Open AccessArticle

The Nonlinear Effects of Digital Finance on Corporate ESG Performance: Evidence from China

Qingmin Yin

^1,2,

Nan Su

^1,2

and

Chenhui Ding

^3,*

Business School, Hohai University, Nanjing 210098, China

School of Economics and Finance, Hohai University, Nanjing 210098, China

Economics and Management School, Dongguan University of Technology, Dongguan 523808, China

Author to whom correspondence should be addressed.

Sustainability 2024, 16(18), 8274; https://doi.org/10.3390/su16188274

Submission received: 13 August 2024 / Revised: 19 September 2024 / Accepted: 22 September 2024 / Published: 23 September 2024

(This article belongs to the Section Economic and Business Aspects of Sustainability)

Download

Browse Figures

Versions Notes

Abstract

Digital finance enhances corporate ESG performance and is essential for achieving sustainable development; however, its consistent effectiveness in improving ESG outcomes remains contested. Using panel data from A-share listed companies on the Shanghai and Shenzhen stock exchanges in China from 2011 to 2021, this study empirically examines nonlinear effects, transmission mechanisms, and moderating factors. The results indicate a U-shaped relationship between digital finance and ESG performance, with a positive impact becoming apparent when digital finance exceeds the threshold of 3.81. Mechanism tests reveal that green technological innovation and public environmental attention are crucial transmission channels for the nonlinear effects. Furthermore, financial regulation levels and environmental uncertainty negatively moderate this relationship, while corporate digital transformation has a positive moderating effect. Further analysis shows that the U-shaped relationship is more pronounced in areas with lesser financial advancement and higher levels of environmental regulation, as well as in non-high-tech industries, non-manufacturing sectors, smaller firms, and companies without political connections. This study provides empirical evidence and policy insights to support the promotion of financial services that better facilitate corporate sustainability.

Keywords:

nonlinear dynamics; financial innovation; sustainable business practices; economic impact analysis

1. Introduction

Recently, worldwide climate catastrophes have created considerable environmental difficulties, positioning sustainable growth as a universal objective. China’s economy has shifted from a period of swift expansion to a stage emphasizing high-quality advancement [1]. Notably, in 2020, the Chinese government introduced the “dual carbon goals”, with ESG serving as a key tool in achieving these objectives. As a result, corporate sustainability and ESG performance have garnered unprecedented attention from both academia and industry [2]. ESG provides a framework for publicly listed companies to implement sustainable development practices, encompassing environmental protection, social responsibility, and corporate governance. It functions as a critical mechanism for companies to pursue sustainability goals. In response, Chinese government authorities have enacted a range of regulations encouraging companies to prioritize ESG practices and actively promote sustainable development. Driven by both policy and practical considerations, ESG has become essential for companies seeking high-quality development. Consequently, improving ESG performance has emerged as a key issue of shared interest across political, academic, and industrial sectors [3].

In the context of the booming digital economy, digital finance has imparted fresh meaning and importance to this hot issue. Specifically, digital finance has opened up new prospects for the growth of China’s financial sector by integrating traditional finance with advancing technologies like big data, cloud services, and AI. Its goal is to facilitate the accessibility and universality of financial services. In addition, it is characterized by easy transactions, varied services, and extensive reach, significantly enhancing financial access and lowering the costs of financing [4]. The ability of enterprises to engage in ESG practices largely depends on market policies and the broader macroeconomic environment. The emergence of digital finance can well help enterprises assume more responsibilities and has great potential to influence enterprise ESG. Then, in the process of advancing “dual carbon” goals, whether digital finance can offer accessible financial services, improve ESG performance, and foster sustainable enterprise growth merits thorough exploration.

The focus of the academic community on digital finance has been substantial, examining its connection with ESG extensively [5,6], yet the precise impact remains contested. Some scholars primarily concentrate on the linear facilitation role, with the existing literature highlighting the positive influence of digital finance on ESG through various channels, including alleviating financing constraints [7], fostering green innovation [8], enhancing corporate reputation, reducing agency costs [9], and improving external oversight [10]. Other academics believe digital finance is not conducive to ESG improvement. They believe enterprises leverage digital finance to reduce financing costs, invest in more equipment [11], and expand financing channels, attracting foreign direct investment that promotes energy-intensive growth [12], which could ultimately lead to environmental degradation in China [13]. Therefore, further research is essential to explore the impact and mechanisms, providing fresh evidence for a deeper insight into this relationship.

The existing literature explores the influence of digital finance on enterprise ESG from both positive and negative aspects, which is limited to the consideration of linear relationship, while few studies pay attention to its nonlinear effect. When the maturity of digital finance remains limited, the initial growth effect may cause environmental pollution, exclusion from traditional finance, and the exacerbation of resource misallocation due to the digital divide. Therefore, the impact is unlikely to follow a simple linear pattern. Based on this, this paper integrates macro-level digital finance advancement with micro-scale corporate ESG performance within a unified research framework to explore whether a nonlinear effect exists. If so, what is this nonlinear relationship? What are the underlying mechanisms? To address these inquiries, this paper conducts empirical analyses based on theoretical foundations and constructs an econometric model using a sample of A-share listed companies in Shanghai and Shenzhen from 2011 to 2021. The study aims to determine a U-shaped nonlinear impact, with financial advancement levels, environmental uncertainty, and enterprise digital transformation acting as moderating factors. Mechanism tests reveal that the U-shaped relationship is driven by increases in corporate green technology innovation and public environmental concern. Furthermore, the impact varies significantly across regions, industries, and firms, extending and deepening the research on this topic.

Overall, the marginal contributions can be summarized as follows: (1) This study attempts to explore relevant research on the influencing factors of corporate ESG performance and to supplement the deficiency of ESG antecedents in the existing literature. It proposes a mechanism through which enterprises can enhance ESG performance finding that development of digital finance exhibits a nonlinear “initial inhibition followed by promotion” effect on ESG outcomes. This finding offers a theoretical foundation and practical guidance for Chinese enterprises to improve ESG performance, attain high-quality growth, and contribute to the national sustainable development strategy. (2) This article expands research on the economic consequences of digital finance by focusing on its relationship with ESG performance, proposing a nonlinear relationship between the two, and filling the research gap on the negative effects of digital finance on ESG performance. The discovery of the “double-edged sword” effect of digital finance aligns with recent studies and contributes to the ongoing discussion about its multidimensional impacts [1]. (3) While most existing research focuses on internal mechanisms, this study incorporates external variables to broaden the research perspective. It analyzes the dual channels of internal green technology innovation and external public environmental concern from the perspectives of innovation theory and behavioral consistency theory, opening the “black box” of the mechanisms. (4) The study examines heterogeneous effects across regional, industrial, and firm levels. By introducing regulatory factors, such as financial regulation, environmental uncertainty, and corporate digital transformation, this paper further explores the contextual factors.

The structure of the article is as follows: Section 2 provides a literature review and presents the research hypotheses. Section 3 introduces the research methodology. Section 4 analyzes the empirical results. Section 5 offers further analysis. Finally, Section 6 summarizes the findings and discusses policy implications and limitations (Figure 1).

2. Literature Review and Hypothesis Formation: Identifying Research Gaps and Constructing a Theoretical Framework

2.1. Digital Finance and Corporate ESG Performance

In recent years, digital finance has rapidly transformed different facets of the financial industry, particularly in China. Traditional financial supply in China has long been mismatched with societal financial needs, both in terms of quantity and structure, resulting in consistently high levels of financial repression among 130 global economies [14]. Digital finance, however, has opened up fresh opportunities for the development of China’s financial sector. As a model that integrates financial development with modern societal needs, it provides essential financial services to small and micro-businesses and low-income groups in less developed and remote regions [15]. This plays a critical role in addressing gaps in financial coverage and improving financing conditions in lagging regions. Digital finance, with Chinese characteristics, acts as a supplement to inadequate financial services for the real economy, stimulating economic growth and offering a beneficial financing environment and resource support for corporate ESG investment decisions.

The existing literature has predominantly focused on the effects of digital finance, producing inconsistent conclusions. On one hand, some studies suggest that digital finance reduces information collection costs and enables precise matching, thereby improving capital allocation efficiency [16]. On the other hand, some scholars take a more negative stance, arguing that during the early stages of digital finance development, a “mission drift” effect occurs, making it difficult to direct funds toward the intended businesses and regions, counter to the original goal of optimizing capital allocation to meet effective demand [17]. Recently, some scholars have suggested that digital finance has a “first inhibit, then promote” impact effect on carbon performance and green total factor productivity [1,18]. These inconsistent findings underscore the necessity for additional exploration into various mechanisms and the consequences of digital finance. While digital finance presents both opportunities and challenges for businesses by enhancing the scope, boundaries, and efficiency of financial services [19], corporate ESG investment relies on the financial system, creating a strong link between the two. Therefore, understanding how digital finance influences corporate ESG performance is increasingly important in this context.

As ESG gains prominence, academics have carried out comprehensive theoretical and empirical studies in this area. However, most existing studies focus on economic benefits, with limited exploration of the underlying mechanisms. Some researchers have examined the impact of external factors, like market competition and environmental regulation [20], as well as internal factors, such as CEO characteristics [21] and shareholder pledging [22], on ESG performance. However, these studies do not sufficiently explain the particular effects of digital finance on ESG performance. Only recently has a potential direct connection between the two garnered academic attention. Since both digital finance and ESG are relatively new fields of study, research has tended to emphasize positive impacts [7,8,10], often overlooking potential nonlinear relationships. In fact, as mentioned earlier, studies on the influence of digital finance have yielded inconsistent outcomes. Based on these, the article delves into the complex relationship and mechanisms between digital finance and corporate ESG performance, aiming to fill the current research gap.

In the early stages of digital finance development, there may be a negative impact on corporate ESG performance because of several factors. Initially, at the onset of digital finance, digitalization may require a substantial amount of public resources, where the scale effect significantly outweighs that of technology, resulting in a certain degree of “green blindness” [23]. The reliance on digital technologies, such as cloud computing, blockchain, and data centers, necessitates the development and operation of an energy-intensive infrastructure [11]. Additionally, technological advancements driven by digitization may compel companies to upgrade their production and operational systems, leading to increased resource depletion and energy consumption, thereby generating adverse environmental externalities.

Secondly, financial exclusion may arise initially. The “dependence” of digital finance on traditional finance perpetuates certain negative effects as the supply and demand conditions of both are highly overlapping. Groups with higher engagement in traditional finance tend to have greater willingness and frequency of participation in digital finance, which creates a path dependence of digital finance on traditional financial systems [24]. However, with the intervention of traditional finance, digital finance’s growth often stems from the usage of digital services by traditional financial service users. This can lead to new forms of financial exclusion, such as “tool exclusion” and “evaluation exclusion”, particularly affecting small and micro-enterprises, low-income groups, and rural residents, who may lack access to internet tools and financial literacy [25]. This exclusion can suppress the motivation for ESG investment.

