle \sum _{\mathrm{j}}{\mathsf{\beta }}_{\mathrm{j}}Controls+{\mathsf{\lambda }}_{\mathrm{i}}+{\mathsf{\mu }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{it}}}$$

(5)

where $du$ is the virtual variable of the processing group, $du$ 1 indicates that the enterprise has carried out digital transformation during the sample period, and $du$ = 0 indicates that the enterprise did not do it. dt is a time dummy variable, $dt$ = 1 when the processing group enterprises implement digital transformation, and $dt$= 0 when the control group enterprises and the processing group enterprises do not implement digital transformation. The empirical test results of multi-stage DID are shown in column (4) of Table 5. The key parameters to be estimated in the model are significantly positive, that is, the core conclusions of this article are still robust and credible after multi-phase DID model identification.

4.3.2. Robustness Test

In this article, in the robustness test, four methods were adopted, such as the hysteresis of the explained variable and the elimination of zero-value samples: Column (1) indicates that the explained variable was introduced into the model for regression with a lag of two periods as explanatory variable, and the model was changed to a dynamic model; column (2) indicates that the explained variable was still used as explanatory variable for regression with a lag of one period; column (3) represents truncation of the core explanatory variable; column (4) indicates that zero samples of DT were eliminated. The results of the above four regressions are shown in

Table 6. Robustness test results.

Variable	(1)	(2)	(3)	(4)
	ESG	L.ESG	ESG	ESG
L2.ESG	0.0532 ***
	(0.0129)
DT	0.0123 **	0.0144 ***		0.0474 **
	(0.0039)	(0.0042)		(0.0171)
DT_w			0.0133 ***
			(0.0040)
controls	YES	YES	YES	YES
code fe	YES	YES	YES	YES
year fe	YES	YES	YES	YES
_cons	−1.1017	−1.2026	−2.3780 **	−2.3018 *
	(0.8266)	(0.8131)	(0.7919)	(0.9685)
N	8612	10,094	11,682	8429
r2	0.0617	0.0550	0.0610	0.0591
F	10.8219	17.1297	23.6177	15.6250
p	0.0000	0.0000	0.0000	0.0000

Standard errors are in parenthesis. *** p < 0.01, ** p < 0.05, * p < 0.1.

. All four regressions passed the significance test, indicating that the above results that prove that digital transformation can positively promote enterprise ESG performance are robust and credible.

4.4. Mechanism Test

According to Equation (2), the mediating effect was examined.

Table 7. Mechanism and regulatory effect test results.

Variable	Mechanism Verification		Moderating Effect Test
	(1)	(2)	(3)
	AC	ESG	ESG
DT	0.0004 *	0.0127 **	0.0151 ***
	(0.0002)	(0.0040)	(3.7286)
AC		1.4934 **
		(0.4658)
RDL			0.5136
			(1.4443)
DT × RDL			0.0535 **
			(2.1521)
controls	YES	YES	YES
code fe	YES	YES	YES
year fe	YES	YES	YES
_cons	0.0900 *	−2.5117 **	−2.4950 ***
	(0.0369)	(0.7931)	(−3.1732)
N	11,682	11,682	11,682
r2	0.1184	0.0636	0.0620
p	0.0000	0.0000	0.0000

Standard errors are in parenthesis. *** p < 0.01, ** p < 0.05, * p < 0.1.

displays the mechanism test results. In the mechanism test, the estimated parameter of DT was significantly positive at the 1% confidence level, suggesting that digital transformation can positively enhance corporate ESG performance by improving corporate absorptive capacity. In conclusion, research hypothesis H2 is validated and confirmed.

4.5. Moderation Effect Test

According to Equation (3), the moderating effect was examined. Table 7 displays the results. In the moderation effect test, the estimated parameter of the interaction term was significantly positive at the 5% confidence level, which means that the regional digitalization level positively moderates the relationship between digital transformation and corporate ESG performance. In conclusion, research hypothesis H3a is validated and confirmed.

