Digital Transformation and Firm ESG Performance: The Mediating Role of Corporate Risk-Taking and the Moderating Role of Top Management Team

Abstract

As digital technology and corporate management increasingly converge, enterprises are actively pursuing digital transformation to enhance their environmental, social, and corporate governance (ESG) performance, thereby seeking to cultivate novel competitive advantages. This paper studies the impacts of risk-taking and top management team (TMT) as distinct mechanisms on the digital transformation and ESG performance of enterprises. The empirical findings demonstrate that digital transformation has a positive influence on corporate risk-taking, as well as further enhancing the ESG performance of enterprises. Additionally, it was found that educational level, as one of the characteristics of TMT, can moderate digital transformation’s impact on ESG performance. This paper enriches relevant research on digital transformation and expands the path for how companies can enhance their ESG performance by digital transformation, which can better empower businesses and contribute to their sustainable development.

1. Introduction

The growing emphasis on sustainable development, underscored by the 2030 Agenda for Sustainable Development, has significantly increased the attention given across various sectors to ESG principles, which are deemed essential for global economic and social sustainability [1]. However, the majority of companies are confronted with challenges like high costs and inadequate returns on investment when improving their ESG performance [2]. Many scholars assert that digital technologies substantially enhance enterprise performance across these ESG dimensions [3]. Digital transformation in enterprises means managers implement digital technologies to fundamentally change the form, function, and structure of an organization. This process reshapes the business to create new value, revolutionizing how it operates, interacts, and competes in a rapidly evolving digital landscape, and creates new value for the enterprise itself [4]. Enterprises adopt digital transformation for a variety of reasons and objectives, each of which aim to harness technology to fundamentally redefine their organizational operations and strategies. Some aim to streamline operations and increase efficiency by transforming internal organizational structures with digital transformation, while others seek to enhance communication and connectivity with external consumers or suppliers. In addition, some companies reshape their existing models through digital platforms to influence industry competition rules and have a broader impact on the entire industry [5].

The digital economy era generates constant changes in the external environment and digital technologies of enterprises. While these changes create more opportunities, they also pose significant challenges to companies. Digital transformation continues to impact various aspects of enterprises’ internal structures and operations. Many enterprise leaders consider digital transformation as their primary challenge, and 70% of them believe that their enterprises have not achieved their expected goals after transformation [6]. Large organizations like GE, Nike, and Burberry, among others, have also experienced varying degrees of failure during their transformation efforts [7]. Typically, the primary objective of business operations is to maximize shareholder value. However, companies that fail in their digital transformation may lack the resources and financial capacity to focus on their ESG performance. Therefore, when companies choose risky strategies like ESG, their risk affordability becomes particularly important.

Enterprise risk-taking is an integral part of enhancing ESG performance and facilitating development [8]. It represents the risks that enterprises are willing to take in pursuit of profit, and the risk affordability of a company is crucial for its long-term performance [9]. Risk-taking capacity is not static; a company’s risk preferences will be constantly adjusted with changes in external and internal environments [10]. Dai et al. [11] argued that digitalization in enterprises not only boosts their innovation capabilities and optimizes resource utilization, but also enhances their overall value. This progression subsequently increases the risk level that companies are likely to undertake for project investments. This paper proposes that, concurrently with digital transformation, companies can effectively enhance their risk-taking capacity, thereby strengthening their ESG performance. Therefore, in this paper, enterprise risk-taking will be studied as a mediating variable, which exists between corporate ESG performance and digital transformation.

In the existing literature, there are divergent views among scholars regarding the connection between ESG performance and digital transformation. Some scholars argue that the impact of digital transformation on ESG follows a U-shaped pattern [12,13], implying that digitalization may have negative effects on an enterprise’s ESG performance after reaching a certain threshold. On the contrary, other scholars believe that digital transformation continues to positively enhance ESG performance [14,15]. Unfortunately, research indicates that the relevancy between digital transformation and corporate ESG performance remains underexplored, with significant theoretical gaps and practical shortcomings in how businesses can boost their ESG rankings through digital transformation processes.

Enterprises can achieve competitive advantages by prioritizing various ESG factors. Executive decisions and behaviors have a significant impact on a company’s direction and performance [16], crucially influencing outcomes in terms of overcoming challenging issues [17]. It is crucial to understand which characteristics best facilitate successful corporate adaptation to these changes [18]. Taking Hyundai Motor’s ESG corporate management as an example, their sales of eco-friendly cars increased by 37% in 2023 compared to 2022 [19]. Within its unique ESG governance framework, Hyundai Motor strengthens its management activities by proactively identifying and mitigating risks related to environmental, social, and governance (ESG) factors. For instance, in 2024, Hyundai Motor partnered with Healthy Seas to address pollution from abandoned fishing nets in Greece, thus promoting marine conservation and advancing the circular economy [20]. Furthermore, building on the scholarly insights of Mirza et al. [21], which emphasize the pivotal role of senior management in advancing digital transformation within organizations, this paper taps into the research gap identified by Alkaraan et al. [22] which underscored the need to explore how decision-maker attributes influence strategic digital choices in enterprises. To address this research gap, this study will examine whether the top management team impacts the ESG performance of enterprises undergoing digital transformation, with the top management team acting as a moderating variable in this context.