Finally, the development of digital finance is contingent upon digital endowment conditions, enterprise adaptive capacity, and the alignment between enterprises and digital technologies. It imposes significant demands on the regional digital infrastructure; if such infrastructure is inadequate to support digital services, it can exacerbate the “digital divide” and lead to a greater mismatch in financial resource allocation [26]. Furthermore, the introduction of emerging technologies and new business models may encounter corporate inertia, with organizations displaying insensitivity due to a lack of awareness and understanding across various departments. As a result, transforming traditional organizational structures and processes necessitates comprehensive, enterprise-wide collaboration and communication. These phenomena require ongoing coordination between digital technology and financial services to be corrected [27].

Although digital finance may have negative effects in its initial stages, as it matures, with the deeper integration of digital technology and broader coverage, it contributes positively to improving corporate ESG performance. Firstly, digital finance helps mitigate information asymmetry, thereby enhancing financing efficiency. Based on credit rationing theory, information asymmetry could constrain enterprise financing. However, the digital nature of finance enables the establishment of risk control and information monitoring systems, improving transparency in the financial market. By leveraging digital channels, it reduces financing costs in the banking and finance sectors [28]. Additionally, digital finance can develop multi-dimensional credit systems based on historical enterprise data, optimizing the credit approval process, shortening review times, improving financing efficiency [7], and encouraging participation in ESG initiatives.

Secondly, digital finance transcends the temporal and spatial constraints of conventional financial services. Resource allocation reaches Pareto optimality under conditions of abundant information and perfect market competition, where factors freely flow. When digital finance reaches a certain level, it then corrects resource misallocation. As its coverage expands, the “long tail effect” becomes increasingly evident, extending services to disadvantaged and vulnerable groups previously excluded from conventional financial systems [24], thereby unlocking new effective demand. Digital finance offers several advantages over traditional financial services, such as improved service efficiency and quality, driven by its low costs, wide reach, and transparency of information [29]. Additionally, it lowers the barriers for businesses to obtain financial services, promotes diversification of financing channels, and enables precise analysis of weaknesses in enterprise financing. By effectively categorizing and allocating financial resources, digital finance reduces the likelihood of resource allocation failures [30].

Finally, digital technologies are highly integrated with production and operational activities. According to the learning curve, as infrastructure development and operations mature, the accumulation of enterprise experience and practice can gradually enhance efficiency, leading enterprises to conclude their break-in and adaptation periods. This facilitates better utilization of digital finance, enabling effective enterprise risk analysis through technologies, enhancing managerial rational decision-making capabilities, and aiding in more accurate credit assessments. It also reduces operational costs and resource wastage [31]. Additionally, digital technologies help break down “data silos”, allowing various scenarios and service platforms to cooperate and share information, thus avoiding redundant construction [32] and aiding enterprises in achieving sustainable development goals. Building on the preceding analysis (Figure 2), the following hypothesis is proposed:

H1.

Digital finance initially suppresses and subsequently promotes corporate ESG performance, exhibiting a nonlinear “U-shaped” relationship.

2.2. Mediating Mechanism

2.2.1. Enterprise Internal Mechanisms

Digital finance fosters green innovation by enhancing resource allocation efficiency and increasing demand for green products. By employing digital technologies, it provides financial services that help alleviate the imbalance between the high demand and limited supply of capital for enterprises’ green innovation efforts. Additionally, digital finance can raise individual incomes, enabling consumers to pursue higher levels of demand for eco-friendly products, which subsequently positively influences firms’ readiness to participate in green innovation [8].

The expansion of enterprise production scale through green technology innovation is likely to require significant financial and material resources, as well as the recruitment of high-tech personnel. This may result in increased input costs, resource wastage, and energy consumption during the production process [33]. Moreover, newly developed green products may experience time lags, and stakeholders may initially be skeptical of a firm’s environmental performance [34], causing investors to hesitate and reduce their investments in the short term. Additionally, the adoption of new environmentally friendly and sustainable manufacturing processes requires companies to readjust their internal governance structures, potentially leading to short-term inefficiencies and disorder [35].

However, with the ongoing advancement of digital finance and gradual maturation of green innovation technologies within companies, the substitution effect will offset the rebound effect [18]. This will improve satisfaction among community stakeholders and internal company members, attracting more financial investors to support enterprises in achieving long-term sustainable development. Further, alleviation of financial constraints will enable firms to better fulfill their environmental governance objectives, improve pollutant treatment efficiency, launch green products, and reduce environmental pollution. This will attract potential consumers, help firms meet corporate environmental responsibilities, and enhance productivity. This will also compensate for environmental conservation costs, improve market profitability, and elevate product quality, thereby giving domestic firms a competitive edge in the global marketplace [36]. Therefore (Figure 3), the following hypothesis is proposed:

H2.

Digital finance has a U-curve impact on corporate ESG performance by improving corporate green technology innovation.

2.2.2. Mechanisms External to the Enterprise

The evolution of digital finance can heighten public environmental concerns. On one hand, mobile payment systems like WeChat and Alipay help reduce the energy consumption associated with cash transactions and aid in the formation of a sustainable consumption ecosystem. Additionally, second-hand trading platforms and similar channels facilitate resource recycling and minimize resource waste. On the other hand, the public’s participation in environmental protection has expanded through digital channels, enabling public concern to influence government policies and collective social action via monitoring, reporting, and public discourse [37]. However, the underdevelopment of digital finance has limited public environmental concern, leading to insufficient network-based public supervision, short timeframes for action, slow government responses, delayed implementation of relevant measures, weak penalties, and ineffective environmental regulation of enterprises. Looking ahead, enterprises and other economic actors are likely to face increased public environmental scrutiny and more stringent external oversight. In response, they may attempt to offset anticipated losses by rapidly increasing carbon emissions in the short term [38].

As digital finance continues to evolve, heightened public concern for the environment, coupled with expanded channels for expressing these concerns, will likely lead to increasingly stringent environmental standards for enterprises. This development will subject companies to greater pressure from public opinion, financial constraints, and more rigorous environmental regulations [39]. To address these challenges, enterprises will need to boost environmental performance through heightened investments in green initiatives and intensified pollution control efforts. Furthermore, according to behavioral consistency theory, individuals’ behaviors across different contexts tend to be similar and stable. Consequently, as enterprises improve their environmental performance, they will also likely advance their corporate social responsibility (CSR) initiatives and optimize corporate governance [40]. Companies aim to build a positive public image and gain social recognition by fulfilling their social responsibilities. Additionally, they seek to optimize management and self-supervision mechanisms through rigorous environmental and social performance assessments, thereby improving their governance structures. Based on this reasoning (Figure 3), we propose the following hypothesis:

H3.

Digital finance has a U-curve influence on corporate ESG performance by improving public environmental concern.

2.3. Moderating Mechanisms

2.3.1. Financial Regulation

Big data and cloud computing technologies have reduced the information asymmetry between borrowers and lenders within digital financial institutions. However, externalities and monopolistic tendencies remain prevalent in the market, leading to inefficient resource allocation. As a result, digital finance must be subjected to financial regulation from the perspective of market failure theory [41].

At low levels of digital finance development, high levels of financial regulation can alleviate the negative effect between digital finance and corporate ESG performance. Regions with stricter financial regulation can promptly update and improve relevant laws and regulations, ensuring comprehensive coverage of financial services. This enhances the operational efficiency and standardization of digital finance, helps maintain financial system stability, and supports the growth of new financial institutions under digital finance, addressing issues such as financial exclusion [41]. Moreover, ESG management and disclosure policies set by regulatory agencies strengthen the supervision and regulation of corporate ESG practices, directing more social capital toward sustainable and environmentally friendly fields.

However, financial regulation may amplify the drawbacks associated with advanced levels of digital finance. In regions with highly developed digital finance, stringent financial regulation imposes more stringent business rules, regulatory standards, and risk control requirements. This creates expectations of tighter regulatory policies, increasing compliance costs for companies, and limiting the scope for digital finance to provide financial resources to enterprises, thus restricting its ability to alleviate financing constraints [14]. Additionally, prolonged stringent financial regulation may stifle financial innovation, leading to stagnation in industry growth [42], thereby dampening the beneficial effects of digital finance on corporate ESG performance (Figure 4).

H4.

Financial regulation negatively moderates the U-shaped relationship between digital finance and corporate ESG performance, making the curve flatter under higher levels of financial regulation (more positive at low levels of digital finance and more negative at high levels).

2.3.2. Environmental Uncertainty

According to dynamic capabilities theory, the environment significantly shapes organizational capabilities. In times of instability and unpredictability, enterprises are compelled to confront environmental uncertainty. This uncertainty complicates their ability to accurately forecast and evaluate forthcoming market and technological developments within their operational environments.

Firms in regions with higher environmental uncertainty, when operating at lower levels of digital finance, rely excessively on public information for investment decisions due to market opacity and managers’ risk aversion [43]. This results in enterprises often imitating the investment decisions of similar firms or those with proven significant effects and benefits, and in such environments, firms also possess greater capabilities and resources to support their decisions. Consequently, they are more inclined to mimic the ESG behaviors of their peers to mitigate potential risks from intense competition [44]. Moreover, to respond to external changes and seize new opportunities and knowledge to address uncertainty, enterprises in uncertain environments recognize and accept the implementation of ESG principles more readily, thus alleviating the inhibitory effect.

However, when the level of digital finance is high, compared to firms facing high environmental uncertainty, those in low-uncertainty environments experience greater transparency. This increased transparency facilitates monitoring by external stakeholders, thereby diminishing managerial opportunism and fostering a long-term developmental focus. It also encourages the willingness of managers to deeply engage in digital finance to fulfill their ESG responsibilities. Conversely, a high degree of environmental uncertainty, characterized by rapid changes in market demands and technological advancements, exacerbates resource constraints and operational risks for manufacturing firms [45], thereby hindering their ability to enhance ESG performance through digital finance (Figure 4). Therefore, the following hypothesis is formulated:

H5.

Environmental uncertainty negatively moderates the U-shaped relationship between the two, making the curve flatter under higher levels of environmental uncertainty (more positive at low levels of digital finance and more negative at high levels).

2.3.3. Digital Transformation

Digital transformation, as a pivotal driver of enterprise innovation, empowers organizations to integrate and reconfigure both internal and external resources, processes, and structures. This catalyzes the enhancement of organizational dynamic capabilities, enabling the acquisition and maintenance of sustainable competitive advantages.

Enterprise digital transformation exhibits a duality. On one hand, during the initial phases of digital finance evolution, digital transformation can intensify the inhibitory effect. Due to organizational rigidity and insufficient flexibility in response to new changes and challenges, enterprises grounded in traditional management models and operational systems may find it difficult to meet the demands of digital transformation [46]. Moreover, constrained by a lower level of digital finance, enterprises might lack the necessary funding, resource base, and capabilities, leading to a transformation dilemma that exacerbates the inhibitory effect on ESG. Additionally, digital transformation enterprises require substantial resources and energy consumption for support, which, under the premise of limited resources, will inevitably squeeze investments in energy conservation and carbon reduction [47]. Furthermore, as digital products serve as intermediate inputs, they provide pollution-intensive intermediate products to non-digital sectors, thereby generating environmental issues.