4.6. Heterogeneity Analysis

This article separates the sample companies into SOES and non-state-owned businesses to perform group regression based on property rights for the heterogeneity analysis. The sample enterprises were then further separated into nine sub-industries, such as the high-end equipment manufacturing industry, for group regression, by the categorization norms of strategic developing industries.

Table 8. Heterogeneity analysis regression results (property rights).

Variable	ESG (State-Owned)	ESG (Private)
DT	0.0164 ***	0.0105 **
	(3.0864)	(2.4165)
controls	YES	YES
code fe	YES	YES
year fe	YES	YES
_cons	−1.5428	−3.1860 ***
	(−1.4689)	(−4.7082)
N	2926	8379
r2	0.0869	0.0721

Standard errors are in parenthesis. *** p < 0.01, ** p < 0.05.

and

Table 9. Heterogeneity analysis regression results (industry).

Variable	(1)	(2)	(3)	(4)	(5)
	The new generation of the information technology industry	High-end equipment manufacturing industry	New materials industry	Biological industry	New energy vehicle industry
	ESG	ESG	ESG	ESG	ESG
DT	−0.0051	0.0240 **	0.0104	0.0160 *	0.0750 *
	(0.0075)	(0.0089)	(0.0081)	(0.0074)	(0.0301)
controls	YES	YES	YES	YES	YES
code fe	YES	YES	YES	YES	YES
year fe	YES	YES	YES	YES	YES
_cons	−1.9721 *	−2.7688	−2.2656	−4.4945 **	−16.3140
	(0.8917)	(1.5179)	(1.6031)	(1.4076)	(8.5966)
N	3817	1688	1568	1842	136
r2	0.0753	0.0671	0.0891	0.0896	0.4395
Variable	(6)	(7)		(8)	(9)
	New energy industry	Energy conservation and environmental protection industry		Digital creative industry	Related service industry
	ESG	ESG		ESG	ESG
DT	0.0130	0.0252 *		0.0325	0.5608
	(0.0102)	(0.0118)		(0.0220)	(0.6900)
controls	YES	YES		YES	YES
code fe	YES	YES		YES	YES
year fe	YES	YES		YES	YES
_cons	−2.7086	−5.6682 *		−1.4455	27.9815
	(1.7140)	(2.3335)		(2.7648)	(24.2249)
N	1276	853		461	41
r2	0.0762	0.1427		0.1447	0.8442

Standard errors are in parenthesis. ** p < 0.05, * p < 0.1.

display the regression results. The empirical findings show that, at least at the 5% confidence level, the positive impact of digital transformation on corporate ESG performance was obvious for both SOEs and non-state-owned businesses, with a more emphatic effect seen in state-owned businesses. Among the nine industries, the promotion effect was more prominent in the high-end equipment manufacturing industry. Conversely, this promotion effect was negative in the new-generation information technology industry. Analysis suggests that the higher digitalization in this industry, coupled with smaller enterprise size and lower social visibility, results in poorer performance in social responsibility fulfillment. In conclusion, research hypotheses H3b and H3c are validated and confirmed.

5. Discussion

This article chooses a sample of 1588 A-share listed companies in strategic emerging industries, selecting ESG rating data, digital transformation data, and relevant financial data from 2011 to 2021 as the sample data. Through employing a two-way fixed effects model, the research empirically investigates how corporate ESG performance is affected by digital transformation and how it works. It examines the moderating effect of regional digitalization level on the relationship between digital transformation and corporate ESG performance and further conducts heterogeneity analysis by property rights and different industries.

Empirical results indicate:

① Digital transformation positively promotes corporate ESG performance. After a series of endogeneity and robustness tests, this conclusion remains valid, confirming research hypothesis H1. Comparisons with relevant literature reveal high consistency with previous studies [18,19,20,21,22,23,24,25,26,27].