Therefore, this study investigates the impact of digital transformation on corporate ESG development by sampling A-share listed companies from 2010 to 2011. With stakeholder theory, this study explores how digital transformation influences corporate ESG performance. The frequency of digitalization-related keywords in annual reports is used as an indicator of the extent of digital transformation within each company. Using a double-fixed effects model, this study empirically analyzes how the extent of corporate digital transformation directly impacts a company’s ESG performance. Subsequently, by integrating the principal–agent theory and the signaling theory, this study identifies the mediating role of risk-taking. Lastly, guided by the upper echelons theory, this study reports that the educational level of the top management team can moderate the relationship between digital transformation and ESG performance. These conclusions remain valid after undergoing robustness tests.

Our study makes three potential contributions. Firstly, grounded in the context of corporate sustainability, we use extensive empirical data to illustrate the pivotal role of digital transformation in enhancing corporate ESG development. This analysis enriches the existing literature on the impact of digital transformation on ESG development. Secondly, the existing research on enhancing ESG performance through internal drivers is limited. Our study provides an in-depth analysis of the mechanism by which the corporate risk-taking affects ESG performance, thus elucidating how digitalization strategies enhance ESG ratings and demystifying the ‘black box’ of this relationship. Thirdly, our research further validates the importance of education levels among top management team members in moderating the relationship between digital transformation and ESG performance. This result confirms that executives with different traits may exert varying effects within this complex mechanism, deepening our understanding of the role of executive characteristics in this domain.

2. Theoretical Background and Hypotheses

2.1. Digital Transformation and ESG Performance

Digital transformation, also known as Industry 4.0, is essential for enabling businesses to effectively navigate through increased volatility and uncertainty [23]. Growing interest among scholars highlights digital transformation as a key catalyst to effectively lead companies in their sustainable development and tackling of challenges, thus enhancing their ESG performance [24,25,26].

The ESG framework, first proposed by the United Nations in 2004, mandates that enterprises align shareholder interests with social responsibility across three dimensions: environmental, social, and corporate governance. This approach emphasizes holistic corporate practices that contribute simultaneously to societal benefits and shareholder value.

With the endorsement of the ESG concept by developed countries, a series of standards and systems related to ESG, such as the ESG disclosure guidelines, have gradually been established. The establishment of these standards and systems aligns well with China’s advocacy for sustainable development and green growth. In addition, academic research indicates that digital transformation holds significant potential for enhancing corporate ESG performance. Wang et al. [14] observed that in the manufacturing sector, a 1% increase in digitalization improved ESG performance by 0.124%; Zhao and Cai [2] proposed that digital transformation can reduce negative ESG impacts and bolster investments in ESG for heavily polluting enterprises, thus enhancing their overall ESG performance, and Lu et al. [27] reported that high-tech enterprises derive more significant ESG benefits from digital transformation. Consequently, more and more Chinese companies are actively improving their ESG performance to catch more social attention [28] and enhance their financial performance [29].

Based on stakeholder theory, digital transformation in enterprises encompasses more than shareholder interests, integrating the concerns of various stakeholders. This transformation employs digital innovations, for example, cloud computing, artificial intelligence and big data, aiming to enhance both the organization’s and stakeholders’ high-quality development. First, digital transformation boosts resource efficiency, enhances pollution control, and promotes energy conservation, elevating the enterprise’s ESG performance. Secondly, it enhances data processing and transparency, reducing information asymmetry and facilitating better ESG performance monitoring. Finally, an enterprise’s financial performance can be improved by digital transformation [14]. Companies with stronger financial performance typically have higher ESG rankings, demonstrating the close relationship between financial performance and ESG [30].

Built upon the above understanding, a research hypothesis is proposed.

H1. 

Digital transformation can promote firms’ ESG performance.

2.2. The Mediating Role of Corporate Risk-Taking

Risk-taking implies a company’s inclination to pay the price while pursuing the benefits, so it is a crucial part of strategic decision-making for enterprises [31]. In line with the principal–agent theory, information asymmetry will lead management teams to be inclined towards more conservative decision-making in order to safeguard themselves, resulting in their reluctance to take risks in pursuing profits and thus, to some extent, reducing the company’s risk-taking. As enterprises advance their digitalization, emerging technologies could facilitate enhanced risk management [11]. In their research, Fischer et al. [32] highlighted that implementing business process management (BPM) in digital transformation enables the continuous improvement of organizational structure, enhancing decision supervision and execution effectiveness while reducing self-serving employee behaviors. Wang et al. [33] confirmed that enterprises leverage digitalization to enhance their informational effects and resource effects, thereby increasing their risk-taking capacity. These findings underscore the relationship between digitalization and corporate risk-taking from the perspective of principal–agent theory. Meanwhile, according to the signaling theory, companies that are successful in digital transformation can send positive signals to society, thereby establishing a favorable corporate image. Plekhanov et al. [34] used an innovative multi-layered framework to confirm that digitalized communication channels improve the interaction between enterprises and the external environment, expanding corporate boundaries to include a wider array of external stakeholders. Enterprises help investors better understand their production processes, internal operations, sales performance, and other information by promptly providing important internal information to external stakeholders. This improves external market support [35]. It can even help withstand shocks in times of crisis by rallying investor confidence [36], thereby improving enterprises’ risk-taking capability. Taking the contrary approach will result in the disclosing of information that may have a negative impact on enterprises.

To sum up, enterprises with a higher degree of digitalization can not only optimize resource allocation but also utilize resources more efficiently. This enhanced capacity allows for the allocation of resources to innovative projects characterized by high uncertainty, long durations, and substantial risks, thus stimulating innovative behaviors within enterprises. Furthermore, the digital economy bolsters the risk-taking of enterprises by enhancing their innovation and financial capabilities. This also promotes the continuity and steady development of enterprises, aligning them with broader ESG goals. By fostering robust digitalization, organizations can better manage and mitigate the risks associated with their operations and strategic initiatives, leading to improved ESG performance [37]. According to the above, the following hypothesis is raised:

H2. 