On the other hand, as the level of development in digital finance increases, it establishes a strong fund foundation and resources for digital transformation. The accumulation of knowledge and resource integration effects from mature enterprise digital transformations enables rapid alignment with stakeholder demands and synergy with the organization’s resource and channel advantages [48]. Leveraging the advantages of digital transformation, enterprises integrate digital technologies into various domains, including internal controls, financial management, information disclosure, and risk management [49]. This enhances the enterprise’s capacity to mine and process information, significantly increases transparency, positively impacts financing, and assists in overcoming challenges such as weak green technological innovation during ESG practice [8]. This provides favorable conditions and technical support for fulfilling ESG responsibilities and impacts corporate ESG ratings. Participants in digital finance can also use digital information to monitor the pre- and post-loan usage of R&D funds, scrutinize green technology innovation projects, and reduce the scope for managerial myopia, thereby greatly enhancing the promotional effect of digital finance on corporate ESG performance (Figure 4). Thus, we formulate the hypothesis:

H6.

Digital transformation has a positive moderating effect on the U-shaped relationship between the two, making the curve steeper under higher levels of environmental uncertainty (more negative at low levels of digital finance and more positive at high levels).

3. Research Design

3.1. Modeling

Compared to linear models, nonlinear model fitting is diverse. In current research involving nonlinear relationships, the quadratic function model is most commonly used. To examine the impacts and mechanisms of digital finance on corporate ESG performance, this section will conduct an empirical analysis following these steps and set up the corresponding econometric model. Firstly, the following equation presents the linear direct impact in a panel two-way fixed effects model:

{E S G}_{i t} = α + β_{1} {D I F}_{i t} + γ {C o n t r o l}_{i t} + δ_{i} + η_{t} + ν_{j t} + ε_{i t}

(1)

In Equation (1), i, t, j represent enterprise, year, and industry separately. DF represents the respective level of digital finance. Control refers to the control variable that considers both the financial characteristics of firms and the economic environment’s impact in which it operates. The regression coefficient of the control variable is

γ

α

is a constant term,

δ, η, ν

denotes individual, time-fixed, and joint industry and time-fixed effects, respectively,

ε

denotes a random perturbation term. Next, referencing the work of Wei-Peng Lin [50] and Haans et al. [51] in the study of nonlinear curve effects, we consider a quadratic function model to capture the nonlinear effects, with the equation extended as:

{E S G}_{i t} = α + β_{1} {D I F}_{i t} + β_{2} {D I F}_{i t}^{2} + γ {C o n t r o l}_{i t} + δ_{i} + η_{t} + ν_{j t} + ε_{i t}

(2)

This setup allows us to observe how variations in the level of digital finance affect the nonlinear relationship with ESG. Specifically, if the regression coefficient

β_{1}

is significantly negative and simultaneously

β_{2}

is significantly positive, this indicates a positive ‘U’-shaped relationship.

Secondly, the nonlinear transmission mechanism is examined, and the corresponding econometric model is tested as:

M_{i t} = α {+ β}_{1} {D I F}_{i t} + γ {C o n t r o l}_{i t} + δ_{i} + η_{t} + ν_{j t} + ε_{i t}

(3)

When

β_{1}

is significant and the digital finance and mediator variables satisfy Equation (3) presenting a linear relationship, the mediator variable as well as the squared term of the mediator variable to Equation (2) are added as follows:

{E S G}_{i t} = α {+ β}_{2} {D I F}_{i t} + β_{3} M_{i t} + β_{4} M_{i t}^{2} + γ {C o n t r o l}_{i t} + δ_{i} + η_{t} + ν_{j t} + ε_{i t}

(4)

M represents a transmission pathway through which digital finance impacts ESG performance. IND refers to the strength of the nonlinear indirect effect of digital finance (DIF) on ESG through the mediator M. The significance and strength of IND can serve as a test for the mediation effect. If IND =

β_{1} β_{4}

> 0 in the regression results of (3) and (4), it indicates that the mediating variable plays a nonlinear mediating role in the U-shaped relationship between digital financial inclusion and corporate ESG.

Finally, the following model is constructed by introducing the interaction terms of the moderating variables with digital finance (including primary and secondary terms) in Equation (2), This model examines how the impact varies under different moderator variables Z. The curvature of the relationship between DIF and ESG depends on the “Curvature” term, which is a function of the moderating factor Z. If the coefficient

β_{4}

is significant, it indicates that the moderator Z can adjust the curvature of the relationship between DIF and ESG:

\begin{array}{l} {E S G}_{i t} & = α {+ β}_{1} {D I F}_{i t} + β_{2} {D I F}_{i t}^{2} {+ β}_{3} Z_{i t} \times {D I F}_{i t} + β_{4} Z_{i t} \times {D I F}_{i t}^{2} + β_{5} Z_{i t} + γ {C o n t r o l}_{i t} + δ_{i} + η_{t} + ν_{j t} + ε_{i t} \\ = α {+ β_{5} Z_{i t} + (β}_{1} {+ β}_{3} Z_{i t}) \times {D I F}_{i t} + {(β}_{2} + β_{4} Z_{i t}) \times {D I F}_{i t}^{2} + γ {C o n t r o l}_{i t} + δ_{i} + η_{t} + ν_{j t} + ε_{i t} \\ = I n t e r c e p t + S l o p e \times {D I F}_{i t} + C u r v a t u r e \times {D I F}_{i t}^{2} + γ {C o n t r o l}_{i t} + δ_{i} + η_{t} + ν_{j t} + ε_{i t} \end{array}

(5)

In Equation (5), if the direction of

β_{1} a n d β_{2}

are consistent with its regression results in (2) and the signs of

β_{2}

and

β_{4}

are the same, this implies that the moderating variable can enhance the U-shaped influence of digital finance on ESG, making it steeper. If the signs are opposite, it results in a weakening effect, making the U-shape flatter.

Further, according to Haans’ test on the moderating effect of the U-shaped curve, to illustrate how the moderating factor influences the turning point of the U-shaped relationship, the turning point is derived by setting the first derivative of DIFto zero (6).

{D I F}^{*} = \frac{- β_{1} - β_{3} Z_{i t}}{2 β_{2} + 2 β_{4} Z_{i t}}

(6)

To illustrate how Z affects the turning point, the following is the derivative of this equation with respect to Z (7):

\frac{δ {D I F}^{*}}{δ Z} = \frac{β_{1} β_{4} - β_{2} β_{3}}{{2 (β_{2} + β_{4} Z_{i t})}^{2}}

(7)

It can be seen that the denominator is strictly greater than zero, so the direction in which the curve shifts depends on the sign of the numerator. When

β_{1} β_{4} - β_{2} β_{3}

> 0, the turning point shifts to the right as Z increases, and vice versa.

3.2. Variables Definition

3.2.1. Explained Variables

Corporate ESG Performance (Environment, Social, and Governance): This study utilizes HuaZheng ESG rating system [52], which is tailored to the dynamics of China’s capital market and accurately reflects the ESG levels of local enterprises. The rating system is developed from a top-down approach, consisting of 16 secondary indicators and 44 tertiary indicators across environmental, social, and governance dimensions. Enterprises are rated on a scale ranging from AAA to C, corresponding to scores from 1 to 9, where elevated ratings reflect enhanced ESG outcomes. This paper posits that the system effectively captures the variations in ESG performance among enterprises and precisely mirrors their capacity for social responsibility.

3.2.2. Core Explanatory Variables

Digital Finance (DF): The data of prefecture-level cities released by the Digital Financial Centre of Peking University (DFC) “Peking University Digital Inclusive Finance Index” are selected to measure the development level of digital finance [53], and logarithmic numbers are taken in order to remove the effect of heteroskedasticity on the digital finance.

3.2.3. Mediating Variables

Green Technology Innovation for Enterprises (GTI): This research quantifies the level of green innovation within companies by using a total number of green invention patents and utility models filed independently by the enterprises within the same year. These patents are chosen as they reflect both environmental friendliness and an enterprise’s innovation capacity.

Public Environmental Concern (PEC): Referencing Wu et al. [54], this study employs the annual average total search index for the term ‘haze’ on Baidu’s search engine (including both PC and mobile platforms) to measure regional public environmental awareness for the current year. The selection of this index is justified by Baidu’s role as the leading Chinese search engine, noted for its comprehensive scope and substantial data access. Relative to other ecological concerns like ‘environmental pollution’, hazy conditions are chosen for their relatively higher perceptibility.

3.2.4. Moderating Variables

Financial Regulatory Intensity (Regulate): This study uses the ratio of regional financial regulatory expenditures to the added value of the financial sector as a proxy for financial regulation [41].

Environmental Uncertainty (EU): Environmental uncertainty stems from external changes that cause fluctuations in a firm’s core business operations, leading to variability in corporate sales revenue. This paper measures it using the standard deviation of a firm’s sales revenue over the past five years, adjusted for industry differences [55].

Enterprise Digital Transformation (DCG): Referring to the thesaurus constructed in the study of Wu [56] for textual analysis of enterprise digital transformation, the index system of enterprise digital transformation is constructed. The indices are then logarithmically transformed and utilized as proxy variables for assessing enterprise digital transformation.

3.2.5. Control Variables

The ESG performance of listed companies is related to their financial characteristics and to the economic environment in which they operate. This study examines factors at both corporate and regional scales, as referenced in previous studies [7,8,10,24]. The detailed labels and definitions for each variable are outlined as follows: (1) Leverage (Lev): Leverage reflects a company’s capital structure and financial risk, represented by the proportion of total liabilities to total assets. This helps assess a company’s financial stability and debt-paying ability. (2) Return on Equity (Roe): Profitability is a key indicator of corporate efficiency and attractiveness to investors, measured by the ratio of net profit to the average owner’s equity. It shows a firm’s ability to generate value for shareholders. (3) Tobin’s Q (TobinQ): Reflects the market’s valuation of a firm compared to its asset replacement cost. A higher Tobin’s Q value might indicate that the market values the company higher than its asset cost, suggesting investment attractiveness. (4) Company Age (Age): The age of a company might affect its resource acquisition, experience accumulation, and market position, calculated by the logarithm of the current year minus the founding year plus one. Older companies might have more stable operating environments and customer bases. (5) Managerial Shareholding (Mshare): The proportion of shares held by management can indicate the alignment of interests with the company. A higher shareholding might encourage management to focus more on long-term development. (6) Cash Flow Ratio (Cash): The company’s cash flow condition is an essential indicator of financial health, represented by the ratio of net cash flow from operating activities to total assets, reflecting a company’s liquidity. (7) Board Size (Board): The size of the board might influence the efficiency and quality of corporate decision-making, measured by the natural logarithm of the number of board members. A larger board might provide more resources and network support. (8) Audit Quality (Big4): Whether being audited by a Big Four firm can serve as a proxy for financial transparency and quality, as Big Four audits are generally associated with high standards of accounting and financial reporting. (9) Government Intervention (Gov): Government fiscal expenditure can reflect the degree of economic intervention by the government. A higher expenditure ratio may indicate a more significant role of the government in economic activities. (10) Openness (Open): The proportion of utilized foreign capital relative to regional GDP can reflect a region’s openness level. Higher openness might facilitate more foreign investment and technological exchanges. (11) Economic Development Level (Pgdp): Per capita regional GDP reflects the economic prosperity and market potential of a region, with higher per capita GDP usually associated with higher consumer spending capability and a more mature market structure.