② Digital transformation positively enhances corporate ESG performance by improving absorptive capacity, suggesting that enhancing absorptive capacity is a key pathway through which digital transformation drives improvements in corporate ESG performance, contributing to long-term competitive advantages and sustainable development. Research hypothesis H2 is supported. However, comparisons with existing research show that previous studies have affirmed the mediating roles of green technology innovation [18], internal information transparency [23], external legitimacy pressure [24], and corporate sustainable development reputation [27] in the link between digital transformation and corporate ESG performance. Some studies also confirm that innovation capability within dynamic capabilities is one of the main channels through which digital transformation affects corporate ESG performance [26]. This study enriches the research on the impact pathways between digital transformation and corporate ESG performance.

③ The regional digitalization level serves as a positive moderator in the interplay between digital transformation and corporate ESG performance, suggesting that a robust regional digital infrastructure significantly bolsters enterprises’ digital evolution and boosts their ESG ratings. Our findings validate research hypothesis H3a. While previous studies have underscored the positive moderating role of regional digitalization between digital transformation and enterprise financial performance [34], our research highlights its pivotal role in influencing the non-economic benefits of digital transformation. Furthermore, while government subsidies and CEO’s career experience are noted as important contextual factors shaping the impact of digital transformation on corporate ESG performance [18], this article augments the understanding of such situational mechanisms, thereby enriching the research landscape.

④ Heterogeneity analysis reveals that the impact of digital transformation on corporate ESG performance varies across different entities. The effectiveness of digital transformation in boosting corporate ESG performance is particularly evident in SOEs and high-end equipment manufacturing industry enterprises. Research hypotheses H3b and H3c are validated. Comparisons with relevant literature show that most scholars believe that digital transformation has differential effects on corporate ESG performance [18,23,24,25,26], but the reasons for these differential effects vary. Specifically, this study finds that the reasons for these differential effects lie in the differences in property rights and sub-industries within strategic emerging industries. The research results on the differential effects caused by the nature of property rights are consistent with the findings of scholars such as Yang Peng [18] and Hu Jie [23]. However, the research results on the differential effects caused by the nature of sub-industries within strategic emerging industries have not been found in the current literature. Additionally, some researchers have found that differences in regional distribution [19,26] and corporate technological levels [25] also lead to these differential effects.

6. Conclusions

This article analyzes the impact and mechanisms of digital transformation on corporate ESG performance using 1588 SEIEs as the research sample. The findings are as follows.

Firstly, digital transformation, to some extent, promotes improvements in corporate ESG performance. Digital transformation enhances not only operational efficiency and management capabilities but also fosters absorptive capacity, enabling better adaptation and response to external environmental changes, thereby improving corporate ESG performance.

Secondly, digital transformation can improve corporate ESG performance by enhancing absorptive capacity. Specifically, the improvement of information technology infrastructure, data analysis, and knowledge-sharing efficiency brought about by digital transformation can promote the enhancement of absorptive capacity, facilitating the effective utilization of knowledge and technology and comprehensive improvement of corporate ESG performance.

Thirdly, it is observed that regional digitalization levels positively moderate the promotion effect of digital transformation on corporate ESG performance. This implies that in regions with higher levels of digitalization, the improvement is more significant. This may be because regions with high digitalization levels have more complete digital infrastructure, abundant digital talents, and more open digital ecosystems, providing better environments and conditions for corporate digital transformation.

Finally, it is noted that this promotion effect is more significant in SOES and high-end equipment manufacturing companies. This may be attributed to SOES having richer resources and policy support during the digital transformation process, enabling faster realization of digital transformation and translation of its effects into improvements in ESG performance. Moreover, because of the traits of the sector, digital transformation in high-end equipment manufacturing companies often brings more significant benefits and competitive advantages, thereby promoting improvements in their ESG performance.