Corporate risk-taking mediates the relation between the enterprise digital transformation and ESG performance.

2.3. The Moderating Role of Top Management Team

Hambrick and Mason [16] stated in their upper echelons theory that the personalities and cognitive habits of TMTs significantly affect their decisions on strategies, particularly in complex scenarios such as digital transformation. According to the theory, the TMT’s background, values, and cognitive complexity shape how they interpret information and make strategic choices that affect the firm’s adoption of digital technologies, which then affect the corresponding performance results. As a result, senior managers are challenged to generate and execute their company’s business strategy, so they need to consider the opportunities and risks they face as they undertake digital transformation.

Indeed, it has been proved that managers’ power influences the behavior and performance of enterprises [38]. Therefore, the top management team is a topic that scholars often focus on in studies involving enterprise reform, transformation, and innovation. Reviewing existing researches, the association between digital transformation and the features of enterprise TMTs has been widely discussed by scholars. Zhang and Jin [39] found that managers with more power tend to prioritize support for innovative behaviors such as digital strategic transformation, considering the alignment with enterprise interests and their own reputation. Mirza et al. [21] believed that individuals in the top management team with academic backgrounds, such as professors or scholars, can promote digital transformation through their additional resources and capabilities. Zhang and Bu [40] argued that top management teams with backgrounds in information technology management have ample professional experience and understanding of advanced technology, so they can recognize the importance of technology and innovation, enabling them to actively participate in and drive the success of enterprises’ digital transformation. Similarly, the link between ESG and the corporate top management team has also been studied. Chen et al. [41] suggested that general managers’ responsible leadership can enhance organizational resilience to improve ESG performance. Kao et al. [42] proposed that managers with higher capabilities possess better thinking and superior resource allocation abilities, and they are more willing to disclose their company’s ESG performance to gain a competitive advantage. He et al. [43] argued that companies’ steps in terms of ESG can increase external supervisory pressure, thereby reducing managers’ misconduct. Such discussions abound, yet there has been no in-depth research on how the top management team moderates the relationship between digital transformation and ESG performance.

In upper echelons theory, the dimensional description of the characteristics of managers is mainly divided into two aspects: one is personality traits, such as values; the other is demographic traits, such as age, gender, and educational background, etc. Among them, the educational backgrounds of the top management team can reflect the knowledge and technical foundations of managers. Educational backgrounds can vary in accordance with different standards, such as differences in the type of education, the level of education, and whether or not they have overseas study experience. This study will set the level of education as the research object.

Digital transformation is a process of reshaping and innovating various aspects of a company, which has far-reaching implications and demands on the roles, skills, and capabilities of the top management team [44]. On one hand, when top management team members have higher levels of education, they typically have longer learning experiences and a higher level of education which can provide them with more resources and networks. Individuals with higher levels of education often possess more specialized and scientific knowledge, enabling them to effectively lead reform and innovation, thus facilitating the digital transformation process. On the other hand, when top management team members have lower levels of education, it signifies that they entered the workforce earlier and have accumulated more extensive work experience and professional skills over time. Their opinions and decisions are often more feasible due to the integration of their practical experience [45]. The different educational levels within the top management team can characterize them with different traits, thereby moderating the process of digital transformation within the company and further influencing its ESG performance.

In summary, we believe that different levels of education may mediate the relationship between digital transformation and ESG performance. Therefore, we propose the following hypothesis:

H3. 

TMTs’ education level moderates the relationship between digital transformation and ESG performance, such that their positive relationship is stronger when the education level is higher.

Our study proposes a theoretical model to assess how digital transformation influences ESG performance, moderated by TMTs’ education level and mediated by risk-taking. We have devised a theoretical framework, as illustrated in Figure 1. This model serves as a roadmap for the ensuing empirical analysis, the details of which follow in subsequent sections.

3. Research Methods

3.1. Data and Research Sample

This paper meticulously investigates the influence of digital transformation within corporations on their environmental, social, and governance (ESG) performance. This analysis is conducted through a comprehensive study of Chinese A-share listed companies, spanning the period from 2010 to 2021. The starting point chosen is 2010, coinciding with the release of the Guidelines for Environmental Information Disclosure of Listed Companies (Draft for Solicitation of Comments) by the former Ministry of Environmental Protection. In accordance with the policy directives outlined, enterprises pay more attention to social responsibility and corporate governance. The data were collected until 2021, as there are significant data gaps for samples after 2021. To ensure the accuracy of the study data, 2021 was chosen as the ending point for available data. To mitigate interference from other factors, the study first excluded samples from the financial industry, followed by samples with “ST” and “*ST” designations which represent companies subject to the warnings of special treatment and delisting risk. Finally, this study encompasses an analysis of 39,448 observations from 5016 listed companies, post-adjustment for extreme values with a 1% winsorization on continuous variables. The data sources include annual reports for digital transformation insights, Bloomberg’s ESG performance metrics, and the China Stock Market & Accounting Research Database (CSMAR) for the educational levels of senior management, as well as financial and governance information. Statistical analysis was conducted using Python 3.11 and Stata 17.0.

3.2. Variables

3.2.1. Dependent Variable

In this paper, the dependent variable is ESG performance. Currently, scholars frequently employ the evaluation scores provided by third-party institutions as a metric to assess the ESG performance of enterprises, ensuring a standardized and objective approach in their research. This study used the Bloomberg ESG disclosure index, which includes comprehensive scores of ESG as well as sub-scores for environmental, social, and corporate governance.