3.3. Data Sources and Descriptive Statistics

This study focuses on Shanghai and Shenzhen A-share listed companies, constructing a panel dataset from 2011 to 2021. The data underwent several processing steps: First, financial enterprises were excluded from the sample. Second, companies that were categorized as ST, PT, or were delisted during the period were also excluded. Third, companies that had initial public offerings during the observed period were excluded. Additionally, samples lacking key or control variables were discarded. Continuous variables were subjected to two-sided winsorization to eliminate the effects of outliers. As a result, a total of 14,300 firm-year observations for 1300 companies were obtained. Data sources include the CSMAR database, the Wind database, the National Bureau of Statistics of China, and the “Peking University Digital Inclusive Finance Index (2011–2021)” report from the Institute of Digital Finance at Peking University. Table 1 presents the descriptive statistics of this study. The maximum ESG value is 8, the minimum is 1, and the standard deviation is 1.021, indicating significant variations in ESG performance among listed companies. The average ESG score is 4.219, indicating that the ESG performance of most enterprises still requires improvement. The mean and standard deviation of digital finance are 5.283 and 0.451, respectively, indicating variations in digital finance development across different cities and years, highlighting a digital finance ‘divide’. The descriptive statistics for all remaining control variables are within a reasonable range and are consistent with previous research.

4. Empirical Results

4.1. Benchmarking Results

Table 2 displays benchmarking results that elucidate the impact of digital finance on firms’ ESG performance. The analysis controls for individual, time, and time-industry fixed effects across models (1)–(3). In the table, R² is a statistical measure used to assess the goodness of fit of a regression model. The value of R² ranges from 0 to 1, with values closer to 1 indicating a higher goodness of fit and effectiveness of the model. Controlling for individual, time, and time-industry fixed effects accounts for unobserved heterogeneity that might influence the dependent variable, reducing potential biases and enhancing the reliability of the variable relationships. Model (1) involves univariate regressions, model (2) incorporates squared regressions of variables and variables, and model (3) includes regressions with control variables. In model (2), the primary coefficient of digital finance on corporate ESG is significantly negative, whereas the quadratic coefficient is significantly positive. The coefficients in model (3), after including control variables, align closely with those in model (2), suggesting robustness. In (3), when the independent variable takes the minimum value, the slope is negative (k₁ =

{2 β}_{2} \times D F {+ β}_{1}

= 2 × 0.211 × 3.057 − 1.608 = −0.317). When the independent variable takes the maximum value, k₂ =

{2 β}_{2} \times D F {+ β}_{1}

= 2 × 0.211 × 5.885 − 1.608 = 0.875, the slope is positive. Additionally, the inflection point of the U-shaped curve is 3.81 (

{- β}_{1}

{2 β}_{2}

= 1.608/(2 × 0.211)), which lies within the sample value range [3.057, 5.885]. This satisfies the conditions for a U-shaped curve, thereby substantiating H1.

Prior to the U-shaped inflection point, corporate ESG performance demonstrates a declining trend with increase in digital finance (DF). This suggests that during the early phases of digital finance evolution, the growth effects may lead to environmental pollution, traditional finance exclusion, and an exacerbation of the digital divide, resulting in a temporary reduction in corporate ESG performance. Beyond the U-shaped inflection point, corporate ESG begins to improve as DF advances, the synergy between digital technologies and financial services is enhanced, internal structures are optimized, and resources are allocated more efficiently, all of which contribute to improvements in ESG performance. This nonlinear relationship aligns with findings from Wang, Zhou, and Bruhn [1,57,58]. Additionally, regression analysis reveals that the gearing ratio, enterprise age, and cash flow ratio all significantly negatively impact corporate ESG. In contrast, corporate profitability and the managerial shareholding ratio exert a significantly positive influence on corporate ESG, enhancing its overall performance.

4.2. Robustness and Endogeneity Treatment

4.2.1. Substitution of Explanatory Variables

This study employs digital finance data from both the current and lagged periods as proxy variables and reruns the regression analysis [59]. The outcomes, detailed in column (1) of Table 3, reveal that the primary term’s coefficients are notably negative, whereas coefficients for the quadratic term are clearly positive. Furthermore, the model achieves a fit value of 0.681, indicating enhanced explanatory power and confirming the robustness of the findings.

4.2.2. Replacing Higher-Order Clustering Robust Criterion Errors

The benchmark regression uses cluster robust standard errors at firm tier. Due to the more robust standard errors reducing biases in statistical inference, they directly impact the significance of the sample regression results. Therefore, the robustness of the benchmark results is tested using higher-dimensional cluster robust standard errors at the industry-firm level [60]. The results, presented in Column (2) of Table 3, show that the regression coefficients remain positive and significant, indicating that different settings of the cluster robust standard errors do not affect the conclusions of this study.

4.2.3. Removing Some of the Effects of Factors

This paper accounts for effects of the 2015 Chinese stock market turmoil, a significant financial event lacking objective variables to measure its extent. Consequently, samples affected by the crash and potential subsequent years are excluded [56], limiting the data regression to the period between 2011 and 2014. Furthermore, due to the distinct advantages and unique characteristics of municipalities directly under central government [61], such as Beijing, Shanghai, Tianjin, and Chongqing, these cities are excluded. The empirical results in Columns (3) and (4) of Table 3 demonstrate a U-shaped relationship between digital finance and corporate ESG, with the conclusions remaining unchanged.

4.2.4. Endogenous Treatment

Considering the potential bidirectional causality between the explanatory and dependent variables and the possibility of omitted variable bias among the selected control variables, this paper employs an instrumental variable approach to estimate the model. A ‘Bartik instrument’ is constructed using the shift-share method, with Bartik_Digital Finance serving as the instrumental variable [62]. The 2SLS regression results are presented in Table 3. Columns (5) and (6), respectively, show the first- and second-stage estimation results. The coefficient of the instrument variable Bartik_Dif is significantly positive. In the Kleibergen–Paap rk LM test, a p-value of 0 indicates strong statistical evidence of a significant correlation between the instrument and the endogenous variable, suggesting that the instrument is valid. In the context of the Kleibergen–Paap Wald rk F statistic, a value greater than 16.38 indicates that the instrument used in the instrumental variable regression is strong, a threshold typically based on critical values for weak identification. The instrument variable meets the requirements of relevance and exogeneity and passes tests for under-identification and weak instrument problems. The coefficients and significance levels of variables in the IV-2SLS estimation results are consistent with those in the baseline regression. Additionally, to mitigate endogeneity issues arising from reverse causality, this study regresses the dependent variable advanced by one period [63], with the results in Column (7) remaining consistent with the baseline regression. These tests confirm the robustness of the empirical results presented in this study.

4.3. Mechanism Identification Test

The aforementioned studies offer substantial empirical evidence, enhancing our comprehension of the effects of digital finance. But the prior section has merely depicted a general influence without delving into the underlying mechanisms. Thus, there is a necessity for further investigation to elucidate the specific channels through which digital finance affects ESG outcomes. In this context, this paper examines two distinct channels: ‘corporate green technology innovation’ (GTI) and ‘public environmental concern’ (PEC).

As shown in Table 4, columns (1) and (3) test the impact of digital finance on the intermediary variables. The coefficients for the first-order terms

β_{1}

are 1.695 and 0.298, respectively, both of which are significantly positive at the 1% level. This indicates that digital finance is positively correlated with both GTI and PEC. Moreover, the coefficients of the primary terms of the mediating variables in columns (2) and (4) are significantly negative, while those of the quadratic terms are markedly positive, indicating both GTI and PEC have a U-shaped relationship on corporate ESG. Furthermore, with IND₁ =

β_{1} β_{4} = 1.695 \times 0.017 > 0

and IND₂ =

β_{1} β_{4} = 0.298 \times 0.007 > 0

, it is confirmed that both channels serve intermediary roles in the nonlinear relationship. These findings support the transmission paths of ‘digital finance → corporate green technology innovation → corporate ESG performance’ and ‘digital finance → public environmental concern → corporate ESG performance’, validating hypotheses H2 and H3. The affirmation of hypotheses H2 and H3 underscores that enhancements in digital finance lead to increases in both corporate green technology innovation and public environmental concern, culminating in a nonlinear impact on corporate ESG.

The characteristics of reverse innovation in corporate green technology, such as high risk of failure, prolonged R&D investment periods, and limited potential yield, are predominantly observed during the initial stages. Additionally, the early phases often see minimal digital finance size and limited funding for green technology development, leading to a more pronounced rebound effect that may temporarily inhibit ESG performance. However, as digital finance continues to evolve, the maturity of technologies, competitive advantages, and green effects become more apparent, enhancing corporate ESG performance. This progression is in line with findings from Xie [64]. Enterprises are advised to navigate the short-term challenges and long-term benefits carefully, minimizing potential adverse effects like resource depletion and governance disorder, to achieve a balance that fosters sustainable development. By persisting through these initial difficulties and continuing to invest in digital finance, firms can develop mature technologies that offer competitive advantages and positive environmental impacts.

Simultaneously, digital finance may initially negatively impact ESG performance due to limited public environmental concern, challenges in expressing public will, lack of regulatory measures, and corporate expectations of stricter future policies, which may increase carbon emissions [38]. Over time, however, rising public awareness has heightened sensitivity and reduced tolerance towards environmental issues. This shift has compelled companies to focus more on their environmental performance, social responsibilities, and governance practices, subsequently enhancing ESG outcomes [40]. It is crucial for companies to perceive public environmental concerns accurately, transform pressure into motivation, and make well-informed, rational decisions that consider the long-term sustainability of the enterprise. This proactive approach can turn potential challenges into opportunities for corporate growth and reputation enhancement.

4.4. Heterogeneity Test

4.4.1. Regional Heterogeneity

This study investigates the effects of regional diversity in financial development and environmental regulation on digital finance. As shown in Columns (1) and (2) of Table 5, the squared term coefficients of digital finance are significant and more pronounced in regions with lower financial development compared to those with higher financial development, suggesting a steeper U-curve in the former regions. The inflection points are 3.81 for the full sample, and 3.47 (calculated by 3.102/(2 × 0.447)) and 4.33 (calculated by 2.756/(2 × 0.318)) for the respective regions. Cities with higher financial development levels may benefit from more robust financial resources and abundant funding [65], contributing to greater stability and reduced volatility. Digital finance has a more pronounced impact on fostering growth in areas with less-developed financial sectors. In these areas, where the financial system is predominantly bank-centered and the banking sector is underdeveloped, there is a stronger dependence on traditional finance. The full implementation of digital finance could compensate for inadequacies in banking development, effectively alleviating constraints on corporate financing and other aspects [14], thus accelerating the approach of the inflection point.

Columns (3) and (4) of Table 5 show that the impact is insignificant in regions with weaker environmental regulatory constraints. However, in regions with stronger regulatory constraints, the impact is more significant, with an inflection point at 3.74 (calculated by 2.745/(2 × 0.367)). This suggests that more intense environmental regulation may accelerate the arrival of the inflection point, shifting the impact of digital finance on firms’ ESG from a reverse inhibitory to a positive promotional effect more quickly. This shift is likely because, in areas with strict environmental regulations, local governments intensify monitoring of corporate pollution and internalize the external costs associated with it [66], leading to increased corporate investment in environmental governance and a heightened focus on balancing environmental concerns with economic efficiency.