The following suggestions are put out in light of these empirical findings:

For enterprises: ① Embrace digital transformation as a strategic imperative for enduring growth. Amidst the dynamic landscape of the digital economy, characterized by swift and continuous technological advancements, businesses must view digital transformation as a cornerstone of their long-term strategic planning. This approach is essential for securing sustainable growth and maintaining a competitive edge in the current cutthroat market environment. The study shows that digital transformation can significantly improve corporate ESG performance, especially for SOEs and high-end equipment manufacturing companies. Therefore, enterprises should formulate differentiated development strategies based on their situations. ② Strengthen the cultivation of absorptive capacity. According to the research findings, enhancements in company ESG performance can be fostered by digitalizing by strengthening absorptive capacity. Enterprises should enhance absorptive capacity while improving digital technology, actively seeking and absorbing knowledge in the environment, applying it to business practices, and thereby improving corporate ESG performance. ③ Strengthen friendly interactions with the government. Regional digitalization levels profoundly affect the process of corporate digital transformation. Enterprises should strengthen interactions with relevant departments, establish good government–business relations, and work together to create a conducive environment for digital transformation.

For the government: ① Improve regional digital infrastructure construction to reduce regional disparities in digitalization levels. Regulatory authorities should continuously improve digital infrastructure construction, enhance regional digitalization levels, and minimize regional disparities in digitalization levels as much as possible, providing more comprehensive support services for corporate digital transformation. ② Enhance support for digital transformation in non-state-owned enterprises. Relevant authorities should enhance policy and financial support for digital transformation in privately owned enterprises, implement tailored policies for enterprises across different sectors, and comprehensively drive corporate digitalization.

The marginal contributions of this study are as follows:

① By introducing absorptive capacity, the study broadens the research on the impact pathways of digital transformation on corporate ESG performance, thereby promoting the development of corporate ESG.

② By introducing regional digitalization levels, the study deepens the research on the situational mechanisms of digital transformation affecting corporate ESG performance, considering the differences in digital development in different regions, thereby facilitating a broader and more nuanced perspective on the impacts of digital transformation across diverse settings.

③ By conducting heterogeneity analysis of sub-industries within strategic emerging industries, the study further enriches the research on the heterogeneity analysis of the influence of digital transformation on corporate ESG performance, providing a theoretical basis and enlightenment for relevant departments and enterprises to promote ESG development and improve ESG performance.

However, this study has the following limitations:

① Theoretical limitations: In terms of impact pathway research, this study only includes absorptive capacity in dynamic capabilities. Future research can simultaneously include innovation capacity, absorptive capacity, and adaptability in dynamic capabilities for more detailed exploration. In terms of moderating effect research, this study only considers the moderating effect of regional digitalization levels, while future research can further consider the moderating influence of elements like the performance of regional ESG. Regarding heterogeneity analysis research, this study only considers the heterogeneity of property rights and sub-industries within strategic emerging industries, while future research can further consider heterogeneities such as data assets and knowledge intensity.

② Sample region limitations: The sample of this study is limited to A-share listed companies in China’s strategic emerging industries. This means that the research results mainly reflect the situation of the Chinese market and specific industries, making it difficult to comprehensively represent the changes in corporate ESG performance during the digital transformation process in other regions and markets globally. Therefore, the universality of the research results across regions or globally is limited.

③ Sample industry limitations: Strategic emerging industries themselves have certain characteristics, and the challenges and opportunities encountered in digital transformation and ESG practices may differ significantly from traditional industries. Therefore, the research results may not apply to companies in other industries, limiting the extrapolation of the research results. Future research could consider expanding the sample selection range.

④ Sample period limitations: Digital transformation is a dynamic process, and corporate ESG performance may change over time. This study’s sample period spans from 2011 to 2021. It is possible that the study’s time frame and data-gathering methods cannot accurately capture the long-term effects of digital transformation on company ESG performance. Future studies might think about extending the sample period.

To encapsulate, this study’s findings reveal the beneficial effects of digital transformation on corporate ESG performance and its mechanisms, providing important theoretical and practical insights for companies undergoing digital transformation. However, due to the regional, industrial, and temporal limitations of the sample companies, the general applicability of the empirical results is constrained. Subsequent studies may delve deeper into the varying degrees of influence that digital transformation exerts on ESG performance for different types of companies and the variations across different regions and industries, thereby providing more targeted recommendations for companies to formulate digital transformation strategies.