3.2.2. Independent Variable

Digital transformation, this study’s core independent variable, is an extremely complex process. All companies listed on the Shanghai and Shenzhen stock exchanges in China publish their annual reports each year. The content of their annual reports reflects the company’s operating conditions over the past year and shows its basic situation, general accounting data, financial indicators, and other relevant information [46]. This paper maintains that the annual report is the summary and guidance report of the enterprise. Therefore, digital transformation, as a major strategy for the high-quality development of enterprises, is easier to quantify in the annual report. Furthermore, this study makes statistics according to the word frequency related to “digital transformation” to reflect the feasibility of the degree of transformation of the enterprise.

In this paper, methodologies are synthesized from previous research, specifically studies by Yang et al., Wu et al. and Wu et al. [46,47,48], to craft a multi-step approach for gauging the extent of digital transformation within enterprises. The process commenced with the aggregation of annual reports from companies listed on the Shanghai and Shenzhen stock exchanges from 2012 to 2021, concentrating on the “Management Discussion and Analysis” (MD&A) section. This section is particularly pivotal as it is traditionally where firms articulate their strategic directions and business performance, thus serving as a rich source of information for our analysis. Secondly, following the study by Chen and Xu [49], this research constructs a keyword database, consisting of 76 keywords, to describe enterprise digital transformation, such as “artificial intelligence”, “business intelligence”, “image understanding” and so on. The third step in the research process involved utilizing Python to extract and clean data concerning specific keywords, remove the stop words, descriptions containing keywords but unrelated to digital transformation, and keywords outside the enterprise from the annual reports that had been previously processed (such as shareholders, customers, etc.). Finally, we quantify the level of digitalization by counting the occurrences of the 76 selected keywords within each company’s MD&A section. This count serves as our indicator of the company’s digitalization level. At the same time, considering the typical “right-skewness” of such data, this study takes their pair numbers to obtain the final index, which is recorded as Dig.

3.2.3. Mediating Variable

In this study, the volatility of corporate earnings is used to measure enterprises’ risk-taking [9]. The higher the volatility value of earnings, the higher the enterprise’s risk-taking ability, and vice versa.

The volatility level of return on assets (Roa) over the observed period is captured to reflect the enterprise’s risk-taking. Roa is calculated by the company’s earnings before interest and tax (EBIT) divided by their total assets (ASSET) at the end of the year. In order to reduce the impact of industry and cycle, this study refers to the practice of Jiang and Chen [50]. Firstly, the Roa of the company is subtracted from the average annual industry average to obtain Adj_Roa (as shown in Equation (1)). Then, the industry classification is refined and samples with only one company in the industry are excluded. Finally, Equations (2) and (3) are applied to obtain the standard deviation and range of Roa calculated on a rolling basis every three years. This study refers to the research method of Faccio et al. [51], and multiplies the result of the sum by 100 to make the result more intuitive, without affecting its significance.

$$Adj\_Ro{a}_{i,t}=\frac{EBI{T}_{i,t}}{ASSE{T}_{i,t}}-\frac{1}{x}{{\displaystyle \sum }}_{k=1}^x\frac{EBI{T}_{i,t}}{ASSE{T}_{i,t}}$$

(1)

$$Risk{1}_{i,t}=\sqrt{\frac{1}{T-1}{{\displaystyle \sum }}_{t=1}^T{(Ad{j_{Roa}}_{i,t}-\frac{1}{T}{{\displaystyle \sum }}_{t=1}^{T}Adj\_Ro{a}_{i,t})}^2}|T=3$$

(2)

$$Risk{2}_{i,t}=Max\left(Ad{j_{Roa}}_{i,t}\right)-Min\left(Ad{j_{Roa}}_{i,t}\right)$$

(3)

In the summation equations, the subscript $i$ represents the enterprise, the subscript $t$ represents the year, the subscript $x$ serves as the upper limit, k is the iterative variable, and T represents the time periods.

3.2.4. Moderating Variable

In this study, the educational differences of team members are reflected by calculating the Herfindahl index. Based on the highest educational level disclosed in the company’s annual report, the top executive teams’ members are categorized and assigned values to. The educational levels of executive members are classified into the following five categories: high school or below (assigned a value of 1), associate’s degree (assigned a value of 2), bachelor’s degree (assigned a value of 3), master’s degree (assigned a value of 4), and doctoral degree (assigned a value of 5).

3.2.5. Control Variables

To improve the accuracy of this study, a series of control variables are considered in the analysis. Each variable has the potential to influence a company’s ESG performance, including the proportion of independent directors (Indep), equity concentration (Top1), asset–liability ratio (Lev), turnover of total capital (Ato), share ratio of institutional investors (Inst), return on equity (Roe), and earnings per share (Eps), with all variables defined in

Table 1. Definition and measurement of variables.

Symbol	Variable	Measurement
ESG	ESG performance	According to Bloomberg ESG rating
Dig	Digital transformation	Ln (number of digital transformation keywords + 1)
Risk	Corporate risk-taking	Measured by Equation (3)
Edu	TMTs’ education level	Education level: high school or below equals 1, junior college equals 2, undergraduate equals 3, master’s degree equals 4, PhD equals 5
Indep	Proportion of independent directors	Number of independent directors/number of directors
Top1	Equity concentration	Percentage shareholding of the largest shareholder
Lev	Asset–liability ratio	Total liabilities/total assets
Ato	Turnover of total capital	Operating income/average total assets
Inst	Share ratio of institutional investors	Total shares held by institutional investors/the outstanding shares
Roe	Return on equity	Net profit/shareholders’ equity
Eps	Earnings per share	Net income/shares of common stocks

.