4.4.2. Industry Heterogeneity

This paper analyzes the impact of industry heterogeneity between the high-tech and manufacturing sectors. The empirical results from Columns (1) and (2) of Table 6 show that digital finance significantly influences ESG performance only within non-high-tech enterprise groups, whereas it does not hold significance in the high-tech group. This divergence may be attributed to the financing challenges in high-tech industries, which typically involve substantial investment, significant risk, short product life cycles, and rapid technological updates. Thus, conventional external financing methods are more challenging and heavily reliant on internal financing, and external financing often fails due to information asymmetry that leads to capital market dysfunction [67,68]. In contrast, digital finance reduces information asymmetry and leverages digitization to reshape business ecosystems. In non-high-tech industries, technology adoption can lead to more direct efficiency gains, providing greater opportunities for digital finance improvements. This facilitates easier achievement of enhanced ESG performance [69]. Additionally, non-high-tech industries, often associated with higher carbon emissions, confront stricter environmental regulations and transparency demands, prompting greater investment in ESG initiatives.

Columns (3) and (4) indicate that the impact is more pronounced in the non-manufacturing sector. This difference may stem from the greater operational flexibility of non-manufacturing firms, allowing them to effectively integrate external resources [70]. This adaptability enables non-manufacturing firms to more rapidly adopt new technologies and more efficiently integrate digital finance solutions into their operations, thereby enhancing resource allocation, reducing costs, and improving ESG outcomes. Additionally, intense market competition compels non-manufacturing firms to proactively digitize and seek new growth opportunities, actively pursuing collaborations with external entities to acquire more complementary resources, which, in turn, facilitates the enhancement of corporate ESG performance through digital finance.

The inflection points for non-high-tech enterprises and non-manufacturing industries are 4.05 (calculated by 3.505/(2 × 0.432)) and 5.35 (calculated by 4.768/(2 × 0.445)), respectively. These shifts to the right may be attributed to the country’s emphasis on advanced manufacturing and high-tech industries, leading to a capital allocation bias. In the short term, this prioritization leads to digital finance exerting a more pronounced inhibitory influence on corporate ESG, thereby delaying the arrival of the inflection points.

4.4.3. Enterprise Heterogeneity

This paper further investigates firm size and ownership structure, which reflect firm heterogeneity. As indicated in Table 7, Columns (1) and (2), digital finance significantly enhances ESG outcomes for small and medium-sized enterprises (SMEs), but it does not have a noticeable effect on larger firms. Several explanations for this discrepancy are presented: Large enterprises, with their robust asset credit, typically receive preferential treatment in the allocation of financial resources, whereas small and micro private enterprises often struggle to secure effective financing due to insufficient collateral and inferior credit scores [14]. Consequently, large enterprises do not depend on benefits derived from the development of digital finance, whereas SMEs become the primary beneficiaries [71]. Moreover, SMEs, which generally lag behind large enterprises in scale and reputation and face intense competition, must continuously innovate and adhere to green development strategies, emphasizing ESG responsibilities to establish a positive social image and attract more societal resources. Thus, digital finance tends to enhance ESG performance more effectively in SMEs than in large companies.

Both formal and informal institutions significantly influence actor behavior, and during China’s transition period, the establishment of political ties with government officials has become a common practice among firms. This is regarded as a crucial informal institutional arrangement that plays a significant role. By forming connections with government officials, firms can gain favors that reduce financing costs, or officials may shield politically connected companies from environmental regulations for their own benefit, often at the cost of environmental responsibility [72]. Additionally, firms without political connections may have less access to traditional forms of capital, making them more reliant on alternative financing options, such as digital finance, to fund their sustainability projects. This reliance makes the impact of digital finance on enhancing their ESG performance more pronounced. Consequently, firms without political connections typically achieve better environmental ratings than those with political connections. According to Columns (3) and (4) in Table 7, this disparity is notably significant in how digital finance impacts firms’ ESG performance.

The inflection points are 4.74 (calculated by 4.353/(2 × 0.459)) and 4.23 (calculated by 2.969/(2 × 0.351)), respectively. These values suggest a slower progression towards the inflection point, which may be attributed to the ‘tail group’ being more susceptible to exclusion from the financial system and facing more stringent financing constraints compared to larger and politically connected firms.

5. In-Depth Exploration: Digital Finance’s Role in ESG Performance via Regulatory Mechanisms

The previous sections discuss and validate the nonlinear impact of digital finance on corporate ESG performance, and further explore whether the relationship between the two is influenced under different circumstances. Table 8 presents the outcomes of regression analyses examining the moderating effects of financial regulatory intensity and environmental uncertainty, detailed in columns (1) and (2), respectively. The coefficients of the interaction term between financial regulatory intensity and the squared term of digital finance, as well as the interaction term between environmental uncertainty and the squared term of digital finance, are significantly negative at −1.702 (p-value is −2.42) and −0.061 (p-value is −2.12), respectively. Using the U-curve measure [51], we first analyze the shift in the curve’s inflection point. If

β_{1} β_{4}

−

β_{2} β_{3}

< 0, the inflection point curve shifts to the left. We then examine the shape of the curve:

{β_{2} β}_{4} =

0.212 × (−1.702) and

{β_{2} β}_{4} =

0.274 × (−0.061) are both significantly less than 0, indicating opposite signs, thus the impact of digital finance on corporate ESG is moderated by the intensity of financial regulation and environmental uncertainty, which have a negative effect on the original U-curve, making it flatter. Hypotheses H4 and H5 are supported.

The empirical results validating the hypothesis indeed reflect the “double-edged sword” effect of financial regulation. Before the turning point, regulation ensures that financial activities adhere to certain environmental and social responsibility standards, effectively regulating the market. However, after the turning point, strict regulation limits the development space and flexibility, suppressing market innovation and vitality, thereby weakening the U-shaped effect [73]. The leftward shift of the inflection point may be due to the fact that stringent financial regulation encourages institutions and enterprises to place greater emphasis on risk management, increasing their willingness to improve risk control and accelerating the improvement of internal systems. This leads to a quicker reduction in negative impacts and an earlier realization of positive effects. Overall, stringent financial regulation enables enterprises to benefit from digital finance more quickly, but it may also limit the maximization of these benefits.

Under high environmental uncertainty, enterprises face rapid changes in market demand and operational risks. The instability of the market and policies may negatively affect the effectiveness of digital finance, weakening its U-shaped impact. The leftward shift of the turning point may be due to the intensified uncertainty, causing enterprises to experience the impact of digital finance on ESG performance earlier and prompting them to take measures in response—this accelerates the arrival of the turning point. This phenomenon reflects the complex influence of uncertainty on corporate decision-making and performance.

Column (3) of Table 8 presents the results of a regression analysis examining the moderating effect of enterprise digital transformation. The coefficient of the interaction term between enterprise digital transformation and the square term of digital finance is significantly positive at 0.124. And

β_{1} β_{4}

−

β_{2} β_{3}

= −0.943 × 0.124 − 0.136 × (−1.151) = 0.04 > 0,

β_{2} β_{4}

= 0.136 × 0.124 > 0, which is significantly greater than 0, indicating the same sign; it indicates that the original U-shaped curve becomes steeper and enterprise digital transformation has a positive moderating effect. The hypothesis H6 has been validated. Digital transformation can increase resource and energy consumption, but with the maturation of technology, it can achieve more efficient access to and integration of green resources [74], effectively enhancing the role of digital finance on corporate ESG. The possible reason for the rightward shift of the inflection point is that the higher the degree of digital transformation, the greater the technological complexity and integration challenges that enterprises need to address. This results in more occurrences of structural reorganization, reduced environmental investments, and increased emissions effects, placing enterprises at a disadvantage in the short term and thereby delaying the arrival of the turning point.

The above results indicate that the government should adjust policies based on market dynamics, understand and predict the turning point, and make timely policy adjustments to avoid overregulation. By encouraging investment-led funding through policies, it can effectively foster the development of green industries. And environmental uncertainty can influence managerial decision-making, and enterprises must enhance their risk management capabilities to balance short-term gains with long-term development. Moreover, enterprises must balance resource investment and long-term benefits. Stakeholders cannot limit the short-term fluctuations caused by digital transformation, while ignoring the efficiency in acquiring and integrating green resources upon deeper engagement, thus losing the double benefits of the combination of digitalization and green innovation. Financial regulation, environmental uncertainty, and corporate digital transformation have significant implications for digital finance and corporate ESG.

6. Conclusions and Implications

6.1. Research Conclusions

This article analyzes the impact, mechanisms, moderating effects, and heterogeneity of digital finance in relation to corporate ESG performance, utilizing data from Chinese A-share listed companies in Shanghai and Shenzhen from 2011 to 2021, along with the Peking University China Digital Inclusive Finance Index. The study reveals that digital finance temporarily suppresses corporate ESG performance in the short term. However, once surpassing the inflection point, it significantly enhances corporate ESG performance. Key findings include:

(1) Short-term Impact and Long-term Enhancement: Initially, digital finance suppresses corporate ESG performance, but beyond the inflection point, it exerts a substantial and positive influence, confirming a nonlinear U-shaped relationship. These results remain robust even after thorough robustness and endogeneity tests, helping firms to enhance their engagement in ESG practices and bolster their commitment to social responsibility. (2) Mechanisms of Influence: Employing theories of innovation and behavioral consistency, this study selects variables from the “internal-external” perspectives of enterprises. It demonstrates that digital finance impacts corporate ESG performance through a U-shaped curve by promoting green technology innovation and enhancing public environmental concern; this confirms the nonlinear mediating roles. (3) Moderating Effects: The research also identifies that external macro-environmental factors, specifically, the levels of financial regulation and environmental uncertainty, negatively moderate the U-shaped relationship. Conversely, internal corporate digital transformation positively moderates this relationship, weakening and strengthening it, respectively. (4) Contextual Variations: Furthermore, the influence varies by region, industry, and firm size. Regionally, its influence is more pronounced in areas with lower financial development and higher environmental regulation. Industrially, the effect is stronger in non-high-tech and non-manufacturing sectors. At the firm level, it is more significant in SMEs and firms without political connections.

6.2. Policy Implications

This paper has significant practical implications based on its empirical results, offering policy insights for accurately grasping the impact of digital finance and effectively utilizing it to enhance corporate ESG performance:

Firstly, the government should continue to advance reforms in conventional financial frameworks and enhance the financial service system. China’s digital finance still faces imperfections in its institutional mechanisms, such as an incomplete credit system for small and micro-enterprises, non-transparent credit processes, and unstable operational systems. These issues may perpetuate the flaws of traditional finance in the initial stages of digital finance development. Financial regulatory authorities need to continue promoting the digitization of traditional financial entities, expand digital finance application scenarios, and establish digital finance technology standards to stimulate iterative innovation and digital technology application upgrades across various sectors. During the initial adjustment and adaptation phase, managers should rely on digital technology while scientifically planning and reasonably arranging infrastructure construction to enhance operational efficiency and reduce carbon emissions. Additionally, it is crucial to strengthen adaptive learning capabilities, to promptly adjust business models, research and development patterns, organizational structures, etc., to adapt to the shift in digital technology paradigms. This can mitigate the conflicts between the new institutional innovation demands triggered by digital development strategies and the old internal institutional norms of enterprises, providing organizational support for improving the effectiveness of resource distribution.