3.3. Empirical Model

To further understand how digital transformation affects ESG performance, with a focus on the impact of the TMT and risk-taking, this study developed a two-way fixed-effect model, as shown in Equation (4). This model was designed to rigorously test the first hypothesis (H1) by assessing the empirical relationship between a firm’s digital initiatives and subsequent ESG outcomes. It takes into account various control variables that could potentially confound this relationship, providing a more precise and holistic view of the direct effects of digital transformation strategies on ESG performance metrics.

$$ES{G}_{i,t}={\alpha }_0+{\alpha }_{1}DI{G}_{i,t}+{\gamma }_{j}C{V}_{i,t}+\mu +\theta +{ϵ}_{i,t}$$

(4)

In this equation, $ESG$ reflects enterprise ESG performance, $DIG$ is the degree of digital transformation of the company, ${\alpha }_0$ represents the intercept, ${\alpha }_1$ denotes the coefficient for the influence of digitalization level on ESG performance, ${\gamma }_j$ represents the coefficients of control variables affecting ESG performance, $C{V}_{i,t}$ represents a set of control variables at the enterprise level, $\mu$ and $\theta$ show the impacts of enterprise $i$ not being observed in year t, and ${ϵ}_{i,t}$ signifies the error term.

To verify the hypothesis H2, Equations (5) and (6) were set up to explore the mediating role of enterprise risk-taking between digital transformation and ESG performance; ${\alpha }_2$ denotes the coefficient for the influence of risk-taking on ESG performance as follows:

$$Ris{k}_{i,t}={\alpha }_0+{\alpha }_{1}DI{G}_{i,t}+{\gamma }_{j}C{V}_{i,t}+\mu +\theta +{ϵ}_{i,t}$$

(5)

$$ES{G}_{i,t}={\alpha }_0+{\alpha }_{1}DI{G}_{i,t}+{\alpha }_{2}Ris{k}_{i,t}+{\gamma }_{j}C{V}_{i,t}+\mu +\theta +{ϵ}_{i,t}$$

(6)

To verify H3, the moderating effect of the educational levels of the top management team on digital transformation and ESG performance is reflected by Equation (7); $EDU$ is the education level of the TMT, and ${\alpha }_3$ denotes the impact of the interaction between $DIG$ and $EDU$ on ESG performance.

$$ES{G}_{i,t}={\alpha }_0+{\alpha }_{1}DI{G}_{i,t}+{\alpha }_{2}ED{U}_{i,t}+{\alpha }_{3}DI{G}_{i,t}\times ED{U}_{i,t}+{\gamma }_{j}C{V}_{i,t}+\mu +\theta +{ϵ}_{i,t}$$

(7)

4. Empirical Results and Analysis

4.1. Descriptive Statistics

Table 2. Descriptive statistics.

Variable	N	Mean	SD	Min	Max
ESG	12,561.000	28.649	9.138	11.719	55.950
LnDT	18,762.000	2.306	1.254	0.693	5.318
Risk2	33,011.000	0.428	2.469	0.015	27.174
Edu	33,299.000	3.377	0.477	2.000	4.500
Indep	37,378.000	37.581	5.325	33.330	57.140
Top1	37,381.000	34.089	14.967	8.413	74.451
Lev	37,500.000	0.436	0.223	0.050	0.975
Ato	37,484.000	0.628	0.444	0.026	2.563
Inst	37,359.000	44.296	24.732	0.326	91.820
Roe	37,207.000	0.047	0.181	−1.182	0.340
Eps	37,495.000	0.391	0.636	−1.700	2.970

displays the descriptive statistical results of the main variables in this paper. According to the results, the average value of the enterprises’ ESG performance is 28.649 and the standard deviation is 9.138. This result reveals that China’s Shenzhen and Shanghai listed companies generally obtain a relatively good ESG performance, but there are still large differences among their uneven development. The standard deviation of enterprise digital transformation is 1.254, and the difference between the maximum value and the minimum value is 4.625, which indicates that the degree of digital transformation of different enterprises varies greatly.

4.2. Correlation Analysis

The correlation analysis of each major variable is summarized in

Table 3. Pearson correlation analysis.

	ESG	LnDT	Indep	Top1	Lev	Ato	Inst	Roe	Eps
ESG	1.000
LnDT	0.032 ***	1.000
Indep	0.069 ***	0.036 ***	1.000
Top1	−0.003	−0.141 ***	0.037 ***	1.000
Lev	0.144 ***	−0.077 ***	−0.007	0.013 **	1.000
Ato	−0.042 ***	−0.020 ***	−0.025 ***	0.096 ***	0.055 ***	1.000
Inst	0.209 ***	−0.132 ***	−0.069 ***	0.492 ***	0.208 ***	0.068 ***	1.000
Roe	0.038 ***	−0.038 ***	−0.026 ***	0.130 ***	−0.219 ***	0.162 ***	0.119 ***	1.000
Eps	0.186 ***	−0.007	−0.009 *	0.133 ***	−0.201 ***	−0.201 ***	0.148 ***	0.633 ***	1.000

Note: *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively.

. The Pearson correlation coefficient between digital transformation and ESG performance is 0.032, which is significant at the 1% level. The data initially reveal that digital transformation has a positive impact on ESG, which is consistent with the expected results. However, the influence of other factors has not been considered in research into the relationship between the two, so further research is needed. Meanwhile, the maximum value of the relationship coefficient between each variable is 0.633, which is less than 0.8 [52], suggesting that there is no serious multicollinearity problem, so it is suitable to perform regression analysis next.

4.3. Baseline Regression Results

The analysis delineated in

Table 4. Baseline regression test results.

	(1)	(2)	(3)	(4)
	Model 1	Model 2	Model 3	Model 4
Variables	ESG	ESG	Risk2	ESG
LnDT	1.505 ***	0.289 ***	0.055 *	0.324 ***
	(15.68)	(4.06)	(1.87)	(4.19)
Risk2				0.045 *
				(1.85)
Indep		0.025	−0.002	0.032 **
		(1.61)	(−0.26)	(1.97)
Top1		0.003	−0.001	−0.003
		(0.28)	(−0.29)	(−0.23)
Lev		−1.037	−1.175 ***	−1.268 *
		(−1.61)	(−4.85)	(−1.83)
Ato		−0.743 **	−0.029	−0.818 ***
		(−2.55)	(−0.26)	(−2.61)
Inst		0.039 ***	0.001	0.036 ***
		(5.63)	(0.46)	(4.97)
Roe		−0.978 *	−0.628 ***	−1.313 **
		(−1.87)	(−3.08)	(−2.08)
Eps		1.058 ***	0.170 **	1.123 ***
		(6.69)	(2.29)	(6.43)
Constant	27.395 ***	17.088 ***	5.516 ***	17.000 ***
	(94.37)	(18.91)	(14.36)	(17.47)
Year	Yes	Yes	Yes	Yes
Ind	Yes	Yes	Yes	Yes
Observations	6690	6661	15,780	5691
R-squared		0.642	0.162	0.646
Number of code	1339	1337	3542	1187
adj_R2		0.551	−0.0821	0.551
F		529.5	124.4	431.7

Note: t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

’s initial columns reveals a positive dynamic between digital transformation efforts, quantified as LnDT, and the enhancement of ESG performance. Model (1) and Model (2) are computed based on Equation (4), and Model 1 does not include the control variables. This positive trajectory is observed consistently, irrespective of whether additional control variables are factored into the regression or not. Specifically, a 1% elevation in digital transformation is associated with an increment of 1.505 and 0.289 in ESG scores in scenarios excluding and including control variables, respectively. These findings underscore the integral role of digital transformation in bolstering ESG outcomes, thereby validating Hypothesis 1 (H1) of this study. Thus, the results suggest that digitalization acts as a catalyst for advancing the ESG ratings within enterprises.

4.4. Mediating Effect Analysis

Following Baron and Kenny’s [53] causal steps approach, we explored the mediating relationships in this study. Model 3 is derived from Equation (5) and Model 4 is derived from Equation (6). The exploration of the mediating role of risk-taking, denoted as Risk2, was conducted to understand its interplay with digital transformation (LnDT) in driving ESG performance. Empirical evidence presented in Model 3 within Table 4 reveals a statistically significant nexus between digital transformation and risk-taking at the 10% significance level. Proceeding to Model 4, this mediating variable, Risk2, demonstrates a positive facilitation in the relationship between digital transformation efforts and ESG outcomes. In conjunction with Hypothesis 1 (H1), this finding elucidates that as organizations’ willingness to embrace risk increases, the influence of digital transformation on enhancing ESG performance is magnified, thereby affirming Hypothesis 2 (H2). This underscores the pivotal role of risk-taking in amplifying the benefits of digitalization on ESG performance.

4.5. The Moderating Effect Analysis

Model 5 is derived from Equation (7), and the analysis in

Table 5. The moderating effect results.

	(1)
	Model 5
Variables	ESG
LnDT	0.064
	(0.65)
Edu	3.986 ***
	(16.20)
DTEdu_c	0.361 *
	(1.82)
Indep	0.095 ***
	(4.96)
Top1	−0.062 ***
	(−7.61)
Lev	6.015 ***
	(11.10)
Ato	−1.176 ***
	(−5.39)
Inst	0.095 ***
	(15.65)
Roe	−5.094 ***
	(−5.71)
Eps	2.595 ***
	(13.18)
Constant	5.739 ***
	(5.16)
Year	Yes
Ind	Yes
Observations	6106
R-squared	0.199
adj_R2	0.198
F	151.8

Note: t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

highlights the moderating role of the top management team’s education level between digital transformation and corporate ESG. The positive and significant coefficient of the interaction term DTEdu_c at the 10% level indicates that as education level of the TMT increases, so does the positive impact of digital transformation on ESG performance, supporting hypothesis H3.

4.6. Robustness Tests

(1) To test the robustness test of main variable, in consideration of the fact that the effect of digital transformation has a certain time lag,

Table 6. Robustness tests results for main effects.

	(1)	(2)
	Model 6	Model 7
Variables	ESG	ESG
lagLnDT	1.311 ***	0.165 **
	(13.36)	(2.27)
Indep		0.022
		(1.36)
Top1		−0.008
		(−0.67)
Lev		−1.130
		(−1.63)
Ato		−0.793 ***
		(−2.62)
Inst		0.041 ***
		(5.46)
Roe		−1.130 **
		(−2.11)
Eps		1.071 ***
		(6.40)
Constant	28.342 ***	17.550 ***
	(96.18)	(18.15)
Year	Yes	Yes
Ind	Yes	Yes
Observations	5920	5902
R-squared		0.627
Number of Code	1262	1262
adj_R2		0.524
F		431.4

Note: t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

was obtained after incorporating explanatory variables with one-lag period and applying them to Equation (4). According to the results of Model 6, when no control variable is added, the regression coefficient of the one-lag digital transformation (LagLnDT) is positive and significant. In Model 7, with the control variable added, the conclusion still holds. These data suggest that digital transformation can still significantly improve enterprises’ ESG performance even if the explanatory variables are processed with one-lag period.