Secondly, the dividends of digital finance development should be leveraged to promote corporate ESG progress. The government should promote the synergy between digital finance and sustainable corporate development, as well as the efficient alignment of financial supply with green development demands. Regulatory bodies should actively utilize the digital finance system, build an appropriate ESG information disclosure system, and enhance transparency in information disclosure. Additionally, environmental or charitable organizations should be supported in promoting ESG policies, encouraging enterprises to enhance their environmental awareness and social responsibility. This will elevate corporate ESG development to a strategic level within enterprises, strengthening the competitive advantages brought by ESG in terms of social resources and financing support, thus further leveraging the advantages of digital finance and unleashing its potential to drive ESG performance. In the future, as reforms in China’s financial sector continue to deepen, digital finance will integrate more profoundly with corporate ESG, achieving a “win-win” situation for economic benefits and ecological protection.

Thirdly, the effects of green technology innovation and public environmental awareness should be fully utilized to achieve sustainable development. Enterprises should optimize the green technology innovation industry chain, comprehensively promote clean production, and establish an eco-friendly production system to minimize energy use throughout manufacturing. There should be focus on the research and development and production of green innovation technologies, targeting green innovation research and development projects, and guiding the integration of digital finance investment and financing services with green innovation research and development, thereby enhancing ESG. Additionally, in the process of guiding corporate ESG behaviors, the government should reduce information asymmetry and maximize the function of public governance as an informal institution. This involves expanding channels for the public to express opinions and suggestions, guiding media, environmental organizations, and the public to demand environmental accountability from enterprises, and regulating ESG behaviors, thus effectively supplementing traditional formal institutions, such as administrative regulations and legal constraints.

Fourthly, the impacts of financial regulatory intensity, environmental uncertainty, and corporate digital transformation should be comprehensively and objectively grasped, with moderate and effective regulation of the development of digital finance. For example, government regulators should develop micro-indicators and business guidelines based on the digital finance development needs of different enterprises, and further establish an information-sharing platform among regulated entities and between regulators and regulated entities to form a regulatory linkage mechanism. This allows for timely adjustments in the direction and focus of regulation, balancing protection and pressure, and giving the market sufficient space and freedom to restore vitality and momentum. Maintaining moderate foresight and continuously adjusting as risks evolve is necessary. In a complex environment of uncertainty, enterprises can establish internal governance and management mechanisms that adapt to the external environment to restrain managers’ self-interested behaviors and minimize the negative impacts brought by environmental uncertainty. Corporate managers need to enhance their professional capabilities, strengthen information exchange with external stakeholders, and improve their ability to utilize information comprehensively. Enterprises should increase investment in digital transformation to leverage the resource and information advantages it brings, enhancing their ability to fulfill social responsibilities. However, digital transformation is not achieved overnight. Corporate managers need to comprehensively consider their own resource endowments and operational management capabilities, and progressively promote digital transformation by targeting weak links to minimize the costs and risks associated with transformation.

Fifthly, differentiated strategies should be adopted based on the heterogeneous impacts. First, in regions with weaker environmental regulations, governments should strengthen environmental monitoring and improve punitive measures. They should fully utilize market regulation to standardize corporate practices and work to evolve an ESG management approach which embodies principles of sustainable development into deliberate actions by decision-makers. Second, the inclusive effects of digital finance across various industries should be continuously strengthened. While digital finance has achieved inter-industry allocation, complete inclusivity has not yet formed across all sectors. Therefore, the design of digital finance products can be tailored to local conditions, forming inclusive products with industry characteristics. Additionally, digital finance has shown considerable effects in solving financing difficulties for SMEs, and the government is supposed to enhance fiscal and policy assistance for SMEs, encouraging them to invest in more ESG projects. Meanwhile, the government should emphasize the role of the market in creating a fair, law-abiding, and transparent business environment, avoiding enterprises focusing on maintaining political relationships at the expense of fulfilling their ESG responsibilities.

6.3. Limitations and Prospects

This paper still has certain shortcomings. Firstly, future research could seek more accurate measurements of digital finance. Additionally, this study uses Huazheng ESG rating data to measure the ESG performance of listed companies. However, as an external third-party rating agency, Huazheng’s ESG ratings may not fully align with the real ESG performance, and not being able to directly study the actual ESG performance of companies is a limitation of this study. Future scholars could improve upon this aspect. Secondly, this paper explores whether there are more scientific methods for testing robustness, mediation, and moderation effects in the analysis of the “U-shaped” relationship, which also remains to be further explored and refined. Thirdly, due to limitations and inadequacies related to factors such as time and resources, the data collection and analysis process might introduce biases into the results. Fourthly, the influence varies by country and region due to differences in economics, culture, and regulation. This research is limited to studying the U-shaped relationship in China and does not include datasets from other countries. Future research could provide a more comprehensive analysis to explore differences and commonalities across different countries.

Author Contributions

All authors certify that they have participated sufficiently in the work to take public responsibility for the content, such as the following: Q.Y.: Conceptualization, methodology, formal analysis, writing—original draft; writing—review and editing, supervision; N.S.: Conceptualization, methodology, formal analysis, writing—original draft; writing—review and editing; C.D.: Data curation, methodology, formal analysis, writing—original draft; writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest

The authors would like to declare that there are no relevant financial or non-financial conflicts of interest to disclose and all the authors have contributed to the papers.