(2) The robustness test of the mediating effect was conducted by the substitution of mediating variable. Using Equation (3), we obtained the variable risk when T = 5. Subsequently, we repeated the previous steps by incorporating this into Equations (5) and (6).

Table 7. Robustness tests results for the mediating effect.

	(1)	(2)	(3)
	Model 8	Model 9	Model 10
Variables	ESG	Risk1	ESG
LnDT	0.289 ***	0.197 **	0.323 ***
	(4.06)	(2.00)	(4.19)
Risk1			0.014 *
			(1.95)
Indep	0.025	−0.006	0.032 **
	(1.61)	(−0.27)	(1.97)
Top1	0.003	0.001	−0.003
	(0.28)	(0.05)	(−0.23)
Lev	−1.037	−3.908 ***	−1.265 *
	(−1.61)	(−4.83)	(−1.82)
Ato	−0.743 **	0.025	−0.817 ***
	(−2.55)	(0.06)	(−2.61)
Inst	0.039 ***	0.002	0.036 ***
	(5.63)	(0.26)	(4.98)
Roe	−0.978 *	−1.538 **	−1.315 **
	(−1.87)	(−2.26)	(−2.08)
Eps	1.058 ***	0.535 **	1.123 ***
	(6.69)	(2.17)	(6.43)
Constant	17.088 ***	14.240 ***	16.994 ***
	(18.91)	(11.11)	(17.47)
Year	Yes	Yes	Yes
Ind	Yes	Yes	Yes
Observations	6661	15,780	5691
R-squared	0.642	0.148	0.647
Number of Code	1337	3542	1187
adj_R2	0.551	−0.100	0.552
F	529.5	111.8	431.7

Note: t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

was obtained by calculating the standard deviation and range of Roa (Risk1) on a rolling basis every five years, while other factors are unchanged. As shown in the results, with the replacement of samples of intermediary variables and the added intermediary variable of corporate risk-taking, digital transformation still positively effects ESG performance. After adding the intermediary variable, the regression coefficient of 0.323 is higher than without the intermediary variable at 0.289, indicating that with the influence of the intermediary role of risk-taking, digital transformation has a stronger impact on ESG performance, which supports the previous conclusion.

(3) In the test of the robustness of the moderating effect, this study also processes the digital transformation with one-lag period and incorporates it into Equation (7) to verify the results of the moderate variables, with the results shown in

Table 8. Robustness tests results for the moderating effect.

	(1)
	Model 11
Variables	ESG
lagLnDT	−0.003
	(−0.03)
Edu	3.619 ***
	(14.11)
LnDTEdu_c	0.404 *
	(1.91)
Indep	0.088 ***
	(4.45)
Top1	−0.062 ***
	(−7.26)
Lev	5.991 ***
	(10.59)
Ato	−1.263 ***
	(−5.58)
Inst	0.097 ***
	(15.44)
Roe	−4.280 ***
	(−4.71)
Eps	2.614 ***
	(12.65)
Constant	7.680 ***
	(6.66)
Year	Yes
Ind	Yes
Observations	5437
R-squared	0.204
adj_R2	0.202
F	138.7

Note: t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

. It is shown that the interaction term LnDTEdu_c is 0.404, which is still significant at the 10% level, so the result is consistent with the previous conclusion in this paper.

5. Conclusions and Discussion

This study delves into the nuanced impact of digital transformation on ESG performance, offering insights that are crucial for both academic research and industry applications. It underscores the importance of ESG performance as a metric for corporate social responsibility and sustainable economic progress. As the digital economy expands, an increasing number of enterprises are embracing digital transformation. Understanding how digital transformation can be leveraged to enhance ESG rankings is vital for guiding sustainable development within China’s rapidly evolving economic landscape. This discussion contributes to a deeper comprehension of the dynamic interplay between digital innovation and ESG achievements. In addition, this study contributes to filling the gap in the existing literature on ESG, which has largely focused on external motivators and micro impacts, by exploring the internal drivers of ESG performance within enterprises. This is particularly relevant in the digital economy’s fast-paced environment.

Based on a dataset spanning from 2010 to 2021 of the Shenzhen and Shanghai A-share listed enterprises, this study explores the impact of digitalization on ESG performance by a double-fixed effects model. It also investigates the mediating influence of risk-taking and the moderating effect of top management education levels on this relationship. The findings, which remain robust after stability checks, underscore the significant role digital transformation plays in enhancing ESG outcomes. This paper contributes to understanding the nuanced dynamics between digital transformation, corporate risk-taking, and the education level of TMTs in advancing ESG goals. First, this study confirms that digital transformation significantly enhances ESG performance by utilizing digital technologies to optimize resource management and improve transparency, thus minimizing information asymmetry. This gain in efficiency not only optimizes investment and outputs but also significantly enhances financial performance. Such improvements in operational and financial domains contribute to better ESG outcomes by fostering more sustainable and socially responsible business practices. Second, this research suggests that digital transformation significantly enhances an enterprise’s capacity for risk-taking, which in turn positively impacts its ESG performance. This improvement in risk tolerance facilitates strategic investments into ESG initiatives, which are crucial for sustainable business practices. By leveraging digital tools and technologies, enterprises are not only able to manage risks better but also commit more effectively to their social and environmental responsibilities. This dynamic underscores the transformative power of digital strategies in aligning business operations with sustainability goals. Third, the success of digital transformation in enhancing ESG performance heavily relies on the involvement of highly educated senior managers. Well-educated leaders bring a wealth of knowledge and robust skill sets that are critical in navigating the complexities of digitalization. Their strategic oversight and problem-solving capabilities are invaluable in overcoming the challenges that arise during digital shifts. Moreover, their educational backgrounds enable them to foster an organizational culture that supports continuous innovation and effective integration of ESG goals into corporate strategies, thereby strengthening the enterprise’s commitment to sustainable practices.

5.1. Theoretical Implications

This paper makes several significant contributions to the extant theoretical frameworks in this field. First, this study demonstrates how digital transformation can enhance ESG performance in Chinese enterprises, which is a priority given the increasing focus on ESG by both the government and stakeholders. Previous research on digital transformation primarily focuses on its economic impacts, such as boosting enterprise value, enhancing competitiveness, fostering innovation, and improving market perception [54], as well as boosting financial outcomes [55]. However, the link between digital transformation and ESG has been underexplored. This study not only solidifies the link between digital transformation and ESG outcomes, but also delves into the mechanisms through which this effect occurs. By incorporating enterprise risk-taking as a mediating variable and executive education level as a moderating variable, this research elucidates the pathways through which digital transformation impacts ESG performance. This approach opens up new perspectives on the dynamic interplay between an enterprise’s digital strategy and its commitment to ESG principles, offering a comprehensive view of the strategic decisions that drive sustainable success. The findings align with recent studies by Zhao and Cai, Lu et al., Wang and Esperança, Ding et al. and Li et al. [2,27,30,56,57], supporting the transformative impact of digital strategies on ESG performance.

Second, this study proves that digital transformation can enhance enterprise risk-taking capability, thereby further influencing its ESG performance. Previous academics have not extensively studied the pathway through which digital transformation affects ESG performance via risk-taking. He et al. [10] demonstrated that pursuing ESG development can reduce risk-taking and lead to more stable development; Dunbar et al. [58] explored the relationships between corporate social responsibility, risk-taking, and CEO incentives, and Liu et al. [31] found that risk-taking mediates the relationship between digital transformation and innovation. As a result, the revelations of this study enrich the pertinent literature in this domain, affirming the existence and benefits of this mechanism.

Finally, this study delves deeper into the influence of top management teams on the effectiveness of digital transformation initiatives and their impacts on ESG performance, following suggestions from existing literature. Recognizing the varied attributes of management teams, which are complex and challenging to quantify, this paper focuses specifically on the education levels of these top managers as a quantifiable and significant factor. This approach allows this study to provide more tangible and reliable insights into the educational backgrounds of top executives to enhance the credibility and outcomes of this research.

5.2. Practical Implications

The research findings provide practical significance and suggestions for enterprise managers and policymakers. First, enterprises should establish a clear strategy for digital transformation. It is of great importance to set well-defined goals for enterprise transformation and make full use of digital tools such as artificial intelligence, data mining, and cloud computing to drive organizational restructuring and innovation across the entire enterprise. During digitalization, simultaneous efforts should be made to improve corporate ESG performance. The aim is to achieve corporate sustainable development. Second, it is crucial to give full attention to the positive impact of digital transformation on corporate ESG performance. On one hand, efforts should be made to strengthen internal governance through the establishment of a robust digital platform to improve collaboration and workflow efficiency. On the other hand, enterprises should put emphasis on digital transformation as they are required to increase external visibility through information-sharing mechanisms to gain stakeholder support. Greater emphasis should be placed on corporate social responsibility fulfilment. Third, it is necessary to construct a well-balanced executive team. Management teams should be allocated based on the characteristics of their executives. In doing so, a balance can be reached between their theoretical knowledge and practical abilities, which improves collaboration efficiency and pushes forward digital transformation.

From the perspective of relevant governmental policymakers, first, the government should lend greater support to corporate digital transformation. By providing guidance to all sectors of society, the government can build a digital infrastructure or offer consulting services regarding digital transformation. The aim is to lay the foundation for corporate digital transformation and lower barriers to entry. Second, ESG-related regulators should expedite the establishment of an ESG information disclosure system. It is equally important to develop an ESG evaluation system based on Chinese characteristics. As they fulfil their regulatory responsibilities, they should simultaneously encourage China-based enterprises to invest in ESG initiatives to promote sustainable development practices.

5.3. Limitations and Future Research

While this paper illuminates the relationship between enterprises’ digital transformation and their ESG performance, it also acknowledges its limitations. There are still some aspects that require further exploration and expansion in future studies. Firstly, when assessing the impact of digital transformation on ESG, this paper only relies on word frequency from text analysis as the measurement of enterprises’ digitalization, and this method has been identified to have limitations. Future research could benefit from incorporating a more comprehensive set of indicators to capture the complexity and breadth of digital transformation.

Secondly, the research sample could be selected with more meticulous efforts. This study’s sample consisted of all listed companies on the Shanghai and Shenzhen stock exchanges, without full consideration of the circumstances faced by enterprises in different industries. Different industries are subject to varying influencing factors, such as market environment, external policies, customer bases, etc., which result in different performance outcomes. As a consequence, it is suggested that future research should categorize industries more finely to explore the development paths and directions of different industries.