References

Zhou, B.; Wang, Y.-L.; Hu, B. The nonlinear effects of digital finance on carbon performance: Evidence from China. J. Innov. Knowl. 2024, 9, 100484. [Google Scholar] [CrossRef]
Gillan, S.L.; Koch, A.; Starks, L.T. Firms and social responsibility: A review of ESG and CSR research in corporate finance. J. Corp. Financ. 2021, 66, 101889. [Google Scholar] [CrossRef]
Wang, Y.; Guo, Y. Firm Digital Transformation and ESG Performance: Evidence from Chin’s A-share Listed Firms. J. Financ. Econ. 2023, 49, 94–108. [Google Scholar]
Tang, X.; Ding, S.T.; Gao, X.; Zhao, T.Q. Can digital finance help increase the value of strategic emerging enterprises? Sustain. Cities Soc. 2022, 81, 103829. [Google Scholar] [CrossRef]
Ahluwalia, S.; Mahto, R.V.; Guerrero, M. Blockchain technology and startup financing: A transaction cost economics perspective. Technol. Forecast. Soc. Change 2020, 151, 119854. [Google Scholar] [CrossRef]
Deng, J.P.; Liu, Y.C. Does digital finance reduce the employment in the finance industry? Evidence from China. Financ. Res. Lett. 2022, 48, 102994. [Google Scholar] [CrossRef]
Li, W.Q.; Pang, W.B. The Impact of Digital Inclusive Finance on Corporate ESG Performance: Based on the Perspective of Corporate Green Technology Innovation. Environ. Sci. Pollut. Res. 2023, 30, 65314–65327. [Google Scholar] [CrossRef]
Mu, W.W.; Liu, K.F.; Tao, Y.Q.; Ye, Y.W. Digital finance and corporate ESG. Financ. Res. Lett. 2023, 51, 103426. [Google Scholar] [CrossRef]
Xue, L.; Dong, J.; Zha, Y. How does digital finance affect firm environmental, social and governance (ESG) performance?—Evidence from Chinese listed firms. Heliyon 2023, 9, e20800. [Google Scholar] [CrossRef]
Ren, X.H.; Zeng, G.D.; Zhao, Y. Digital finance and corporate ESG performance: Empirical evidence from listed companies in China. Pac.-Basin Financ. J. 2023, 79, 102019. [Google Scholar] [CrossRef]
Sadorsky, P. The impact of financial development on energy consumption in emerging economies. Energy Policy 2010, 38, 2528–2535. [Google Scholar] [CrossRef]
Ali, S.; Waqas, H.; Ali, S. Analyzing the Dynamics of Energy Consumption, Liberalization, Financial Development, Poverty and Carbon Emissions in Pakistan. Environ. Sci. Econ. 2015, 5, 166–183. [Google Scholar]
Zhang, Y.J. The impact of financial development on carbon emissions: An empirical analysis in China. Energy Policy 2011, 39, 2197–2203. [Google Scholar] [CrossRef]
Liu, X.; Wu, F.; Ye, X. Influence of Fintech on Corporate Finance Constraints. Financ. Forum 2022, 27, 22–31. [Google Scholar] [CrossRef]
Qian, H.; Tao, Y.; Cao, S.; Cao, Y. Theoretical and Empirical Analysis on the Development of Digital Finance and Economic Growth in China. J. Quant. Technol. Econ. 2020, 37, 26–46. [Google Scholar]
Siddik, M.N.A.; Kabiraj, S. Digital Finance for Financial Inclusion and Inclusive Growth. In Digital Transformation in Business and Society: Theory and Cases; George, B., Paul, J., Eds.; Springer International Publishing: Cham, Switzerland, 2020; pp. 155–168. [Google Scholar]
Xing, Y. The Effectiveness and Realization of Inclusive Finance: Review and Enlightenment. Stud. Int. Financ. 2015, 32, 24–36. [Google Scholar]
Yu, Y.; Zhang, Q.; Song, F. Non-Linear Impacts and Spatial Spillover of Digital Finance on Green Total Factor Productivity: An Empirical Study of Smart Cities in China. Sustainability 2023, 15, 9260. [Google Scholar] [CrossRef]
Tang, Y.; Geng, B.Y. Digital finance development level and corporate debt financing cost. Financ. Res. Lett. 2024, 60, 104825. [Google Scholar] [CrossRef]
Vallentin, S. Governmentalities of CSR: Danish Government Policy as a Reflection of Political Difference. J. Bus. Ethics 2015, 127, 33–47. [Google Scholar] [CrossRef]
Aabo, T.; Giorici, I.C. Do female CEOs matter for ESG scores? Glob. Financ. J. 2023, 56, 100722. [Google Scholar] [CrossRef]
Huang, Q.; Li, Y.; Lin, M.; McBrayer, G.A. Natural Disasters, Risk Salience, and Corporate ESG Disclosure. J. Corp. Financ. 2022, 72, 102152. [Google Scholar] [CrossRef]
Yang, Z.; Gao, W.; Han, Q.; Qi, L.; Cui, Y.; Chen, Y. Digitalization and carbon emissions: How does digital city construction affect china’s carbon emission reduction? Sustain. Cities Soc. 2022, 87, 104201. [Google Scholar] [CrossRef]
Wu, X.; Ren, B.; Liu, Y. Financial Development and Common Prosperity—A Study on the Interaction Effect between Traditional Finance and Digital Finance. Inq. Into Econ. Issues 2023, 44, 125–141. [Google Scholar]
Fernández-Olit, B.; Martín, J.; Porras, E. Systematized literature review on financial inclusion and exclusion in developed countries. Int. J. Bank Mark. 2019; ahead-of-print. [Google Scholar]
Jiang, H. Research on the impact of digital finance on the efficiency of urban green economy. Soft Sci. 2022, 36, 37–43. [Google Scholar]
Duan, J.; Gao, W. Research on the impact of digital finance development on the total factor productivity of enterprises. Contemp. Financ. Econ. 2022, 450, 51–62. [Google Scholar]
Lyu, Y.; Ji, Z.; Zhang, X.; Zhan, Z. Can Fintech alleviate the financing constraints of enterprises? Evidence from the Chinese securities market. Sustainability 2023, 15, 3876. [Google Scholar] [CrossRef]
Liang, B.; Zhang, J. Can the Development of Digital Inclusive Finance Stimulate Innovation?—Evidence from Chinese Cities and SMEs. Mod. Econ. Sci. 2019, 41, 74–86. [Google Scholar]
Xiao, H.; Yang, Z. Can Digital Inclusive Finance Promote Enterprise Digitization? Jilin Univ. J. Soc. Sci. Ed. 2022, 62, 106–126+233–234. [Google Scholar]
Yang, X.; Han, Q. Nonlinear effects of enterprise digital transformation on environmental, social and governance (ESG) performance: Evidence from China. Sustain. Account. Manag. Policy J. 2024, 15, 355–381. [Google Scholar] [CrossRef]
Li, Y.; Wang, L.; Su, J.; Tang, Y.; Wang, P. “Credit Use” and Cost of Discredit: The Supreme People’ Court and Zhima Credit Cooperating to Punish “Lao Lai”. In Inclusive Finance in China; Li, Y., Wang, L., Eds.; Springer: Singapore, 2021; pp. 321–355. [Google Scholar]
Birol, F.; Keppler, J.H. Prices, technology development and the rebound effect. Energy Policy 2000, 28, 457–469. [Google Scholar] [CrossRef]
Torelli, R.; Balluchi, F.; Lazzini, A. Greenwashing and environmental communication: Effects on stakeholders’ perceptions. Bus. Strategy Environ. 2020, 29, 407–421. [Google Scholar] [CrossRef]
Huang, H.; Lai, M.; Lin, L.; Chen, C. Overcoming organizational inertia to strengthen business model innovation: An open innovation perspective. J. Organ. Change Manag. 2013, 26, 977–1002. [Google Scholar] [CrossRef]
Rao, S.; Pan, Y.; He, J.; Shangguan, X. Digital finance and corporate green innovation: Quantity or quality? Environ. Sci. Pollut. Res. 2022, 29, 56772–56791. [Google Scholar] [CrossRef]
Mao, X.; Wang, R. Digital finance and green development—Empirical evidence from 286 cities in China. Financ. Forum 2023, 28, 69–80. [Google Scholar]
Sinn, H.-W. Public policies against global warming: A supply side approach. Int. Tax Public Financ. 2008, 15, 360–394. [Google Scholar] [CrossRef]
Zheng, S.; Wan, G.; Sun, W.; Wan, D. Public Appeal and Urban Environmental Governance. J. Manag. World 2013, 6, 72–84. [Google Scholar]
Tao, Y.; Hou, W.; Liu, Z.; Yang, Z. How can public environmental concerns enhance corporate ESG performance?—Based on a dual perspective of external pressure and internal concerns. Sci. Sci. Manag. S. & T. 2023, 44, 1–28. [Google Scholar]
Tang, S.; Wu, X.; Zhu, J. Digital Finance and Enterprise Technology Innovation: Structural Feature, Mechanism Identification and Effect Difference under Financial Supervision. J. Manag. World 2020, 36, 52–66+9. [Google Scholar]
Li, X.; Cui, S.; Lai, X. Can Digital Finance Increase the Value of Listed Companies?: Theoretical Mechanism Analysis and Empirical Test. J. Tianjin Univ. Financ. Econ. 2020, 40, 83–95. [Google Scholar]
Du, X. How the market values greenwashing? Evidence from China. J. Bus. Ethics 2015, 128, 547–574. [Google Scholar] [CrossRef]
Adhikari, B.K.; Agrawal, A. Peer influence on payout policies. J. Corp. Financ. 2018, 48, 615–637. [Google Scholar] [CrossRef]
Wang, W.; Sun, Z.; Wang, W.; Hua, Q.; Wu, F. The impact of environmental uncertainty on ESG performance: Emotional vs. rational. J. Clean. Prod. 2023, 397, 136528. [Google Scholar] [CrossRef]
Mikalef, P.; van de Wetering, R.; Krogstie, J. Building dynamic capabilities by leveraging big data analytics: The role of organizational inertia. Inf. Manag. 2021, 58, 103412. [Google Scholar] [CrossRef]
Constantinides, P.; Henfridsson, O.; Parker, G.G. Platforms and Infrastructures in the Digital Age. Inf. Syst. Res. 2018, 29, 381–400. [Google Scholar] [CrossRef]
Chen, Y.; Zuo, S. The Influence of Digital Transformation of Circulation Industry on Regional Green Economy Efficiency and its Mechanism. J. Commer. Econ. 2023, 42, 31–34. [Google Scholar]
Hinings, B.; Gegenhuber, T.; Greenwood, R. Digital innovation and transformation: An institutional perspective. Inf. Organ. 2018, 28, 52–61. [Google Scholar] [CrossRef]
Lin, W.; Feng, B. Curvilinear Effect and Statistical Test Method in the Management Research. Nankai Bus. Rev. 2022, 25, 155–166. [Google Scholar]
Haans, R.; Pieters, C.; He, Z.-L. Thinking about U: Theorizing and Testing U- and inverted U-shaped Relationships in Strategy Research. Strateg. Manag. J. 2016, 37, 1177–1195. [Google Scholar] [CrossRef]
Xie, H.; Lv, X. Responsible Multinational Investment: ESG and Chinese OFDI. Econ. Res. J. 2022, 57, 83–99. [Google Scholar]
Guo, F.; Wang, J.; Wang, F.; Kong, T.; Zhang, X.; Cheng, Z. Measuring China’s digital financial inclusion: Index compilation and spatial characteristics. Q. J. Econ. 2020, 19, 1401–1418. [Google Scholar]
Wu, L.; Yang, M.; Sun, K. Impact of Public Environmental Attention on Environmental Governance of Enterprises and Local Governments. China Popul. Resour. Environ. 2022, 32, 1–14. [Google Scholar]
Shen, H.; Yu, P.; Wu, L. State Ownership, Environment Uncertainty and Investment Efficiency. Econ. Res. J. 2012, 47, 113–126. [Google Scholar]
Wu, F.; Hu, H.; Lin, H.; Ren, X. Enterprise Digital Transformation and Capital Market Performance: Empirical Evidence from Stock Liquidity. J. Manag. World 2021, 37, 130–144+110. [Google Scholar]
Wang, Y.; Zhang, Y.; Li, J. Digital finance and carbon emissions: An empirical test based on micro data and machine learning model. China Popul. Resour. Environ. 2022, 32, 1–11. [Google Scholar]
Bruhn, M.; Love, I. The Real Impact of Improved Access to Finance: Evidence from Mexico. J. Financ. 2014, 69, 1347–1376. [Google Scholar] [CrossRef]
Liu, S.; Yang, S.; Liu, J. Can Local Fiscal Self-sufficiency Improve Public Goods Supply Efficiency? Financ. Trade Res. 2021, 32, 52–68. [Google Scholar]
Han, C.; Chen, Z.; Wang, Z. Study on the Mechanism of Firms’ Pollution Reduction Under the Constraint of Energy-saving Target. China Ind. Econ. 2020, 10, 43–61. [Google Scholar]
Song, M.; Zhou, P.; Si, H. Financial Technology and Enterprise Total Factor Productivity—Perspective of “Enabling” and Credit Rationing. China Ind. Econ. 2021, 39, 138–155. [Google Scholar]
Bartik, T.J. How Do the Effects of Local Growth on Employment Rates Vary with Initial Labor Market Conditions? Upjohn Institute Working Paper No. 09-148; W.E. Upjohn Institute for Employment Research: Kalamazoo, MI, USA, 2009. [Google Scholar] [CrossRef]
Liu, J.; Xia, Q.; Xu, Y.; Hou, X. Industrial Technical Complexity, the Government Subsidy and the Green Technology Innovation Incentives. Nankai Bus. Rev. 2024, 27, 94–103+149+104–105. [Google Scholar]
Xie, Z.H.; Wu, R.; Wang, S.J. How technological progress affects the carbon emission efficiency? Evidence from national panel quantile regression. J. Clean. Prod. 2021, 307, 127133. [Google Scholar] [CrossRef]
Shen, H.; Kou, H.; Zhang, C. An Empirical Study of Financial Development, Financing Constraints and Corporate Investment. China Ind. Econ. 2010, 28, 55–64. [Google Scholar]
Huang, D.; Xie, H.; Zhou, M. Environmental Regulation and Enterprise ESG Performance under the Background of Dual Carbon Practice—Empirical Study Based on both Macro and Micro Mechanisms. J. Shanxi Univ. Financ. Econ. 2023, 45, 83–96. [Google Scholar]
Bertoni, F.; Colombo, M.G.; Croce, A. The Effect of Venture Capital Financing on the Sensitivity to Cash Flow of Firm’s Investments. Eur. Financ. Manag. 2010, 16, 528–551. [Google Scholar] [CrossRef]
Revest, V.; Sapio, A. Financing technology-based small firms in Europe: What do we know? Small Bus. Econ. 2012, 39, 179–205. [Google Scholar] [CrossRef]
Yang, Y.; Han, J.M. Digital transformation, financing constraints, and corporate environmental, social, and governance performance. Corp. Soc. Responsib. Environ. Manag. 2023, 30, 3189–3202. [Google Scholar] [CrossRef]
Li, Q.; Liu, L.; Shao, J. The Effects of Digital Transformation and Supply Chain Integration on Firm Performance: The Moderating Role of Entrepreneurship. Bus. Manag. J. 2021, 43, 5–23. [Google Scholar]
Wan, J.; Zhou, Q.; Xiao, Y. Digital finance, financial constraint and enterprise innovation. Econ. Rev. 2020, 41, 71–83. [Google Scholar]
Xiao, G.; Shen, S. To pollute or not to pollute: Political connections and corporate environmental performance. J. Corp. Financ. 2022, 74, 102214. [Google Scholar] [CrossRef]
Zhang, Y.; Zhou, Y. The Impact of Digital Finance Development on the Operational Risks of Rural Financial Institutions: A Study Based on the Moderating Effect of Financial Supervision Intensity. Chin. Rural Econ. 2022, 38, 64–82. [Google Scholar]
Xiao, R.; Wang, R.; Qian, L. The Non-linear Impact of Digitalization Level on Corporate Carbon Performance: The Mediating Effect of Green Technology Innovation. Sci. Technol. Prog. Policy 2023, 40, 96–106. [Google Scholar]

Figure 1. Research framework.

Figure 2. Direct influence analysis diagram.

Figure 3. Mediation mechanism analysis diagram.

Figure 4. Regulatory mechanism analysis diagram.

Table 1. Descriptive statistics results.

Variable	N	Mean	sd	max	min
ESG	14,300	4.219	1.021	8.000	1.000
DF	14,300	5.283	0.451	5.885	3.057
Lev	14,300	0.443	0.198	1.056	0.007
Roe	14,300	0.071	0.131	1.536	−4.857
TobinQ	14,300	1.909	1.335	31.400	0.627
Age	14,300	2.895	0.359	3.738	0.693
Mshare	14,300	7.761	14.967	62.917	0.000
Cash	14,300	0.050	0.068	0.516	−0.744
Board	14,300	2.163	0.196	2.890	1.099
Big4	14,300	0.083	0.276	1.000	0.000
Gov	14,300	0.156	0.055	0.675	0.044
Open	14,300	0.004	0.003	0.035	0.000
Pgdp	14,300	11.409	0.526	13.056	9.091

Table 2. Benchmark regression results.

	(1)	(2)	(3)
	ESG	ESG	ESG
Df	0.003 *	−1.646 *	−1.608 *
	(1.81)	(−1.93)	(−1.99)
Df²		0.228 **	0.211 **
		(2.36)	(2.21)
Lev			−0.512 ***
			(−4.26)
Roe			0.269 ***
			(3.23)
TobinQ			−0.002
			(−0.13)
Age			−0.355 *
			(−1.90)
Mshare			0.011 ***
			(4.73)
Cash			−0.299 ***
			(−2.76)
Board			−0.058
			(−0.65)
Big4			0.096
			(1.12)
Gov			−0.437
			(−0.65)
Open			0.714
			(0.15)
Gdp			0.002
			(0.03)
_cons	3.491 ***	6.521 ***	8.119 ***
	(8.68)	(3.25)	(3.85)
Individual Fixed	YES	YES	YES
Time × Industry Joint Fixed	YES	YES	YES
Time Fixed	YES	YES	YES
R²	0.649	0.649	0.658
N	14,300	14,300	14,300

Notes: ***, ** and * indicate significant at the 1%, 5%, and 10% levels, respectively. Figures in () are the t-values.

Table 3. Robustness and endogeneity test results.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
	ESG	ESG	ESG	ESG	Df	ESG	FESG
Df		−1.608 **	−1.280 *	−1.243		−3.569 **	−1.469
		(−2.31)	(−1.69)	(−1.61)		(−2.30)	(−1.52)
Df²		0.211 **	0.128 **	0.179 *		0.437 **	0.188 ***
		(2.39)	(2.19)	(1.83)		(2.57)	(2.66)
MLDf	−2.499 *
	(−1.68)
MLDf²	0.312 *
	(1.83)
Bartik_Df					0.768 ***
					(36.10)
Kleibergen–Paap rk LM Test					p-value = 0.0000
Kleibergen–Paap Wald rk F Test					48.791 > 16.38
_cons	10.070 ***	8.119 ***	9.448 ***	7.958 ***	4.416 ***	11.635 ***	7.160 ***
	(2.92)	(5.26)	(6.00)	(4.06)	(74.62)	(2.95)	(3.09)
Individual Fixed	YES	YES	YES	YES	YES	YES	YES
Time × Industry Joint Fixed	YES	YES	YES	YES	YES	YES	YES
Time Fixed	YES	YES	YES	YES	YES	YES	YES
R²	0.681	0.658	0.774	0.664	0.996	0.061	0.687
N	12,921	14,300	5167	11,230	12,921	13,000	12,923

Notes: ***, ** and * indicate significant at the 1%, 5%, and 10% levels, respectively. Figures in () are the t-values.

Table 4. Empirical results of intermediation mechanisms.

	(1)	(2)	(3)	(4)
	GI	ESG	PEG	ESG
Df	1.695 ***	0.242 **	0.298 **	0.085 **
	(9.39)	(2.30)	(2.59)	(2.36)
Df²
SA
GI		−0.275 ***
		(−3.00)
GI²		0.017 ***
		(3.36)
upin
PEG				−0.097 **
				(−2.42)
PEG²				0.007 *
				(1.93)
_cons	0.117	4.827 ***	1.031	5.068 ***
	(0.08)	(3.78)	(1.34)	(3.72)
Individual Fixed	YES	YES	YES	YES
Time × Industry Joint Fixed	YES	YES	YES	YES
Time Fixed	YES	YES	YES	YES
R²	0.925	0.659	0.973	0.658
N	14,215	14,215	14,215	14,215

Notes: ***, ** and * indicate significant at the 1%, 5%, and 10% levels, respectively. Figures in () are the t-values.

Table 5. Empirical results on regional heterogeneity.

Category	Low Finance	High Finance	Low Regulation	High Regulation
	(1)	(2)	(3)	(4)
Dep. Var	ESG	ESG	ESG	ESG
Df	−3.102	−2.756 **	−1.058	−2.745 *
	(−1.59)	(−2.03)	(−0.93)	(−1.77)
Df²	0.447 *	0.318 **	0.196	0.367 *
	(1.74)	(2.03)	(1.40)	(1.77)
_cons	4.960	11.235 ***	5.785 **	8.989 **
	(1.18)	(3.31)	(2.11)	(2.38)
Individual Fixed	YES	YES	YES	YES
Time × Industry Joint Fixed	YES	YES	YES	YES
Time Fixed	YES	YES	YES	YES
R²	0.710	0.689	0.678	0.721
N	2275	11,778	11,067	2824

Notes: ***, ** and * indicate significant at the 1%, 5%, and 10% levels, respectively. Figures in () are the t-values.

Table 6. Empirical results on industry heterogeneity.

Category	Non-High Tech	High-Tech	Non-mfg	MFG
	(1)	(2)	(3)	(4)
Dep. Var	ESG	ESG	ESG	ESG
Df	−3.505 ***	0.009	−4.768 ***	−0.487
	(−2.97)	(0.01)	(−3.33)	(−0.50)
Df²	0.432 ***	0.012	0.445 **	0.118
	(2.78)	(0.09)	(2.46)	(0.90)
_cons	12.378 ***	4.602 *	17.131 ***	4.824 **
	(4.86)	(1.79)	(5.26)	(2.16)
Individual Fixed	YES	YES	YES	YES
Time × Industry Joint Fixed	YES	YES	YES	YES
Time Fixed	YES	YES	YES	YES
R²	0.719	0.615	0.719	0.631
N	6468	7700	5229	8931

Notes: ***, ** and * indicate significant at the 1%, 5%, and 10% levels, respectively. Figures in () are the t-values.

Table 7. Empirical findings on firm heterogeneity.

Category	LSE	SME	Non-Pol Conn	Pol-Conn
	(1)	(2)	(3)	(4)
Dep. Var	ESG	ESG	ESG	ESG
Df	0.307	−4.353 ***	−2.969 ***	0.461
	(0.29)	(−3.42)	(−2.90)	(0.35)
Df²	−0.000	0.459 ***	0.351 ***	−0.016
	(−0.00)	(2.81)	(2.62)	(−0.10)
_cons	5.482 **	13.148 ***	11.676 ***	3.383
	(2.22)	(4.37)	(4.71)	(1.12)
Individual Fixed	YES	YES	YES	YES
Time × Industry Joint Fixed	YES	YES	YES	YES
Time Fixed	YES	YES	YES	YES
R²	0.667	0.716	0.695	0.745
N	6988	6986	9293	4702

Notes: ***, ** indicate significant at the 1%, 5% levels, respectively. Figures in () are the t-values.

Table 8. Empirical results of regulating mechanisms.

	(1)	(2)	(3)
	ESG	ESG	ESG
Df	−1.607	−2.181 **	−0.943
	(−1.57)	(−2.49)	(−1.11)
Df²	0.212 *	0.274 ***	0.136 **
	(1.89)	(2.73)	(2.37)
supF	−32.053
	(−0.34)
Df × supF	15.078
	(0.39)
Df² × supF	−1.702 **
	(−2.42)
EU		−1.211 *
		(−1.72)
Df × EU		0.548 *
		(1.92)
Df² × EU		−0.061 **
		(−2.12)
dcg			2.666
			(1.55)
Df × dcg			−1.151 *
			(−1.71)
Df² × dcg			0.124 *
			(1.90)
_cons	8.126 ***	9.403 ***	6.617 ***
	(2.99)	(4.10)	(3.06)
Individual Fixed	YES	YES	YES
Time × Industry Joint Fixed	YES	YES	YES
Time Fixed	YES	YES	YES
R²	0.658	0.660	0.659
N	14,300	14,300	14,300

Notes: ***, ** and * indicate significant at the 1%, 5%, and 10% levels, respectively. Figures in () are the t-values.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Share and Cite

MDPI and ACS Style

Yin, Q.; Su, N.; Ding, C. The Nonlinear Effects of Digital Finance on Corporate ESG Performance: Evidence from China. Sustainability 2024, 16, 8274. https://doi.org/10.3390/su16188274

AMA Style

Yin Q, Su N, Ding C. The Nonlinear Effects of Digital Finance on Corporate ESG Performance: Evidence from China. Sustainability. 2024; 16(18):8274. https://doi.org/10.3390/su16188274

Chicago/Turabian Style

Yin, Qingmin, Nan Su, and Chenhui Ding. 2024. "The Nonlinear Effects of Digital Finance on Corporate ESG Performance: Evidence from China" Sustainability 16, no. 18: 8274. https://doi.org/10.3390/su16188274

APA Style

Yin, Q., Su, N., & Ding, C. (2024). The Nonlinear Effects of Digital Finance on Corporate ESG Performance: Evidence from China. Sustainability, 16(18), 8274. https://doi.org/10.3390/su16188274

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

The Nonlinear Effects of Digital Finance on Corporate ESG Performance: Evidence from China

Abstract

1. Introduction

2. Literature Review and Hypothesis Formation: Identifying Research Gaps and Constructing a Theoretical Framework

2.1. Digital Finance and Corporate ESG Performance

2.2. Mediating Mechanism

2.2.1. Enterprise Internal Mechanisms

2.2.2. Mechanisms External to the Enterprise

2.3. Moderating Mechanisms

2.3.1. Financial Regulation

2.3.2. Environmental Uncertainty

2.3.3. Digital Transformation

3. Research Design

3.1. Modeling

3.2. Variables Definition

3.2.1. Explained Variables

3.2.2. Core Explanatory Variables

3.2.3. Mediating Variables

3.2.4. Moderating Variables

3.2.5. Control Variables

3.3. Data Sources and Descriptive Statistics

4. Empirical Results

4.1. Benchmarking Results

4.2. Robustness and Endogeneity Treatment

4.2.1. Substitution of Explanatory Variables

4.2.2. Replacing Higher-Order Clustering Robust Criterion Errors

4.2.3. Removing Some of the Effects of Factors

4.2.4. Endogenous Treatment

4.3. Mechanism Identification Test

4.4. Heterogeneity Test

4.4.1. Regional Heterogeneity

4.4.2. Industry Heterogeneity

4.4.3. Enterprise Heterogeneity

5. In-Depth Exploration: Digital Finance’s Role in ESG Performance via Regulatory Mechanisms

6. Conclusions and Implications

6.1. Research Conclusions

6.2. Policy Implications

6.3. Limitations and Prospects

Author Contributions

Funding

Institutional Review Board Statement

Informed Consent Statement

Data Availability Statement

Conflicts of Interest

References

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI